UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analysis of the sustainability and management of the talun-kebun system of West Java, Indonesia Christanty, Linda 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1989_A1 C47.pdf [ 14.83MB ]
Metadata
JSON: 831-1.0098259.json
JSON-LD: 831-1.0098259-ld.json
RDF/XML (Pretty): 831-1.0098259-rdf.xml
RDF/JSON: 831-1.0098259-rdf.json
Turtle: 831-1.0098259-turtle.txt
N-Triples: 831-1.0098259-rdf-ntriples.txt
Original Record: 831-1.0098259-source.json
Full Text
831-1.0098259-fulltext.txt
Citation
831-1.0098259.ris

Full Text

ANALYSIS OF T H E SUSTAINABILITY AND MANAGEMENT OF T H E TALUN-KEBUN SYSTEM OF WEST JAVA, INDONESIA by LINDA CHRISTANTY B.Sc, University of Gadjah Mada, Indonesia, 1972 Dra., University of Gadjah Mada, Indonesia, 1976 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF GRADUATE STUDIES Interdisciplinary Department Resource Management Program We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA April, 1989 ©Linda Christanty, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of In Vefd ISC i p) j n 8 fy £hj The University of British Columbia Vancouver, Canada Date DE-6 (2/88) i i A B S T R A C T Bamboo talun-kebun is one of the traditional agroforestry systems in West Java, Indonesia, in which annual food crops are alternated with a fallow period of bamboo and trees. Its overall pattern is similar to that of traditional shifting cultivation, but the length of the fallow period is much shorter and there is a deliberate selection of species planted in each stage. Because of increasing population pressure, easier access to fertilizer and pesticides, and aspirations for a higher standard of living, some farmers are intensifying the cropping, shortening the fallow stage, or replacing the bamboo with other cash crops. These changes raise concerns about the maintenance of site productivity. This study was undertaken to analyze the sustainability of the bamboo talun-kebun under the current management practice by examining the biogeochemistry of five major nutrients (N, P, K, Ca, and Mg) in various stages of the cycle, and by using this information to assess the effects of extending the period of cropping on production. The energy efficiency and some other aspects of the talun-kebun management practice were also examined. Plant biomass and the accumulation and distribution of nutrients (N, P, K, Ca, and Mg) in plants and in mineral soils were determined over a complete bamboo talun-kebun rotation cycle to provide the data needed for the biogeochemical analysis. The dynamics of litterfall, forest floor, and soil organic matter mass and nutrients were studied. The inputs and outputs of nutrients to and from the system in precipitation, soil leaching, fertilizer, biological fixation, and losses in the burning of forest floor and slash were measured and/or estimated. Approximately 96.5 t ha-1 of biomass was accumulated and 62.8 t ha-1 were removed from the site over a six year talun-kebun rotation cycle. Biomass removals during the first and second year cropping of food species accounted for 17 and 11% of the total removals, respectively. The overall removals represented the loss of 53% i l l of N, 43% of P, 43% of K, 46% of Ca, and 42% of Mg accumulated in the live plant biomass over the rotation cycle. An input-output balance of nutrients over a complete talun-kebun rotation cycle indicated that harvest removals was the major pathway for nutrient loss from the system. The nutrient budget analysis of the cycle indicated the importance of the fallow stage to the long-term sustainability of the system. Nutrients were accumulated in the forest floor and in plant biomass, while minor losses from the system occurred only through leaching. Therefore, shortening the fallow stage might lead to a depletion of nutrients in the mineral soil. A tabular analysis to assess the consequences for nitrogen uptake and biomass production of extending the period of mixed cropping and reducing the length of the fallow emphasized the importance of the fallow stage for the maintenance of site productivity. Extending the period of mixed cropping without increasing the use of fertilizer would lead to a depletion of the site nutrients and a decrease in biomass production, which would be followed by a decrease in litter production and in turn would reduce the amount of nutrients returned to the soil. Although sale of the products from the talun-kebun was not the main source of income for the farmers, the costs of production were low (limited to labor cost), and the system was economically profitable. The talun-kebun proved to be an energy efficient cropping system. The ratio of net energy output/input of food crop production was 38:1. i v T A B L E O F C O N T E N T S Page A B S T R A C T ii T A B L E O F C O N T E N T S iv LIST O F T A B L E S . . xii LIST O F F I G U R E S xvii A C K N O W L E D G M E N T S xx C H A P T E R 1. G E N E R A L I N T R O D U C T I O N 1 1.1. Background 1 1.2. Key Questions and Research Approach 4 1.3. Thesis Objectives 6 1.4. Organization of the Thesis 6 C H A P T E R 2. A N I N T R O D U C T I O N T O B A M B O O TALUN-KEBUN 8 2.1. Historical Development 8 2.2. Current Practice 10 2.3. The Bamboo Talun-kebun 10 2.4. Plant Composition and Vegetation Structure 11 2.5. Sequence of Cultural Activities 13 2.6. Summary and Discussion 17 C H A P T E R 3. D E S C R I P T I O N O F T H E S T U D Y A R E A 20 3.1. The Study Regions 20 3.1.1. Climate 20 3.1.2. Geology and soil 21 3.1.3. Vegetation 23 3.1.4. Land use . 23 3.1.5. Population 23 V 3.2. The Study Area 24 C H A P T E R 4. A C C U M U L A T I O N A N D DISTRIBUTION O F B A M B O O BIOMASS IN T H E TALUN-KEBUN S Y S T E M 29 4.1. Introduction 29 4.2. Objectives . 30 4.3. Literature Review 31 4.3.1. Studies of bamboo biomass 31 4.3.2. Biomass estimation 34 4.4. Methods 36 4.5. Results and Discussion 39 4.5.1. Culm density, distribution, and mortality in various talun-kebun stages 39 4.5.2. Regression equations to estimate living biomass 46 1. Aboveground and rhizome biomass 46 2. Coarse root biomass 48 4.5.3. Estimated Biomass 48 1. Aboveground biomass 48 2. Belowground biomass . 51 A. Rhizome biomass 51 B. Coarse and fine root biomass . . 55 4.5.4. Biomass Distribution 56 4.6. Summary and Conclusions 57 C H A P T E R 5. A C C U M U L A T I O N A N D DISTRIBUTION O F A G R I C U L T U R A L C R O P A N D W E E D BIOMASS IN T H E TALUN-KEBUN S Y S T E M 61 5.1. The First Year Cropping 61 5.1.1. Introduction 61 5.1.2. Objectives 64 v i 5.1.3. Study Sites 64 5.1.4. Methods 64 5.1.5. Results 66 1. Crop biomass .66 A. Cucumber 66 B. Bitter solarium 67 C. Hyacinth bean 67 2. Biomass removal 71 5.2. The Second Year Cropping 73 5.2.1. Introduction 73 5.2.2. Objectives .74 5.2.3. Study Sites 74 5.2.4. Methods 74 5.2.5. Results and Discussion 75 1. Cassava biomass 75 2. Biomass removal 80 5.3. Weeds 81 5.3.1. Introduction 81 5.3.2. Objectives 82 5.3.3. Study Sites 82 5.3.4. Methods 82 5.3.5. Results and Discussion 83 1. Weed biomass 83 2. Biomass removal 85 5.4. Summary and Conclusions 86 v i i C H A P T E R 6. L ITTER PRODUCTION, F O R E S T F L O O R MASS, AND SOIL ORGANIC M A T T E R DYNAMICS IN T H E TALUN-KEBUN SYSTEM 88 6.1. Introduction 88 6.2. Literature Review 88 6.3. Study Area 91 6.4. Methods 91 6.4.1. Litterfall sampling 91 6.4.2. Forest floor sampling 91 6.4.3. Mineral soil organic matter sampling 92 6.5. Results and Discussion 93 6.5.1. Litter Accumulation 93 1. The cropping stage 93 2. The fallow stage 94 6.5.2. Forest Floor Dynamics 98 6.5.3. Litter Removals 99 6.5.3. Soil Organic Matter Dynamics 99 6.6. Summary and Conclusions 100 C H A P T E R 7. T H E B IOGEOCHEMISTRY OF BAMBOO TALUN-KEBUN 104 7.1. Literature Review 105 7.1.1. The late stage of the fallow 105 7.1.2. The clearing and burning of the vegetation 107 7.1.3. The cropping stage 108 7.1.4. The recovery stage of fallow vegetation 109 7.2. The Overall Biomass Accumulation and Dynamics in the Talun-kebun System I l l 7.2.1. Biomass accumulation I l l 7.2.2. Biomass removals 115 v i i i 7.3. Nutrient Inventory and Dynamics in Plants 116 7.3.1. Introduction 116 7.3.2. Objectives 118 7.3.3. Methods 119 7.3.4. Results and Discussion 119 1. Nutrient concentrations in plants 119 A. Crop plants 119 B. Bamboo 120 2. Nutrient Accummulation in live plant biomass at various stages of the talun-kebun cycle 120 3. Removals of nutrients in live plant biomass at various stages of the talun-kebun cycle 126 4. Albizia and other trees 128 7.4. Nutrient Inventories in the Forest Floor/ Ectorganic Layer and in the Mineral Soils 128 7.4.1. Introduction 128 7.4.2. Methods 130' 7.4.3. Results and Discussion 131 1. Forest floor nutrients 131 2. Soil texture, bulk density, and pH : 132 3. Mineral soil nutrients 134 7.5. Nutrient Dynamics and Inputs-Outputs Balance in the Talun-kebun system 139 7.5.1. Nutrients in litterfall . , . . 140 7.5.2. Internal Conservation of Nutrients in Plants ;.. 142 7.5.3. Nutrients in precipitation and throughfall 142 1. Methods . . 143 i x 2. Results and Discussion 144 7.5.4. Nutrient Uptake by Plants 146 7.5.5. Nutrients in Fertilizers 147 7.5.6. Nitrogen Fixation 147 7.5.7. Nutrient losses in leaching and surface runoff 150 1. Methods 150 2. Results and Discussion 151 7.5.8. The loss of nutrients from the slash (bamboo leaves and forest floor) during and after the burn 152 7.6. The Overall Biogeochemical characteristics of the Bamboo Talun-kebun 152 7.7. Summary and Conclusions 156 C H A P T E R 8. A N A L Y S I S O F S O M E A S P E C T S O F T H E M A N A G E M E N T O F T H E TALUN-KEBUN S Y S T E M 164 8.1. Introduction 164 8.2. Objectives 164 8.3. Study Sites 167 8.4. Methods 167 8.5. Results and Discussion 168' 8.5.1. Household Background 168 8.5.2. Cropping Pattern 169 8.5.3. Farmers' Motivation 169 8.5.4. Labor use in the Talun-kebun system 170 A. Work activities and the division of labor between men and women 170 B. Time spent and work distribution throughout the year 175 X 8.5.5. Energetic Aspects 179 A. Human energy expenditure 179 B. Energy outputs 190 C. Energy efficiency 192 8.5.6. Economic of the Talun-kebun products 194 A. Costs of producing the first and second year crops 194 B. Returns from the harvest products 195 8.5.7. Pest and Conservation Aspects 198 8.5.8. Future Trends 200 8.6. Summary and Conclusions 202 C H A P T E R 9. B I O G E O C H E M I C A L A S S E S S M E N T O F T H E SUSTAINABILITY O F P R O D U C T I O N IN T H E B A M B O O TALUN-KEBUN . . . 204 9.1. Introduction 204 i • 9.2. Overall Input-Output Balance of Five Major Nutrients (N, P, K, Ca, and Mg) in the Different Stages of the Talun-kebun System 204 9.3. Compartmental Analysis of the Inputs and Outputs of Five Major Nutrients (N, P, K, Ca, and Mg) in the Different Stages of the Talun-kebun System 207 9.3.1. The Mature Talun Stage 208 9.3.2. Clearing, Burning, and Hoeing 209 9.3.3. The Cropping Stage 209 9.3.4. The Early Fallow Stage 210 9.3.5. The Sustainability of the Talun-kebun System 211 9.4. Nutrient Uptake-Production Relationships 222 9.5. Implications of Alterations of the Traditional Cropping Practice for Crop Production: A Tabular Analysis 225 9.5.1. Extending the period of mixed cropping 226 x i 9.5.2. Extending the period of cassava cropping 236 9.6. Discussion and Conclusions 240 C H A P T E R 10. O V E R A L L S U M M A R Y A N D C O N C L U S I O N S 241 R E F E R E N C E S 246 A P P E N D I C E S 253 x i i LIST O F T A B L E S Table Page 4.1. The density and dimensions of bamboo culms in various stages of the talun-kebun system in Selaawi hamlet, Soreang district, West Java, Indonesia 40 4.2. Equations for biomass (kg ha-l) of each component of bamboo in different stages of the talun-kebun system in Selaawi hamlet.Soreang district, West Java, Indonesia 45 4.3. Aboveground biomass accumulation (t ha-l) of each bamboo component and its percentage distribution (%) in various ages of bamboo fields in Selaawi hamlet, Soreang district, West Java, Indonesia 47 4.4. Live rhizome biomass (t ha-l) in various ages of bamboo fields in Selaawi hamlet, Soreang district, West Java, Indonesia 50 4.5. Coarse and fine root biomass in various ages of bamboo fields in Selaawi hamlet, Soreang district, West Java, Indonesia 53 4.6. The biomass and density of fine roots at various soil depths of a mature bamboo talun in Selaawi hamlet, Soreang district, West Java, Indonesia ; 54 4.7 The estimated biomass (t ha-l) of each bamboo component over the age sequence of a talun-kebun. The number in brackets represents the percentage of total biomass contributed by that component 59 5.1. Living biomass of cucumber grown in mixture during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period 66 x i i i 5.2. Living biomass of bitter solanum grown in mixture during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period 67 5.3. Living biomass of hyacinth bean grown in mixture during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period 68 5.4. Total biomass removed during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass removed 72 5.5. Biomass of a first cassava crop grown during the second year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period ; . 76 5.6. Biomass of a second cassava crop grown during the second year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period 77 5.7 Total biomass of cassava removed during the second year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia 78 5.8 Weed biomass over a six-year bamboo talun-kebun rotation cycle in Selaawi hamlet, Soreang district, West Java, Indonesia 82 6.1. The monthly litterfall of hyacinth bean and cassava during the cropping stage of a bamboo talun-kebun in Selaawi, Soreang, West Java, Indonesia .92 xiv 6.2. Seasonal litterfall during the early and mature fallow of a bamboo talun-kebun in Selaawi, Soreang, West Java, Indonesia 96 7.1. Accumulation of live plant biomass at various stages of the talun-kebun cycle. The numbers in brackets are the data expressed as a percentage of the total biomass for that stage ' 113 7.2. Removal of live plant biomass (t ha-1) at various stages of the talun-kebun cycle. The numbers in brackets are the removals as a percentage of the total removal at that stage 116 7.3. Removal of nutrients in live plant biomass (t ha-1) at various stages of the talun-kebun cycle. The numbers in brackets are the removals as a percentage of the total removal at that stage 126 7.4. N, P, K, Ca, and Mg contents in forest floor/ectorganic layer at various stages of the talun-kebun rotation cycle 132 7.5. Available phosphorus contents in mineral soils at various stages of a talun-kebun cycle in Soreang, West Java, Indonesia 138 7.6. The amount of exchangeable nutrients in mineral soils at various soil depths and at various stages of the talun-kebun cycle in Soreang, West Java, Indonesia 139 7.7. Nutrient content (kg ha-1) of aboveground litterfall at various stages of a talun-kebun cycle in Soreang,. West Java, Indonesia 141 7.8. The nutrient content of precipitation and throughfall under various stages of the talun-kebun cycle in Soreang, West Java, Indonesia (December 1983-December 1984). The number within brackets represents net leaching / leaf wash 145 7.9. The annual uptake of nutrients in various stages of the talun-kebun rotation cycle 148 7.10. The leaching losses of nutrients in various stages of the talun-kebun rotation cycle 149 XV 7.11. Sources of data for the conceptual model ofthe biogeochemistry of the talun-kebun system shown in Figure 7.8. Qualitative estimates of the accuracy of data are given 154 8.1. Work activities during clearing and cropping of a talun-kebun in Soreang, West Java, Indonesia 171 8.2. The time spent, the relative difficulty, and the estimated energy expenditure in field activities during the two-year cropping of a talun-kebun cycle (Based on the Malaysian calorie expenditure values) 185 8.3. The time spent, the relative difficulty, and the estimated energy expenditure in field activities during the two-year cropping of a talun-kebun cycle (Based on Uhls calorie expenditure values) 186 8.4. Energy outputs in a two year cropping stage of the talun-kebun system 191 8.5. Energy efficiency ratios of various inputs and outputs for the two year cropping stage of a talun-kebun cycle 193 8.6. The average costs and benefits over a six-year talun-kebun production cycle in Soreang, West Java (data given on a per ha basis over the whole cycle) 197 9.1. The input-output balance of N, P, K, Ca, and Mg in a six year talun-kebun rotation cycle. Statements about the reliability of the various estimates are given in Table 7.11 205 9.2. Estimates of various transfers of N into and out of the vegetation, forest floor, and upper mineral soil in various stages of the talun-kebun cycle 212 9.3. Estimates of various transfers of P into and out of the vegetation, forest floor, and upper mineral soil in various stages of the talun-kebun cycle 214 xvi 9.4. Estimates of various transfers of K into and out ofthe vegetation, forest floor, and upper mineral soil in various stages of the talun-kebun cycle 216 9.5. Estimates of various transfers of Ca into and out of the vegetation, forest floor, and upper mineral soil in various stages of the talun-kebun cycle 218 9.6. Estimates of various transfers of Mg into and out of the vegetation, forest floor, and upper mineral soil in various stages of the talun-kebun cycle 220 9.7. Nutrient uptake-production relationships for various species grown at various various stages of the talun-kebun cycle (the number within brackets represents percentage distribution of uptake at that particular stage) 223 9.8. Tabular analysis of the implications of an alternative cropping sequence for crop and bamboo production, based on the estimated amount of N available for plant uptake and biomass production in various stages of the scenario. This scenario involves two years of mixed species cropping, one year of cassava cropping, and three years of bamboo fallow 229 9.9. The effect of extending the cassava cropping on the amount of total N available for plant uptake and biomass production in various stages ofthe talun-kebun cycle . 237 x v i i LIST OF FIGURES Figure Page 2.1. Diagramatic summary of the bamboo talun-kebun cropping cycle in Soreang, West Java, Indonesia 16 3.1. Location of the study area 26 3.2. The layout of experimental plots in TK-2 . 27 3.3. The layout of five study fields representing various stages of a bamboo talun-kebun cycle. Contour lines indicate relative altitude (m) 28 4.1. The growth of culms after clearcutting 31 4.2. Distribution of bamboo culms per diameter classes in 500 m 2 plots of various talun-kebun stages 43 4.3. Distribution of bamboo culms per height classes in 500 m 2 plots of various talun-kebun stages 44 4.4. Culm development and mortality on the four plots 45 4.5. The percentage distribution of rhizomes at various ages after clearcutting a mature bamboo talun 53 4.6. Rhizome biomass in various ages of the talun-kebun cyle 54 4.7. The percentage distribution of bamboo biomass at 16, 24, 36, and 72 months. Values in brackets are percentage of total biomass 60 5.1. Spatial arrangement of agricultural crops in the first year cropping stage of the talun-kebun in Soreang, West Java, Indonesia 65 x v i i i 5.2. Total biomass (t ha - 1) of cassava in the second and third year of cropping in a bamboo talun-kebun in Soreang, West Java, Indonesia 76 6.1. The estimated values of bamboo litterfall over a six year talun-kebun rotation cycle in Selaawi hamlet, Soreang district, West Java, Indonesia 97 6.2. Soil organic matter concentration (%) in the upper 25 cm of mineral soil at various stages of a. talun-kebun cycle in Selaawi hamlet, Soreang district, West Java, Indonesia 102 6.3. Soil organic matter content in the upper 25 cm of mineral soil at various stages of a talun-kebun cycle 103 7.1. The percentage accumulation of live plant biomass at various stages of the talun-kebun system. Values in brackets are percentage ofthe total biomass 114 7.2. Accumulation of N, P, K in live plant biomass at various stages of the talun-kebun cycle 123 7.3. Accumulation of Ca and Mg in live plant biomass at various stages of the talun-kebun cycle 124 7.4. Nutrient accumulation in live plant biomass at various stages of the talun-kebun cycle 125 7.5. Total N concentration in soil in various stages of a talun-kebun cycle 135 7.6. Total N in the forest floor / ectorganic layer and in the mineral soils under various stages of a talun-kebun rotation cycle 136 7.7. The uptake of the major macronutrients by crops, bamboo and weeds at various stages of the talun-kebun rotation cycle 149 7.8. Flowchart of compartment and processes in a conceptual model of the biogeo-chemistry of the talun-kebun agroforestry 158 7.9. Flowcharts of major compartments and transfer processes of nitrogen in the five stages of the talun-kebun agro-forestry system 159 7.10. Flowcharts of major compartments and transfer processes of phosphorus in the five stages of the talun-kebun agro-forestry system 160 7.11. Flowcharts of major compartments and transfer processes of potassium in the five stages of the talun-kebun agro-forestry system 161 7.12. Flowcharts of major compartments and transfer processes of calcium in the five stages of the talun-kebun agro-forestry system 162 7.13. Flowcharts of maj or compartments and transfer processes of magnesium in the five stages of the talun-kebun agro-forestry system 163 8.1. Five common cropping systems for the kebun stage of a talun-kebun cycle in in Soreang,. West Java, Indonesia (thickened lines indicate major crops) The percentage of the 40 farms investigated using each of these systems is shown 168 8.2. The percentage of time spent in each activity over the first year of the talun-kebun rotation cycle 177 8.3. The percentage of time spent in each activity over the second year of the talun-kebun rotation cycle ,. . 178 8.4. The human energy inputs for each activity during the two year cropping cycle of a hectare talun-kebun (based on the Malaysian calorie expenditure figures) 187 8.5. The human energy inputs for each activity during the two year cropping cycle of a hectare talun-kebun (based on the Uhls calorie expenditure figures) 188 XX A C K N O W L E D G M E N T S I would like to express my highest gratitude to Dr. J.P.Kimmins, Faculty of Forestry, University of British Columbia (UBC), my research supervisor, for his patience, guidance, and assistance during the long preparation of this thesis. I am also grateful for his encouragement and friendship in difficult times during my stay in Canada. Thanks are also due to Drs. M.C.Feller, L.M.Lavkulich, and H.E.Schreier for their useful comments. The support given by Dr O.Soemarwoto, Director of the Institute of Ecology, Padjadjaran University (UNPAD), Bandung, Indonesia, is gratefully acknowledged. I appreciate the help of J.Iskandar, H.Y.Hadikusumah, Y.Yuhana, M.R.Noerdin, D.Hermanto, and H.Sukandar for their help in the field data collection. I am indebted to Lorens Kerans for the encouragement and moral support during difficult times, and to M.A. Laturiuw and his family for their helping hands during the final year of my stay in Canada. A special thanks also goes to Norman Price for exchanging ideas and for his useful suggestions for the thesis. Financial support for this study was given by UBC, the Ford Foundation, EMDI-CIDA scholarship, the Institute of Ecology, and my family, especially from my brother. My final and most deeply felt appreciation goes to my husband Igie, for his understanding, encouragement, and moral and financial support during all these years; and to my baby Christy for all the time I did not spend with her while working on the thesis. xxi This thesis is dedicated to Gerry Marten and Raymond Atje who made me believe in myself 1 C H A P T E R 1 G E N E R A L I N T R O D U C T I O N 1. l . B A C K G R O U N D Java, the centre island of Indonesia, has to support a population of approximately 120 million (BPS, 1985) on a land area of only 132,000 km2. This population, the majority of whom obtain their livelihood from intensive dryland and irrigated rice cultivation, is growing at an annual rate of 2%. The combination of high population density and a predominantly agrarian life style has resulted in serious land shortages for several decades, and with the continued growth of the population these shortage are likely to get worse. Java has been known for its fertile and productive volcanic soils, but as the intensity of cultivation has increased to support the increasing population, nutrient depletion and land degradation have become a problem in many parts of the island. These conditions are particularly severe in upland areas, where deforestation and soil erosion have occurred extensively. This in turn affects the stability and sustainability of lowland agricultural ecosystems (McCaulley, 1985). Recognizing the importance of upland management in maintaining the sustainability of the lowland systems, the Indonesian government has developed strategies for soil conservation and upland. management, such as reforestation of state lands and "regreening" (aforestation) of private lands. Unfortunately, cooperation of upland farmers in the implementation of these strategies, particularly in the regreening of private lands, has not always been easy to obtain (McCaulley, 1985). The increasing demand for the production of more food has become a major constraint on the efforts of the Indonesian government to promote the conversion of the less sustainable seasonal crop gardens into the more 2 stable and sustainable untilled forest gardensa. Furthermore, there has been a lack of upland community involvement in problem identification and in the design of alternative strategies for land management. In communities where traditional customs are strictly applied, acceptance or rejection of new land management practices depends largely on the perceptions by the community of the proposed system (Marten, 1986). No matter how good the new technology seems to be, it will not be accepted unless it is considered to be appropriate from the farmers' perspective. Accordingly, improvements to, or modifications of, traditional practices tend to have a better prospect for implementation than totally new land management systems. Such modifications require a thorough understanding of the present land management practices and how they might best be improved. Only on the basis of such understanding can one develop crop production systems that are more productive, are compatible with the maintenance of soil fertility, and are still acceptable to the local community. About 31% of the total land area of 4.4 million ha in West Java is used for dryland/upland agriculture, and about 25% for irrigated rice cultivation. Agricultural activity on about 56% of the land base contributes up to 50% of the regional income. However, in spite of the economic importance of the agricultural sector, most farmers belong to the lowest income group. This situation is caused by the small size of landholdings, coupled with improper upland management practices. So far, most of the effort to increase production has focused on increasing productivity of lowland agriculture, particularly irrigated ricefields; there has been little work done on developing the potential of upland agriculture. Considering the extent and importance of upland areas in West Java, there is an urgent need to develop a land management system which serves both production and conservation functions satisfactorily and which optimizes sustained upland productivity. a Forest gardens are uncultivated gardens which consist of more than 100 trees per hectare 3 Because of the need to produce both food and wood in a sustainable manner, an integrated system which combines agricultural and tree crops seems to be the most appropriate method of upland management in West Java. This system is known as agroforestry (Bene et al, 1978; ICRAF, 1982). While use of the term agroforestry is relatively recent, the practice of combining tree and food crops has existed for many centuries as the traditional land use method of rural people in most tropical countries (Finnegan, 1981). Traditional methods have had a high level of social acceptance, and incremental modifications of such methods are believed to be the easiest and least risky way to promote rural development in the tropics (Marten, 1986). Talun-kebun is one of the traditional crop production systems which has been widely practised by upland farmers in West Java for at least six generations. It is defined as an upland land use system in which annual food or cash crops are alternated sequentially with tree crops. Its overall pattern is similar to that of traditional shifting cultivation, but the length of the fallow period is much shorter and there is a deliberate selection of species used in each stage. This system seems to have the potential to resolve the conflict between the desire of the upland farmers to produce more food and cash rewards, and the government's conservation objectives. The talun-kebun system produces both food and wood, and it sustains the productive capacity of the site. However, with increasing population pressure and aspirations for a higher standard of living, the talun-kebun practice is now being changed. This change raises concerns about both the short term and the long term effects on site productivity. In order to identify the extent to which this ecologically-stable traditional system can be modified without causing site degradation, a thorough understanding of the major ecological processes that have rendered the system sustainable is required. In addition, socio-economic and cultural aspects which determine its 4 practice also need to be explored. 1.2.KEY Q U E S T I O N S A N D R E S E A R C H A P P R O A C H The present study was undertaken to examine the nutrient cycling and production aspects of the bamboo talun-kebun in West Java, Indonesia, and to investigate some of the long-term consequences of altered land-use practices in the study area. Two key questions were identified: 1. To what extent can the traditional agroforestry system be altered without significant reduction in land productivity? 2. If the traditional bamboo talun-kebun practice must be changed, what new management practices should be applied to ensure sustained production, and what are the socio-economic implications of these changes? The optimum way to answer these questions would be to study the processes that occur in the system over a period of several management cycles (20-30 years); it is highly desirable that such long term experiments be set up. However, the direct, empirical approach would require an extended time period and extensive financial resources. It would not be appropriate either for the current study or for the present Indonesian situation, where answers are required much sooner (within 5-10 years). Therefore, an indirect approach was chosen, involving several subsidiary questions. K E Y Q U E S T I O N #1 The first key question was divided into three subsidiary questions: 1. What is the temporal pattern of crop productivity, and what changes in biomass occur during a complete cycle ofthe bamboo talun-kebun? 2. How are the site nutrient resources redistributed during the different stages of the cycle? 5 3. How are the biomass and nutrient patterns of each stage changed by departure from the traditional practice, and what is the implication of these changes for future productivity? A traditional scientific investigation involving a static inventory and an empirical quantification of the temporal patterns of a number of nutrient, biomass and production parameters in each stage of a complete rotation cycle can help in the development of an understanding of the overall character and basic processes of the talun-kebun system. The thesis research employed this traditional scientific approach to develop a basic conceptual model of the functioning of the bamboo talun-kebun. This conceptual model of nutrient and biomass dynamics over a complete rotation cycle was used to guide field data collection, the data being used in a simple, conventional, tabular analysis of the consequences of changing the talun-kebun system. K E Y QUESTION #2 In order to answer the second key question, the following three subsidiary questions were advanced: 1. What is the energy efficiency of the bamboo talun- kebun in comparison with traditional shifting cultivation? 2. What is the economic yield of the bamboo talun- kebun, and what are the economic consequences of changing the traditional pattern? 3. What factors influence the farmers' decision in selecting their cropping pattern, and what measures are taken to maintain crop production? Energy efficiency was estimated by measuring the energy input-output balance of the talun-kebun. The energy inputs to the system consist of human energy expenditures for field and related activities, while the energy outputs are represented by the caloric value of harvested products (crops and mature bamboo 6 culms). Data on the economic aspects of the talun-kebun system were gathered by documenting the costs of labor and other economic inputs to the system, and the economic return of the harvested products (yields). In conjunction with the on-farm data collection, an interview survey was done to gather data on the economic and management aspects of the talun-kebun system. Detailed information on the methodology is presented in Chapter 8. 1.3.THESIS OBJECTIVES The main objectives of the thesis are: 1. To describe and quantify the biomass accumulation and organic matter dynamics in the crop production system that constitutes the bamboo talun-kebun. 2. To quantify the distribution and dynamics of macronutrients in the bamboo talun-kebun crop production system. 3. To describe the management aspects and to assess the economic and energy efficiency of this crop production system. 4. To develop conceptual models of biogeochemical cycles in each stage of a complete rotation cycle; and to calculate, using the conventional tabular nutrient budget approach, the nutrient balance of an alternative crop strategy. 1 .4 . 0 R G A N I Z A T I O N OF THE THESIS Chapter 1, the present chapter, has presented the background to the study and the research approach and objectives of the thesis. Chapter 2 summarizes the major features of the bamboo talun-kebun system (plant composition and vegetation structure, and the sequence of cultural activities) and presents a review of relevant 7 literature. The study area is described in Chapter 3. Chapter 4 examines the biomass production of bamboo, and describes the development of biomass regression equations for various components of bamboo. The biomass production of agricultural crops, and of weeds, are presented in Chapter 5. Chapter 6 discusses litterfall, forest floor, and soil organic matter dynamics. The overall nutrient inventories and dynamics are described in Chapter 7, which covers the overall biomass accumulation and dynamics in the bamboo talun-kebun cycle, the nutrient inventories in plants and in the mineral soils, and some aspects of the nutrient dynamics, including aboveground litterfall, accumulation of the forest floor, and various inputs to and removals/outputs from the system. Chapter 8 is concerned with yield, economic and energy aspects, and Chapter 9 presents a biogeochemical assessment of the sustainability of the bamboo talun-kebun system. The latter involves a tabular analysis of the site nutrient balance under various alternative management scenarios. The overall summary and conclusions of the thesis are covered in the last chapter (Chapter 10). 8 C H A P T E R 2 A N I N T R O D U C T I O N T O B A M B O O TALUN-KEBUN 2.1 .HISTORICAL D E V E L O P M E N T Historically, the Sundanese people of West Java employed three types of agricultural practice: conventional shifting cultivation in the uplands, the tipar system (semipermanent dryland rice cropping system) in the lowlands, and the talun system (the growing of perennial crops) around villages (Terra, 1953). Conventional shifting cultivation involved two successive cropping stages: the humal ladang (swidden growing of dryland rice) and the kebun (swidden growing of annual crops and tubers after the rice). After two or three years of cropping, crop productivity usually started to decrease (this is assumed to be due to the decline in soil fertility) and the land was then abandoned and left fallow for several years to restore soil fertility (Terra, 1953). The land could revert to secondary forest or to mixed talun or old ladang. The latter consisted of a mixture of fruit trees which were planted during the cropping stage (Terra, 1958). After the introduction of agricultural systems from Central Java in the early 1700's, wet rice cultivation was developed in the lowland areas of West Java and the tipar system gradually fell into disuse. The wet rice cultivation system was also adopted by upland farmers, and shifting cultivation only retained its traditional dominance in areas where topography and soil were unsuitable for growing irrigated rice (McCauley, 1982). In areas where wet rice cultivation was practised, most of the family's carbohydrate requirements was supplied from the ricefields, and the huma was converted to kebun (an area for growing other types of food crop) which was commonly planted with cash crops (Soemarwoto et al, 1985; Christanty et al, 1986). The rapid growth in the local population and the migration of people from 9 Central Java to West Java over the past 300 years have resulted in a significant increase in population density in West Java over this period. Where shifting cultivation was still practised, it was gradually intensified to fulfill the needs of this growing population. Prolonged cropping for several years with a shortened fallow period frequently resulted in poor vegetation development and growth because of a deterioration in the physical condition and nutrient status of the soil. The successional processes of natural fallow establishment were not sufficiently rapid for this intensified land use, and, therefore, a planted fallow was developed to maintain the sustainability of the system (Soemarwoto et al, 1985). Talun is one of the planted fallow systems which has been widely practised in the upland areas of West Java. It usually involves a mixture of perennial crops, such as fruit trees, bamboo, fuelwood crops and fast growing tree species. After several years, the bamboo, fuelwood crops, and some of the fast growing trees are harvested by clearcutting, and the cleared space is then used as a kebun for growing a mixture of annual food crops ("cropping"). After two or three years of kebun, the sprouting of new bamboo culms and the re-growth of young trees eliminate the space for cropping, and the land is permitted to revert to uncultivated woody vegetation until this is ready for the next harvest. As the rotation consists of a talun-kebun-talun cycle, the system is called the talun-kebun system (Christanty et a/,1981; Soemarwoto et al, 1985). 10 2.2. C U R R E N T P R A C T I C E a Talun-kebun is usually practised on private lands. Most farmers own several parcels of land and clear only one parcel annually in order to maintain a continuous cropping activity. Sometimes the parcels are located adjacent to each other, but in most cases they are scattered and located in different hamlets or even in different villages. Landless farmers and farmers who own very little land are commonly engaged in share-cropping activity, in which they rent land parcels for cropping and give 50% of the crop yield to landowners as payment. Share-cropping is one of the traditional employment-sharing mechanisms in Indonesia which has endured over several centuries, particularly in rice cultivation systems (Booth & Sundrum, 1981). In the talun-kebun activities, share-cropping is practised during the two year cropping period. After harvesting the bamboo by clearcutting, landowners, whose fields are located in other hamlets or other villages, rent out their land to share-croppers from surrounding areas. The share-croppers are responsible for all the work needed during the cropping period. They also provide all inputs (seeds, fertilizers, pesticides, and labor). After two years of cropping, the cultivation activities cease and the land is permitted to return to bamboo cover. 2.3. T H E B A M B O O TALUN-KEBUN Based on its dominant species, talun in West Java can be divided into three types: woodlot (dominated by a mixture of fuelwood and timber species); permanent mixed talun (dominated by a mixture of fruit trees and perennial cash crops); and bamboo talun (dominated by bamboo species with scattered trees between the bamboo clumps). The bamboo talun constitutes the best example of a talun-kebun rotation a Based on Christanty et al, 1986 11 cycle. Since most of the marketable production in a permanent mixed talun is derived from fruit, tree cutting is rarely done and therefore there is no clear rotation cycle between talun and kebun. Clear-cutting is practised in the woodlot only if the trees are uniform in age. Otherwise, selective cutting is employed as the main harvest method. In contrast, harvesting in a bamboo talun is usually done by clear-cutting. Bamboo culms are clear-cut, leaving the rhizomes alive but reduced in vigour. After clear-cutting, the field is cultivated in preparation for the kebun stage. Cropping is usually abandoned after two years because of competition from the bamboo regrowth, weed invasion, and the experience that productivity declines due to lowered soil fertility if food cropping is continued. The field is allowed to revert to the talun stage for three to five years after which it is ready for the next clearing and cropping. A complete rotation cycle takes about five to eight years, depending on the type of plant species grown and the inherent fertility of the soil. 2.4.PLANT COMPOSITION AND VEGETATION STRUCTURE The bamboo talun of West Java is dominated by various bamboo species, among which awi ater {Gigantochloa ater) and awi gombong (Gigantochloa verticillata) are the most common species. Although they are not dominant, other species, such as awi tali (Gigantochloa apus) and awi haur Bambusa vulgaris) are found frequently. Albasiah (Albizia falcataria), peteuy (Parkia speciosa), sugar palm (Arenga pinnata) and some fruit trees, such as mango (Mangifera indica) and durian {Durio zybethinus), are usually found scattered between the bamboo clumps. The mixture of perennial trees and bamboo forms a multilayered canopy structure. The uppermost canopy layer is usually occupied by Albizia, Parkia and Arenga. Mango and other fruit trees dominate a second layer, while the lowest layer is occupied by bamboo. Since bamboo has a very dense canopy structure, there is 12 practically no undergrowth in a mature bamboo talun', however, scattered weeds are sometimes found in the open spaces between clumps, where sunlight is still able to penetrate. Hyacinth bean (Dolichos lablab) is the main crop planted during the first year of the kebun stage. It is usually intercropped with cucumber (Cucumis sativus) and black nightshade or bitter solanum (Solanum nigrum). Crops are usually grown in rows along the contour lines, with different species planted in alternate rows. Other crops commonly planted in the kebun are: sweet basil (Ocimum basilicum), bitter melon (Momordica charantea), and chili pepper {Capsicum frutescens). Maize (Zea mays) and cassava (Manihot esculenta) are frequently planted along the edges of the kebun. The lowest layer of canopy stratification during this first year of the kebun stage is occupied by cucumbers, which creep along the ground, supressing the growth of weeds. Various vegetables (e.g. bitter solanum, chili pepper and sweet basil) occupy the middle layer, while the upper layer is occupied by hyacinth bean which grow up bamboo poles, and by fruit trees , young Albizia, and other trees which were left uncut during the clearing of the bamboo culms. Weed growth during the first cropping period is minimized by manual weeding before the mature bamboo talun is cut, and by regular manual weeding before the harvest period of each agricultural crop species. However, weeds have invaded and occupied most of the area by the end of the first year of cropping. These are killed during soil hoeing before the planting of the second year crop. The second year of the kebun stage is characterized by row cropping of cassava mixed with young bamboo regrowth and some scattered trees. Three canopy layers are formed. The upper layer is occupied by tree species, followed by young bamboo in the middle layer, and cassava in the lowest layer. These three canopy layers are spatially separated, with the bamboo being limited to a number of 13 clumps, and the cassava between these clumps and the trees. No weeding is done, and consequently weeds quickly occupy the spaces between rows and become abundant by the end of the cropping period. After harvesting the cassava (about two years after the bamboo was clearcut), there is no further cultivation of the field, permitting the bamboo to regrow until it is ready for the next harvest. Weeds are gradually shaded out as the bamboo dominates the site, and are almost completely eliminated from the sites by the mature stage of the bamboo talun. 2.5.SEQUENCE OF CULTURAL ACTIVITIES b The sequence begins with preparation for the planting of food crops. This preparation, which includes litter raking, piling and burning, and soil hoeing, takes place in the mature bamboo stage, before the bamboo is cut (the work is more pleasant in the cool shade of the bamboo). The mature bamboo culms and mature Albizia are often clearcut, but the most recently grown culms (which are still soft and have a high moisture content) are left uncut until they have become dry by exposure to the sun. Harvest residue (leaves and small branches) from the clearing is piled, dried and burned. Ash from the burning of both litter and harvest residue is collected and placed in a small hut (to protect it from rain), to be mixed with livestock manure and used for fertilizer during cropping. The mature bamboo culms and the stems of Albizia are removed from the field for sale (mostly as building materials). The dried young culms are cut and set in rows (with a defined spacing within and between rows) for use as supporting poles for hyacinth beans. Planting holes for bean seeds are prepared at the base of the bamboo poles, and shallow furrows are made between the rows for cucumber and bitter solanum. These preparations for planting take about two to three months (depending on the size of the clearing and the availability of laborers) at the end of the dry season (August-October). b Based on Christanty et al, 1986 and direct observation in the field 14 The first year cropping sequence starts with the planting of hyacinth bean seeds after the first rain (end of October /early November). Three seeds are planted in each hole (two holes and six seeds per pole), and the hole is covered with a mixture of ash and manure. Cucumber seeds are planted in the furrows a week after planting the hyacinth beans and also fertilized with ash and manure. Seedlings of bitter solanum are transplanted from a nursery area three weeks after planting the cucumber. The second fertilization, consisting of N P K (nitrogen, phosphorus, and potassium) mixed with ash and manure, is applied two or three weeks after planting. The hyacinth bean receives an additional fertilization with urea after another six weeks. Before the introduction of N P K and urea to the area, only ash and manure were used for fertilization. Weeding is done regularly (once or twice a month) before each harvest, and the weeds are composted in the field. The first product to be harvested is cucumber, about forty days after planting. Cucumber harvest continues every three to five days for up to two months. Bitter solanum can be harvested weekly, starting about a month after it has been transplanted, for a period of three to four months. The hyacinth beans are harvested after the bean plants have scenesced and the pods have dried. This harvest, which occurs six or seven months after planting, ends the first year cropping sequence. The field is rested for two or three months until the dry season ends, and then the soil is hoed to remove weeds and to prepare the field for the second year of cropping. The main crop grown in the second year is cassava. The planting of cassava cuttings usually takes place after the first rain. Cassava cuttings are planted at 1 X 1 m2 spacing. This species does not require intensive maintenance, and fertilization and weeding are rarely done. The cassava is ready for harvest after eight or nine months depending on the variety planted. The second year cropping ends after the cassava harvest. 15 The bamboo clumps start to sprout "grass stage" shoots from the underground rhizomes during the first year of cropping, and small culms are produced during the second year. At the end of the second year, the shade cast by the bamboo and other perennials limits the further growth of agricultural crops, and the field is permitted to revert to the talun stage, which will be ready for the next clearcut harvest after three to six years of bamboo fallow. The complete sequence of a talun-kebun rotation is shown in Figure 2.1, Figure 2.1 Diagramatic summary ot the bamboo talun-kebun cropping cycle in Soreang, West J a v a , Indonesia 17 2.6 .SUMMARY & DISCUSSION The bamboo talun-kebun rotation pattern has been practised by upland farmers in West Java for at least six generations, but with increasing population pressure and aspirations for a higher standard of living, this traditional practice is now being changed. External market demand and higher prices for cash crops provide a motivation for farmers to switch from bamboo to more valuable crops, such as cloves (Zyzygium aromaticum) and oranges (Citrus sp.). Unfortunately, while the gross income may be higher, these two crops require high fertilizer and pesticide inputs, as well as pest and disease control. Consequently, farmers face higher production costs and higher risks, particularly if the more intensive input requirements are not met (Christanty et al, 1986). Since most farmers gain their farming knowledge by tradition and from trial and error, adequate extension education and training on basic land management should be provided in any attempt to improve their practice or to introduce new species. The use of the talun-kebun rotation system provides an opportunity for landless farmers to be engaged in share-cropping activities. However, if the bamboo is replaced by fruit trees or cloves, the system will become more permanent without a clear short-term rotation pattern. As a result, share-cropping is unlikely to be practised. Unfortunately, the percentage of landless farmers in Java is high, and is still continuing to increase. In 1973 3.2% of the 8.8 million farmers in Java were landless, but this percentage had risen to 14.9% by 1980 (Soemarwoto, 1979; Inside Indonesia, 1985). The percentage of farmers occupying less than 0.5 ha also increased during this period, from 45.7% to 63%. Switching to off-farm employment will be very difficult since most farmers do not have the necessary skills. Consequently, these landless and poor farmers will be forced to find other land for 18 their cultivation, which in most cases will result in further clearing of the dwindHng area of forest land. The importance of bamboo in maintaining the fertility and productivity of land is reflected in the Indonesian saying "Without bamboo, the land dies". Bamboo plays a key role in restoring soil fertility through the accumulation of organic matter and nutrients during the fallow period. In fact, the color of the upper mineral soil is often used by farmers as a guide as to the timing of the next cropping. If the color of the mineral soil below the litter layer is already black or dark, farmers consider that the field is ready for the next period of cropping. Bamboo conserves moisture through the formation of a thick organic layer due to the slow decomposition of its litter, and it also minimizes surface runoff and erosion. In spite of the advantage of the traditional talun-kebun, farmers are shortening the fallow period and intensifying the cropping because of the increasing population and economic pressure. Consequently, there has been a decline in soil organic matter accumulation (based on personal observation), which in turn reduces soil fertility. Although the reduction of fallow period organic matter accumulation could be compensated for by higher fertilizer input to the system, few farmers can afford to buy the extra fertilizer. Reduced fertility leads to a decrease in productivity during the successive years. The consequences of reduced fallow periods are even worse when practised on poor or dry soils. The bamboo talun-kebun rotation cycle has sustained itself for at least several centuries with minimum external inputs of nutrients (the use of N P K fertilizers and urea is relatively recent). Intensive research is needed to identify the sensitivity of this land use system to significant changes before they occur on a wide scale. The objective ofthe research should be to assess the validity ofthe traditional "Without bamboo the land dies" saying. We cannot afford to wait to find out if it is 19 true, because the potential social and environmental consequences are unacceptable. The research should produce the knowledge and "decision support tools" that are needed to assess the risk and predict the consequences of altering the traditional talun-kebun system. This thesis contributes to this area of research. 20 C H A P T E R 3 D E S C R I P T I O N O F T H E S T U D Y A R E A 3.1.THE S T U D Y R E G I O N S The study was conducted in Soreang district, located about 23 km southwest of Bandung (Figure 3.1). The district is located at elevations between 725 m and 1100 m. About 64% of the total land area of 6,425 ha is used for upland agriculture. Bamboo talun-kebun is a typical upland land use practice in the district. The study area consisted of a survey area, where interviews were undertaken to gather information on socio-economic and management aspects of the bamboo talun-kebun system, and a field study area where research sites were established to collect data on biomass and nutrients in the bamboo talun-kebun rotation cycle. Two villages in the district (Sadu and Sukajadi) were selected for the survey. Bamboo talun-kebun is the dominant land-management practice in these two villages. About 59% of a total land area of 431 ha in Sadu, and approximately 74% of the 542 ha total land area in Sukajadi are used for bamboo talun-kebun. 3.1.1.CLIMATE The annual rainfall of the area ranges from 2000 mm to 2100 mm with a wet period that averages 9 months (October until June) and a dry period of 3 months (July until September). Monthly mean daily minimum temperature ranges from 24°C to 26°C, and monthly mean daily maximum temperature ranges from 29oC to 31°C (Village Statistics, 1980). 21 3.1.2. G E O L O G Y A N D SOIL Geologically, the area belongs to the central section of the Bandung zone, which was formed in the middle and upper pleistocene (van Bemmelen, 1949). The bedrock is primarily quarternary volcanics and tertiary basalts and andesites. West Java is characterized by a series of quarternary volcanoes, several of which are the explosive type and have periodically deposited volcanic ash over much of the region (Caroll, 1970). Over the past century, the study area has been affected by five volcanic eruptions: Mt.Guntur in 1887, Mt.Galunggung in 1918/1920, Mt.Papandayan in 1924/1927, Mt.Tangkuban Perahu in 1935 (van Bemmelen, 1949), and the latest eruption of Mt.Galunggung in 1982. A thin layer (1-2 cm) of volcanic ash from the 1982 eruption was still found on the soil surface of the mature talun sites at the beginning of the field study in August 1983. The soils of the study region, which are classified as Ultisols (Reddish Brown Latosol), are derived from intermediary volcanic tuff in the uplands, and associated low-humic gley and gray alluvial soils in the lowlands (LPT, 1966; USDA, 1975). 3.1.3. V E G E T A T I O N Although the area belongs to the medium elevation tropical mountain zone, little of the original forest vegetation remains because of land clearing for agriculture. Most of the land is privately owned and cultivated, and talun has become the most important component of upland land use. The taluns are dominated by various bamboo species, albasiah (Albizia falcataria) and suren (Toona sureni). Although uncommon, several other forest tree species, such as talingkup (Claoxylon polot), mindi (Melia azedarach), ki meong (Mallotus moluccanus), kayu afrika (Maesopsis emanii), mara (Macaranga tanarius), Omalanthus populneus, angsret (Spathodea campanulata), and hantab (Sterculia javanica), can still be found (Hadikusumah, 1982). A dense shrub thicket of ki rinyuh (Chromolaena odorata), harendong bulu 22 A dense shrub thicket of ki rinyuh (Chromolaena odorata), harendong bulu (Clidemia hirta), harendong (Melastoma malabathricum), putri malu (Mimosa pudica), and putri malu (Mimosa invisa) mixed with variousweed species, such as jarong lalaki (Achyrantkes aspera), babadotan (Ageratum conyzoides), babadotan (Ageratum houstonianum),jukut kakawatan (Cynodon dactylon), teki (Cyperus iria), teklan (Eupatorium riparium), and alang-alang (Imperata cylindrica) cover the open areas during fallow (Hadikusumah, 1982). Fruit trees, such as jackfruit (Arthocarpus heterophyllus), papaya (Carica papaya), langsat (Lansium domesticum ), horse mango (Mangifera foetida), mango (Mangifera indica), banana (Musa paradisiaca), rambutan (Nephelium lapaceum), and avocado (Persea americana) are commonly planted between bamboo clumps or Albizia trees (Sirie, 1981). Other tree or palm species, such as peteuy (Parkia speciosa), jengkol (Pithecelobium dulce), coconut (Cocos nucifera) and sugar palm (Arenga pinnata) are also found in the taluns. Coffee (Coffea canephora) is frequently planted as living hedges. Recently, cloves (Syzygium aromaticum) and oranges (Citrus sp) have been promoted in the area (Sirie, 1981; Hadikusumah, 1982). Agricultural crops, such as hyacinth bean (Dolichos lablab), bitter melon (Momordica eharantea), black nightshade (Solanum nigrum, bitter solanum, wonderberry), cucumber (Cucumis sativus), chili pepper (Capsicum frutescens), sweet basil (Ocimum basilicum), green bean (Phaseolus vulgaris), corn (Zea mays), and cassava (Manihot esculenta) are the major upland crops of the study area (Christanty et al, 1981; Hadikusumah, 1982). 23 3.1.4. L A N D U S E About 53% of the total area of 6,425 ha in Soreang is used for talun-kebun, 38% is used for ricefields, and only about 10% is used for homegardens and housing compounds (District data, 1980). This proportion varies from village to village depending on topography, altitude,, soil, and economic activity (Christanty et al, 1986). For example, 37% of the total land in Sadu is used for ricefields, while ricefields in Sukajadi, which is located at a higher altitude, only cover about 13.5% of the total land area. On the other hand, talun-kebun in Sukajadi covers approximately 74% of the total land in comparison to only 59% of the land area in Sadu (Village data, 1980). There is no significant difference in the average size of ricefield per household between Sadu and Sukajadi (about 0.084-0.085 ha), but the average area of land that is used for talun-kebun in the two villages is significantly different. In Sadu, the average area is 0.11 ha per household, while in Sukajadi the average area for talun-kebun is 0.46 ha. 3.1.5. P O P U L A T I O N The total population of Soreang, which was 73,550 in 1980 (46% male and 54% female), had an annual growth rate of 2% (District data, 1980). About 60% of the population belonged to the working age classes (age 10-55). The average population density was 1,145 persons/km2, ranging from 640-2,500 persons/km2. The population density varied from 2068 persons/km2 in Sadu to 640 persons/km2 in Sukajadi (Village data, 1980). Approximately 58% of the population was engaged in agricultural activities, but only 36% of the farmers owned their agricultural land. There were about 11,545 sharecroppers and 14,316 landless laborers, who represented 29% and 35% of the total farmers in the area, respectively (District data, 1980). 24 3.2.THE S T U D Y SITES Sites representing different stages of the talun-kebun cycle were selected for the quantification of biomass and nutrient accumulation over a management cycle. The sites were located in Selaawi hamlet, Sukajadi village, approximately 1100 m above sea level (Figure 3.1). All sites were comparable in soil condition and slope (which averaged 150 and varied little within the sites). The solum was approximately 60-75 cm deep, and was underlain by yellowish tuff. The soil surface in the mature talun plot before clearing was covered with a layer (1.5 cm depth) of moist leaf litter and fine roots. The litter layer could be separated into fresh litter, fragmented litter and decomposed litter, the latter being mixed with a thin layer of volcanic ash. The upper 5 cm of the A l horizon was a friable,granular, dark brown (7.5 YR 3/3) silty clay, containing many fine roots mixed with decomposed litter. This layer was separated by a sharp, smooth boundary to a dark reddish brown (5 YR 3/2) subangular blocky silty clay with many fine to medium roots. This second layer extended to a depth of 25 cm where it was separated by a gradual smooth boundary from a B i horizon which was dark reddish brown (5 YR 3/3), with a clay texture, a subangular blocky structure, many medium roots, and a mean depth of 20 cm. The B i layer merged gradually with a reddish brown (5 YR 4/4), moist, strongly subangular blocky clay B 2 horizon (45-75 cm), which contained very few roots. In the talun-kebun system, the litter layer is normally removed from between the bamboo clumps during the preparation of the site for first year cropping, and the upper 30 cm of mineral soil is mixed by hoeing. This destroys the boundary between A and B horizons, and causes the upper soil layers to become very friable. Study plots were established in August 1983 in six adjacent fields all of which had generally similar soil and topographic conditions (except for varying 25 litter layer and upper mineral soil condition as explained above), but which represented various stages of the talun-kebun cycle. There were five study plots: TK-1 (30mxl40m), which had a slope gradient of 15° and a southeasterly aspect, was a mature talun field (i.e. it carried a closed-canopy crop of 4-year old bamboo growth and was 6 years old since the previous clearcutting). Part of TK-1 (30mx70m) was cleared in early September 1983 and used for first year cropping. This area was subsequently known as TK-2, on which a series of experimental subplots using different agricultural crop mixtures and fertilization levels were set up in a randomized block design Figure 3.2). TK-3 (20mx35m), the plot which was used for second year cropping (with cassava), had a northwesterly aspect with a slope gradient of 15°. The soil condition was similar to TK-1, but there was no litter layer on the soil surface and no clear boundary between A and B horizons because of soil hoeing. TK-4 (20mx50m), located next to TK-2, was representative of an early bamboo fallow, i.e. the year after the cassava cropping. The soil and slope conditions were similar to TK-2. To study the effect of having a second year of cassava cropping (ie. a third successive year of cropping) instead of a first year bamboo fallow, a small plot TK-5 (7mx7m) adjacent to TK-3 was established on a field that had been scheduled for early fallow after two years of food crops. The layout of these plots is shown in Figure 3.3. Figure 3.1. Location of the study area block I S l o c k 2- 8 lodc . a B l o c k , <f Mixed F - c u c u m b e r - KyQaV\t»i bcon M i x e d U F - c u c u m b e r - hyaa'nHi M«xed P _ cucumber - SOlaHiinn - nyacii+k baw Mono U F - n y a c m + n b*«r\ Mono up Mono f Mono F Mono cacu»Hrj«r UF M o n o S o l a n u m Of-Mono Solanum P M o n o F Mono J o l a n o m UP Mono UP-. h y a c i n t h be«n M n c e d P - c u c u m b e r - solonom - hyoaVrt-Vi bean Mono P Mono F r i y a c i n r h b e a * M i / e d U P - cu c u m b e r - s o l a n o n * - hyadriiM bean Mono P M u e d U P . c u c c / m b a r _ j o l a n o r n - hyaa'nttt l*«n Mixed P - c u c u m b e r . h y a c i n t h b e a n Mono Solano** u F Mono Solano 01 F Mono p l a n u m UF (Vlono P MoAO F Mono U P Mono u p Mono t o l a n o n i P F M o n o U F M o n o U P r^ qcurm b « a ^ U F - c u w w b e / _ S o k n O w _ hyodH-rh UF: unftrti'l"'***4. Fj'g*ne 5.2.. TWe layoo*- o f Operinn«ik»l plots 'n TK - 2 . T K - 1 : mature talun T K - 2 : f irst year of cropping T K - 3 : second year of cropping T K - 4 : early fallow TK- 5 : third year of cropping N f o 10 2 0 30 4 0 50 i i i i i Figure 3.3 The layout of five study fields representing various stages of a bamboo talun-kebun cyc le . Contour lines indicate relative alt i tude. 29 C H A P T E R 4 A C C U M U L A T I O N A N D D I S T R I B U T I O N O F B A M B O O BIOMASS IN T H E TALUN-KEBUN S Y S T E M 4. l . I N T R O D U C T I O N a Bamboob is considered to be one of the noble plants in China. It has a long history and has occupied a place of pride, as expressed eight hundred years ago by Pou-Sou-Tung, a Chinese poet, who wrote that "A meal should have meat, but a house must have bamboo. Without meat we become thin; without bamboo we lose serenity and culture". It was estimated in 1980 that China produces about 3.5 million tons of bamboo annually, which is 35% of the global annual bamboo production of about 10 million tons. In addition to its social stature in China, bamboo plays a very important role in the daily life of rural people in all South and Southeast Asian countries. Although it is often called the "poor man's timber", bamboo has an age-old connection with the material needs for housing, fishing, weaponry, musical instruments, handicrafts, furniture and landscaping, and it has been closely interwoven with the life of the people. For example, the Vietnamese proverb "the bamboo is my brother" reflects the high place of bamboo in the Vietnamese society and culture. The same expression is given in the Indonesian saying "without bamboo the land dies". Bamboo is even considered as the life-blood of the rural Indonesian. It was estimated in 1980 that the total bamboo demand in Indonesia was about 6 million pieces (equal to 3.3 million metric tonnes annually), and that nearly 30% of the bamboo produced was consumed for housing and other household needs. a Unless otherwise noted, the Introduction is based on Sharma (1980) b Botanical terminology for bamboo is given in Appendix 1. 30 The importance of bamboo in providing basic human needs has been recognized for millenia, yet research activities on bamboo in most Asian countries (except Japan, China, India, Burma and Taiwan) are still very limited. Only recently have other countries become interested in bamboo research. So far, most research in Indonesia has focused on certain species growing in bamboo forest, on silvicultural practices in bamboo plantations, and on bamboo technology for industrial purposes. Very little attention has been given to village bamboo, although more than 95% of the bamboo used in Indonesia is obtained from individual farmlands or village homesteads. Although studies on bamboo ethnobotany and bamboo taxonomy have been conducted in recent years (Widjaya, 1980), there is a lack of documentation of the patterns of biomass accumulation, of production and yield, and of organic matter dynamics over an age sequence of bamboo under traditional cultivation practice. Bamboo fields in West Java are usually managed under the talun-kebun agroforestry system (as noted in Chapter 2), which is based on a short-rotation cycle between bamboo and various agricultural crops. The system has been in use for centuries, and local farmers believe that the bamboo stage of this cycle is the key to the long-term sustainability of the system. Considering the importance of bamboo for rural housing and other village structures, and since bamboo in the village is usually planted on marginal lands or hilly areas, a study of the cultivation and management aspects of village bamboo in relation to long term sustainability of bamboo yield and soil fertility needs to be conducted. 4 .2 .0BJECTIVES The objective of this part of the study was to quantify the biomass accumulation, culm diameter and height development, and culm mortality of bamboo over a complete talun-kebun rotation cycle. These data become the basis for 31 describing the biogeochemistry of the cycle (Chapter 7), and provide input data for the uptake-production relationships discussed in Chapter 9. 4 .3 .LITERATURE R E V I E W 4.3.1. S T U D I E S O F B A M B O O BIOMASS Bamboo has distinctly different growth characteristic in comparison to other woody plants. The size of culms is a function of the age of culm initiation rather than its age since initiation. The longer the time between a previous clearcut and culm initiation, the larger the culm. The oldest culms (culms formed soon after the previous clearcut) are thus the smallest. This is because a bamboo sprout (a culm) emerges from the ground with a diameter which will remain constant until it dies (McClure, 1966). The sprout grows very rapidly and attains its full height in a few weeks. Thereafter, the diameter and height of the individual bamboo culms remain unchanged with age (Figure 4.1.). Since rhizomes of the clump-forming species are short and immediately turn upward to form new culms, the photosynthate for growth of the new culms is mostly supplied by the assimilates of the mother culms, and the size of the new culms is more affected by the quality (size and vigour) of the mother culms than by rhizomes Figure 4.1. The growth of culms after clearcutting. 32 (Ueda, 1960). Therefore, under natural condition, a clump which contains large-size culms will develop large-size new culms and a clump with small-size culms will develop small-size new culms. Ueda (1960) used fresh mass to express the mass of a culm. Fresh mass of culms increases parabolically with increasing culm diameter. However, the relationship between culm diameter and culm biomass is also influenced by the thickness of the culm walls. For example, the fresh mass of a culm of Dendrocalamus strictus or Melocana bambusoides is greater than that of Phyllostachys sp. of the same diameter, because the culm walls of the former two species are thicker (Ueda, 1960). Oven dry mass is commonly used in the more recent biomass studies for bamboo than fresh mass which was used in the earlier studies (Sabhasri, 1978; Uchimura, 1981; Wang, 1981). Senescence and leaf-fall of bamboo leaves is closely related to the seasonal growth of bamboo. Ueda (1981) showed that senescence and leaf-fall for Phyllostachys bambusoides occurs during the season of maximum growth of the rhizomes. Senescence and leaf-fall in the clump-forming species of bamboo in tropical regions usually occurs at the end of the dry season, but new foliage is produced in several weeks (Ueda, 1960). The quantity of leaves per culm has been reported to be greatest for two-year-old culms (age since culm initiation) and it decreases proportionately with age for culms older than three years (Uchimura, 1980). The ratio of fresh mass of leaves to the fresh mass of culms varies depending on the the culm diameter. Culms with smaller diameter tend to have a higher ratio than culms with larger diameter. The ratio ranges from 0.05 to 0.2 for bamboo species of the monopodial type and from 0.03 to 0.08 for bamboo species of the sympodial type (Uchimura, 1980). Published information on the root systems of bamboo is very limited. Ueda 33 (1960) reported that the number of roots per culm varies according to the size and the age of the culm and the condition of the soil. He found a significant reduction in the number of fibrous roots per culm of Phyllostachys edulis and Phyllostachys reticulata when the culms became older than 6-7 years after the initiation. From their study on the extent and subterranean distribution of the root system of Bambusa tulda, White and Childers (1945 in McClure, 1966) found that 83% of the roots were concentrated in the upper 30 cm of the soil; 12% between 30-60 cm; 4% between 60-90 cm, and only 1% between 90-120 cm below the surface. Bamboo plants have two below-ground organs: roots and rhizomes. The latter are important in storing and translocating nutrients, as well as for propagation (Ueda, 1960; Hsiung, 1984). Ueda (1960) found in bamboo forest in Japan, that in Phyllostachys sp. the annual total length of the newly grown rhizomes was equal to the total length of the dead rhizomes. Therefore, the total length of the live rhizomes in bamboo groves remained constant. Rhizomes can be categorized into short-lived rhizomes Give for 4 or 5 years) and long-lived rhizomes (still alive after 8 years). From his study of the growth of short-lived rhizomes in Pleioblastus pubescens, Ueda (1960) found that two-year-old rhizomes were the most vigorous, while the four year-old rhizomes had already lost some of their vitality. Other studies have shown that the viability and growth of rhizome buds decreases with an increase in age (Ueda, 1960; Banik, 1981). A similar pattern was found in the growth of long-lived rhizomes of Phyllostachys sp. Ueda (1960) showed that 2 to 6 year old rhizomes of Phyllostachys reticulata produced a greater number of large-size culms than those produced by older rhizomes. From his study of Phyllostachys bambusoides, Uchimura (1980) found that rhizomes older than 3 years produced buds with reduced vigor and shoots that did not grow as tall. This condition relates to the changes in nutrient content of the 34 old rhizomes. Ueda (1960) reported that soluble nitrogen and phytine were higher in 2-5 year old rhizomes than in the older ones. The size of the rhizome also determines the number and size of culms produced. In most cases, rhizomes of species with large diameter culms are larger than those of species producing smaller diameter culms. The number of new culms that develop from a clump varies by species, soil and climatic conditions, harvesting technique and light intensity. For example, a decrease in photosynthate supply after clear cutting may suppress the production or growth of well developed culms for a year or more (Ueda, 1960). Kondas (1981) distinguished between daughter rhizomes, mother rhizomes, grandmother rhizomes, and great-grandmother rhizomes in the rhizomes of clump forming species. He found that the daughter rhizomes were the most active during the early succession of a bamboo clump, and their growth and development were greatly influenced by the mother rhizomes. The older rhizomes also had a collective influence but in a diminishing manner. It was not clear, however, at what stage of succession the influence of the older rhizomes became insignificant. 4.3.2.BIOMASS E S T I M A T I O N Clearcutting of bamboo clumps is the most common method of harvesting a mature bamboo talun, and this permits the collection of data on the biomass accumulated at the time of harvest. This could be accomplished by weighing the entire biomass, but such a total inventory would be very laborious and time consuming, and a total inventory approach is inappropriate for use in estimating the biomass of bamboo clumps at younger stages of the talun-kebun cycle. Consequently, regression equations based on the destructive sampling of all the culms in a few clumps at each stage of the cycle were prepared and used to estimate the total biomass of bamboo components at the different stages. Equations for predicting tree biomass are frequently based upon tree height 35 (H) and diameter at breast height (DBH). Although predictive equations obtained from D B H and H have been considered appropriate in estimating stem biomass, these two variables are not always sufficient for estimating the biomass of crown components (Satoo & Madgwick, 1982). Variables such as crown length, sapwood basal area and diameter at the base of the life crown should be included in estimators of crown components. The logarithmic relationships between component biomass and tree diameter are commonly used to estimate the biomass of aboveground tree components, because they reduce the variance associated with successive increases in tree size (Tritton & Hornbeck, 1982). However, use of logarithmic regression can result in biased estimates which tend to underestimate the true value unless a correction factor is used (Krumlik, 1974). Correction methods have been developed by Baskerville (1982), Mountford & Bunce (1973), and Satoo & Madgwick (1982). However, Madgwick (1983) found that the bias and variability of estimates were greater when correction factors were used in the calculation, as compared with logarithmic regressions without the correction factor. Therefore, the question as to whether or not biomass estimates obtained using logarithmic equations should be corrected still remains unresolved (Ferreira, 1984). Wang (1981) found that DBH2xHt was strongly correlated with bamboo dry weights, and therefore considered it as the best independent variable to be used in developing regression equations for predicting bamboo biomass. A logarithmic relationship was recommended to reduce variance associated with successive increases in both diameter and height (Uchimura, 1980; Wang, 1981). There have been very few studies on the use of regression equations to estimate root (belowground) biomass. A logarithmic equation using D B H and tree height as independent variables has been used to estimate the biomass of coarse roots of trees (Santantonio et al, 1977; Gholz et al, 1979). 36 4.4 . M E T H O D S The initial biomass sampling for mature bamboo talun was in October -1983 before the site (TK-1) was cleared in preparation for cropping. Additional sampling of clumps in various stages of the talun-kebun cycle was done in August 1984 and August 1986 to prepare the regression equations needed to estimate biomass over the entire age sequence. The agec of culms included in the regressions varied from 16 to 72 months. A total inventory was made of the clumps and individual culms therein on four subplots of 500 m2 (25x20 m) in each of 4 age classes (16, 24, 36, and 72 months) to provide data on the number of clumps per plot and the number of culms per clump at each stage of the rotation cycle. Diameter at breast height and total height were measured for each culm in each plot for subsequent use in combination with biomass regression equations. With an initial spacing of about 4.0 x 5.0 m, the number of bamboo clumps ranged from 400 to 550 with an average of 500 clumps per hectare. However, the number of culms per clump varies considerably, depending on the soil, water, light, and nutrient conditions (Yadav, 1963). My observations in various talun-kebun fields suggest that the types of crops grown during the cropping period, the frequency of soil hoeing, and the length of a complete rotation affect the growth rate of bamboo in the year after harvest. Moreover, the farmers continuously prune the bamboo sprouts during the cropping stages, particularly during the first year of cropping. This in turn inhibits the growth of new rhizomes and new culms. The biomass of branches prunned during this period is included in the calculation of biomass removals in Chapter 7. Four clumps were destructively sampled for biomass evaluation for each of c Age in months since clearcutting the previous mature bamboo talun 37 the four age classes, resulting in a total of 16 clumps being sampled for the age range of 16 to 72 months since clearcutting. The number of culms per clump ranged from 4 to 25 depending on the age of the clump (increasing number of culms as age increases). Each bamboo clump was hand-excavated by loosening the soil around the clump and then pulling the entire clump over manually (this required up to 5 people per clump). The felled clumps were then subdivided into culms, branches, foliage, and rhizomes. Total height and diameter at breast height were measured for each culm sampled, and the fresh mass of each biomass component of each culm was determined separately in the field. Five randomly chosen subsamples of each type of biomass component from each age class were weighed and then oven dried to a constant weight at 800C to determine an age-specific (time since clearcut) fresh mass to dry mass conversion factor. Dry mass was calculated for culms, branches, foliage, and rhizomes for each culm. Biomass equations (weight/individual bamboo culm) were obtained for each component using a stepwise regression technique. The (DBH) 2x ht was used as independent variable (Wang, 1981). The equations so obtained were used in conjunction with the stand data on numbers, diameters, and heights of culms to estimate biomass on a per hectare basis over the age sequence of the talun-kebun rotation cycle. All of the rhizomes produced before the clear-cutting of the mature bamboo stand in this study were categorized as mother rhizomes. Therefore, rhizomes in this study were differentiated into live mother rhizomes, dead mother rhizomes, live new rhizomes (those produced after the clearcutting), and dead new rhizomes. The proportion of mother rhizomes, live new rhizomes, and dead new rhizomes was obtained by calculating their percentage values per clump. The proportion varied with clump age and there were five replications in each age. 38 Since the culms above the live mother rhizomes were already cut during the clearcutting, only the relationship between live new rhizomes and (DBH)2xH was used in the regression equation to estimate rhizome biomass. Therefore, the results of this equation only represented the estimated biomass of the live new rhizomes. The biomass of the live mother rhizomes was then estimated based on the ratio between the live mother rhizomes and the live new rhizomes. To calculate the living biomass of rhizomes, the biomass value of the live mother rhizomes obtained here was added to the estimated biomass of the live new rhizomes obtained from the regression equation. Due to the limitation on time and labor, coarse root (>2 mm) samples were bulked per clump (4 clumps per age). Fresh mass of each bulked coarse root sample was determined, and subsamples were taken and oven dried to obtain a conversion factor to calculate dry mass of the coarse roots. Simple regression equations were then developed for the relationship between total coarse root biomass and other biomass components per clump (ie. coarse root/culm biomass ratios). Fine roots (<2 mm) were sampled at various depths: 0-5 cm, 5-25 cm, 25-45 cm and 45-75 cm. Nine mineral soil-root samples were taken randomly using a 5.2 cm diameter steel tube corer. In addition, nine samples of forest floor-root mats were collected using a 25x25 cm2 block excavation. Sampling was limited to the mature talun stage because most of the bamboo fine roots are killed by soil hoeing during the preparation for cropping, and it is not known how rapidly the bamboo fine roots reoccupy the site. Following collection, mineral soil-root samples were washed and decanted over a 0.2 mm sieve to separate root material from mineral soil. Other roots besides the bamboo roots (which had a different color) were discarded. Forest floor-root mats were suspended in water and roots were separated manually from other 39 organic materials. Fresh mass and dry mass were determined for each sample. Dry mass was then used to calculate root mass on a per hectare basis. Based on the assumption that the ratio of foliage biomass to fine root biomass remained constant over the age sequence (Kimmins, personal communication), this ratio was used to estimate fine root biomass at 16, 24, and 36 months. Because of the continuous disturbance by soil hoeing during the cropping period, fine roots did not have any chance to develop in the upper soil layer (the first 25 cm depth). Therefore, the estimation of fine root biomass in 16 and 24 months old field was only done for 25-45 cm and 45-75 cm layers. 4.5.RESULTS AND DISCUSSION 4.5.1. CULM DENSITY, DISTRIBUTION, AND MORTALITY IN VARIOUS TALUN-KEBUN STAGES. Table 4.1 summarizes the density and dimensions of bamboo culms on the four study plots, based on the 500 m2 plot inventory data. These data are examined in more detail in Figures 4.2 and 4.3. Over the time period covered in this study the variation in both diameter and height increased with age after clearcutting as newer, larger culms were added faster than the older, smaller culms which died. In much older bamboo stands there is less variation in culm size. Figure 4.2A illustrates the distribution of bamboo culms by diameter classes in the 500 m2 plots on the four study sites. The diameter of bamboo culms on the 16 month since clearcutting site ranged from 0.5 cm to 3.5 cm (see Table 4.1). The range increased with increasing age of the field, although the number of culms of smaller diameter decreased due to natural mortality of the old culms, which are always smaller in size than the new culms. For example, no culms less than 1.5 cm D B H and 4.5 m height were found in the 72 month old site. The diameters of bamboo culms at 72 months ranged from 1.5 cm to 12.5 cm in comparison to the 40 range of 0.5 cm to 6.5 cm in the 36 month field. The distribution of immature bamboo culms at each stage is slightly different from that of the mature culms. Immature culms are usually produced with a bigger diameter than the mother culms. Figure 4.2B shows that immature bamboo from the 72-month old site had a diameter range of 4.5 cm to 15.5 cm in comparison to the range of 0.5 cm to 7.5 cm from the 36-month old site, and of 0.5 cm to 5.5 cm range from the 24-month old site. Figure 4.2C illustrates the overall distribution of bamboo culms per diameter class at various talun-kebun stages. The first 16 months after clear cutting is the recovery period for the bamboo, and only small culms are produced during this period. Bamboo culms recorded in the 24 month site showed only a slight increase in culm diameters in comparison to the 16 month site, and the 36 month site showed only modest increases in culm size in comparison with the 24 month site. Only in the 72 month site was there evidence of a rapid size increase. Part of this pattern in size differentiation can be explained by the change in culm number. There is a smaller percentage change in culm density from 36 to 72 months than from 24 to 36 months ,ie. 24% in comparison to 86%, respectively. As noted earlier in this chapter, bamboo has distinct growth characteristics. The diameter of an individual bamboo culm is already determined when it sprouts from the ground, and it will remain constant as the culm ages (McClure, 1966). Culm diameter increases as sprouting time since clearcutting increases. Therefore, bamboo culms in the 72-month site can be differentiated into various ages according to their diameter. Some culms that are still alive at 72 months were produced during the first 16 months after clear cutting, some sprouted within 24 months, others were produced between 24 months and 36 months, and the rest sprouted during the period between 36 months and 72 months after clearcutting. At this latter stage, there were only 20% as many bamboo culms less than 3.5 cm D B H 41 (those produced during the first 16 months) as in the 36-month site. Reduction in the number of small-size culms occurs as a result of natural mortality which may be due to light and nutrient competition with larger-size culms in the clump. Ueda (1960) stated that the growth of the new culms in a clump-forming species is assisted by the assimilate from the mother culms. Since the maximum diameter of individual culms is attained three years after culm initiation (Wang, 1981), bamboo culms produced during the period of 36 months and 72 months have larger diameters than those produced in the former period. The number of culms larger than 3.5 cm diameter in the 72-month site was 224% higher than that in the 36 month site. The distribution of bamboo culms per height class in the four study plots is illustrated in Figure 4.3. The distribution is similar to that for diameter. New culms in the clump-forming species grow on the periphery of the clump surrounding the old shorter culms which are usually concentrated in the center (Ueda, 1960). New culms are taller than the old culms; consequently, old culms in a dense clump are frequently shaded out by the new culms. The data show, however, that some culms less than 3 m were still alive in the 72 month field (Figure 4.3A). Figure 4.4 summarizes culm development and mortality on the four study plots. Culm mortality was low in the periods from 16 and 36 months, and was observed only for culms less than 1.5 cm D B H and shorter than 3 m height. New culms were produced with a bigger diameter (Figure 4.4A) or greater height (Figure 4.4B). Culm mortality increased with increasing age. Figure 4.4.A shows that about 60% of the culms less than 5.5 cm D B H died during the period of 36 to 72 months. A trend similar to this is shown in Figure 4.4B, where 53% of the culms shorter than 9 m height during the period of 36 to 72 months were shaded out by taller culms. Table 4.1. The density and dimensions of bamboo culms in various stages of the talun-kebun system in Selaawi hamlet, Soreang district, West Java, Indonesia. Age Total culms Mature culms Immature culms (months) density (# ha-l) dbh range (cm) ht range (m) density (# ha-1) dbh range (cm) ht range (m) density (# ha-1) dbh range (cm) ht range (m) 16 1340 0.5-3.5 1.5-6.0 1060 0.5-3.5 1.5-4.5 280 0.5-3.5 1.5-6.0 24 2940 0.5-5.5 1.5-7.5 2460 0.5-5.5 1.5-7.5 480 1.5-5.5 1.5-7.5 36 5480 0.5-7.5 1.5-12.0 4340 0.5-6.5 1.5-12.0 1140 1.5-7.5 3.0-10.5 72 6820 0.5-15.5 1.5-21.0 5840 1.5-12.5 1.5-16.5 980 4.5-15.5 7.5-21.0 4^  70 .5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 155 16.5 80 Diameter Class (cm) Figure 4.2 Distribution of bamboo culms per diameter classes in 5 0 0 m 2 plots of various talun-kebun stages 70 r • «* • Y i i O i i — r j i i i * " a — i i i i 1.5 3J0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 / \ B. Immature bamboo culms 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.022.5 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 Height C lass (m) Figure 4 .3 Distribution of bamboo culms per height classes in 5 0 0 m 2 plots of various talun-kebun stages A. Based on diameter c lasses 16-24 months 24 -36 months 36-72 months 80 r .5 2.5 4.5 6.5 .5 2.5 4.5 6.5 8.5 .5 2.5 4.5 6.5 8.5 10.5 12.5 14.5 16.5 diameter class B. Based on height c lasses new culm old culm dead culm height class Figure 4.4 Culm development and mortality on the four study plots Cn 46 4.5.2.REGRESSION EQUATIONS TO ESTIMATE LIVING BIOMASS 1. ABOVE GROUND AND RHIZOME BIOMASS Logarithmic regression equations were developed to estimate the dry weights of various bamboo components (culms, foliage, branches and rhizomes). Three independent variables: D B H (diameter at breast height), H (height), and (DBH)2xH were tested. Variable selection using the scatter diagram and the backwards option provided by the MIDAS statistical package (Fox and Guire, 1976) for each component indicated that (DBH)2xH was the best independent variable to be used in the regression equations, as has also been found in other bamboo studies (Uchimura, 1980; Wang, 1981). Wang (1981) also recommended the use of a logarithmic relationship to reduce the variance associated with successive increase in both diameter and height. Since culm diameter and height are determined by age of the bamboo clump, separate equations for each age class are preferable (Wang, 1981). Immature culms were separated from the mature culms in the development of the regression equations due to their significant differences in the thickness of the culm wall as well as in the production of foliage and branches. The equations for component biomass of immature culms were developed based upon a sample of 12 individual culms taken from various fields. The number of mature culms per clump varied with age of the clump (the number increases with increasing age). Final equations for component biomass at various age classes are presented in Table 4.2. Table 4.2. Equations for biomass (kg ha"1) of each component of bamboo in different stages of the talun-kebun system in Selaawi hamlet, Soreang district, West Java, Indonesia. Age & Equations r 2 standard # of component error observation 16 months: culm log y= -2.4289 + 0.76389 log d2h 0.93 0.17 15 foliage log y= -3.6360 + 0.70123 log d2h 0.96 0.12 15 branch log y= -3.4280 + 0.58566 log d2h 0.87 0.18 15 rhizome logy= -3.1688 + 0.78345 log d2h 0.90 0.21 15 24 months: culm logy= -2.1122 + 0.62390 log d2h 0.96 0.15 15 foliage log y= -3.3232 + 0.62321 log d2h 0.96 0.14 15 branch log y= -3.3978 + 0.62103 log d2h 0.97 0.13 15 rhizome log y= -2.7436+ 0.62182 log d2h 0.91 0.22 15 36 months: culm log y= -1.4239 + 0.48918 logd2h 0.96 0.09 15 foliage log y= -2.9284 + 0.57607 log d2h 0.90 0.18 15 branch log y= -2.8933 + 0.54484 log d2h 0.92 0.16 15 rhizome log y= -2.5712 + 0.54517 log d2h 0.93 0.15 15 72 months: culm log y= -2.2042 + 0.64613 log d2h - 0.000096606 d2h 0.99 0.08 15 foliage log y= -3.9530 + 0.64706 log d2h - 0.000086076 d2h 0.99 0.07 15 branch log y= -3.5331 + 0.60495 log d2h 0.99 0.08 15 rhizome logy= -3.4203 + 0.65078 log d2h - 0.000108400 d2h 0.99 0.08 15 immature bamboo: culm log y= -2.3484 + 0.52015 log d2h 0.98 0.06 12 foliage log y= -3.9530 + 0.63939 log d2h 0.92 0.14 12 branch log y= -5.2343 + 3.7666632 log d2h 0.82 0.28 12 rhizome • log y= -1.6847 + 0.50402 log d2h 0.96 0.08 12 D = Diameter at breast height (cm); H = Total culm height (m). 48 2.COARSE ROOT BIOMASS Simple regression equations for coarse root biomass were developed based on the relationship between total coarse roots and other component biomass per clump. It was found that coarse root biomass showed the best relationship to rhizome biomass. Due to the Umited number of samples (only 4 clumps per age class), there was no differentiation into age classes in developing the equations. The equation selected to estimate per clump coarse root biomass is a simple power regression equation: y= 0.079 * x l . l which can be expressed in a logarithmic equation as In y=-2.538 + 1.1 In x r2 = 0.98 where x is the rhizome dry mass and y is the coarse root dry mass. 4.5.3.ESTIMATED BIOMASS 1 .ABOVE GROUND BIOMASS Biomass of each above-ground bamboo component in each plot was calculated by applying the regression equations to the inventory data on culm diameter and height. Calculated biomass values for individual culms were summed to obtain stand values on a 500 m2 plot basis. The results were then multiplied by 20 to give the stand values per hectare. The estimated biomass for each bamboo component over an age sequence is presented in Table 4.3. The biomass per ha of each component increased with increasing field age due to the increase with age in both the number of culms per hectare and the biomass per culm. Table 4.3. Aboveground biomass accumulation (t ha"1) of each bamboo component and its percentage distribution (%) in various ages of bamboo fields in Selaawi hamlet, Soreang district, West Java, Indonesia. Age (months) Culm density (# ha-l) Aboveground biomass (t ha 1 ) Culm (SD) (%) Branch x (%) (SD) Foliage x (%) (SD) T o t a l (SD) b) 16 24 36 72 1340 2940 5480 6820 0.4 (0.05) .2.7 (0.16) 9.2 (0.37) 34.4 (1.50) (71) (66) (65) (76) 0.1 (0.01) 0.7 (0.04) 2.4 (0.14) 6.0 (0.15) (16) (16) (17) (13) 0.1 (0.07) 0.7 (0.05) 2.6 (0.18) 4.7 (1.24) (13) (18) (18) (11) 0.6 (0.07) 4.1 (0.25) 14.2 (0.70) 45.1 (1.80) (100) (100) (100) (100) 50 The culm biomass at 16 months was small (0.4 t ha - 1), but the value increased quite rapidly to 2.7, 9.2, and 34.4 t ha-1 at the ages of 24, 36 and 72 months, respectively. The low value at 16 months resulted from the continuous disturbance of soil during the first 12 months by hoeing, weeding and food crop harvesting, which in turn influenced the growth of rhizomes, roots and the sprouting ability of the buds. Since the photosynthate needed for the growth of new culms is supplied by the assimilation of mother culms, rapid growth only occurred after the field had become fully occupied by a population of mother culms at 24 months. Wang (1981) reported an above-ground biomass of 43.23 t ha- 1 for a 5 year Moso-bamboo stand in Taiwan, and an estimated range of 25 t ha- 1 to 73 t ha- 1 was reported for 8 to 10 year old bamboo of the genus Phyllostachys (Uchimura, 1980). Culm biomass values estimated in this study are lower than those of the Moso-bamboo stand, but are within the range reported by Uchimura (1980). Branch and foliage biomass per ha increased with increasing field age. Values for branch biomass increased from 0.11 ha- 1 at 16 months to 6.0 t ha- 1 at 72 months. Foliage biomass increased from 2.6 t ha- 1 to 4.7 t ha- 1 between 36 and 72 months. This compares with values of 2.5 t ha- 1 to 4.7 t ha- 1 for the foliage biomass of mature Sasa as found by Oshima (1961 in Wang, 1981). The distribution of total aboveground biomass at 72 months was quite similar to values reported by Wang (1981) for a Moso-bamboo stand. Culms represented 65% to 76% of aboveground bamboo biomass in various ages. Values for the proportion of aboveground biomass in branches and foliage were higher for 36 months (17% for branches and 18% for foliage) than for 72 months (13% for branches and 11% for foliage). 51 2. B E L O W G R O U N D BIOMASS A . R H I Z O M E BIOMASS Figure 4.5. illustrates the percentage distribution of rhizomes at various ages after clearcutting a mature bamboo stand. The percentage of live mother rhizomes per clump decreased with age after clearing, while the percentage of dead mother rhizomes per clump increased (Figure 4.5A). Rhizomes of clump forming bamboo species are usually short-lived, and the mother rhizomes lose vitality and vigor after the new culms reach maturity (4 or 5 years after clearcutting). By then, most mother rhizomes were dead. All mother rhizomes found at 72 months after clearing were dead and they accounted for 10% of the total rhizomes per clump. The percentage of new rhizomes per clump was low until 36 months after the clearcutting. The poor growth of new rhizomes was likely due to the continuous disturbance during the cropping period. The assimilate of the mature culms is important for the growth of new rhizomes and for the production of new culms. Following clearcutting, the production of new rhizomes depends entirely on the photosynthates and nutrients stored in the mother rhizomes. Once new rhizomes were established and had developed new culms, the growth of the subsequent culms depended on the assimilate of the new culms and not on the mother rhizomes. The farmers continuously prune branches and foliage of bamboo sprouts to prevent shading during the two year cropping period. As a result, the growth of new rhizomes and new culms was inhibited until after harvesting the last crop (cassava) at 24 months. The bamboo was then able to increase its production of photosynthate. This situation was reflected in the increasing percentage of live new rhizomes with age after clearcutting (Figure 4.5B). Although a portion of the new rhizomes failed to develop and died during their growth, the percentage of dead new rhizomes per clump was low. 52 Rhizome biomass was partitioned into different components over the various stages of the talun-kebun (Figure 4.6). Biomass of live mother rhizomes decreased, while the biomass of live new rhizomes increased with increasing field age. Live new rhizome biomass at 16 months was small, but increased rapidly thereafter. In contrast, the highest biomass of live mother rhizomes was found at 16 months, dropping dramatically by 36 months and becoming zero by 72 months after clearcutting. The biomass values for total live rhizomes were obtained from the summation of live new rhizome and live mother rhizome biomass at each age. By 36 months the value for total live rhizome biomass decreased by 33% due to the mortality of mother rhizomes. With the high production of live new rhizomes during the fallow period (from 36 to 72 months), however, the value of live rhizome biomass almost doubled (Table 4.4). Table 4.4. Live rhizome biomass (t ha-1) in various ages of bamboo fields in Selaawi hamlet, Soreang district, West Java, Indonesia. Age Rhizome biomass (t ha - 1) (months) Mother rhizome New rhizome T o t a l 16 8.4 0.2 8.6 24 5.1 1.4 6.5 36 2.1 3.7 5.8 72 - 10.5 10.5 53 A. Mother rhizomes live mother rhizome dead mother rhizome 2 0 4 0 6 0 age (months) Figure 4 . 5 The percentage distribution of rhizomes at var ious ages af ter c l e a r c u t t i n g a m a t u r e bamboo talun ] dead new rhizome live new rhizome live mother rhizome dead mother rhizome Figure 4 . 6 Rhizome biomass in various ages of the talun-kebun cycle. 55 B. COARSE AND FINE ROOT BIOMASS Table 4.5 shows the estimated values of live coarse and fine root biomass in the four plots. Since the coarse roots which were produced by the mother rhizomes during the previous rotation were already dead after 16 months, they were excluded from the calculation. Table 4.5. Coarse and fine root biomass in various ages of bamboo fields in Selaawi hamlet, Soreang district, West Java, Indonesia. Age (months) Coarse root biomass (t ha-l) Fine root biomass (t ha-l) 16 0.0 0.1 24 0.2 0.8 36 0.7 7.0 72 2.1 18.9 The total fine root biomass in the mature bamboo field (72 months) was approximately 18.9 t ha-l. The fine root biomass values in this study are higher than the total fine root biomass in the 8-year-old regrowth of tropical rain forest in Costa Rica (Berish, 1982), but are within the range reported for tropical forests (Santantonio et al, 1977). Although fine roots consisted of both live and dead roots, my observations in the field suggest a rapid decomposition of bamboo fine roots (within several months); therefore, it was assumed in this study that fine root biomass was mostly composed of live roots. The fresh mass of total roots (coarse and fine roots) in a 4-year-old stand of Phyllostachys reticulata ranged from 2 to 211 ha-l, and a higher range (15 to 35 t ha-l) W as reported in a 5 year stand of Phyllostachys edulis (Ueda, 1960). With a ratio of dry mass to fresh mass of approximately 0.7, the dry mass values of total 56 roots ranged from 1.4 to 14.7 t ha"1 for Phyllostachys reticulata and from 10.5 to 24.5 t ha-l for Phyllostachys edulis. Thus, the range of 16 to 25 t ha-l obtained in this study is higher than that of Phyllostachys reticulata, but similar to the range found for Phyllostachys edulis. Table 4.6 summarizes the biomass and density of fine roots in various soil depths. The densities of surficial root mats and of fine roots in the 0 to 5 cm soil layer were high, but dropped quite rapidly with increasing depth. Although the density of surficial root mats was high, root mats only contributed 5% of the total fine root biomass. Table 4.6. The biomass and density of fine roots at various soil depths of a mature bamboo talun in Selaawi hamlet, Soreang district, West Java, Indonesia. Soil depth (cm) Root density (kg m-3) Root biomass (t ha-l) % by depth root mats 8.59 -10.26 0.9 ± 0 . 1 5 0- 5 9.05 - 14.50 5.1 + 1.0 27 5-25 2.62- 4.58 7.5 ± 1.2 40 25-45 1.16- 2.92 4.4 + 1.0 23 45-75 0.19- 0.49 0.9 + 0.5 5 T o t a l 18.9 + 1.9 100 * mean + 95% confidence interval 4.5.4. B I O M A S S D I S T R I B U T I O N Table 4.7 provides a summary of the estimated biomass values of each bamboo component over the age sequence. Aboveground biomass accounted for only 6% of total bamboo biomass at 16 months, but the percentage value increased significantly to 34% at 24 months. After 57 36 months aboveground biomass contributed over 50% to the total biomass. The proportion of total bamboo biomass allocated to the aboveground system at 72 months was within the range of 50 to 60% given by Uchimura (1980) for mature stands of Phyllostachys sp. Culms contributed 33 to 45% of the total biomass between the age of 36 and 72 months. Figure 4.7. illustrates the distribution of bamboo biomass between aboveground and belowground components at 16, 24, 36, and 72 months. Unlike the distribution pattern of the aboveground biomass, the proportion of the total biomass contributed by belowground components decreased with increasing age. It accounted for about 94% of the total bamboo biomass at 16 months, but represented only 41% at 72 months. 6. S U M M A R Y A N D C O N C L U S I O N S Accumulation and distribution of bamboo biomass was studied over an age sequence of talun-kebun fields located at Selaawi Hamlet, Sukajadi Village, Soreang District, West Java, Indonesia. Biomass of each living bamboo component (including rhizome and root systems) was determined as a basis for describing bamboo growth responses to site disturbance during a complete rotation of a talun-kebun cycle. Culm diameter, height development, and mortality were also quantified to provide information needed for various modeling activities to be presented later in Section IV. Results from this chapter lead to the following conclusions: 1. The biomass of each aboveground component increases with increasing age due to the increase with age in both the number of culms per hectare and the biomass per culm. 2. The continuous disturbance of soil by hoeing, weeding and food crop harvesting during the cropping period suppressed the growth of 58 rhizomes, roots and the sprouting ability of the buds. 3. Mother rhizomes contributed significantly to the belowground biomass until 36 months after the clearance, but were completely replaced by live new rhizomes by the age of 72 months. 4. Surficial root mats and fine roots which were concentrated in the upper 25 cm of the soil layer after 72 months were killed by soil hoeing during preparation for cropping. The establishment of new root mats and upper layer fine roots did not start until 36 months after the clearance. 5. The distribution of total bamboo biomass between aboveground and belowground components varied with age. Aboveground biomass only represented 6% in comparison to 94% contributed by the belowground biomass at 16 months. By the end of the mature talun, however, aboveground biomass accounted for 59% compared with 41% contributed by belowground biomass. Table 4.7. The estimated biomass (t ha -1) of each bamboo component over the age sequence of a talun-kebun. The number in brackets represents the percentage of the total biomass contributed by that component. Age Aboveground biomass Belowground biomass Total (months) biomass Culm Branch Foliage Total Rhizome Roots* Total 16 0.4 0.1 0.1 0.6 8.6 0.1 8.7 9.3 (4) (1) (1) (6) (93) (1) (94) (100) 24 2.7 0.7 0.7 4.1 6.5 1.0 7.5 11.6 (23) (6) (6) (35) (56) (9) (65) (100) 36 9.2 2.4 2.6 14.2 5.8 7.7 13.5 27.7 (33) (9) (9) (51) (21) (28) (49) (100) 72 34.4 6.0 4.7 45.1 10.5 21.0 31.5 76.6 (45) (8) (6) (59) (14) . (27) (41) (100) coarse roots + fine roots (6.1) roots Figure 4 . 7 The percentoge distribution of bamboo biomass at 16, 24, 36 , and 72 months. Values in brackets are percentage of total biomas. 61 CHAPTER 5 ACCUMULATION AND DISTRIBUTION OF AGRICULTURAL CROP AND WEED BIOMASS IN THE TALUN-KEBUN SYSTEM The two year cropping stage is characterized by mixed cropping of various annual crops in a temporal pattern that ensures a continuing flow of crop production throughout the year. This chapter is divided into three sections. The first section describes biomass accumulation and yields of cucumber, bitter solanum, and hyacinth bean grown in mixture during the first year of cropping. The second section deals with biomass accumulation and yields of cassava during the second year of cropping. In order to evaluate changes in biomass and yields if cropping is prolonged, an assessment of the biomass accumulated during an extra year of cassava planting was conducted in a nearby field. The last section of the chapter discusses biomass accumulation of weeds over a talun-kebun rotation cycle. Weeds often become a problem during the cropping stage, and previous studies have reported that weed growth is considered to be one of the primary reasons for field abandonment in shifting cultivation (Joachim & Kandiah, 1948; Seavoy, 1973). The results of this chapter are used to evaluate the farmers' management strategy in designing the cropping sequence of a talun-kebun cycle. 5.1. THE FIRST YEAR CROPPING 5.1.1.INTRODUCTION The main crops grown during the first year cropping of a talun-kebun system in Soreang are hyacinth bean (Dolichos lablab), cucumber (Cucumis sativus), and 62 bitter solanum (Solanum nigrum, black nightshade). The planting arrangement of these crops has been described in Chapter 2. Hyacinth bean has been widely used for food, cattle feed and soil improvement in Angola, Egypt and other African countries, as well as in Brazil (Schaaffhausen, 1962). In Indonesia, particularly in West Java, hyacinth bean is grown for food. The pods are economically valuable and are known to be a good source of protein; they have been reported to contain 20-28% protein (Schaaffhausen, 1962). As a nitrogen-fixing species, hyacinth bean is commonly used as green manure. A report from Sudan (Neme, 1955 in Schaaffhausen, 1962) showed that incorporating hyacinth bean into the soil as green manure increased maize yield from 2.4 t ha-l to 4.1 t ha-l. Luck (1965) reported 220 kg ha-l N was produced by leaf decay of hyacinth bean when planted as green manure. This figure, however, will be significantly reduced if hyacinth bean is planted as a food crop. Keya (1977) found that about 60-90% of the nitrogen accumulated in leguminous crops may be removed in harvest. He stated that in legumes with 100 kg ha- 1 yr-1 nitrogen fixation capacity, only 5-20 kg would benefit the successive crop. Schaaffhausen (1962) also pointed out the role of the deep tap roots of hyacinth bean in bringing up nutrients from the lower soil depths to the surface. These nutrients otherwise would not be available to other annual crops. Since it takes about eight months from the time of planting hyacinth bean until it can be harvested, farmers usually intercrop this species with cucumber and bitter solanum. In Indonesia, cucumber is usually cultivated in dry rice-fields or on upland fields, but it is also commonly intercropped with beans in the talun-kebun system. Since cucumber reaches maturity in only 6-8 weeks after sowing the seeds, and since the entire cucumber cropping period is 12 weeks, there is not thought to be significant interspecies competition between cucumber and the bean. Cucumber is also widely cultivated in slash and burn agriculture (jhum system) in India (Toky 63 & Ramakrishnan, 1980). After harvesting the fruits, farmers usually leave the shoots and roots to decompose in the field. Bitter solanum is characteristic of West Java. This species is also known as black nightshade, which is often considered to be poisonous by Europeans; however, the varieties grown in Indonesia are harmless (Ochse, 1977). People in West Java use the fruits as a vegetable, while in West Africa people use the leaves as spices for cooking. Bitter solanum is propagated by seeds, and the average yields are 12-15 t ha-iyr-l of fresh products (Tindall, 1983). This species is shade intolerant, and its production is significantly reduced by shading (Omta & Fortuin, 1978). The yield of each of the species grown in a mixed cropping is lower than that produced in a monoculture of that species when it alone fully occupies the area. However, previous studies have shown that the total yield of a mixed culture may be considerably higher than that of any one of its component species grown in monoculture over the same area (Janzen, 1952; Donald, 1963; Trenbath, 1981). The difference in the timing of soil nutrient utilization seemed to be a significant factor for obtaining higher yields of crop mixtures (Trenbath, 1974; Andrews & Kassam, 1976; Kass, 1978; Willey, 1979a; 1979b). Unlike conventional shifting cultivation practice, farmers use fertilizer inputs during the first year of cropping. They believe that the use of animal manure and inorganic fertilizer in addition to using ash from the burning of bamboo litter and slash can increase yields of the vegetable crops, particularly the yields of hyacinth bean. This part of the chapter discusses biomass accumulation and distribution of the first year crops grown in mixture, as usually practised by the farmers. The results will be used to estimate nutrient distribution during the first year cropping. 64 5.1.2.0BJECTIVES The objective of this part of the study was to quantify the temporal patterns of biomass accumulation of the first year crops grown in mixture, as usually practised by the farmers. These data were used as the basis for describing nutrient dynamics during the cropping stage (Chapter 7), and provided the input data for the uptake-production relationships presented in Chapter 9. 5.1.3.STUDY SITE The study area has been described in Chapter 3. 5.1.4.METHODS A series of experimental subplots using different agricultural crop mixtures and fertilization levels was set up in a completely randomized block design. Since this chapter is focused on the traditional management pattern, only the results of the traditional mixed cropping with fertilizer application are presented here. The treatment plots in TK-2 were arranged in a completely randomized block design, with four replications of each of the treatments (Figure 3.2). The individual sub-plots for mixed cropping were 12 x 10.5 m (126 m2), with 1.75 m spacing within a row and 2 m spacing between rows for hyacinth beans. Cucumber and bitter solanum were planted at 40 cm spacing between individual plants within a row. The layout is illustrated in Figure 5.1. 65 Figure 5.1. Spatial Arrangement of Agricultural Crops in the First Year Cropping Stage of the Talun-kebun in Soreang, West Java, Indonesia o o o o A * A * A * A * A * A * A A * A * A * A * A * A * A A * A * A * A * A * A * A A * A * A * A * A * A * A 0 0 0 0 A * A * A * - A * A * A * A A * A * A * A * A * A * A A * A * A * A * A * A * A A * A * A * A * A * A * A o Hyacinth bean A Cucumber * Bitter solanum Planting of hyacinth bean was done on November 15, 1983. Biomass sampling was done at 45, 62, 84, 125, and 180 days after planting. Three individual plants per subplot were sampled, which resulted in a total of 12 individual plant samples per age. Cucumber seeds were planted a week after planting the hyacinth bean. Cucumber fruits can be harvested regularly for about 6 weeks, starting at 45 days after planting the seeds. Therefore, individual plant sampling was done at 22, 26, 66 32, 38, 45, and 62 days after planting. Three replicate plants from each subplot were sampled on each date, which resulted in 12 individual plant samples per age. After 4 weeks of growth in the nursery, bitter solanum seedlings were transplanted to the experimental plots. Biomass sampling was started 6 weeks after transplanting at the age of 70, 86, 100, 130, and 160 days. Three plants were sampled in each of the four subplots, resulting in a total of 12 samples per age. Each individual plant sample was separated into foliage, stem, root, flower, and fruit or pod. The fresh mass of each component was determined directly in the field. Each sample was then oven dried to a constant mass at 80°C to determine dry mass, which was then used to calculate biomass on a per hectare basis. 5.1.5.RESULTS 1. CROP BIOMASS A. CUCUMBER Table 5.1 shows that the biomass (dry mass) of each vegetative component (roots, stems, foliage) increased with increasing age until the crops reached full maturity at 45 days. Foliage comprised about 70-85% of the total biomass at 22 to 38 days, but the percentage values dropped significantly by 64 days when most of the biomass was accumulated in fruits. Roots made up about 2-9% and stems contributed 2-12% of the total biomass during the cucumber cropping period. About 60% of the total biomass was accumulated in fruit biomass at 45 days and the percentage increased slightly to 68% at 64 days. The total fruit biomass (expressed in dry mass harvested over a cropping period was 303.9 kg ha- 1. The fruits were harvested regularly every 3 days from 38 days onwards, the values of fruit biomass at 45 and 64 days being the cummulative values of fruits harvested during the period of 38 to 45 days, and 45 to 64 days, respectively. 67 B. B ITTER SOLANUM The biomass of each component of the bitter solanum is presented in Table 5.2. Root and stem biomass increased with increasing age. Roots accounted for 12-23%, while stems represented 32-67% of the total biomass during the cropping period. Foliage biomass increased from 32 kg ha-l at 70 days to 90 kg ha-l at 100 days, but declined rapidly thereafter, to 11 kg ha-l by the end of the cropping period (160 days). The standard deviation of the foliage biomass value at 160 days was higher than the mean value due to the fact that 25% of the plants were leafless on that date (most leaves had fallen to the ground). The harvest of bitter solanum fruits started at 86 days and continued regularly every two weeks until 160 days. Therefore, the fruit biomass values presented in Table 5.2. are the values accumulated upto 86 days, then between 86-100 days, 100-130 days, and 130-160 days. The highest dry mass value of fruit biomass was achieved at 100 days (196.1 kg ha-l). The biomass decreased to 119.9 kg ha-l at 130 days and dropped significantly to 22 kg ha- 1 at 160 days. The total fruit biomass harvested during the cropping period was 365.5 kg ha- 1. C. HYACINTH B E A N Table 5.3. summarizes the biomass of each component of hyacinth bean. The root and stem biomass increased with increasing crop age. The maximum foliage biomass was obtained at 125 days, but dropped dramatically at 160 days, because most of the leaves were dry and had fallen prior to harvest. In contrast to cucumber and bitter solanum, harvesting the pods of hyacinth bean was only done at 180 days, at which time the entire plants were pulled out of the ground. The total fruit biomass harvested was 5020.8 kg ha-l. Table 5.2. Living biomass of bitter solanum grown in mixture during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period. Age (days) Roots Stems Biomass (kg ha" Foliage 1 dry mass)* Flowers Fruits** T o t a l 70 21.8 + 12.3 (22) 42.7+ 24.3 (43) 32.0 + 12.0 (32) 1.6 + 0.9 (2) 1.2+ 1.5 (1) 99.3 + 48.4 (100) 86 29.0 + 13.3 (19) 56.7+ 28.3 (36) 46.0 + 17.0 (29) 2.1 + 1.3 (1) 23.8+ 5.7 (15) 157.6 + 59.8 (100) 100 64.0 + 32.0 (12) 164.0+ 71.4 (32) 90.2 + 37.3 (17) 3.2 + 2.1 (1) 196.1 + 38.8 (38) 517.5 + 158.1 (100) 130 79.8 + 18.7 (18) 205.0 + 136.9 (46) 34.8 + 23.0 (8) (28) 119.9 + 20.3 (100) 443.3 ± 183.0 160 82.5 + 24.4 (23) 240.0 + 103.6 (68) 11.3 + 13.4 (3) - 22.0 + 11.1 (6) 355.6 + 131.9 (100) *+mean + 95% confidence interval cumulative biomass of fruit harvested within a sampling period Table 5.1. Living biomass of cucumber grown in mixture during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period. Age (days) Roots Stems Biomass (kg ha ' 1 dry mass)* Foliage Flowers Fruits** T o t a l 22 1.0 + 0.3 (9) 1.3 + 0.5 (12) 8.4+ 0.4 (79) - - 10.7+ 4.1 (100) 26 1.4 + 0.5 (6) 2.1 + 1.3 (8) 21.2 + 7.3 (86) - - 24.7+ 8.1 (100) 32 2.8 + 1.0 (7) 2.8 + 1.0 (7) 33.9 + 17.8 (79) 3.0 + 2.1 (7) - 42.5 + 20.5 (100) 38 4.7 + 2.1 (5) 5.0 + 2.3 (5) 69.6 + 33.5 (71) 5.1 + 2.8 (5) 13.2 + 0.3 (14) 97.6 + 35.0 (100) 45 5.9 + 1.7 (3) 7.0 + 2.0 (3) 74.9 + 23.6 (32) 4.9 + 2.3 (2) 140.0 + 19.6 (60) 232.7 + 38.7 (100) 64 4.4 + 1.4 (2) 5.7 +1.7 (3) 57.6 + 14.4 (26) 4.6 + 1.7 (2) 150.7 + 33.2 (67) 223.0 + 29.5 (100) * mean + 95% confidence interval cumulative biomass of fruit harvested within a sampling period Table 5.3. Living biomass of each component of hyacinth bean grown in mixture during the first year cropping of a bamboo talun-kebun in Soreang, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period. Age (days) Roots Stems Biomass (kg ha"1 Foliage dry mass)* Flowers Pods** T o t a l 45 10.3+ 2.5 (11) 35.5+ 15.0 (37) 49.3+ 38.7 (52) - - 95.1+ 51.7 (100) 62 16.3+ 4.7 (5) 88.3 + 40.9 (28) 207.5+ 88.0 (67) - - 312.3+ 128.7 (100) 84 41.7 + 13.9 (4) 358.3 + 150.3 (36) 587.3 + 200.6 (59) 9.3 + 4.2 (1) - 996.6+ 351.9 (100) 125 96.7 + 22.5 (3) 1474.9 + 270.2 (47) 945.7 + 214.7 (30) 12.3 + 3.9 (1) 639.3 + 313.0 (20) 3168.9 + 704.0 (100) 180 103.9 + 20.8 (1) 1847.3 + 404.4 (25) 373.7 + 144.3 (5) - 5020.8 + 1342.6 (69) 7345.7 + 1585.6 (100) mean + 95% confidence interval standing crop of pods. Pod harvest only occurred on day 180 71 2. BIOMASS R E M O V A L The biomass removed during the first year cropping consisted of harvested materials, such as fruits and pods, and part of the harvest residues (i.e. roots, stems, leaves). Since roots, stems, and leaves of cucumber decompose rapidly after the harvest period, farmers usually leave all cucumber residues on the ground. Therefore, 303.9 kg ha-l of fruit was the only biomass removed from the field during the cucumber cropping. As the harvest residues (stems and roots) of bitter solanum decompose slowly, these materials were usually removed from the field. Leaves had mostly fallen to the ground by the time these residues were removed, but if not the residues were shaken vigorously so that almost all of the leaves fell off and remained on the site. Therefore, the total biomass removal during the bitter solanum cropping consisted of fruits, stems and roots. The total biomass removed was approximately 684.3 kg ha-l. The end of the first year cropping was determined by the harvest of hyacinth bean pods. Since the plants grew up bamboo poles, harvesting was done by pulling out the entire plants together with the poles. The bean pods were then separated from the other components which were piled in a corner of the field and permitted to decompose. No attempt was made to utilize these residues or redistribute them over the field. The total harvest of pods was 5020.8 kg ha-l and the total residue biomass (roots, stems, leaves) removed to the corner of the field was 2324.9 kg ha-l, a total of 7345.7 kg ha-l. The summary of the total crop biomass removed during the first year cropping is presented in Table 5.4. Table 5.4. Total biomass removed during the first year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass removed. Crop species Harvest products Biomass removed (kg ha" 1 dry mass)* Harvest residues T o t a l Cucumber Bitter solanum 303.9 ± 34.4 (4) 362.0 ± 52.5 (4) Hyacinth bean 5020.8 ± 1342.6 (60) 322.3 ± 116.8 (4) 2324.9 ± 527.4 (28) 303.9 ± 34.4 (4) 684.3 ± 115.9 (8) 7345.7 ± 1585.5 (88) T o t a l 5686.7 ± 1310.5 (68) 2647.2 + 520.7 (32) 8333.9 ± 1526.6 (100) * mean + 95% confidence interval 73 5.2.THE S E C O N D Y E A R C R O P P I N G 5.2.1.INTRODUCTION Cassava is the most important root crop and the fourth most important source of food energy in the tropics (Cock, 1982). It is estimated that cassava has become the basic food source for more than 300 million people in the tropics (Nestel 1973 in Howeler, 1985).Indonesia is the main cassava producing country in Asia with a harvested area of 1.34 million ha and a total production of 14.2 million tonnes in 1984 (CBS, 1985). Cassava is the third staple food after rice and maize in Indonesia and is very important as supplemental meal for most people in rural areas. Cassava grows relatively well on infertile soil, but it extracts large amounts of nutrients from the soil and may exhaust the soil nutrient reserves unless adequately fertilized. Howeler (1980) reported that 2.3 kg of N, 0.5 kg of P, and 4.1 kg of K is removed per tonne of cassava tubers harvested. These values would increase to 4.9 kg of N, 1.1 kg of P, and 5.8 kg of K if all above ground parts (i.e. including harvest residue) were removed. Den Doop (1937 in Howeler, 1985) found that three consecutive cassava plantings without K-fertilization resulted in a decrease in yield from 15 t ha- 1 in the first year to 4 t ha-l in the third year. Fertilization in cassava cropping is only done on commercial monoculture cassava farms, and has very rarely been practised by subsistence farmers. Cassava is usually planted as the last crop during the cropping stage of a talun-kebun in Soreang. The tubers can be harvested after 8-10 months and the field is then left uncultivated to revert to bamboo talun. Farmers believe that two consecutive plantings of cassava is not economically sound because of the declining yield associated with both a decline in soil fertility and the shading and other competitive effects of the rapidly developing young bamboo canopy. This part of the chapter deals with the biomass accumulation in cassava 74 grown during the second year (traditional) and in an experimental third year- of cropping. The results will be used to assess nutrient budgets for the traditional cropping sequence of a talun-kebun cycle. 5.2.2.0BJECTIVES The objective of this part of the study was to quantify biomass accumulation and distribution during cassava cropping. These data were used as the basis for describing nutrient dynamics during the second year cropping and the changes that occur during a third, experimental, year of cropping. 5.2.3.STUDY SITES The study area has been described in Chapter 3. The plot for the cassava cropping during the second year cropping was known as TK-3, and the plot for the third year cropping was known as TK-5. The lay out of both plots was illustrated in Figure 3.3. 5.2.4.METHODS Stem cuttings, 25 cm long, were planted in TK-2 and TK-3 with a l x l m spacing. The planting dates were November 4, 1983 for TK-3 and November 23, 1983 for TK-5. As the local farmers never use fertilizers for cassava cultivation in the talun-kebun, no fertilization treatment was applied in this study. Five randomly chosen individual plants were harvested per sampling date from TK-3 at 2.5, 4, 5, 7, and 9 months after planting. Considering the smaller size of TK-5 (7x7 m2), only three randomly chosen individual plants per sampling date were harvested at 2, 2.5, 3, 4, 5, 7, and 9 months, after planting. Each sample plant was separated into foliage, stem, old stalks (stem cutting), and root (including tuber). The fresh mass of each component was determined directly in the field. Each sample was then oven dried to a constant mass at 8fjOC to determine dry mass, which was then used to calculate biomass on a per hectare basis by determining average mass per plant and multiplying this by the number of plants per hectare. 5.2.5.RESULTS A N D DISCUSSION l . C A S S A V A BIOMASS Figure 5.1. illustrates the total biomass of cassava in the second and third year of cropping (first & second year of cassava growth). Cassava growth was slow during the first 3 months, but rapidly increased from 4 to 9 months. The total biomass of cassava grown in the third year of cropping was 45% lower than in the second year cropping. Similar trends were found by Uhl (1980) in San Carlos where the yield declined by 32% between the first and second year cropping, and by Tondeur (1956 in Nye and Greenland, 1960) in the Belgian Congo where he recorded a 33% decline in cassava yield between year one and year two. Shading effect of young bamboo culms in the third year seemed to be one of the main cause of the decline in biomass production. Shading influences yield and growth of cassava. Okoli & Wilson (1986) found that cassava grown under 50% shade showed a 60% yield reduction in comparison to that grown without shade. 76 Figure 5.2. Total biomass (t ha"1 dry mass) of cassava in the second and third year of cropping in a bamboo talun-kebun in Soreang, West Java, Indonesia. age £morrfchs_) x x : second year of cropping . : third year of cropping The biomass of cassava tubers started to increase rapidly 4 months after planting (Table 5.5). Tubers represented 12-33% of the total biomass during the early growth (2.5-4 months), but 49-68% at 5 to 9 months. The biomass of the old stalks increased slightly from 239 kg h a - l at 2.5 months to 302 kg ha- 1 at 5 months, and finally decreased to 230 kg ha- 1 at 9 months. Ezedinma (1981) reported that old stalks provided an alternative sink for assimilates during the growth period of cassava. Stem and foliage biomass increased steadily with increasing plant age. Part 77 of the foliage biomass was periodically returned to the soil as litterfall. Stems contributed 17-20% of the total biomass during the 9 month cropping period, while the percentage of foliage to total biomass decreased from 38% at 2.5 months to only 8% at the end of the cropping (9 months). The biomass of each component of cassava grown in the third year of cropping is presented in Table 5.6. The biomass of cassava tubers at the time of harvest in the third year of cropping was 60% lower than the comparable value for the second year of cropping (1862 kg ha-l in comparison to 4585 kg ha-l). This supported den Doop's finding (1937 in Howeler, 1985) that a 64% decline in yield occurred when cassava was planted in consecutive years without fertilizer. Plant growth was slow during the early stage of the third year cropping. The stem biomass at 2.5 months was only 66 kg ha-l in comparison to 131 kg ha-l during the second year of cropping. However, the stem biomass increased rapidly from 4 months to the time of harvest (9 months). Stems contributed 9-31% of the total biomass during the third year of cropping. The foliage biomass at 2.5 months was only 26% of the comparable value during the second year of cropping, but it increased with increasing crop age. The total biomass of cassava at the end of the third year cropping was 3.7 t ha-l, while the total biomass at the end of the second year cropping was 6.7 t ha-l. Table 5.5. Biomass of a first cassava crop grown during the second year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total biomass at that sampling period. Age Biomass (kg ha"1 dry mass)* (months) Roots & tubers Old stalks Stems Foliage T o t a l 2.5 92.2+ 38.9 239.4+ 54.0 130 .8 + 31.8 277.8+ 69.6 740.2 + 158.2 (13) (32) (18) (37) (100) 4 501.0 + 25.8 309.6 + 119.8 325 .2 + 47.3 402.2+ 60.8 1538.2+ 61.2 (33) (20) (21) (26) (100) 5 1114.0 + 276.8 301.0 + 70.7 373 .0+ 73.6 495.2 + 58.8 2283.2 + 293.4 (49) (13) (16) (22) (100) 7 2088.8 + 611.1 302.2+ 64.4 303 .6 + 178.7 578.6+ 39.0 3713.2 + 699.0 (56) (8) (22) (14) ' (100) 9 4585.0 + 854.5 230.2+ 65.5 1350. .0 + 262.2 551.8 + 104.0 6716.8 + 694.9 (68) (4) (20) (8) (100) * mean + 95% confidence interval —j CO Table 5.6. Biomass of a second cassava crop grown in the third year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. The number in brackets represents a percentage of total weight at that sampling period. Age (months) Roots & tubers Old stalks Biomass (kg h a - 1 dry mass)* Stems Foliage T o t a l 2 17.9+ 7.9 (7) 206.0+ 16.5 (72) 27.1+ 4.3 (9) 33.3+ 10.1 (12) 284.4+ 15.4 (100) 2.5 36.0+ 14.9 (9) 233.0 + 59.8 (57) 65.7+ 7.0 (16) 72.3+ 28.6 (18) 407.0+ 29.9 (100) 3 170.3+ 67.6 (23) 264.3 + 46.0 (35) 115.6+ 29.1 (15) 202.0+ 38.3 (27) 752.3 + 177.3 (100) 4 319.0+ 19.0 (32) 310.3 + 67.8 (31) 146.7+ 40.8 (15) 217.7+ 44.8 (22) 993.7 + 164.0 (100) 5 847.7 + 110.4 (49) 313.0 + 88.2 (18) 318.7+ 35.9 (18) 267.0+ 20.0 (15) 1746.3+ 60.0 (100) 7 1081.7 + 329.1 (43) 312.3 + 137.0 (12) 787.3+ 68.5 (31) 339.7+ 50.3 (14) 2521.0 + 541.7 (100) 9 1861.7 + 127.1 (50) 316.7 + 63.1 (9) 1161.7 + 232.2 (31) 369.0 + 114.3 (10) 3709.0 + 352.3 (100) * mean + 95% confidence interval 80 2. BIOMASS R E M O V A L Harvesting of cassava tubers was done by pulling out the entire plants; therefore, there is total biomass removal during the harvest period. Howeler (1980) stressed the importance of returning stems and leaves to the soil to reduce soil nutrient depletion after harvest; however, farmers in Soreang usually removed all materials from the field since no further cultivation was performed after cassava harvest. The leaves were usually used for cattle feed and the stems were cut and used as stem cuttings for cultivation in other fields. The upper part of the stems were left to decompose in the corner of the field. A summary of the total biomass removed during cassava cropping is presented in Table 5.7. Table 5.7. Total biomass of cassava removed during the second year cropping of a bamboo talun-kebun in Selaawi hamlet, Soreang district, West Java, Indonesia. Cropping period Biomass removed (kg h a - 1 Harvested Harvest products residues** dry mass)* T o t a l 2 n d yr cropping (IStyr cassava) 4855.0 + 854.5 (68.3) 2131.8 + 233.4 (31.7) 6716.8 + 694.4 (100) 3rdyr cropping (2ncTyr cassava) 1861.7 + 127.1 (50.2) 1847.3 + 231.2 (48.8) 3709.0 + 352.3 (100) * mean + 95% confidence interval includes biomass of the old stalk 81 This shows that harvested products (tubers) constituted 68% of the total biomass removed during the first year of cassava cropping, but only 50% during the second year of cassava cropping. The net removal of biomass (excluding old stalks) was 6.5 t ha-l m the first year of cassava cropping and 3.4 t ha- 1 in the second year of cassava cropping. This resulted in a total removal of 9.9 t ha- 1 over the two consecutive years of cassava planting. 5.3. W E E D S 5.3.1 .INTRODUCTION Weeds constitute a major problem for most upland farms in the tropics, regardless of the cropping practice (Okigbo, 1978; Kamanoi et al, 1983). Weeds are also common in most shifting cultivation fields, particularly after the first year (Kelly, 1975; Norman, 1979). Several investigators have concluded that the abandonment of the cropfield to fallow in shifting cultivation is associated with the difficulty of controlling weeds (Joachim & Kandiah, 1948; Seavoy, 1973; Norman, 1979). Weeds are common in the talun-kebun, particularly during the cropping stage and the early period of bamboo regrowth. They are eliminated from the later stages of the bamboo talun by the shading effect of the bamboo canopy and the accumulation of surface Utter. Manual weeding is done during the early growth of the first year crop and prior to harvesting the hyacinth bean. Weeding in cassava fields is rarely done as farmers believe that once established the cassava canopy will suppress weed growth or not be adversely affected by it. This part of the chapter discusses weed biomass accumulation over a bamboo talun-kebun rotation cycle. 82 5.3.2.0BJECTIVES The objective of the weed biomass study was to quantify biomass accumulation in weeds during a bamboo talun-kebun rotation cycle. These data will be used to estimate nutrient accumulation in weeds and the return of these nutrients to the soil as weeds are composted or left to die on the ground. 5.3.3.STUDY SITES The biomass of weeds was studied in five plots: TK-1 (mature talun), TK-2 (first year cropping), TK-3 (second year cropping), TK-4 (early fallow), and TK-5 (third year cropping). All of these study plots were described in Chapter 3. 5.3.4.METHODS In all cases except the first year cropping, weed biomass was quantified by total weed harvest on five l x l m2 subplots per plot. Weed biomass during the first year cropping was estimated by manually harvesting all of the weeds in the four 126 m2 mixed cropping subplots of TK-2. Although manual weeding was also done in other treatment subplots of the first year cropping plot, only the results of biomass measurement in the mixed cropping subplots were reported in this part of the chapter. Weed sampling was done early in the cropping stage and at the end of the cropping prior to harvesting the hyacinth bean (i.e. at 2 and 6 months after planting the beans). The total weed biomass during the first year cropping was calculated by summing the weed biomass collected during these two sampling periods. Weeding is not normally practised during cassava growing, but, nevertheless, weed biomass was estimated prior to harvesting the cassava (9 months after planting) in five 1 m2 randomly-located subplots in the second and third year cropping fields (TK-3 and TK-5). 83 Similarly, five 1 m 2 subplots were randomly located in the early fallow field (TK-4) for assessment of weed biomass. Sampling was done at the end of the early fallow (36 months after clearcutting the mature talun). In the mature talun, weeds only grew in patches where light penetrated the bamboo canopy. Such patches covered approximately 11% of the area of TK-1. Ten randomly-located 1 m2 subplots were sampled for weed biomass in these weed patches. Sampling was done at the end of the mature talun prior to clearing the bamboo (at 72 months after the previous clearcutting). The total fresh mass of weed harvested in each plot/subplot was determined directly in the field. Five subsamples of 100 gram fresh mass each per age of field were taken to the laboratory, and oven-dried to a constant mass at 80°C to obtain a dry mass-fresh mass conversion. The dry mass was used to calculate biomass per hectare. 5.3.5.RESULTS A N D DISCUSSION 1. W E E D BIOMASS Table 5.8. summarizes the weed biomass over a bamboo talun-kebun rotation cycle. Weed biomass during the cropping stage was higher than the biomass during the fallow. It seemed that weed growth was influenced by the light intensity that reached the ground and by belowground competition with food crops and bamboo. The decrease in the average light intensity was accompanied by a decrease in the percentage weed cover and a reduction in weed biomass. The dominant species of weeds varied between the different stages, because each species required different amounts of light for its growth. Sun-loving species dominated the cropping stage, but only shade-tolerant species survived during the fallow stage. A list of the dominant species at each stage is presented in Table 5.8. Table 5.8. Weed biomass over a six-year bamboo talun-kebun rotation cycle in Selaawi hamlet, Soreang district, West Java, Indonesia. Rotation stage Months after clear-cutting % light inten-sity** % weed cover Dominant weed species Biomass* (kg ha-l dry mass) 1 s t yr cropping 8 78-94 (89) 80 Ageratum conyzoides Ageratum houstonianum 1780.8 ± 446.4 2ndyr cropping 20 57-84 (64) 68 Paspalum conjugatum Eupatorium riparium Cyperus iria 1481.4 + 166.3 3rdyr cropping 32 47-69 (52) 62 Eupatorium riparium Cynodon dactylon Imperata cylindrica 1188.4 ± 336.4 early fallow 36 38-64 (44) 58 Achyranthes aspera Cynodon dactylon Synedrela nudiflora 1076.2 + 506.2 mature talun 72 5-13 (8) 11 Clidemia hirta Melastoma malabatricum 214.6 + 50.9 * mean + 95% confidence interval ** the number in brackets represents the average value of light intensity 85 2. BIOMASS R E M O V A L Complete manual weeding is done by farmers during the first year of cropping. Weeds are then piled in a corner of the field and left to decompose. Although the nutrients contained in these weeds will be used up by bamboo and redistributed over the field through bamboo litterfall, the weed piles were considered as an output from the field during this particular period of the cycle. The weed biomass removed during the first year cropping was 1781 kg ha-l. There is normally no weeding during the second and third years of cropping (if a third is done); weeds would normally grow, die and decompose in the field. Therefore, no weed biomass removal was recorded for these periods. The fallow period is considered as the recovery period of a talun-kebun cycle. No disturbance and no weeding occurs during this period. Weeds are simply shaded out by bamboo and decompose on the site; no biomass is removed from the site. However, all weeds are killed and removed from the mature talun site during litter raking in preparation for clearcutting. Some of the nutrient content of these weeds is returned to the soil in the form of ash from the burned slash (litter, slash, and weeds). Therefore, all of the weed biomass accumulated during the mature talun stage was counted as a temporary removal from the site. The biomass removed was 215 kg ha-l. Farmers do not permit weeds to become a major biomass component of the talun-kebun cycle. Weeds were quantified simply to establish the magnitude of their biomass relative to the other plants in the cycle, and the magnitude of any nutrient removals that accompany weeding. Weeds constitute a declining percentage (from 9% to 0.2%) of the total plant biomass during the first year of cropping, second year of cropping, early fallow, and mature fallow. The nutrient removals in weed biomass will be quantified in Chapter 7. 86 5.4. S U M M A R Y A N D C O N C L U S I O N S Accumulation and removal of biomass during the cropping stage of a bamboo talun-kebun cycle was studied in various fields located in Selaawi Hamlet, Sukajadi Village, Soreang District, West Java, Indonesia. The main crops grown during the first year of cropping were: cucumber, bitter solanum, and hyacinth bean. Cassava was grown as the second year crop. In order to evaluate changes in biomass and yields if cropping is prolonged, an extra year of cassava cropping was also studied. Weed biomass was also studied because weeds often become a problem during the cropping stage. The following conclusions can be made from the results of this chapter: 1. Fruit and pod biomass constitute the highest percentage of total crop biomass accumulated during the first year of cropping. 2. Approximately 8.3 t ha-l of crop biomass was removed during the first year of cropping. 3. Practising continuous cassava cropping without any fertilization caused a significant yield decline. The harvest of cassava products in the third year of cropping was 60% lower than the comparable value for the second year. 4. The total biomass removed during the second year of cropping was 6.7 t ha-1 (including 0.2 t ha-l of old stalks). Cassava tubers comprised 68% ofthe total biomass removed; the rest was harvest residues. 5. Farmers are aware that using more than two years of cropping within a six year talun-kebun rotation will result in a declining yield. Their conclusion was supported by the experiment on the third year cropping. 6. Weed biomass during the cropping stage is higher than the weed biomass during the bamboo fallow, because weed growth is influenced by the light 87 intensity that reaches the ground. 7. Manual weeding in the first year of cropping removed 1.7 t ha-l of weeds. Weeds were also killed and removed from the field during the litter raking at the end of the mature talun stage. The biomass removed at this stage was 0.21 ha-l. 8. The total crop and weed biomass removed during the two year cropping period was approximately 16.7 t ha-l. Weeds made up 12% of this removal. CHAPTER 6 LITTER PRODUCTION, FOREST FLOOR MASS, AND SOIL ORGANIC MATTER DYNAMICS IN THE TALUN-KEBUN SYSTEM 6.1.INTRODUCTION As noted in Chapter 2, the talun-kebun agroforestry system is a shorter-rotation version of shifting cultivation, in which the fallow period is dominated by bamboo. The local farmers in West Java believe that the bamboo stage is the key to long-term sustainability of the talun-kebun. The ability of bamboo to maintain soil fertility on marginal agricultural land is attributed to the build up of "dark soil" (humus) in the upper soil layers, which may be a result.of the continuous deposition and decomposition of dead fine roots during the several years of bamboo fallow. Moreover, bamboo is also important because the accumulation of undecomposed leaf and sheath litter over the years is the source of the ash which is used to fertilize the annual food crops. This chapter describes litterfall, litter accumulation in the forest floor, litter removal by raking, and soil organic matter dynamics in the talun-kebun rotation cycle. The results will be used as input data for the description of the biogeochemistry of the talun-kebun system in Chapter 7. 6.2.LITERATURE REVIEW The importance of litter production as a major pathway for transferring organic matter and nutrients from vegetation to soil has been recognized in many studies (Vitousek & Sanford, 1986; Vogt et al, 1986). The amount of litter produced by an ecosystem is influenced by climate, plant age, site condition and plant species characteristics (Bray & Gorham, 1964). Aboveground litter production in tropical rain forests is usually much higher than in temperate and boreal forests, and higher than in tropical deciduous or semideciduous forests (Tsutstumi et al, 1983). Various studies have shown that tropical plantations usually have a lower aboveground litter production than natural tropical forests (Lundgren, 1978). The biomass of the forest floor is often studied in conjunction with litterfall studies. The forest floor may include recent litterfall (L horizon), decomposing organic matter (F horizon), and well humified organic matter (H horizon) above the mineral soil (Vogt et al, 1986), or some or all of these materials may be rapidly incorporated into the mineral soil by animal activity. The amount of forest floor accumulation is more strongly determined by the type of tree (evergreen or deciduous) than by climatic factors or latitude (Vogt et al, 1986). Rogers & Westman (1977) noted that stand cultural practices, soil type and the age of the stand influenced the accumulation of forest floor biomass. Vogt et al (1986) found that fine root mass contributed significantly to the forest floor biomass. Fine root turnover contributes from 29 to 255 kg ha-l year-1 of organic matter to the forest floor and upper soil horizon in tropical broadleaf evergreen forest. Cuevas and Medina (1983, in Vogt et al, 1986) suggested that high root mass and turnover in the forest floor of tropical evergreen forests increased the quantity of organic matter to be decomposed. Ewel (1976) pointed out that the litter layer on the soil surface acts as an input-output system, receiving inputs of organic matter and nutrients from the vegetation, losing biomass by decomposition, and supplying nutrients to the mineral soil and roots. He stressed the importance of litter accumulation and decomposition on the soil surface as the key for restoring fertility during the fallow period of shifting cultivation systems. A review of studies of litter production in bamboo forests and plantations showed that total aboveground litterfall ranges from 6.6 t h a - 1 yr-l (for bamboo forest under thinned tropical forest) to 10.6 t ha-l yr-1 (for bamboo forest under thinned monsoonal forest; Rozanov & Rozanova, 1964). Seth et al (1963) reported an aboveground leaf litterfall production of 3.2 t ha-1 in a bamboo plantation in India. Dali (1980) reported an annual leaf litterfall of 1.3 t ha-l i n a Bambusa bamboos plantation in South Sulawesi, Indonesia. Scientific documentation of the forest floor biomass of bamboo stands is not available in the literature, but I observed a thick layer of leaf litter on the soil surface of the mature bamboo talun. Yadav (1963) pointed out that planting bamboo as an understorey in teak plantations helped to preserve the fertility of the soils because its leaf litter tends to accumulate as forest floor. Most studies of litter production in shifting cultivation systems have focused on the early secondary growth (ie. the fallow) period; none of them has discussed litter production during the cropping period. However, in a study of a tropical successional ecosystem in Costa Rica, Brown (1982) reported a litterfall of 2.5 t ha-l yr-1 for maize monoculture and 4.2 t ha-l yr-i for cassava. Studies of soil organic matter in shifting cultivation in Guatemala showed that the organic matter and soil nitrogen content increased slightly after burning as a result of the addition of partly burned vegetation to the soil, but decreased gradually with cultivation (Sanchez, 1976). Nye & Greenland (1964) found that the loss of topsoil organic matter in shifting cultivation in Ghana was rapid during the first year of cropping, but was slower in the second year. Their data indicated that the changes in soil organic matter content usually were limited to the uppermost soil layer, except when the effect of crop root decomposition was evident in the subsoil. After several years of fallow, however, the organic matter content increased gradually but reached a new equilibrium level slightly below that of the virgin forest. 6.3.STUDY A R E A The litterfall, forest floor, and soil organic matter study was conducted in the same area as the biomass study. The area is described in Chapter 3. 6 .4 .METHODS 6.4.1. L I T T E R F A L L S A M P L I N G The litterfall study was started in January, 1984. Aboveground litterfall was collected from TK-1 (mature talun), TK-2 (first year cropping), TK-3 (second year cropping), and TK-4 plots (early fallow), using four Utter traps of lm2 that were placed randomly at 50 cm above the ground in each plot. Each trap consisted of a square bamboo frame with a screen bottom and 10 cm tall sides. Litter collection in TK-1 and TK-4 was conducted over a twelve month period. The collection in TK-2 and TK-3 was done over periods of 6 and 8 months, respectively, according to the harvest schedule of the agricultural crops. The litter from each trap was collected monthly and weighed to determine fresh mass. Dry mass was obtained by oven drying the samples at 80°C. 6.4.2. F O R E S T F L O O R S A M P L I N G The forest floor in TK-1 was sampled randomly in fifteen 25x25 cm2 quadrats prior to clearing the field in preparation for cropping. Samples were divided into fresh Utter, fragmented litter and finely decomposed litter (humus). The samples were weighed to determine fresh mass and bulked in groups of 5 by layer (i.e. 3 bulk samples per layer) for dry mass determination. Forest floor sampling in other plots (TK-2, TK-3 and TK-4) was done with 5 replications of 25x25 cm2 quadrats in each plot. Sampling in TK-2 and TK-3 was done prior to harvesting the beans and the cassava, respectively. Forest floor sampling in TK-4 (early fallow) was done at 36 months after clearcutting. Additional forest floor sampling was done in 5 randomly located 25x25 cm2 quadrats in nearby bamboo fields of 48 and 60 months after clearcutting. Fresh mass of each sample was determined and the samples were oven dried to a constant mass at 80°C to obtain dry mass. Dry mass was used to calculate the mass of the forest floor on a per unit area basis. 6.4.3 .MINERAL SOIL O R G A N I C M A T T E R S A M P L I N G The mineral soil was sampled in TK-1, TK-2, TK-3, and TK-4 at the beginning and at the end of each stage. Soil was sampled at various depths at three locations in each plot. The samples from each location consisted of a composite of 5 cores (5.6 cm diameter), i.e. one from a randomly chosen point, and one from each of four points located 2 meters away in the cardinal directions. Sampling depths were 0-5 cm, 5-25 cm, 25-45 cm, and 45-75 cm. Each composite sample was well mixed and oven dried at 55°C. The samples were then ground by using a mortar and pestle, and passed through a 2 mm sieve. Fine roots were already removed, but it was difficult to separate most of the dead fine roots, (which were dry, broken and very small in size (<1 mm)) from the soil particles. The organic matter content was determined on subsamples that were finely ground and passed through a 0.25 mm sieve. The determination of soil organic matter was carried out by measuring total organic carbon (using the wet combustion method) and multiplying the result by 1.72, assuming that organic matter contains 5 8 % carbon (Hidayat, 1978; Mongi et al, 1979). The analyses were conducted by staff members of the National Institute of Chemistry, Bandung, Indonesia. The results were then converted to field mass per unit area by using soil bulk density data (see Chapter 3). 6.5.RESULTS AND DISCUSSION 6.5.1. LITTER ACCUMULATION A.THE CROPPING STAGE Because the leaf litterfall of cucumber and bitter solanum decomposed very rapidly (within a week after reaching the ground), and because the height of these two species was lower than 50 cm, only leaf litterfall of the hyacinth beans was collected during the first year of cropping. Consequently, the value of total litterfall from the first year crops estimated in this study was lower than the actual value in the field. The monthly litterfall of hyacinth beans and cassava is presented in Table 6.1. Table 6.1. shows that leaf litterfall increased with increasing age ofthe crop. The total leaf litterfall during the bean cropping period was approximately 1.3 t ha-l. The total cassava litterfall, which consisted of leaves and petioles, was 1.71 ha- 1. 94 Table 6.1. The monthly litterfall of hyacinth beans and cassava during the cropping stage of a bamboo talun-kebun in Selaawi, Soreang, West Java, Indonesia. Age* (months) Month Litterfall (kg ha' 1)** Hyacinth beans Cassava 2 January 1984 80.0 + 21.2 70.0 + 21.2 3 February 1984 205.0 + 70.6 110.0 + 34.9 4 March 1984 212.5 + 70 175.0 + 37.1 5 April 1984 287.5 + 111.8 227.5 + 37.9 6 May 1984 487.5 ± 1 0 4 . 7 232.5 + 61.1 7 June 1984 - 280.0 + 78 8 July 1984 - 285.0 + 98.8 9 August 1984 - 332.5 ± 121.5 T o t a l *** 1272.5 1712.5 * Age of crops at the commencement of litter sampling. Litter traps were set in the field a month after planting the crops and litterfall was collected monthly until the final harvest (6 months for hyacinth beans, and 9 months for cassava). ** Mean + 95% confidence interval. Based on four traps. *** The total litterfall value for the first year (hyacinth beans) and the second year (cassava) cropping stages. B . T H E F A L L O W S T A G E Bamboo litterfall collected in this study consisted of leaves and sheaths, because dead branches were taken away by local farmers to be used as fuelwood. Table 6.2. presents the biomass of seasonal litterfall during the early fallow and mature bamboo talun stages. The highest litterfall values were obtained during the peak of the dry season (July-October), whereas the lowest values were recorded during the peak of the rainy season (November-February). Ueda (1960) has pointed out that bamboo leaves usually fall when they are between 12 and 18 months old, and are quickly replaced by new leaves. The low litterfall values during the peak of the rainy season reflects a period of new foliage production. The total annual leaf litterfall was 2.0 t ha-l during the early fallow (sampling commenced 36 months after clearcutting) and 3.5 t ha-l at the end of the (sampling commenced 36 months after clearcutting) and 3.5 t ha" 1 at the end of the mature bamboo talun ( sampling commenced 72 months after clearcutting). These values are equal to approximately 78% of the foliage biomass at each of the two stages. Assuming that the proportion of leaf litterfall to foliage biomass was constant between 36 months and 72 months, annual leaf litterfall for the periods commencing 48 and 60 months after clearcutting is estimated to be 2.9 and 3.2 t ha-l, respectively (Figure 6.1.). According to the data, sheath litterfall accounted for only 8% of the total aboveground litterfall. The standard deviation of sheath litterfall was high because sheath litterfall was not evenly distributed in the field, the majority of sheaths falling directly beneath the clumps. Because the litter collectors could not be placed directly beneath the culms, values obtained in this study are thought to underestimate the actual sheath litterfall in the field. However, most of the sheaths that fell directly beneath the clumps were caught between culms and did not reach the ground; therefore, only those which fell between clumps contributed to the forest floor mass. The measured sheath annual litterfall was 0.2 t ha- 1 in the period commencing at 36 months, and 0.3 t ha- 1 in the period commencing at 72 months. Based on the assumption that the percentage of sheath to total litterfall over age remained constant, sheath litterfall in the period commencing at 48 and 60 months was estimated at 0.3 t ha-l and 0.3 t ha-l, respectively. Seth et al (1963) found that branch litterfall in a Dendrocalamus strictus plantation in India was equal to 25% of its leaf litterfall. Applying the same percentage, I estimate the annual branch litterfall to have been 0.5 t ha- 1, 0.7 t ha- 1, 0.8 t ha- 1, and 0.9 t ha-l for the periods commencing at 36, 48, 60, and 72 months, respectively. The estimated total annual aboveground litterfall during the fallow 96 removed from the field, and used as fuelwood. Therefore, only the leaf and sheath litterfall contributed to the forest floor mass. Table 6.2. Seasonal litterfall during the early and mature fallow of a bamboo talun-kebun in Selaawi.Soreang, West Java, Indonesia. Age* Month Litterfall (kg ha (months) Foliage Sheath T o ta 1 36 January 85.0 ±20.4 22.5 ±25.7 107.5 ±20.2 February 97.5 ±21.7 17.5 ±34.3 117.5 ±56.2 March 125.0 ±39.6 0 125.0 ±39.6 April 157.5 ±37.8 0 157.5 ±37.8 May 182.5 ±54.5 32.5 ±41.9 215.0 ±63.3 June 177.5 ±54.5 32.5 ±38.7 210.0 ±71.1 July 202.5 ±65.7 0 202.5 ±65.7 August 267.5 ±57.9 0 267.5 ±57.9 September 275.0 ±70.4 30.0 ±37.5 305.0 ±58.5 October 290.0 ±33.0 32.5 ±41.9 322.5 ±57.4 November 82.5 ± 16.7 12.5 ± 4.5 95.0 ± 38.0 December 90.0 ±31.0 0 90.0 ±31.0 T o t a l 2032.5 180 2212.5 72 January 182.5 ±50.2 0 182.5 ±50.2 February 190.0 ± 69.3 45.0 ± 53.4 235.0 ± 114.4 March 252.5 ±94.5 0 252.5 ±94.5 April 285.0 ±60.7 40.0 ±46.0 325.0 ±101.4 May 310.0 ±61.9 0 310.0 ±61.9 June 265.0 ±53.4 50.0 ±65.0 315.0 ±104.5 July 397.5 ±114.1 0 . 397.5 ±114.1 August 412.5 ±110.9 50.0 ±37.5 462.5 ±141.7 September 497.5 ± 147.4 57.5 ±42.6 555.0 ±122.5 October 530.0 ±115.2 47.5 ±64.7 577.5 ±130.1 November 162.5 ±71.2 40.0 ±28.8 202.5 ±42.6 December 170.0 ±115.6 0 170.0 ±115.6 T o t a l 3472.5 330 3802.5 * Age of the field after the previous clearcutting at the start of the sampling period ** Mean + 95% confidence interval 5 O 12 24 36 48 60 72 age (months) = estimates based on biomass regression (Chapter 4) O = estimates based on litter collection data \7 - estimates based on foliage biomass data and foliage biomass/litterfail ratios Figure 61 The estimated values of bamboo leaf litterfall over a six year talun-kebun rotation cycle in Selaawi, Soreang, West Java, Indonesia 6.5.2.FOREST F L O O R D Y N A M I C S In contrast to the bamboo litter, crop litter decomposed rapidly; therefore, the accumulation of an ectorganic layer during the cropping stage was only temporary. The mass of this ectorganic layer prior to harvesting the hyacinth beans at the end of the first year of cropping was 0.8 t ha-l. The crop litter decomposed rapidly and was mixed with the mineral soil by soil hoeing during the preparation for cassava cropping. The mass of the ectorganic layer during cassava cropping was estimated to be 1.1 t ha-l of cassava litter plus an additional 0.2 t ha-l of bamboo leaf litterfall blown in from the adjacent bamboo stand. The cassava litter decomposed rapidly and had disappeared by the early fallow stage. The mass of the forest floor during the fallow stage was 2.2 t ha-l, 5.5 t ha-l, 9.9 t ha-l, and 13.5 t ha-l, at 36, 48, 60, and 72 months, respectively. The forest floor at 36 months consisted/almost entirely of intact litterfall. Both fragmented litter and intact litter were observed at 48 and 60 months, but these two forest floor components were not separated in the samples from the fields of these ages. In contrast, the forest floor in the mature bamboo talun stage was separated into L (intact litter), F (fragmented litter) and H (well humified material not combined with mineral soil). The mass of these three forest floor components was 4.7 t ha-l, 6.0 t ha- 1, and 2.8 t ha-l for L, F, and H layers, respectively. The accumulation of forest floor over the study period indicated a low rate of decomposition of bamboo Utter. Litter decomposition rate was not quantified in this study, but DaU (1980) reported an annual decomposition rate of 5% for leaf litter in a Bambusa bamboos plantation in South Sulawesi, Indonesia. The forest floor mass at 72 months (13.5 t ha-1) exceeded the total 99 months should have been 11.31 ha" 1. Thus, it seems that fine root death must have contributed at least 2.2 t ha- 1 in 72 months. However, because fine root decomposition was apparently rapid (all the fine roots killed by hoeing decomposed within the two year cropping period, and possibly within the first year), fine root death may have contributed as much as 3.5-7 t ha- 1 over the whole period (see Appendix 2), and was probably the major contributor of dark colored humic materials to the upper mineral layers during the fallow period. 6.5.3. LITTER REMOVALS Litter raking was done at the end of the mature bamboo talun stage. All intact litter (L layer) and approximately two thirds of the fragmented litter (F layer) were raked, piled, and burned. Although this activity takes place within a field, for the purposes of this study, litter raking is considered to be a temporary removal. Ash from the burn is collected and put in a small hut until it is mixed with compost and returned to the field as a fertilizer during the vegetable cropping. Approximately 8.7 t ha-l 0 f bamboo litter was removed at the end of the fallow (72 months after the previous clearcutting). The fallen dead branches were collected, removed from the field, and used as fuelwood. The estimated total dead branch removal over a rotation cycle (72 months) was 2.9 t ha- 1. The branch removal is counted as a net removal. 6.5.4. SOIL ORGANIC MATTER DYNAMICS Organic C concentrations in the study fields ranged from 3.6 to 4.5% within the upper 5 cm, and from 2.9 to 3.5% within 5-25 cm depth of the mineral soil. Coefficient of variability ranged from 4 to 11% within each field (representing various talun-kebun stages. Yadav (1963) has reported a range of 1.6 to 3.4% for organic carbon concentration in the upper 20 cm of alluvial soil under Bambusa arundinacea 100 plantation and a range of 0.8 to 1.4% under Dendrocalamus strictus plantation (ranges from 0.8 to 1.4%) in India. The areas occupied by B.arundinacea have higher rainfall than those occupied by D. strictus (2300 mm in comparison to 1400 mm, annually). The high concentration of soil organic carbon in the upper mineral soil was also reported from various shiftiing cultivation areas of the tropics (Brams, 1971; Sanchez, 1976; Zinke et al, 1978; Tulaphitak et al, 1983). Sanchez (1982) has found that there is no significant difference in the soil organic carbon concentrations and contents between the tropical and temperate region soils. He suggested that the lack of moisture during dry period in the tropics has an effect on biological activity in the soil similar to that of low temperature during the winter in temperate region. Areas with heavy rainfall do not have the temperature and moisture problems, but these areas are usually covered by tropical rainforests, in which the biomass and organic matter production are higher than those in temperate forests. Therefore, the rapid decomposition of soil organic matter in the humid tropics resulted in a similar equilibrium content of organic matter between the tropic and the temperate regions. The concentrations and the total inventories of soil organic matter in the surface 25 cm at various stages of the talun-kebun cycle are illustrated in Figure 6.2. and Figure 6.3. There was no significant change in soil organic matter concentration and content at the 25-75 cm depth throughout the whole cycle (Appendix 3) so this depth is not discussed further. Litter raking and soil hoeing killed all fine roots that occupied the surface 25 cm of the mineral soils. The incorporation of the remaining partly-decomposed (fragmented) litter and soil humus during the hoeing, and the accumulation of dead fine roots after hoeing caused a major addition of soil organic matter in the surface 25 cm two months after clearing and hoeing (Figure 6.2.). A subsequent decrease of 101 soil organic matter concentration and content in the surface 25 cm was recorded during the two year cropping stage (Figures 6.2 and 6.3). Sanchez (1976) stresses that organic matter mineralization increases upon cultivation and exposure, and results in a decline of organic matter content during the cropping stage. The decrease in soil organic matter content during the two year cropping period in this study was probably related to the decomposition of dead fine roots and mineralization of soil humus. By the end of the first year fallow, however, the mineral soil organic matter (TK-4 in Figure 6.3.) started to increase again. This increase was associated with the increasing mass of dead mother rhizomes from the previous clearing, the turnover of bamboo fine roots, and the buildup of a new bamboo litter layer. Assuming that only 5% of the leaf and sheath litter decomposed and contributed to soil humus annually, most of the recovery of soil organic matter in the fallow period was probably due to fine root death and decomposition. Figure 6.3. shows that there is an increase of approximately 7 t/ha of soil organic matter in the surface 25 cm of soil within a four year fallow (from the end of the second year cropping until the end of the mature talun). u d) +J a id tn u o H •H O n Tk-1 & Tk-2 Tk-3 A B month after clearing _ l & hoeing Tlc-4 C Soil depth 0- 5 cm 5-25 cm A. Mature talun, cleared & cropped B. Cassava f i e l d (after hoeing and prior to harvest) C. Early fallow (0-1 year fallow) Figure 6.2. Soil organic matter concentration (%) in the upper 25 cm of mineral s o i l at various stages of a talun-JceJbun cycle in Selaawi hamlet, Soreang d i s t r i c t , West Java, Indonesia. soil organic matter It ho " ' ) CD o r~ 0 0 o T™ - J O c 3 lb c 3 O o at to v> a 2 o _ 3 2 <o Q O 3 § 3 = a (D — a a 3 < a o c in c a 01 0> O O O o —r— 31 31 11 II o rt • -» 2. S. 2. o" * 3 - 3 - O o o a - i a u> * ^ < -» a w a 3 a. ^ a _ a O —r-o a a . 3 1 < * \ \ . - M, »m m J 3E2 ro a 3 3 o a. 3 > n SOI ho ths a LJ a a pu a de ing c -* Ol o •a - i o <t i 1 sr ro •» to Ol o •a Q Ol o 5" •o 3 a » C o -i a. 3 3 3 <o COT 104 C H A P T E R 7 T H E B I O G E O C H E M I S T R Y a O F B A M B O O TALUN-KEBUN S Y S T E M There are three major nutrient compartments within the talun-kebun system: the vegetation, the forest floor (ie. the ectorganic surface layer of the soil), and the mineral soil. Nutrients are transferred or exchanged between these compartments by various processes, including foliar leaching, litterfall, litter decomposition, and nutrient uptake by plants. Inputs of nutrients to the system occur in the form of precipitation, biological fixation of nitrogen, mineral weathering, and inputs from upslope in the form of seepage, runoff or erosion. Nutrients removals from the system include losses in the smoke (gasification of N & S, and flyash), harvesting of plant biomass, downslope runoff or erosion, soil leaching and possibly denitrification. Because the magnitude and importance of these compartments and processes vary over the rotation cycle, the biogeochemistry of the talun-kebun system is examined separately for four stages: the mature bamboo talun stage, the clearing and burning stage, the cropping (kebun) stage, and the site recovery stage (the early fallow stage). The objective of this chapter is to summarize the quantity of nutrients in each major compartment, and the rates of some of the transfers between the compartments. The data presented in this chapter will be used in Chapter 9 to evaluate the sustainability of productivity in the talun-kebun system under altered cropping practice. The chapter is divided into five parts. The first part reviews the biogeochemical cycling of nutrients in the various stages of the shifting cultivation cycle as basic background for a discussion of the biogeochemistry of the talun-a The biogeochemistry of an ecosystem refers to the distribution and dynamics of nutrients within that ecosystem and between that ecosystem and other ecosystems. 105 kebun system. The second, which is a synthesis of the results of Chapter 4 and Chapter 5, discusses the overall biomass accumulation and dynamics in the bamboo talun-kebun cycle. The third and fourth parts deal with nutrient inventories in plants, and in the forest floor/ectorganic layer and mineral soil, respectively. The last part deals with the nutrient dynamics, including aboveground litterfall and various inputs to and removals/outputs from the system, and a summary of the biogeochemistry ofthe different stages of the bamboo talun-kebun system. 7. l . L I T E R A T U R E R E V I E W There were relatively few detailed studies of nutrient cycling aspects of shifting cultivation in tropical ecosystems until the mid 1960's, but since then there has been increasing interest in this topic (Nye & Greenland, 1960; Sanchez, 1976; Kundstadter et al, 1978; Norman, 1979; Jordan, 1987). There have been no detailed biogeochemical studies of nutrient cycling in the talun-kebun, but because of the similarity of this system to shifting cultivation, the literature on the latter provides a useful background for this chapter. There are four fairly distinct patterns of biogeochemical cycling during the shifting cultivation cycle. These occur during the following stages: 1) The late stage of the fallow; 2) the clearing and burning of the vegetation; 3) the cropping stage; and 4) the recovery stage of fallow vegetation. 7.1. l . T H E L A T E S T A G E O F T H E F A L L O W The biogeochemical cycle during this stage is comparable to the direct nutrient cycle of a natural forest ecosystem (Stark & Jordan, 1976). Nutrients in vegetation are returned to the soil through litterfall and foliar leaching, and a concentration of fine roots at the soil surface in combination with rapid decomposition ensures prompt and efficient recovery of nutrients by the plants (Sanchez, 1976; Norman, 1979). 106 Nutrient changes in the soil are closely linked with changes in soil organic matter and soil humus, which help to hold exchangeable cations, store important amounts of phosphorus and sulphur, and act as a nitrogen reservoir (Nye and Greenland, 1960). Organic matter is added to the soil as aboveground litter and by the death of fine roots. The annual rates of litterfall range from 5.5 to 15.3 tons ha-l in the tropics, as compared with 1.0 to 8.1 tons ha-l in temperate forests (Ewel, 1968). Decomposition studies in Guatemala (Ewel, 1968) showed that approximately half ofthe dry matter in the litter was mineralized within the first 8 to 10 weeks, after which the rate decreased. There has been no scientific documentation of the annual production of dead roots. A mat of fine roots frequently develops within the Utter layer and at the mineral soil surface, and it has been demonstrated (Stark & Jordan, 1978) that such root mats act as an efficient nutrient recovery mechanism which minimizes leaching losses to the subsoil in tropical forest. Deep-rooted species of the fallow vegetation play an important role in transferring nutrients from the subsoil to the surface soil where they are more available to superficially-rooted plants. Although the majority of the nutrients are retained in the vegetation, there is some enrichment of the topsoil through the accumulation of litter, and this accumulation often compensates for leaching losses and vegetation uptake (Jordan, 1985). Nutrient losses during the late fallow stage may occur through runoff, erosion and/or soil leaching, but they are thought to be more than balanced by gains from precipitation, biological fixation, upslope runoff/erosion, and mineral weathering. Therefore, after a sufficiently long period of fallow, the exchangeable nutrients in the topsoil will be restored nearly to their original levels (Norman, 1979). 107 7.1.2.THE C L E A R I N G A N D B U R N I N G O F T H E V E G E T A T I O N Disturbance of the tight biogeochemical cycle that characterized the mature fallow stage occurs during the clearing and burning of fallow vegetation. When vegetation is burned, nearly all the nitrogen and sulphur in the burned materials are lost to the atmosphere as oxides, while most of the phosphorus, potassium, calcium, and magnesium remain in the ash (Norman, 1979; Jordan, 1985). The presence of basic cations in the ash causes a dramatic increase in soil pH and in exchangeable calcium, potassium, and magnesium levels. This increase is followed by a gradual decrease during the cropping stage due to leaching and nutrient uptake by the crop. The level of available phosphorus in the soil is increased by clearing and burning as a result of the phosphorus content of the ash. Popenue (1960) found that the Bray-extractable phosphorus in the upper 5 cm layer of a Guatemalan Inceptisol increased about four times following clearing and burning, remained at this level for about six months, and then decreased to a level twice the original value at the end of one year of cropping. The decrease in available phosphorus that accompanies cropping is probably due to the removals in the harvested crop. Burning has little effect on soil organic matter. Changes in organic carbon and nitrogen are limited to the consumption of non-humified material at the soil surface (Nye & Greenland, 1964; Sanchez, 1976). Although the litter layer is largely destroyed by the burn, there is no loss of humified organic matter from the mineral soil (Nye & Greenland, 1960; Sanchez, 1976). If the heat energy released during the burn is sufficient to raise the soil temperature and cause partial sterilization of the soil (Nye & Greenland, 1960; Ahn, 1974), there may be an initial decrease in the microbial population. This is followed by an increase to higher levels than before the burn, usually with a modified species composition. This in turn may increase the rate of mineralization. 108 7.1.3 .THE C R O P P I N G S T A G E The rapid mineralization of organic matter and the addition of ash after clearing and burning provides a significant increase in soil pH and in available nutrients to the first crop planted in shifting cultivation on most tropical soils (Sanchez, 1976). However, yields gradually decline with successive croppings, which may due to one or more of: increases in pests, diseases, and weeds; topsoil erosion; deterioration in the physical condition or nutrient status of the soil; decline in soil pH; and changes in the numbers and composition of soil flora and fauna (Ashby & Pfeiffer, 1956; Nye & Greenland, 1960; Ahn, 1974; Moody, 1974; Norman, 1979; Aranson et al, 1982). Of these potential causes, declines in crop yields have most frequently been attributed to the decline in soil fertility, which may result from a decline in the amount of humus in the soil, changes in the physical condition of the soil, and/or changes in the nutrient status (Nye & Greenland, 1960; Kelly, 1975). The decline in humus content during the cropping period is rapid during the first year and then becomes progressively slower. Nye & Greenland (1960) suggested that the loss of humus may be less serious than a change in humus quality, even over a cropping period of ten or more years. Nutrient levels in the soil decline during the cropping stage because of their removals in crop harvest, and because of losses by erosion and leaching. Unlike the trees in a natural forest, field crops do not transfer nutrients from the subsoil to the surface because of their shallow rooting. Consequently, there can be net losses of nutrients due to leaching to below the shallow root zone during cropping. Until the first crop is fully established on cleared land, rapid organic matter decomposition and the physical impact of large raindrops on the soil can cause degradation of soil structure (Cunningham, 1963). A consequence of this on sloping sites can be potentially serious erosion (Norman, 1979). Phosphorus availability increases after burning, but decreases during the 109 cropping stage due to crop uptake, sorption by the mineral soil, and a decrease in availability, the latter being particularly important when iron and aluminum oxides are present and there is a drop in soil pH. Although some of the ash is washed away (surface erosion or soil leaching) by the first rain, the burning of a forest fallow adds considerable quantities of nutrient cations to the exchange complex of the soil (Webster & Wilson, 1966). During cropping the quantities of exchangeable cations tend to decline because of leaching and removal by the crop, but the losses are compensated for to some extent by cation release from nonexchangeable forms (Nye & Greenland, 1960). 7.1.4.THE R E C O V E R Y S T A G E O F F A L L O W V E G E T A T I O N The early recovery stage of a forest fallow is marked by rapid nutrient accumulation in the vegetation (Norman, 1979) and in the topsoil (Nye & Greenland, 1960). The increase of nutrients in the topsoil is associated with an increase in soil humus, which in turn depends on the aboveground litter production, fine root turnover, and the decomposition of organic materials. Since nitrogen is lost during the burn, the build up of organic nitrogen from the aboveground litterfall in the topsoil is important during the recovery stage of the fallow vegetation. The increase in N parallels to the increase in humus (Nye & Greenland, 1960; Sanchez, 1976). The increase in soil organic phosphorus during the recovery stage also depends on the amount accumulated in the vegetation. Only a slight increase in the exchangeable potassium, calcium and magnesium has been recorded during the early recovery stage. 110 7.2.THE OVERALL BIOMASS ACCUMULATION AND DYNAMICS IN THE TALUN-KEBUN SYSTEM 7.2.1.BIOMASS ACCUMULATION The accumulation of live plant biomass at various stages of the talun-kebun system is presented in Table 7.1. Aboveground biomass during the first year cropping stage was dominated by hyacinth bean, while the belowground biomass. was dominated by the mother rhizomes of bamboo. The high percentage value of the rhizomes is the result of the killing of virtually all the bamboo fine roots by hoeing. Weed biomass was not recorded separately for above and below-ground components in this study, so the data on total weed biomass was subdivided using ratios of belowground to aboveground from Uhl (1981): 0.26 for forbs and 0.20 for grasses. Based on this ratio, weeds contributed 6.7% to the aboveground biomass and 2.4% to the belowground biomass during the first year of cropping. Weed biomass decreased with increasing age after clearcutting. To eliminate the shading effects of bamboo grass shoots on the food crops, farmers continuously prune the shoots during the first year cropping stage. The total biomass of bamboo grass shoots removed in the weeding activity was 2.6% of the total biomass accumulated in this stage. Because cassava was the only crop species grown during the second year of cropping, and because it is a root crop, it contributed a substantial proportion of the belowground biomass (primarily tubers) in that year. This belowground biomass contributed 22.8% to the total biomass accumulation in the second year of cropping in comparison to 10.4% contributed by the aboveground biomass of cassava. Small bamboo culms were being produced by the second year of cropping, and these increased the contribution of the aboveground biomass of bamboo to approximately 20% of the total biomass accumulation in that year, almost twice the contribution of cassava aboveground biomass. However, the live belowground I l l biomass of bamboo decreased to 39% of the total biomass (from a value of 48% during the first year), because 20% of the mother rhizomes had died by the end of this stage. About 49% of the total biomass was accumulated in the aboveground bamboo biomass, and 47% was accumulated in the belowground bamboo biomass during the early fallow stage. The value of the aboveground biomass increased to over 50%, but the value of the belowground bamboo biomass decreased to 41% by the end of the mature fallow (bamboo talun). The percentage accumulation of live plant biomass at various stages of the talun-kebun system is illustrated in Figure 7.1. Approximately 98% of the total crop biomass during the first year cropping stage was accumulated as aboveground biomass and only 2% as belowground biomass. In contrast, about 70% of the total cassava biomass was accumulated as belowground biomass and only 30% in the aboveground biomass. The belowground biomass during the first year cropping stage was dominated by the mother rhizomes (93% of the total belowground biomass). Although the percentage biomass of mother rhizomes during the second year cropping stage was lower than that during the first year cropping stage, mother rhizomes still played an important role as a storage for nutrients. Mother rhizomes comprised 28% of the total live biomass during the second year cropping stage. Live new rhizomes and fine roots made up 11% of the total biomass accumulated in this period. Fine roots had the highest percentage of any of the belowground biomass components during the early and mature fallow stage. New rhizome biomass increased in importance relative to the total biomass, while mother rhizomes continued to lose their vigor during the early fallow stage. The biomass of the mother rhizomes continued to decline and had been completely replaced by live new rhizomes by the end of the mature fallow (mature bamboo talun). 112 The total biomass accumulation13 over the 6 year talun-kebun rotation cycle was 96.5 t ha-l, consisting of 59.6 t ha-l of aboveground biomass and 36.9 t ha-l of belowground biomass. The total accumulation of biomass in food crop species during the two year cropping period constituted 14.5% of the total biomass accumulation over the entire cycle. Weeds contributed 3.5% of the total biomass, and the rest consisted of bamboo biomass. Total ecosystem net primary productivity (standing crop of biomass + litterfall, ignoring fine root turnover) was 11.7 t ha-l for the first year of cropping, 16.41 ha-l for the second year of cropping, and a total of 85.9 t ha-l (including branch litterfall) over the four years of fallow (litterfall data are presented in Chapter 6). b The data presented here are: the bamboo and weed biomass at the end of the fallow + the crop, weed and aboveground bamboo biomass at the end of the first year of cropping + the crop and weed biomass at the end of the second year of cropping + weed biomass at the end of the early fallow. Table 7.1.Accumulation of live plant biomass at various stages of the talun-kebun Scle. The numbers in brackets are the data expressed as a percentage of e total for that stage. Stage Species Biomass (t ha"1 dry mass) Aboveground Belowground T o t a l ]St Cucumber 0.241 (1.3) 0.01 (0.1) 0.25 (1.3) yr Bitter crop- solanum 0.451 (2.3) 0.10 (0.5) 0.55 (2.6) ping Hyacinth bean 7.201 (36.2) 0.10 (0.5) 7.30 (36.7) Weeds 1.332 (6.7) 0.47 (2.3) 1.80 (9.0) Bamboo 0.502 (2.5) 9.50 (47.7) 10.00 (50.2) T o t a l 9.72 (48.9) 10.18' (51.1) 19.90 (100.0) 2 n d Cassava 2.101 (10.6) 4.60 (23.2) 6.70 (33.8) yr Weeds 1.15 (5.8) 0.35 (1.8) 1.50 (7.6) crop- Bamboo 4.10 (20.7) 7.503 (37.9) 11.50 (58.6) ping T o t a l 7.35 (36.4) 12.45 (63.6) 19.70 (100.0) early fallow Weeds Bamboo 0.88 14.20 (3.1) (49.3) 0.22 13.503 (0.9) (46.9) 1.10 27.70 (4.0) (96.2) T o t a l 15.08 (52.4) 13.72 (47.6) 28.80 (100.0) mature fallow Weeds Bamboo 0.15 45.10 (0.2) (58.7) 0.05 31.503 (0.1) (41.0) 0.20 76.60 (0.3) (99.7) T o t a l 45.60 (58.9) 31.55 (41.1) 76.80 (100.0) 1 the maximum observed biomass of these species 2 the total biomass that was removed by weeding/pruning 3 mother rhizomes + live new rhizomes + fine roots (see Chapter 4) 114 60r 40 20 • « > o .a a o E o H "5 o •1 * d iri io ci CM q ro o oi 1-above ground: crops weeds bamboo CM CO in 20-C 3 O o 4 0 • CM 60 |St y r -cropping (yr I) o aj CM O r-' ao CM CM CO CM CM CM CM early w fallow (yr 3) in to m CM mature fallow (yr 6) 2 n d yr cropping ( yr 2) bamboo below ground: mother rhizomes live new rhizomes fine roots weed roots crop roots and tubers Figure 7.1 The percentage accumulation of live plant biomass at various stages of the talun-kebun system. Values in brackets are percentages of the total biomass. 115 7.2.2. BIOMASS R E M O V A L S The removals of live plant biomass at various stages of the talun-kebun cycle are presented in Table 7.2. Approximately 10.8 t ha-l and 6.7 t ha-l of biomass were removed during the first and second year cropping, respectively. There were no removals during the early to mid fallow stages (year 3 to the end of year 5 after clearcutting the previous bamboo talun): The highest removals occurred at the end of the mature fallow stage (the end of year 6), when all bamboo culms were cleared prior to cropping. Approximately 45.3 t ha-l of biomass was removed from the field at that time, out of which 0.2 t ha-l W as weed biomass and 4.7 t ha-l was bamboo foliage. Weeds were removed during the litter raking, piled together with bamboo litter, and burned. Foliage was also piled and burned after the removal of the culms during the clearing of the mature talun. The ash from both piles was mixed and returned to the soil during fertilization of the first year crops. The total removal of biomass during a 6 year talun-kebun rotation cycle was 62.8 t ha-l, of which 17% was removed during the first year cropping stage, 11% was removed during the second year cropping stage, and about 72% was removed during the clearance of the mature bamboo talun at the end of year 6. This overall removal represents 61% of the total biomass accumulation or 58% of the total net production. In addition to this removal, an estimate of 2.9 t ha-l 0 f dead branches was removed during four years of fallow to be used as firewood. 116 Table 7.2.Removal of live plant biomass (t ha*1) at various stages of the talun-kebun cycle. The numbers in brackets are the removals as a percentage of the total removal at that stage. Stages Species Biomass removals (t ha -1 dry mass) Aboveground Belowground Total jst Cucumber 0.3 (2.8) 0.3 (2.8) year Bitter crop- solanum 0.6 (5.6) 0.1 (0.9) 0.7 (6.6) ping Hyacinth bean 7.4 (68.5) 0.1 (0.9) 7.5 (69.5) Weeds 0.5 (4.6) 1.3 (12.1) 1.8 (17.0) Bamboo 0.5 (4.6) - 0.5 (4.7) T o t a l 9.3 (86.1) 1.5 (13.9) 10.8 (100) 2nd Cassava 2.1 (31.3) 4.6 (68.7) 6.7 (100) year Weeds - - -crop- Bamboo - - -ping T o t a l 2.1 (31.3) 4.6 (68.7) 6.7 (100) fallow1 Weeds _ Bamboo - - -T o t a l mature Weeds 0.2 (0.3) 0.1 (0.1) 0.3 (0.4) fallow Bamboo 45.1 (99.6) - 45.1 (99.6) (yr6) T o t a l 45.3 (99.9) 0.1 (0.1) 45.3 (100) i.e., the period of yr 3, yr 4, and yr 5. 7.3. NUTR IENT INVENTORY AND DYNAMICS IN PLANTS 7.3.1.INTRODUCTION Knowledge of the inventory and dynamics of nutrients in plant components at various stages of the talun-kebun is important for an assessment of changes in the site nutrient capital and future site productivity under alternative cropping strategies. The nutrient content of each plant component is usually determined by multiplying the biomass by the average nutrient concentration in that particular 117 component (Vitousek & Sanford, 1986). Ueda (1960) studied N, P, and K concentration in culms, leaves, branches, and rhizomes of bamboo. Except for K, there was no significant difference in the nutrient concentrations of the clump forming species (Leleba multiplex) and the single culm species (Phyllostachys reticulata). Leaves had higher concentrations than culms, branches, and rhizomes. For example, the N concentration in Leleba leaves was nine times, eight times, and four times higher than the concentration in culms, branches, and rhizomes, respectively. Except for rhizomes, the effect of age on nutrient concentration in various components of bamboo is still not well documented. Ueda (1960) has reported that N concentrations in rhizomes of Phyllostachys edulis decreases with increasing rhizome age as follow: 1 year old > 2-5 year old > 6-10 years old. The highest P concentration, however, is found in the 2-5 years old rhizomes. Nutrient concentrations in cassava vary between plant components and with the age of both tissue and the plant (Howeler & Cadavid, 1983). The concentration of most nutrients is highest between 2-4 months, within which time the cassava shows the highest growth rate and nutrient demand. N, P, and S concentrations in cassava plants are highest in leaves, followed by stems, and tubers, while K, Ca, and Mg concentrations tend to be higher in stems than in leaves, and lowest in the tubers (Howeler, 1985). Initially, most nutrients accumulated by cassava are in leaves and the stem, but are translocated to tubers in the latter stage of the growth cycle. Ca and Mg accumulate more in stem than roots (Howeler & Cadavid, 1983). Removal of accumulated nutrients in the tuber harvest may deplete nutrient reserve in the soil, particularly for K. Most studies of nutrients in the first year crop species (hyacinth bean, cucumber, and bitter solanum) have focused on the calorie and protein contents; 118 none of them has discussed nutrient concentrations of each crop component. Schaafhausen (1962) reported briefly the N, P, and K concentrations in hyacinth beans without separating the plant components. He found that the mean concentrations for N, P, and K in the whole plant grown as pulse were 2.61, 0.15, and 1.97%, respectively. 7.3.2. OBJECTIVES The objective of this part of the study was to quantify the contents of five major nutrients (N, P, K, Ca, and Mg) in each plant component. These data are used to describe nutrient accumulation and removals in various stages of the talun-kebun rotation cycle. 7.3.3. METHODS Samples for chemical analysis were the same as those used for biomass measurement (see Chapter 4 and Chapter 5). Dry mass of all samples was calculated after constant mass was obtained in an oven at 80°C. Plant tissues (leaves, stems, roots, rhizomes, tubers, flowers, fruits, and pods) were ground in a Wiley mill to pass a 1 mm sieve. N in plant tissues was determined by the semi-micro Kjehdahl digestion method. Phosphorus was determined colorimetrically after digestion with nitric-perchloric acid (Johnson & Uhlrich, 1959; Yoshida et al, 1976). Ca and Mg were determined by atomic absorption spectrophotometry following dry ashing, and K was determined by flame photometry (Yoshida et al, 1976). Except for bamboo samples, three replicate analyses were run for each sample. Due to the large amount of samples, four replicate analyses were run for each bamboo sample. The analyses were done by staff members of the National Institute of Chemistry, Bandung, Indonesia. 119 7.3.4.RESULTS AND DISCUSSION 1. NUTRIENT CONCENTRATIONS IN PLANTS A. CROP PLANTS N, P, Ca, and Mg concentrations in cucumber plants were highest in foliage, followed by stem and roots, while K concentration was highest in roots and lowest in foliage. Except for P and K concentrations in the roots, nutrient concentrations increased with increasing age until the plants reached maturity, and then started to decline. The concentration of P increased with increasing plant age, while K concentration decreased. The concentration of all nutrients in cucumber flowers and fruits was higher than stem and roots, but lower than foliage. Nutrient concentrations in bitter solanum were highest in foliage and lowest in roots. The concentration of all nutrients increased during the growth period (until 100-130 days after planting), but had begun to decline at the end of the harvest. Flowers had a higher nutrient concentration than fruits. The N and Ca concentrations of hyacinth bean were highest in the foliage, while the highest P and Mg concentrations were found in the stems. Roots showed the highest K concentration. Ca and Mg concentrations in all components increased with increasing plant age. P and K showed the highest concentrations in young plants and decreased thereafter. N concentration in the foliage decreased with age, while the concentration in the stems increased. The mature bean pods had the highest N concentration in comparison with the N concentrations in other components at the time of harvest. N and P concentrations in cassava plants were highest in foliage, followed by stems and tubers. K concentration was high in all components. Stem showed the highest concentration of Ca and Mg. All nutrient concentration decreased with increasing age. Nutrient concentrations in weeds growing during the cropping stages were 120 higher than those growing during the fallow. This condition might be related to differences in the species at the different stages, to differences in the percentage of light intensity that penetrated to the ground, and to differences in soil fertility. B. BAMBOO Nutrient concentrations varied among various tissues as well as with time after planting. Bamboo foliage contained the highest concentrations of all nutrients studied (N, P, K, Ca, and Mg) than the other components. Culms had lower nutrient concentrations than foliage and branches. The concentrations in roots was lower than those in rhizomes. Nutrient concentrations in all components decreased with increasing age. The means and standard deviation of nutrient concentration for each species are presented in Appendix 4. 2. NUTRIENT ACCUMULATION IN LIVE PLANT BIOMASS AT VARIOUS STAGES OF THE TALUN-KEBUN CYCLEc Figures 7.2 and 7.3 illustrate the accumulation of N, P, K, Ca, and Mg in the above- and below-ground components of live plant biomass at various stages of the talun-kebun cycle. Approximately 77% of the nitrogen in the plants during the first year cropping stage was accumulated in the aboveground, and only 1% in the belowground crop biomass. Cassava accumulated 43% of the total biomass nitrogen during the second year of cropping, of which 17% was in the aboveground materials and 26% was in tubers. N accumulation in aboveground bamboo biomass increased significantly during the fallow period, from 123 kg ha-l at the early fallow to 244 kg ha-l by the end of the mature fallow. Approximately 48 kg ha-l (11% of the total c Only mean of the nutrient contents of the live plant biomass are included in the text. The standard deviation associated with the biomass and nutrient concentration data from which these means are derived are presented in Appendix 5. 121 biomass N) was stored in mother rhizomes during the first year of cropping; however, this amount decreased substantially during the cassava cropping and during the fallow stage. New rhizomes and roots became increasingly important from the end of the cropping stage onwards, and contributed approximately 54% of the total biomass N content by the end of the mature fallow. Weeds accumulated 10 to 20% of the total biomass nitrogen during the two year of cropping and early fallow stages, but this value became negligible at the end of the mature fallow stage. 33% of the total biomass P during the first year of cropping was found in bamboo mother rhizomes; 53% was found in the crops. Although the belowground biomass of cassava was much higher than the aboveground biomass, the P content of the belowground cassava was lower than that of the aboveground biomass due to the low concentration of P in the tubers. Aboveground bamboo stored 1 kg ha-l of P during the first year cropping. This value increased to approximately 54.4 kg ha-l by the end of a six year rotation. Approximately 230.8 kg ha- 1 of K was accumulated during the first year of cropping, of which 55% was in crop biomass, 12% was found in weeds, 2% was accumulated in aboveground bamboo grass-shoots, and 31% was stored in the mother rhizomes. In contrast to the higher accumulation of K in the aboveground than in the belowground parts of other stages, more K was stored in the belowground part during the second year of cropping (57% of the total biomass K). Crop plants contained 73% of the total calcium and 78% of the total magnesium in plant biomass during the first year of cropping, most of which was stored in the aboveground part. Despite the large belowground biomass of cassava, the Ca content in this component was only half of that in the aboveground part. Mg content in the aboveground cassava was lower than its content in the belowground part. Ca and Mg accumulation in other stages was similar to the accumulation 122 pattern of other nutrients. Weeds contributed 5 to 16% of the total nutrient accumulated in plant biomass during the cropping stages, 2 to 6% during the early fallow stage, and only 0.1 to 0.3% at the end of the rotation cycle. Figure 7.4. summarizes the nutrient accumulation in live plant biomass in the four stages of the talun-kebun cycle. Nitrogen and potassium constituted the major part of the nutrient accumulation in all four stages. Nutrient accumulation during the first year of cropping was higher than in the second year of cropping. The accumulation of nutrients in the plants was rapid after the cropping ceased at the end of the second year. Approximately 541.7 kg ha-l of nutrients was accumulated in the first year fallow, and about 1118.1 kg ha-l was reached after four year of fallow. Kalpage (1974) has noted that the rate of accumulation of nutrients in the forest vegetation in a shifting cultivation practice is high in the early years of fallow, which reflects the period of maximum biomass accumulation. 4 0 0 392.2 2 0 0 • o xz 0 2 00 tZ777l 100 r 5 0 • 114.7 i 77.974.3 55.4 70.5 98 .4 131.2 B C D Nitrogen 50 54.6 20.2 25.2 14.1 7Z& 11-5 13 I 18.9 3 0 0 200 o •*= 100 100 3 4.7 A B C D 2 0 0 Phosphorus 277.7 148 103.4 78.7 49.7 99.9 142 A B C D Potassium crops | | weeds f23 bamboo above ground below ground mother rhizome lllllll new rhizome bamboo roots weed roots crop roots |St year cropping 2 n ( * year cropping • A B C D early fal low mature f a l l o w Figure 7.2 Accumulation of N , P, K in live plant biomass at various stages of the talun-kebun cycle. M CO 105 100 54.3 50 42 28.4 22 16.4 17.8 20.4 5 0 35.1 A B C D Calcium 150 100 'o 5 0 a 50 121.5 49.6 49.9 29.3 22.6 26.7 25.8 45.8 A B C D 100 L Magnesium above ground below ground 1:: j crops | | weeds I'.'H bamboo mother rhizome new rhizome bamboo roots | | weed roots V/A crop roots |St year cropping 2 n d year cropping early fallow mature fallow A B C D Figure 7.3 Accumulation of Ca ond Mg in live plant biomass at various stages of the talun-kebun cycle.  1500 r l o J C i IOOO c o u 819.9 c 0> 500 487.8 1062.3 507.4 TuT stage mm N • P Ca I : I s t year cropping H : 2 n d year cropping UT: early fallow E T : mature fallow Figure 7.4 Nutrient accumulat ion in live plant biomass at various stages of the talun-kebun cycle. 126 3. REMOVALS OF NUTRIENTS IN LIVE PLANT BIOMASS AT VARIOUS STAGES OF THE TALUN-KEBUN CYCLE Table 7.3 shows removals of nutrients in live plant biomass at various stages of the talun-kebun cycle. Table 7.3. Removal of nutrients in live plant biomass (kg ha-l) at various stages of the talun-kebun cycle. The numbers in brackets are the removals as a percentage ofthe total removal at that stage. Stages Species Nutrient removals (kg ha"1) N P K Ca Mg s^t Cucumber 14.4 2.2 12.8 1.7 1.7 year (4) (11) (9) (4) (3) crop- Bitter 17.3 1.3 18.0 3.1 2.5 ping solanum (4) (6) (12) (8) (5) Hyacinth 305.6 12.4 85.6 30.2 33.6 bean (79) (61) (57) (74) (67) Weeds 34.1 3.4 28.1 3.4 10.8 (12) (17) (19) (8) (21) Bamboo 5.2 1.0 7.2 2.5 1.8 U> (5) (3) (6) (4) T o t a l 376.6 20.4 149.1 41.0 50.4 (100) (100) (100) (100) (100) 2nd Cassava 74.4 6.4 78.2 16.6 13.4 year (100) (100) (100) (100) (100) crop- Weeds - - - - -ping Bamboo - - - - -T o t a l 74.4 6.4 78.2 16.6 13.4 (100) (100) (100) (100) (100) Fallow1 Weeds Bamboo - - - - -T o t a l . . . Mature Weeds 1.7 0.2 1.9 0.3 0.7 fallow (1) (0.4) (1) (0.3) (1) Bamboo 113.4 54.4 293.4 104.8 121.0 (99) (99.6) (99) (99.7) (99) T o t a l 115.1 54.6 295.3 105.1 121.7 (100) (100) (100) (100) (100) 1 the period of yr 3, yr 4, and yr 5. 127 A total of 463 kg ha" 1 of N (excluding the N content of cassava stalk) was removed from the field during the two year cropping stage, consisting of 378.9 kg ha- 1 net removal (harvested food products and firewood) and 84.1 kg ha- 1 temporary removal (composted dead plant material). No removals occurred during the fallow stage, but at the end of the mature talun the bamboo was clearcut and approximately 113.4 kg ha- 1 of N was removed. Therefore, the total removals of N in live plant biomass over a 6 year talun-kebun rotation cycle was 566.1 kg ha-l, of which 67% was removed during the first year cropping stage, 13% was removed in cassava harvest, and the other 20% was removed during clearcutting the mature talun. P removals were much lower than N removals. Only 81.4 kg ha- 1 of P was removed in a 6 year rotation cycle. Approximately 27% of this was removed in crop harvest (first year crops and cassava), 5% in weeds, 1% in bamboo grass shoots, and 67% in bamboo harvest. Potassium had the second largest nutrient removal during the 6 year talun-kebun cycle. The harvested products (fruits and pods) of the first year crops removed 103.1 kg ha-l of K or 22% of the total K removed during the 6 year cycle. Approximately 49.1 kg ha- 1 of K was removed in the harvest of cassava tubers during the second year cropping stage. Considering the tuber production of 6.7 tonnes, cassava tubers extracted an average of 7.3 kg ha- 1 of K in per tonne drymass of harvested tubers. This value was within the range of 1.5 to 9.9 kg ha-l removed in per tonne of harvested cassava tubers reported by Howeler (1980). Bamboo harvest removed 293.4 kg ha- 1 of K including 52.6 kg ha- 1 in foliage which was removed temporarily. In contrast to other nutrients, the removals of Ca and Mg in harvest residues during the first year cropping stage were relatively high, accounting for 54% and 51% of their total removals during that stage, respectively. 128 The overall nutrient removals accounted for 72% of N, 66% of P, 62% of K, 75% of Ca, and 75% of Mg accumulated in the live plant biomass over the 6 year talun-kebun rotation cycle. Net removals represented 78 to 83% of the total removals, the balance being removed and then returned as compost. 4. ALBIZIA A N D O T H E R T R E E S Albizia and other tree species are found at low density (less than 200 trees per hectare) in the bamboo talun-kebun. These trees were not cut during the clearcutting of bamboo in the study fields; therefore, they were not counted in biomass accumulation and removals. The contribution of Albizia to the biomass accumulation of the talun-kebun was probably minor since they only represented a small percentage of canopy space and basal area in the system, but Albizia may contribute significantly to the biogeochemistry of the system by fixing N. The lack of data on Albizia and other tree species is believed to have resulted in a small underestimate of biomass accumulation values, but probably a somewhat more serious underestimate of the nitrogen budget of the system. 7.4. N U T R I E N T I N V E N T O R I E S I N T H E F O R E S T F L O O R / E C T O R G A N I C L A Y E R A N D I N T H E M I N E R A L SOILS . 7 . 4 . 1 . I N T R O D U C T I O N Knowledge of the amount and the distribution of nutrients in the forest floor/ectorganic layer and in various soil depths under different talun-kebun stages is important for an understanding of the biogeochemical changes that occur during the rotation cycle. Sanchez (1976) noted that soils undergo a series of changes in chemical properties when cleared, burned, cropped, and abandoned to forest regrowth. There has been a common belief that tropical soils deteriorate rapidly both physically and 129 chemically upon exposure following devegetation, and that soils are depleted of nutrients during the cropping phase of shifting cultivation. Watters (1971) claimed that nutrient depletion is the major reason for the rapid decline in productivity observed during successive years of cropping in shifting cultivation. Recent investigators, however, have found that despite declining crop yield, nutrient stocks in the soil of cultivated plots often remain higher than those in undisturbed forests. For example, in a study on a reddish yellow latosol in Ghana, Nye and Greenland (1964) found that exchangeable cations and total nitrogen increased after burning and then decreased, but after two years the levels were still higher than those before clearing. From his study on a series of slash and burn sites in the upper Orinoco region of Venezuela, Harris (1971) found that the mean values of organic carbon, available phosphorus, and exchangeable calcium and magnesium in all cultivated plots were higher than those in undisturbed forest. Zinke et al (1978) in Thailand found that exchangeable calcium and magnesium were highest in recently cut and burned plots and remained higher throughout a 10 year rotation cycle than in a nearby old growth forest that served as a control. It is not really surprising that these results were obtained because: 1. Mature vegetation removes much of the available nutrients from the soil and stores them in biomass or holds them in a tight cycle. 2. After cutting and burning, much of the nutrients so held are released to the soil. The latest results of a study on shifting cultivation in San Carlos, Venezuela (Jordan, 1987) showed that there was no sharp decline in total soil nutrient stocks during cultivation, and Jordan pointed out that the decline in available nutrients, not total nutrients, might have been the critical factor. He explained further that stocks of nutrients showed relatively little change, because nutrients that leached out of the soil were replaced by nutrients that leached into the soil from ash and 130 decomposing slash. It is the total site budget and the temporal pattern of gains and losses that is important, not just the "snapshot" inventory that is quantified in so many studies. However, such inventory data are needed for the evaluation of such temporal budgets, and for use in models of nutrient cycling. The objective of this part of the study was to quantify the inventories of five major nutrients (N, P, K, Ca, and Mg) in the forest floor/ectorganic layer and in various soil depth under different talun-kebun stages. The data are used later in this chapter to illustrate the temporal changes in these two nutrient compartments under the talun-kebun cycle. 7.4.2.METHODS The sampling method for forest floor/ectorganic material has already been described in Chapter 6. The samples were chemically analyzed using the same methods as used for plants (Part 7.3). The mineral soil was sampled in TK-1, TK-2, TK-3, and TK-4 at the beginning and at the end of each stage. Soil was sampled at various depths at three locations in each plot. The samples from each location consisted of a composite of 5 cores. Sampling depths were 0-5 cm, 5-25 cm, and 45-75 cm. Each composite sample was well mixed, oven dried at 105°C, and ground, using a mortar and pestle, to pass a 2 mm sieve. Soil texture was determined by the determination of particle size distribution using the pipette method (Hidayat, 1978). Bulk density was determined at each depth by taking four volumetric cores at each location, 5.4 cm in diameter, coated with saran-resin, oven drying at 105°C, weighing, then dividing each soil volume by its mass (Brasher et al, 1966). Soil pH was determined in water and 1.0 M KC1 using a soihliquid volume ratio of 1:2.5 (Hidayat, 1978). Total N was determined by a semi-micro Kjeldahl technique (Bremmer, 1965; 131 Hidayat, 1978). Available soil P was extracted using Bray-Kurtz II extractant solution, followed by colorimetric determination (Hidayat, 1978). Exchangeable cations in the soil were extracted using ammonium acetate at pH 7.0. Ca and Mg were determined using an atomic absorption spectrophotometer, and K was determined using a flame spectrophotometer (Hidayat, 1978). 7.4.3.RESULTS AND DISCUSSION 1.FOREST FLOOR NUTRIENTS Table 7.4. summarizes the N, P, K, Ca, and Mg contents in the forest floor/ectorganic layer of various stages of the talun-kebun rotation. Relative nutrient concentrations were N > Ca > Mg > K > P for the cropping stages, N > K > Mg > Ca > P during the early fallow, and N > K > Ca > Mg > P in the mature fallow (Appendix 5). Concentrations of N and Ca were higher in the cropping stages than those in the fallow. K concentration was highest during the first year of cropping, followed by that of the early fallow and the mature talun. The lowest concentration of K was found in the ectorganic layer during the second year of cropping. Mg con-centration was high until the end of the early fallow but then decreased substantially during the mature fallow. P concentration was highest during the early fallow stage. Accumulation of N, P, K, Ca, and Mg in the forest floor reached its maximum at the end of the mature talun, because forest floor mass was the highest during this period. Forest floor nutrients at the end of the mature talun (72 months after the previous clearcutting) were approximately 3 to 4 times higher than those during the early fallow stage (36 months after the previous clearcutting). The maximum total forest floor N, P, K, Ca, and Mg contents were 105.6, 11.5, 69.3, 27.5, and 24.6 kg ha-l, respectively. The values for N, K, and Mg were within the range of nutrient contents of litter layers observed in tropical forests in Zaire, Ghana, Guatemala, and Puerto Rico as summarized by Sanchez (1973). The 132 Ca content of the forest floor at the end of the mature talun in this study was below the range of tropical forests reported by Sanchez (33.1 kg ha-l in comparison to a reported range of 45-220 kg ha-l), while the P content (11.5 kg ha-l) was higher than the reported range (1-7 kg ha-l). Although the nutrient concentrations in the ectorganic layers during the two year cropping stages were higher than those during the fallow, the rapid decomposition of the ectorganic materials in these stages resulted in a lower accumulation of ectorganic materials (see Chapter 6). As a result, the nutrient content of the ectorganic layer in the two cropping stages was relatively low in comparison to that of the mature talun. Table 7.4. N, P, K, Ca, and Mg contents in forest floor/ectorganic layer at various stages of the talun-kebun rotation cycle. Stages N Nutrient content (kg ha"1) P K Ca Mg Mass (t ha-l) First year of cropping 21.3 1.0 2.6 14.9 4.2 0.8a Second year of cropping 25.3 0.9 4.9 6.8 4.4 1.3b Early fallow 36.2 3.3 18.3 7.0 7.5 2.6c Mature fallow (mature talun) 105.6 11.5 69.3 27.5 24.6 13.5d a accumulation of hyacinth beans litter " cassava litter and bamboo litter that was flown into the field c intact bamboo litter d surface litter, fragmented, litter and very decomposed litter. 2. SOIL T E X T U R E , B U L K D E N S I T Y , A N D p H There were no significant differences in soil texture under various stages of the talun-kebun cycle. The soils have a high clay content (71-84%), 14-24% silt, and 2-7% sand. Lundgren (1978) considered soil texture as a comparatively stable 133 property which is unlikely to undergo changes as a result of vegetation changes. Soil bulk density increased with depth. There is a slight change of bulk density of the upper soil layers (0-5 and 5-25 cm depths) after clearing and hoeing a mature talun. The bulk density decreased from 1.18 to 1.12 at 0-5 cm depth, and from 1.22 to 1.18 at 5-25 cm depth. Lundgren (1978) pointed out that unlike texture, soil bulk density was directly affected by changes in land use and vegetation. The mean bulk density over a talun-kebun rotation ranges from 1.12 to 1.18 and from 1.18 to 1.22 at 0-5 and 5-25 cm depths, respectively. The analysis of variance (ANOVA) for the bulk density data, however, showed that there were no significant differences in soil bulk density under various stages of the talun-kebun cycle (Appendix 6). The above data show that the physical properties of soils under various stages of the talun-kebun cycle do not differ significantly from one another. Therefore, any differences between stages in soil nutrient concentrations, biomass, and nutrient content in plant biomass are due to management (treatment) effects. The soils in the study sites are quite acid with pH ranging from 5.0 to 5.6 when measured in H2O and 3.8 to 4.2 when measured in KC1. Soil pH in the top 0-25 cm depth increased after clearing and burning but decreased gradually with time. Sanchez (1976) found that the pH of acid soils leveled off faster than the less acid one. For example, Brinkman and Nascimento (1973 cited in Sanchez, 1976) observed that the pH of yellow latosol (Oxisol) topsoils increased from 3.8 to 4.5 with burning, but decreased to the original level within about 4 months. The pH of the upper soil layer in the study area increased from 5.0 before clearing to 5.6 after clearing and burning, decreased gradually during the cropping stage, and leveled off to 5.1 by the end of one year fallow. The acid conditions of the soil in the study area seemed to be responsible for the low availability of 134 phosphorus in the soil. 3. M I N E R A L SOIL N U T R I E N T S The concentration of total N in the top 0-5 cm decreased slightly from the mature talun to two months after clearing and hoeing, but the concentration in the 5-25 cm depth increased (Figure 7.5). This probably reflects the removal of the forest floor, and the killing of bamboo fine roots, respectively. The N concentration in both depths had decreased by the end of the first year cropping. A similar trend was found during the second year of cropping; with a small decrease in total N concentrations over the year. There was a slight increase of N concentration within the first year of fallow, which probably resulted from the increase in deposition of above-ground litterfall and the death fine roots. The CV (coefficient of variation) of N concentrations of the mature talun field was 14% and 16% in the surface 0-5 cm and 5-25 cm depths, respectively. The value increased after clearing and hoeing (20% and 25%, respectively), which showed that soil disturbance (hoeing) increased the spatial variability of soil N. The CV in the surface 0-5 cm increased to 30% at the end of the first year cropping due to mixed cropping and fertilization; however, there was less variability in the 5-25 cm depth (only 13%). The spatial variability in the second year of cropping and in the early fallow stage was quite low (ranging from 10-14% in 0-5cm and 5-25 cm depths). Figure 7.6. illustrates the changes of N contents in the forest floor/ectorganic layer and mineral soil under various stages of the talun-kebun cycle. Total N in the upper 25 cm of mineral soil varied between 6,000 and 7,000 kg ha-l throughout the six year talun-kebun rotation cycle. This was approximately 60 times higher than the N held in the aboveground bamboo and the N content of forest floor in the mature talun stage. Apparently much of this is in relatively stable organic nitrogen, because the food crops responded well to N fertilization, and sustained cropping is not possible without sustained N inputs in fertilizer. 0 . 4 0 r 10 12 TK-1 STK-2 I s t year cropping T K - 3 2 n d year cropping 24 36 •> months after clearing and hoeing T K - 4 I s t year fallow soil depth: 0 - 5 cm 5- 25 cm Figure 7.5 Total N concentration in soil in various stages of a talun-kebun cycle. 100 Z 90 • o o ° c *- o ° s I o o 40 30 20 1400 1600 9 1800 >» o "= 2000 o a - 5000 6000 7000 soil surface forest f l o o r / ectorganic layer •mineral soil 0 - 5 c m depth tx£1 mineral soil 5 - 2 5 c m depth 0 10 20 3 6 mature talun Ist year cropping 2nd year cropping ist y e a r talun (early fallow) months after clearing and hoeing Figure 7.6 Total N in the forest f loor/ ectorganic layer and in the mineral soils under various stages of a talun- kebun rotation cycle. 137 Sanchez (1979) stressed that in moist tropical forest ecosystems, the bulk of the N (70-80%) was in the soil and not in the live biomass. Therefore, changes in the available N are more important than changes in total N for assessing changes in soil fertility in relation to the decline in crop productivity when cultivation is prolonged. Unfortunately, due to time and manpower limitations, the availability of N in the soil was not analyzed in this study. Variability of the available P concentrations during the cropping stages was higher than in the fallow stages (the CV was 21% in comparison to 13% in the latter stage). A similar pattern was found for K, Ca, and Mg concentrations. Beckett and Webster (1966) have stated that row cultivation, the growth of row of tree crops, and fertilizer application tend to superimpose additional heterogeneity on soil chemical properties. The Bray-extractable phosphorus in the surface 5 cm layer of a Guatemalan Inceptisol increased about four times after burning, remained at this level for about 6 months, and it was still twice the original value at the end of 1 year (Popenoe, unpublished in Sanchez, 1976). After clearing, burning and hoeing, the available P in the surface 5 cm layer of the mature talun field in my study area only increased by 45% and the level declined to 92 and 75% of the original level after the first and second year of cropping, respectively, and still remained at that level by the end of the first year fallow. Sanchez (1976) found that the available phosphorus level of a soil increases upon clearing and burning because of the phophorus contents in the ash. Since burning in the talun-kebun was done in piles, the ash was not distributed on the surface and only returned to the soil during as fertilizer mixture during the first year of cropping. There was a significant reduction in the content of the available P in 5-25 and 25-45 cm of mineral soil depths during the cassava cropping. Sanchez (1976) has suggested that the decrease in available P with cropping is probably due to 138 sorption by the mineral soil and/or crop removal. Table 7.5 summarizes the changes of the available phosphorus in mineral soils at various stages of the talun-kebun cycle. Table 7.5. Available phosphorus contents in mineral soils at various stages of a talun-kebun cycle in Soreang, West Java, Indonesia. Soil depths (cm) Mature talun Available P content (kg ha'1) After End ofthe End ofthe clearing first year second year & hoeing cropping cropping After one year of fallow 0-5 5.1 7.4 4.7 3.8 3.8 (2.1)* (2.8) (1.7) (1.2) (1.4) 5-25 10.5 12.0 9.5 6.4 6.3 (5.0) (5.1) (3.9) (2.6) (2.9) 25-45 4.9 6.1 5.8 3.4 4.1 (1.8) (2.7) (2.3) (2.0) (1.6) * The number between brackets represents standard deviation Quantities of exchangeable mineral elements in the mineral soils were present in the order Ca > Mg > K (Table 7.6.). The exchangeable K content increased by 25% after the first year of cropping which may due to the application of the fertilizer, but followed by a significant decrease after the cassava cropping. This was probably related to crop uptake and soil leaching. Exchangeable Ca and Mg still increased after the first year of cropping, but this was followed by a gradual decrease during the cassava cropping. The values of the exchangeable nutrients still declines during the one year of fallow. 139 Table 7.6. The amounts of exchangeable nutrients in mineral soils at various soil depths and at various stages of the talun-kebun cycle in Soreang, West Java, Indonesia. Stages Soil depths (cm) Soil exchangeable nutrients (kg ha-l) K Ca Mg Mature talun * 0-5 5-25 25-45 100.0 254.4 131.4 334.3 1447.7 1377.6 196.8 754.1 532.9 After clearing and hoeing* 0-5 5-25 25-45 87.6 307.6 229.9 321.1 1683.5 1461.6 198.1 748.6 532.9 End of the first year of cropping* 0-5 5-25 25-45 74.3 375.4 131.4 364.8 1664.0 1377.6 187.8 780.8 532.9 End of the second year of cropping** 0-5 5-25 25-45 66.9 222.6 132.4 372.4 1610.4 1371.6 201.7 714.4 537.1 After one year of fallow*** 0-5 5-25 25-45 60.0 159.0 131.4 356.5 1594.1 1360.8 184.7 694.6 522.7 ** Plot TK-1 & TK-2 Plot TK-3 *** Plot TK-4 7.5. N U T R I E N T D Y N A M I C S A N D I N P U T S - O U T P U T S B A L A N C E LN T H E TALUN-KEBUN S Y S T E M This part ofthe chapter describes the biogeochemical cycle (plant-soil-plant), the biochemical cycle (internal conservation in the plants), and the geochemical cycle (input-output balance) of the talun-kebun system. The biogeochemical cycle involves uptake by the vegetation, losses from the vegetation by litterfall (above & below-ground), defoliation, and rainwash, and recovery by the plants of nutrients released by decomposition and mineralization. Although burning is not normally considered to be part of the biogeochemical cycle, 140 it is an important mechanism in increasing the availability of nutrients in soil organic matter (Nye and Greenland, 1960). The biochemical cycle (internal conservation of nutrients witliin the plants) is an important mechanism by which plants can sustain growth on nutrient poor soils (Kimmins, 1987). Nutrients are usually transferred from older to younger leaves before leaf shedding, and similar transfers may occur when other tissues senesce. The geochemical cycle consisted of inputs of nutrients to the system, i.e. precipitation, fertilizers application, and biological nitrogen fixation. Losses of nutrients from the system occur through harvested materials that are removed from the fields, by leaching to below the root zone, downslope runoff or erosion, and by losses of N to the atmosphere during the burn (or by denitrification or other gaseous losses). 7.5.1.NUTRIENTS IN L I T T E R F A L L The return of organic matter and nutrients to the upper soil layers by litterfall and fine root death during the fallow period is the most important mechanism of the restoration of soil fertility in agroforestry or shifting cultivation systems (Ewel, 1976; Nye & Greenland, 1960). Fine root turnover was not quantified in this study. Nye & Greenland (1960) estimated an annual return of 199, 7.3, 68, 206, and 45 kg ha-l of N, P, K, Ca, and Mg, respectively, in leaf fall of a mature forest at Khade, Ghana. Swift et al (1981) recorded an annual input of 92, 6, 30, 140, and 27 kg ha-l of N, P, K, Ca, and Mg, respectively, in leaf litterfall of a five-year bush-fallow in Ibadan, Nigeria. In this part of the chapter, data are presented on nutrient transfer (N, P, K, Ca, and Mg) of aboveground litterfall in various stages of the talun-kebun system. Litterfall sampling was discussed in Chapter 6. Litterfall samples were chemically analyzed using the same methods as those for plants (Section 7.3). 141 Table 7.7 summarizes the nutrient content (kg ha"1) of aboveground litterfall at various stages of a talun-kebun cycle. Table 7.7. Nutrient content (kg ha-l) 0 f aboveground litterfalll at various stages of a talun-kebun cycle in Soreang, West Java, Indonesia. Stage2 N Nutrient content (kg ha-l) P K Ca Mg First year cropping 42.0 (1.4)3 1.5 (0.2) 4.8 (0.8) 34.1 (7.4) 4.5 (1.4) Second year 48.5 (1.8) 1.1 (0.4) 10.4 (0.1) 10.8 (1.7) 6.1 (1.6) Early fallow 28.2 (4.2) 3.7 (1.1) 26.4 (6.0) 7.5 (2.2) 7.7 (1.8) Mature talun 47.2 (8.5) 5.5 (2.2) 37.9 (6.2) 10.5 (2.3) 10.7 (1.6)-1 Cropping and fallow stages at the commencement of litter sampling 2 Numbers within brackets represent standard deviation. Table 7.7 shows that the contents of the five nutrients in litterfall for crops (hyacinth beans and cassava) was in the order N > Ca > K > M g > P, while for bamboo was in the order N > K > M g > Ca > P. The N content of aboveground litterfall of bamboo in the mature fallow was 1.7 times that of the early fallow. Bamboo leaf and sheath litterfall contributed 28.2 kg ha-l yr-1 of N during the early fallow and 47.2 kg ha-l yr-i during the mature fallow. Seth et al (1963) reported an annual return of 32.1 kg ha- 1 of N litterfall in a Dendrocalamus strictus plantation in India; this is within the range of bamboo N-litterfall values of the early fallow and the mature fallow of the talun-kebun cycle. The nutrient content of branch litterfall was estimated separately from the leaf and sheath litterfall because dead branches were periodically collected by the farmers, removed from the field and used as fuel wood, and therefore they did not 142 contribute to the forest floor nutrients. The nutrient content of branch litterfall was 2.2, 1, 2.9, 0.9, and 0.6 kg ha-l yr-l during the early fallow stage, and 2.8, 1.1, 5.1, 1, and 0.9 kg ha-l yr-l in the mature talun for N, P, K, Ca, and Mg, respectively. 7.5.2. INTERNAL CONSERVATION OF NUTRIENTS IN PLANTS Seth et al (1963) noticed that the concentration of N in the litterfall of Dendrocalamus strictus was the same as the N concentration of its foliage. A similar finding was reported by Nye (1961) from his study on broad leaved species in Ghana. These findings suggested that, in contrast to most temperate plant species, there may be little or no translocation of nitrogen out of leaves prior to abcission in tropical species (Ewel, 1968). In agreement with these studies, the concentration of N in leaf litterfall of bamboo in this study did not show any significant decreases in comparison to its foliage N. A similar pattern was found for P and Ca, but there was a 12% and 40% decreases in concentration of K and Mg, respectively, which may reflect internal cycling and foliar leaching of these nutrients. 7.5.3. NUTRIENTS IN PRECIPITATION & THROUGHFALL Rainwater collected under forest canopies (throughfall) often contains higher chemical concentrations than that in the open. Precipitation washing over foliage picks up exudates or dry fallout which has settled on the foliage, and some nutrients are leached out of the leaves (Nye, 1961; Kenworthy, 1971; Manokaran, 1980; Bruijnzel, 1982). However, Jordan et al (1980) reported a reverse pattern from •the Amazonian rain forest, where the concentrations of Ca and P in throughfall were actually less than in precipitation collected in the open. They argued that these nutrients were intercepted in the canopy by algae and lichens growing on leaf surfaces, and that this is a nutrient conserving mechanism in these nutrient-limited ecosystems. 143 Since bamboo has a closed canopy structure, it is important to quantify nutrients in throughfall to estimate losses by rainwash. In this part of the study, data are presented on nutrients in precipitation and in throughfall at various stages ofthe talun-kebun rotation cycle. The results were used to calculate nutrient losses from the vegetation by rainwash. 1 .METHODS The collectors for throughfall and bulk precipitation volume measurements consisted of V-shaped trough type raingauges of 140 cm long and 10 cm wide, which was connected by plastic tubing to plastic containers (Wiersum et al, 1979). Throughfall and precipitation for chemical analysis were collected using polyethylene funnels containing a plug of glass wool, and polyethylene containers. Three collectors for throughfall were installed at three randomly chosen locations in each of the fields (TK-1, TK-2, TK-3, and TK-4), 1 meter above soil surface to avoid soil splash. One polyethylene collector was installed near each throughfall collector. Collectors for precipitation were placed in an open field (grass cover with no tall vegetation). In addition to the existing precipitation collectors, a standard weather bureau rain gauge was installed to calibrate results of the trough type rain gauge against the standard rain gauge. Throughfall and precipitation water in each collector were measured after each major rain storm from December 1983 until April 1984, and samples from the polyethylene collectors were taken every second week. This frequency of sampling was dictated by the short rainy season during the study period. The collectors were rinsed with distilled water after each collection to prevent contamination between collection. Samples for chemical analyses were collected in polyethylene bottles, which had been rinsed with dilute hydrochloric acid and distilled water. Precipitation data for May-November 1984 were obtained from the local weather station, and the throughfall data for this period were estimated using 1 4 4 regression equations developed based on the relationship between precipitation and throughfall data for the December-April period at various stages of the talun-kebun rotation. Water samples collected in the field were brought to the laboratory in Bandung after each collection. Chemical analyses were conducted by staff members of the National Institute of Chemistry, Bandung, Indonesia. pH was measured soon after the samples were received, and the samples were then stored at 00C prior to analyses. The pH was measured using a temperature-compensating OrionR model 801 Ion-analyzer, equipped with reference electrode and pH electrode. Analyses for calcium, magnesium, and potassium were performed using a Varian Techtron model 1200 Atomic Absorption Spectrophotometer using an air-acetylene flame. Ammonium ion and phosphate were measured on a Technicon Autoanalyzer IlR system using standard methods. Nutrient contents in precipitation and throughfall for the months of May-November 1984 were calculated by using average concentrations from the months of December-April. Considering that nutrient concentrations in precipitation and throughfall during the dry period are usually higher than those of the wet period (Manokaran, 1981; Bruijnzeel, 1983), the results will underestimate the actual values for May-November. 2 .RESULTS A N D DISCUSSION Total rainfall for the 12 month period (December 1, 1983 until November 30, 1984) was 2096 mm. Throughfall under various stages of the talun-kebun system accounted for 62 to 95% of the precipitation. The value varied from 62 to 78% during the bamboo stage. Higher values were observed during the two year cropping stage (approximately 86 and 95% of the precipitation were collected as throughfall during the first and second year of cropping, respectively). The chemical content of precipitation reported here represents bulk 145 The chemical content of precipitation reported here represents bulk precipitation (mixture of rain and dry fallout), because the collectors were continuously open to the atmosphere. Precipitation during the one year period was estimated to contain 1.6, 2.0, and 0.6 kg ha- l of K, Ca, and Mg, respectively. N and P were not detectable in the precipitation water. Table 7.8. summarizes the nutrient content of precipitation and throughfall under various stages of the talun-kebun cycle. Table 7.8. The nutrient content of precipitation and throughfall under various stages of the talun-kebun cycle in Soreang, West Java, Indonesia (December 1983-December 1984). The number within brackets represents net leaching/ leaf wash. N Nutrient (kg ha'1) P K Ca Mg Precipitation - - 1.6 2.0 0.6 Throughfall: First year cropping 1.1 (1.1) 0.8 (0.8) 2.5 (0.9) 2.1(0.1) 0.7 (0.1) Second year cropping 1.0 (1.0) 0.9 (0.9) 3.8(2.2) 3.1(1.1) 1.4 (0.8) Early fallow 0.6 (0.6) 0.8 (0.8) 9.3 (7.7) 1.5 (-0.5) 4.1 (3.5) Mature talun 0.5 (0.5) 0.8 (0.8) 8.7 (7.1) 1.5 (-0.5) 3.9 (3.3) Throughfall contained higher nutrient quantities than rainwater for all stages of the talun-kebun rotation cycle, the difference between precipitation and throughfall being attributed to "rainwash". Rainwash contributed 0.5-1.1 kg ha-l of N03-N during the various stages of the talun-kebun cycle. Approximately 0.8-0.9 kg ha-l of P was returned to the forest floor/ectorganic layer in "net canopy wash" over a one year period. Most of these transfers occurred during the period of heavy rain (December-April). 146 Table 7.8 showed that 2.2 kg h a - 1 of K was added to the rainwater after hitting the cassava leaves, and 7.1-7.7 kg ha-l of K was contributed by the bamboo canopy during the fallow stage. Mg content in throughfall under bamboo (during the early fallow and mature talun stages) was nearly seven times the value in precipitation. There was negligible leaching of Ca throughout the cycle, except for cassava stage. 7.5.4. N U T R I E N T U P T A K E B Y P L A N T S Uptake of nutrients is equal the sum of the nutrients in litterfall and canopy wash and the net change in plant nutrient content (Bruijnzel, 1983; Tsutsumi et al, 1983; Kimmins, 1987). Figure 7.7 shows the nutrient uptake (the contribution of fine root production and turnover was not included in the estimates) by crops and bamboo at various stages of the talun-kebun system. The first year crops have a much higher N, P, and K uptake in comparison to cassava and bamboo. Hyacinth beans dominate nutrient uptake of the first year crops. The value for N excludes N fixation by the hyacinth beans. Bamboo shows a low P and Mg uptake during its early stage, but the uptake increases with increasing culm age. The uptake of nutrients by weeds is highest during the first year cropping stage, but decreases toward the mature talun stage. The uptake patterns reflect species differences. Table 7.9 summarizes the annual uptake of nutrients at various stages of the talun-kebun rotation. Uptake from different soil layers was assumed to be in proportion to the vertical distribution of fine roots. Therefore, the quantity of fine roots in the forest floor (root mats) and in mineral soil, as compared to total fine roots, was used to aportion uptake between forest floor and mineral soil in the mature bamboo stages. Uptake during the early fallow and the cropping stages was assumed to be directly 147 from the 0-25 cm depth of mineral soil. 7.5.5. N U T R I E N T S LN F E R T I L I Z E R S In contrast to conventional shifting cultivation, fertilizer was used intensively during the first year of cropping. Therefore, nutrient inputs in fertilizer have to be considered in the biogeochemical cycle ofthe system. Fertilizer was applied in the form of an ash and manure mixture, plus some commercial N P K and urea. Nutrient content in the fertilizer was measured by analysing the N, P, K, Ca, and Mg concentrations in ash-manure mixture using the methods described in Section 7.3.3, and by using the known concentration of N, P, and K in synthetic fertilizer and multiplying the results with the amount of each type of fertilizer used. The ash and manure mixture contributed 95.5, 10.4, 89.5, 28.9, and 51.1 kg ha-l of N, P, K, Ca, and Mg, respectively. Approximately 83, 12, and 12 kg ha-l of N, P, and K were added to the soil by the application of commercial N P K and urea. 7.5.6. N I T R O G E N F I X A T I O N Nitrogen fixation is considered as an important input to the talun-kebun system because of the use of leguminous species in the system (hyacinth beans in the first year, and some Albizia trees). Dennis (1977) estimated fixation of 14-19 kg N ha-l yr-l by hyacinth beans grown on unlimed nitrosol in Ghana. He found that the estimate increased to 31-33 kg ha-l yr-l 0 n limed soil. Considering the use of fertilizer during the first year of cropping, the mean value of the higher range (32 kg ha-l yr-l) was used as an estimate of nitrogen fixation by hyacinth beans in the talun-kebun. There has been no published measurement of nitrogen fixation rates by leguminous tree species, such as Albizia which is commonly found scattered between the bamboo clumps. Nair (1982) and Roskoski (1986, personal 148 communication) estimated values of approximately 150 kg N ha" 1 yr-l for leguminous trees grown in pure stands. Since Albizia in the talun-kebun grows in mixture with bamboo clumps and the number per.hectare is low (< 100 trees ha-1), an estimate of 10% of the above value (15 kg N ha-l yr-l) seems to be reasonable as an estimate of its nitrogen fixation rates. Table 7.9. The annual uptake of nutrients in various stages of the talun-kebun rotation cycle. Stages N Nutrient uptak P e (kg ha^yr-l) K Ca Mg First year cropping 418.8 23.7 164.3 90.7 57.9 Second year cropping 193.7 23.0 147.7 48.1 49.0 Early fallow 128.2 37.1 152.5 48.3 57.7 Mature talun 91.0 24.7 127.7 38.5 48.6 Nitrogen Phosphorus Potassium Calcium Magnesium 500 D J= CP JC o o 300 o. 3 O 3 C C < 100 • 50 30 10 2 0 0 r 120 40 I E I I I I I f f i ii m iz 100 60 . 20 100 60 40 • 4 I H E 1 I I 1 LLL1I c r ° P [ [ weeds bamboo I first year cropping II second year cropping HI early fallow J2 mature talun Figure 7. The uptake of the major macronutrienta by crops, bamboo and weeds at various stages of the talun-kebun cycle 150 7.5.7. NUTRIENT LOSSES LN LEACHING AND SURFACE RUNOFF Studies of nutrient leaching in various shifting cultivation systems have shown that there is an increase in leaching losses following clear-cutting (Toky & Ramakhrisnan, 1981; Mishra & Ramakhrisnan, 1983; Jordan, 1987). Since shifting cultivation is usually practiced on steep slopes, clear-cutting followed by burning and cropping also increases nutrient losses in run-off water (Toky & Ramakhrisnan, 1981; Mishra & Ramakhrisnan, 1983). The leaching pattern in the talun-kebun is predicted to be similar to that of the shifting cultivation. 1. METHODS Six ceramic suction cup soil water samplers (4.8 cm diameter, 2 bar, Model 1900 from Soil Moisture Equipment Company) were installed in each of TK-1 and TK-2 at two depths, 25 and 75 cm (three replicates for each depth). The samplers were inserted into holes excavated using a 5 cm diameter insertion tool (Soil Moisture Equipment Company) nearby the location for throughfall collection. Each hole was backfilled with its own soil, and then tamped. Because of the limited number that were available, only four samplers were installed at each of 25 and 75 cm depths in each of TK-3 and TK-4. Water samples were collected in acid-washed plastic bottles on average every 15 days over a 5 month-period, ending in April 1984. The samples were chemically analyzed using the same methods as those for precipitation and throughfall. To estimate the volume of soil water flux, a water balance was calculated as: D = P - (R + E + S) where D= drainage P= precipitation R= runoff E= evapotranspiration S= change in soil moisture content between the sampling date Assumptions were as follows: 1. no change in soil moisture content between 151 December and April, 2. runoff equals to zero, 3. evapotranspiration was estimated to be 50 percent of the total precipitation in all stages, because both crops and bamboo achieved canopy development during the measurement period (Chang, 1968; Wiersum, 1979). This, however, was considered as a conservative assumption 2. R E S U L T S A N D DISCUSSION Soil solution flow for the dry months was assumed to be zero as little precipitation was recorded for this period. Table 7.10 summarizes the amount of nutrient losses in leaching under various talun-kebun stages. From these data it is apparent that soil leaching was a relatively unimportant mechanism on my study plots. Table 7.10. The leaching losses of nutrients, in various stages of the talun-kebun rotation cycle. Nutrient losses (kg ha - 1 yr-l) Stages N P K Ca Mg First year of cropping 0.3 0.3 2.8 7.7 6.1 Second year of cropping 0.2 0.2 3.7 6.8 5.9 Early fallow 0.04 0.2 1.8 3.9 2.7 Mature talun 0.04 0.04 0.4 2.0 1.2 Toky and Ramakhrisnan (1981) studied the leaching losses from the shifting cultivation fields in Northern India. They estimated leaching losses of 1.1, 0.02, 0.5, 2.7, and 0.9 kg ha-l yr- 1 during the early fallow, and of 0.5, 0.01, 0.2, 1.5, and 0.5 kg ha-lyr-l in a mature fallow for N, P, K, Ca, and Mg, respectively. These low values supported my findings that soil leaching was considered as unimportant mechanism, particularly in the fallow stages. The leaching losses during the 152 cropping stage were 9.2, 0.07, 13.7, 4.6, and 2.3, for N, P, K, Ca, and Mg in the shifting cultivation area, respectively, in comparison to 0.3, 0.3, 2.8, 7.7, and 6.1 kg ha-l yr-l as reported from my study area. 7.5.8. T H E L O S S O F N U T R I E N T S F R O M T H E S L A S H (BAMBOO L E A V E S A N D F O R E S T F L O O R ) D U R I N G T H E B U R N The quantities (kg ha-l) of nutrients lost during the burn were calculated by substracting the nutrient contents in slash Qitter and bamboo foliage) before the burn and those in the ash after the burn. Based on this method, approximately 84.5, 1, 4.5, 3.4, and 1.9 kg ha-l of N, P, K, Ca, and Mg were lost during the burn. These amounts represented 69, 11, 8, 13, and 3%, respectively, of the total amounts of these nutrients in the forest floor and fresh bamboo slash that was piled during the litter raking (surface litter + two-third of the fragmented litter) after the clearcut. The ash from the burn was mixed with manure (goat dung) and applied to the soil during the first year cropping. The amount of nutrients in this mixture was discussed in part 7.5.5 above. 7.6. T H E O V E R A L L B I O G E O C H E M I C A L C H A R A C T E R I S T I C S O F T H E BAMBOO TALUN-KEBUN Previous parts of the chapter have established the pattern of accumulation and removals of N, P, K, Ca, and Mg in and from live plant biomass at various stages of the talun-kebun rotation cycle (Part 3), quantified the N, P, K, Ca, and Mg contents in the forest floor/ectorganic layer and in the mineral soils (Part 4), and described the dynamics and inputs/outputs of those nutrients over a rotation cycle (Part 5). For a complete understanding of the biogeochemistry of the talun-kebun cycle, all of these data must be combined into an overall model, which is the focus of this part of the chapter. The conceptual model of the overall biogeochemical characteristics of a talun-kebun system is presented in Figure 7.8. The system is shown as a series of major 153 compartments (double boxes), minor compartments (single boxes), transfer pathways (lines and diamonds joining the boxes), and inputs/outputs (circles). The biogeochemical processes that occurred during the bamboo talun-kebun rotation cycle were described in the introduction of this chapter. Table 7.11 reminds the reader of the data sources for these components and transfer processes, and presents a statement of the level of confidence (non-statistical) concerning the accuracy of the data or estimates. The biogeochemical cycle during the mature bamboo talun is comparable to that of the late stage of the fallow in shifting cultivation cycle (Part 7.1.1), and the processes that occur during the clearing of bamboo and the burning of litter and harvest residue are basically similar to the clearing and burning stage of shifting cultivation (Part 7.1.2). Except for the use of chemical fertilizers as additional inputs during the cropping stage of the talun-kebun, most of the processes during this stage were described in Part 7.1.3. The recovery stage of fallow vegetation in shifting cultivation (discussed in Part 7.1.4) is similar to the processes that take place during the early fallow stage of the talun-kebun. Although the bamboo talun-kebun is not as closed a biogeochemical system as the tropical rain forest it replaces, it was hypothesized at the start of this study that losses of nutrients from the system would be balanced by nutrient inputs to the system; as a consequence, it was hypothesized that the system is quite sustainable. Figures 7.9-7.13 present calibrated flow-chart of the biogeochemistry of N, P, K, Ca, and Mg in five different stages that make up a complete rotation cycle of a bamboo talun-kebun: the mature bamboo talun, clearing of bamboo and burning, first year of cropping, second year of cropping, and the early fallow stage. An analysis of the sustainability of production if the traditional talun-kebun were to be changed to provide a different mix of products is presented in Chapter 9. 154 Table 7.11. Sources of data for the conceptual model of the biogeochemistry of the talun-kebun system shown in Figure 7.8. Qualitative estimates of the accuracy ofthe data are given. COMPARTMENTS DATA CONFIDENCE IN DATA / ESTIMATE Vegetation - measurements of the contents of N, P, K, Ca, and Mg in live plant biomass of various talun-kebun stages (part 7.3) Forest floor/ - measurements of forest floor/ectorganic layer Ectorganic N, P, K, Ca, and Mg at various talun-kebun layer stages (part 7.4) Mineral soil - measurements of total N, available P, K, Ca. and Mg in the upper 25 cm of soil in various stages of the talun-kebun cycle (part 7.4) Harvest residue - measurements of the contents of N, P. K, Ca, and Mg in plant residues after each harvest (part 7.3) Ash - measurements of the contents of N, P, K. Ca. and Mg in ash from burning slash (litter and bamboo foliage) during the clearcutting of a mature bamboo talun high high high moderate moderate PROCESSES DATA CONFIDENCE Inputs Precipitation Biological fixation Fertilization Runoff erosion Mineral weath-ering • measurements of the amount of precipitation in the study area during the rainy period and a year precipitation data from the local weather station • measured concentrations of nutrients (N, P, K, Ca, and Mg) in precipitation during the rainy season • average concentrations of nutrients (N, P, K. Ca, and Mg) during the rainy season were used as average values for other months (part 7.5.3) • estimate N-fixation by Albizia was estimated from the fixation rates by leguminous trees (Nair, 1982) multiplied by 10% (considering a very low number of Albizia trees, i.e. < 100 trees per hectare) • N-fixation by hyacinth bean was taken from Dennis (1977) - measurements of the amount of nutrients in the ash and manure mixture, and the amount in commercial fertilizer (part 7.5.5) . estimate: assumed to be zero because overland flow hardly everoccurs in mature talun and the clearing site is surrounded by talun. No sign of soil movement was observed in any of the plots in any of the stages - no data. Assumed to be very slow over the time scale covered in this study moderate (might underesti-mate the values of nutrient con-tents during the dry period) low high moderate low 155 PROCESSES DATA CONFIDENCE Outputs Market • measurements of the amount of the contents of N, P, K, Ca, and Mg in the harvested products Residue output Output to the atmosphere Runoff and erosion - measurement of the amount of the contents of N, P, K. Ca, and Mg in the harvest residues that were taken away from the fields - measurements of the amount of nutrients in the slash before the burn substracted by the amount of nutrients in the ash after the bum - estimate: assumed to be zero because observation showed no sign of soil movement high high moderate Leaching losses - measurements of the nutrient concentrations in soil flux below 25 cm plus estimates of volume of drainage water moderate to low Trati-sfiir Foliar leaching Litterfall Plant uptake Decomposition - measurements of the amount of nutrients in precipitation substracted by the amount of nutrients in throughfall (part 7.5.3) - measurements of the amount of nutrients in litterfall at various stages of the talun-kebun (part 7.5.1) • summation of nutrients in litterfall. foliar leachine and the net change in plant nutrient contents (part 7.5.4) - estimated at 5% for bamboo (Dali, 1980) to low moderate high moderate to high moderate Confidence in the precision of the data defined on the basis of coefficient of variation of data presented in earlier chapters. CV < 20 Confidence high CV 20-40 Confidence moderate CV > 40 Confidence low Confidence in the accuracy of the data or estimate based on considerations of the adequacy of the methods used to produce the data or estimate. 156 The accumulation and removals of biomass, and the inventory of five major nutrients (N, P, K, Ca, and Mg) in plants, litterfall, forest floor, and in mineral soils were quantified at various stages of the talun-kebun cycle in Selaawi hamlet, Sukajadi village, Soreang district, West Java, Indonesia to obtain the data needed to explain the biogeochemistry of the system. Nutrient transfers in various input and output pathways were also measured or sought from the literature. The following conclusions are made based on the results of the data presented in this chapter: 1. The total biomass accumulation over a six year talun-kebun rotation cycle was 96.5 t ha-l, consisting of 59.6 t ha-l of aboveground biomass and 36.9 t ha-l of belowground biomass. 2. The total removal of biomass during a six year talun-kebun rotation cycle was 62.8 t ha-l, of which 17% was removed during the first year cropping stage, 11% was removed during the second year cropping stage, and about 72% was removed during the clearance of the mature bamboo talun at the end of year 6. 3. The accumulation of five major nutrients in live plant biomass was 0.82, 0.49, 0.51, and 1.06 t ha-l during the first and second year of cropping, and in the early and mature fallow, respectively. 4. The overall nutrient removals accounted for approximately 53% of N, 43% of P, 43% of K, 46% of Ca, and 42% of Mg accumulated in the live plant biomass over the six year talun-kebun rotation cycle, in which net removals represented 77 to 85% of the total removals (because of the returns of compost). 6. The accumulation of N, P, K, Ca, and Mg in the forest floor reached its maximum at the end of the mature talun (72 months after the previous 157 clearcutting). 7. The total N in the mineral soil was much higher than the N content in the plants and in the forest floor. 8. The available P in the surface 5 cm of mineral soil under the mature talun increased slightly after clearing and hoeing, but decreased to 92 and 75% of its original value in the first and second year cropping stages. 9. Cassava cropping decreased the content of the exchangeable K in the mineral soil. 10. The contents of the five major nutrients in the litterfall of crops (hyacinth beans and cassava) was in the order N > Ca > K > Mg > P, while for bamboo was in the order N > K > Mg > Ca > P. 11. The first year crops, particularly hyacinth bean had a greater uptake of N, K, and Ca in comparison to cassava and bamboo. 12. Fertilizer was an important input during the first year of cropping. It accounted for 64, 140, 93, 111, and 166 % of N, P, K, Ca, and Mg output, respectively, during this period. 13. Approximately 69,11, 8, 13, and 3% of N, P, K, Ca, and Mg, respectively, in the slash (forest floor and bamboo leaves) were lost during and after the burn. 14. Leaching losses were very small in comparison to losses in harvest removals. Figure 7.8. Flowchart of compart mental and processes i n a conceptual model of the biogeochemistry of the Talun-Jcebun Agroforestry system of West Java, Indonesia Mature talun Clearing, burning,and hoeing First year cropping 245.9 4.6 O 105.6 6642 2 0 . Second year cropping Early fallow 46.8 27.6 176.3 1 1< 15x 5 IS Figure 7.9. Flowchart of major compartments and transfer processes of Nitrogen i n the five stages of the Talun-kebun Agroforestry system Mature talun Clearing, burning, and hoeing Fi r s t year cropping Figure 7.10 Flowchart of major compartments and transfer processes of Phosphorus i n the five stages of the ralun-Jfcebun Agroforestry system Figure 7.11 Flowchart of major compartments and transfer processes of Potassium i n the five stages the Talun-kebun Agroforestry system Mature talun Clearing, burning, and hoeing F i r s t year cropping °1 12 3 h L - O - O Second year cropping Early fallow Figure 7.13 Flowchart of major compartments and transfer processes of Magnesium i n the five stages the Talun-kebun Agroforestry system Mature talun Clearing, burning, and hoeing F i r s t year cropping Figure 7.12 Flowchart of major compartments and transfer processes of Calcium i n the five stages of the Talun-kebun Agroforestry system 164 CHAPTER 8 ANALYSIS OF SOME ASPECTS OF THE MANAGEMENT OF THE TALUN-KEBUN SYSTEM 8.1.INTRODUCTION Recent changes in socio-economic conditions (such as increasing population pressure, increasing demands for cash crops, declining availability of labor, and easy access to fertilizer inputs) have had a considerable influence on the talun-kebun system. In West Java, there has been a tendency on the larger farms to shorten the fallow period and to replace bamboo with cash crops in order to intensify the use of the land and to increase the production per unit area. However, this trend has not yet been adopted on the majority of the small farms because of the lack of resources and capital inputs. Most traditional farmers are not commercially oriented. They value security and stability more than profit, and they have been unwilling to take unnecessary risks, such as cash investment and new cropping systems, unless they are convinced that the new changes will be successful and will improve their income (Steiner,1984; Marten, 1986). Their traditional cropping practice has been well-adapted to the socio-economic situation of smallholdings and has been passed to them by their parents and grandparents. They are very cautious and prudent in adopting any new innovations. Therefore, any efforts to develop or to improve traditional practice must be made in collaboration with the farmers, and should be based on a sound understanding of the existing systems. 8.2.0BJECTIVES The objectives of this part of the study were: 1, to determine the motivation of farmers to practice the talun-kebun cycle and to select particular crop species; 2, to study the use of labor during the cycle; 3, to quantify the energy inputs to and outputs from the system; 4, to study the economic returns; 5, to consider pest and 165 conservation aspects of the system; and 6, to predict the future trends in the development of the talun-kebun cycle. The findings are used to describe and evaluate the efficiency of the traditional system and recent modifications thereof. 8.3.STUDY SITES The survey area covers two villages: Sadu and Sukajadi, as described in Chapter 3. 8.4JVCETHODS The data were collected by interviewing farmers. The majority of these farmers had not completed elementary school and none of them kept written records of their farming activities or financial transactions. Thus, the data gathered in this survey were based on the farmers' memory and on direct observations of the farmers' fields. Data were collected from farmers in one or both of two types of interview. General interviews were given to 40 randomly selected households which own talun-kebun fields in the two villages. Detailed interviews were conducted with 20 of these households. All of the latter were actively involved in cropping during the period of the study. The questionnaires for both types of interview are presented in Appendix 7. The interview included questions on the following topics: General interview - household background; - land tenure and land use; - cropping pattern; - labor allocation; - pest management and soil conservation; - the advantages and disadvantages of the practice; 166 - attitude toward "improved" practices. Detailed interview - cropping activities; - labor use and time allocation; - labor costs; - material expenditures; - cash income; - marketing; - future expectations. In order to obtain accurate data on time and energy expenditures, these parameters were quantified in the field in addition to interviewing the farmers. The amount of work and labor used during the clearing of my experimental area TK-1 and the cropping in TK-2 and TK-3 fields were documented and assumed to be a representative sample of these kinds of activity. All data were calculated on a per hectare basis. 8.5 .RESULTS A N D D I S C U S S I O N 1 . H O U S E H O L D B A C K G R O U N D Approximately 17.5% of the heads of the households never attended school and were illiterate, 62.5% had not finished elementary school (only attended school until grade 2, grade 3, or grade 4), and only 20% had graduated from elementary school. Of the housewives, 30% were illiterate, about 55% were literate but had not completed elementary school, while the rest (15%) had completed 6 year elementary school. Farming has been the main occupation for the interviewees. Approximately 50% of them lived on farming alone, 42.5% lived on farming and market trading, and 7.5% lived on farming and working as hired labor. Family size ranged from two 167 to nine; this resulted in a total of 172 persons in 40 households. Of this total, 60% belonged to the working age classes (10-60 years old), 5% were over 60 years old, and the rest were under 10 years old. 2. CROPPING PATTERNS a Figure 8.1. illustrates the common cropping patterns during the kebun stage of a. talun-kebun cycle in Soreang, West Java. The major crops grown during the first year cropping stage were hyacinth beans (Dolichos lablab), cucumber (Cucumis sativus), and bitter solanum (Solanum nigrum). Chili pepper (Capsicum frutescens), and eggplant (Solanum melongena) were sometimes planted as additional crops. If the rainy season was long enough, beans (Phaseolus vulgaris) and cowpea (Vigna sinensis) were frequently used as short-season "catch crops" (crops that use residual moisture and fertility) after the harvest of hyacinth bean. Cassava was then grown as the sole crop during the second year of cropping. 3. FARMERS' MOTrVATION The talun-kebun system has been practised in the area since the early 1900s. As has already been described in Chapter 2, the system was developed from conventional shifting cultivation. There were various reasons that motivated farmers to continue practising the talun-kebun rotation: 1. It has been proved to be beneficial. This seemed to be the primary motive for most farmers to continue to practise the talun-kebun rotation system (accounted for 42.5% of the interviewees). a Although it has been described in Chapter 2, the cropping patterns are described again in detail for the socio-economic and energy analysis purposes. 168 First Year I Second Year A S 0 N D J F M A M J J A S 0 N D J F M A M J J / ^ / / c u / 3 7 . 5 % BS 7 / CA / n / HB / 20.0 % nr HB / / ,cu / / BS / / CP / / EP / 25.0 % CA HSI j HB / / c u / 17.5 % / CP / ZiiZ HB= hyacinth beans CU - cucumber BS- bitter solanum CA-cassava BE= beans C0= cowpeas CP= chili pepper EP= eggplant Figure 8.1 Five common cropping systems for the kebun stage at a talun-kebun cycle in Soreang, West Java, Indonesia (thickened lines indicate major crops). The percentage of the 40 farms investigated using each of these systems is shown. 169 2. Tradition. Approximately 27.5% of the interviewees chose tradition as their motive for continuing the talun-kebun practice. Steiner (1984) argued that tradition is the easiest answer, and that it is cited mainly by those farmers who are not able to express themselves well enough to identify the more fundamental reasons behind their attitude toward certain practices. 3. Good for soil fertility. About 25% of the interviewees believed that the rotation between two years of cropping and four years of bamboo fallow was good for maintaining the fertility of soils in the talun-kebun fields. 4. Lack of capital. This reason was given by 20% of the interviewees as their motive for continuing their current practice. They were aware of the higher economic return if they replace bamboo with cash crops, but they lacked the necessary capital to undertake the change. 5. Lack of knowledge of innovations. Some farmers (5% of the interviewees) felt that they lacked adequate knowledge about new technology or new crops, and therefore it would be safer to continue practising their existing agricultural system than to replace it with the new one. Regarding their decision to employ multicropping during the first year cropping stage, the majority of farmers stated tradition as their primary motive. Again, this answer seemed to be the simplest answer that might occur to them. On the basis of his study, Steiner (1984) assumed that farmers were convinced that their practice was rational under prevailing conditions. Higher output per unit land area was another motivation that was given frequently by farmers as their primary 170 motive for intercropping. 4.LABOR USE IN THE TALUN-KEBUN SYSTEM A.WORK ACTrVTriES AND THE DIVISION OF LABOR BETWEEN M E N & WOMEN Most of the labor requirement of the talun-kebun system is concentrated at the end of the rotation cycle just prior to cropping and during the first year cropping stage. The labor requirement can be categorized according to four activities: clearing, preparation for cropping, cropping, and harvesting. Clearing normally started before the rainy season (between August and October). The first activity was clearing the undergrowth, cutting the small bamboo branches, and raking the bamboo litter, which was usually done by both men and women. Soil hoeing, which was done after the clearing and raking was completed, was considered as very demanding work and was usually done by men. The mature bamboo culms were then clearcut, while the most recently grown culms were left uncut until they became dry by exposure to the sun (for about 3-4 weeks) and were then cut during the preparation for cropping. The mature culms were removed from the field (sold to middlemen who came from the district), while the residue (leaves, sheaths, and small branches) was piled, dried, and burned. Data from the interviews and from the observations during the clearing of TK-1 showed that 90% of the total labor time used during the clearing stage was done by men, the other 10% being accounted for by women who participated in clearing the undergrowth and raking the bamboo litter. The preparation for cropping began with the cutting of the recently grown bamboo culms (which by this time have dried), and then setting them in rows for use as supporting poles for hyacinth beans. As a security measure, and to show a clear demarcation with the neighboring fields, bamboo fences were built around the field. All of these activities were done by men. Women participated in clearing the 171 remaining slash and leaf litter from the field after all the bamboo poles were set, and then men loosened the soil using small hoes. Male labor constituted about 80% of the total labor time used during the preparation for planting. The cropping activity commenced with the planting of hyacinth bean seeds. Holes were made at the base of the bamboo poles, and shallow furrows were made between the rows for cucumber and bitter solanum or other vegetables. Seed planting was mostly done by women, the other activities being done by men, including the transplanting of bitter solanum or other vegetables. Depending on the species planted, fertilization was done (by either men or women) two or three times during each cropping period. Weeding was done two months after planting the hyacinth beans, and the soil was heaped up around the plants. Women did most of the weeding, while men heaped up the soil. Women contributed 48% of the total labor time during the cropping in comparison to 52% contributed by men. Both men and women participated in harvesting the crops. The first harvest started with cucumber, about forty days after planting. Harvesting continued every three to five days (up to 11 times) until the plants died. Cucumber harvest did not require much labor and was usually done by women. Bitter solanum was harvested weekly, starting about two months after transplanting. Depending on the availability of labor, men sometimes participated in the harvest of bitter solanum. A complete weeding was done by both men and women to clear paths between rows before harvesting the hyacinth beans. Bean harvest was done by picking the pods off the vines after pulling down the bamboo poles. Men pulled down the poles and women picked up the pods. The bamboo poles were then cut up by men and sold as firewood. The beans were peeled and the seeds were sold in the market, while the empty pods were thrown away. Women contributed about 73% of the labor time needed for harvesting. If there was enough rain, the soil was hoed again and used to grow catch 172 crops (beans and/or cowpeas). Men did the hoeing and setting of small poles for the crops, while women participated in planting and harvesting. The land was then left idle until the end of the dry season (September or October), and then it was hoed again in preparation for the second year crop (cassava). All the field work during the second year of cropping (i.e., hoeing, planting of cassava cuttings, and harvesting the tubers) was done by men. No fertilization or weeding was done during the cassava cropping. Cassava was sold as raw tubers without further processing. After the cassava harvest the fields were usually abandoned for three to four years until the bamboo culms were ready to be harvested again. Almost no labor input was needed during the fallow (bamboo talun) stage. Therefore, farmers usually maintained several fields in different stages of the cycle in order to produce a crop of each of the food species annually. Although they may already belong to the working age group, children up to the age of 12 or 13 years rarely engaged in the talun-kebun activity because the work was considered too heavy for them and the location of the field was often quite far from their house. Moreover, most of their time was occupied by school work. Data from the interviews and the direct recording of field activities showed that non-family labor was used for clearing and most of the preparation for cropping (96% of the total labor needed during this stage). Cropping was done by both family labor (70%) and non-family labor (30%). Harvesting of cucumber and bitter solanum were usually done by family labor, but the harvest of hyacinth beans required the use of non-family labor (50%). All activities during the second year of cropping (cassava cropping) were done by family labor. Table 8.1. illustrates work activities related to clearing a mature talun and the subsequent two year cropping period. 173 Table 8.1. illustrates work activities related to clearing a mature talun and the subsequent two year cropping period. Table 8.1. Work activities during clearing and cropping of a talun-kebun in Soreang, West Java, Indonesia. LCLEARING 1. Nyacan cutting undergrowth and small bamboo branches 2. Ngeduk: raking the bamboo litter into piles 3. Mencug: hoeing the soil 4. Nuar : clearcutting the bamboo culms 5. Ngaduruk: burning bamboo slash and litter piles 6. Ngangkut lebu: collecting ash from the burn piles and storing it in a hut II. PREPARATION FOR CROPPING 1. Nanceb tuturus : cutting the bamboo culms reserved for poles and setting them in rows 2. Mager: building fences 3. Ngararad : clearing the remaining slash and leaf litter from the field 4. Ngeprak and ngipuk : loosening the soil using a small hoe and preparing seedbeds for vegetables (i.e., bitter solanum) m. CROPPING 1. Melak roay : planting hyacinth beans seeds 2. Mupuk roay : fertilizing the beans by covering the seeds with ash and compost mixture 3. Malintang : mounding the soil between the rows to make shallow furrows for growing cucumber and other vegetables 4. Mupuk bonteng : fertilizing the cucumber by covering the seeds with ash and compost mixture 5. Melak bonteng : planting cucumber seeds 6. Mupuk roay kadua kali : fertilizing the beans for the second time using N-P-K fertilizer and then covering it with ash and compost mixture 7. Mupuk bonteng kadua kali : fertilizing the cucumber for the second time using N-P-K fertilizer and then covering it with ash and compost mixture 8. Melak leunca : transplanting bitter solanum from the seedbeds to the furrows, alternated with cucumber plants 9. Ngoyoslnyaeur : loosening the soil and heaping it up around each plant 10. Mupuk urea : fertilizing the beans with urea 11. Ngored: weeding IV. HARVESTING 1. Panen bonteng : harvesting the cucumber 2. Panen leunca : harvesting the bitter solanum 3.' Ngored : weeding to clear paths between rows before harvesting the hyacinth beans 4. Panen roay : harvesting the hyacinth bean 5. Ngabereskeun tuturus : cutting up the poles to be sold as fuelwood 6. Ngupas roay : peeling the hyacinth beans to collect the seeds for sale V. CASSAVA CROPPING 1. Mencug : hoeing the soil in preparation for cropping 2. Melak : planting cassava cuttings (taken from other fields) 3. Panen : harvesting cassava tubers by pulling out the entire plants 175 B. TIME SPENT AND WORK DISTRIBUTION THROUGHOUT THE YEAR The common daily work period in the talun-kebun was 5-6 hours per person per hectare, but certain activities such as crop planting and fertilization only took 1-2 hours. Each harvest of cucumber and bitter solanum required 1 hour and 1.5 hours, respectively. Data from the interviews showed that the total time spent by male labor during clearing and the two years of cropping ranged between 2,500-3,000 hours per hectare. About 25% of this was spent during the clearing, 45% during the first year cropping stage, and 30% during the second year cropping stage. Female labor accounted for approximately 700-1,000 hours per hectare during the clearing and first year of cropping. Only 7% of the hours they contributed was devoted to clearing, and 15% for the preparation for cropping. The rest (78%) was spent for the actual cropping and harvesting. They did not contribute any labor to the field work during the second year of cropping. The above estimates of total labor time spent by both males and females did not include the time spent for walking to and from the field and for marketing the products. Non-family laborers usually lived in housing compounds near the fields (about 5-10 minutes walking distance); however, the owner and other family members were often from different hamlets and they had to walk about 30-45 minutes each way to and from the field. Travel time during the clearing and the two year cropping stage ranged from 250-300 hours for male family laborers, in comparison with only 70-90 hours for male non-family laborers. Female family laborers spent 310-370 hours and female non-family laborers spent 10-20 hours in travel during the clearing and the first year cropping stage. The time spent walking is used in the calculation of energy expenditures later in this chapter. The time spent transporting products to the market was not included in the 176 calculation of the total time spent during the talun-kebun cycle. Norman (1979) suggested that if the purpose of the analysis is to compare the efficiency of the talun-kebun system with other cropping systems, estimates of input should be limited to net inputs related to crop production, and that marketing should be excluded from the calculation. Figures 8.2. and 8.3. summarizes the percentage time spent on each activity, from clearing to the harvesting of the first and second year crops. September, when the clearing took place, was the peak of activity for male laborers. Hoeing was the most time-demanding activity. The field was then left for 2-3 weeks to let the soil and remaining bamboo culms dry, with the result that there was only 25% as much activity in October as in September. Female labor did not contribute to any activity in October. Both sexes invested a considerable proportion of the total time in November, and then the activity declined again in December as the planting was completed and the farmers waited for the harvest. Female laborers were more active than male laborers during the January to March harvest period. There was no activity in the field during April. Bean harvest took place in May which also determined the end of the first year cropping stage. If catch crops were planted after the bean harvest, May and June would then be devoted to growing and maintaining these crops. Harvest usually took place 50 days after planting. Activities for the second year cropping stage did not begin until the following rainy season. Usually, soil hoeing was done in July or August and the planting in September or October, depending on the timing of the first rain. There was no activity thereafter until the cassava was ready for harvest, which usually took place in July of the following year, about 9-10 months after planting. 177 30 25 2 2 0 I 5 10 A. Male Labor 888$ Sep Oct Nov Dec Jan Feb Mar Apr May B. Female Labor • • • e • • * 4 Sep Oct Nov Dec Jan Feb Mar Apr May collecting ash from the burn cutting and burning soil hoeing litter raking clearing the undergrowth fencing cutting and setting bamboo poles mounding fertilizing planting loosening soil and preparing seed beds clearing the remaining slash and bamboo litter |L|L-: heaping soil harvesting weeding cutting up and binding used bamboo poles for firewood peeling beans Figure 8.2 The percentage of time spent in each activity over the first year of the talun-kebun rotation cycle 45 40 • Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul soil hoeing planting harvesting Figure 8.3 The percentage of time spent in each activity over the second year of the talun ~ kebun rotation cycle 179 5. E N E R G E T I C A S P E C T S A. H U M A N E N E R G Y E X P E N D I T U R E There are two components of energy expenditure: basal metabolic energy expenditure, which continues all the time whether the person concerned is working or not, and net energy expenditure, which is the additional energy expended in work. The total is known as gross energy expenditure (Norman, 1979). Although gross energy expenditure is commonly used in the calculation, Norman (1979) pointed out that if the purpose of the study is to compare the energy output/input ratios of different cropping systems, the appropriate denominator is net energy expenditure. On the other hand, if the purpose is to compute the total energy balance of the farm worker, the appropriate denominator for the calculation of output/input should be his/her total gross energy expenditure. As one of the objectives of this study was to compare different cropping systems, energy expenditures discussed in this chapter are the net energy expenditures. The rate of human energy expenditure varies with body weight, sex, and the task performed (Noraini, 1976 (cited in Koh, 1978); Norman, 1978). Two sets of calorie expenditure values per body mass were used as a basis for estimating the energy expended for various activities performed during the first two year cycle of the talun-kebun. The first set of values used were from the National Health Centre, Kuala Lumpur. This data set was chosen because of the physical similarity between the Indonesian and the Malaysian people. In my study, the average mass of a male laborer was 53 kg and the average mass of a female laborer was 48 kg. Therefore, the caloric expenditure values per male laborer per hour were estimated at 92 kcalh for light work, 100 kcal for moderate work, and 118 kcal for heavy work, and the values per female laborer were 71, 79, and 93 kcal per hour, respectively. These b 1 Meal = 1,000 kcal; 1MJ = 239 kcal 180 values, however, did not reflect a significant difference between the calories expended during heavy work and during light work. Therefore, a second set of values was used for comparative measurements. The second set of values was based on Uhl (1980), who classified work into light, intermediate, demanding, and very demanding. He used calorie expenditure values of 50, 100,150, and 250 kcal per 45.4 kg human body mass per hour for light, intermediate, demanding, and very demanding work, respectively. There was no differentiation of calorie expenditure values between male and female laborers. Considering that the physical labor productivity of women was lower than that of men, I used 80% male adult equivalents for female adult laborers. This number agrees with the equivalent value used by Lagemann (1977) in his study in Eastern Nigeria. Accordingly, the caloric expenditure values for female labor were estimated at 40, 80, 120, and 200 kcal per 45.4 kg human body mass per hour for light, intermediate, demanding, and very demanding work, respectively. Data from the interviews showed that a total of 1,500-2,000 hrs of male labor and 700-1,000 hrs of female labor were devoted to clearing, preparation for cropping, cropping, and harvesting activities during the first year talun-kebun cycle. Male laborers spent 240-260 hrs, while female laborers spent 330-360 hrs for travel. Based on the Malaysian classification, the work done by male laborers during the first year cropping stage (excluding travel) consisted of 85% heavy work and 15% moderate work. Approximately 25% of the work done by female labor was categorized as heavy work, 70% was considered moderate, and 5% was light work. All of these activities amount to an estimated energy expenditure (inputs) of 170-240 Meal ha-l for male labor (i.e., 115 kcal per work hour) and 55-85 Meal ha-l (i.e., 82.5 kcal per work hour) for female labor. The energy expended for travel for both male and female labor was estimated each 22-26 Meal ha-l. 181 Less time was invested during the second year cropping stage. The total time spent for hoeing, planting, and harvesting was estimated at 650-950 hrs ha-l, which corresponded to an estimated energy expenditure of 75-115 Meal ha-l. All of these activities were done by male labor. Travel during this period required approximately 14-16 Meal ha-l. When using Uhl's classification, about 30% of the work done by male labor during the first year cropping stage (excluding travel time) was classified as very demanding, 45-50% was demanding, and 20-25% was intermediate. Approximately 25-27% of the work done by female labor was considered demanding, 70-75% was intermediate , and 3-5% was light. The estimated energy expenditure during the first year cropping stage was 300-400 Meal ha-l for male labor (i.e., 198 kcal per work hour) and 65-95 Meal ha-l (i.e., 95 kcal per work hour) for female labor. Each sex spent about 15-16 Meal ha-l for travel during the first year cropping stage. 45% of the work done during the cassava cropping was very demanding, and the rest (55%) was demanding. The total energy expended during this period was 180-200 Meal ha- 1, while the energy expended for travel ranged from 9.5 to 10.5 Meal ha- 1. Results from direct observation of labor activities in the field study are presented in Table 8.2. and Table 8.3. Both tables show complete figures for the time spent, the relative difficulties, and the estimated energy expended by male and female labor in field activities undertaken during the first two year cycle of the talun-kebun. All activities, except for travel, are listed in the chronological order of their occurrence and all figures are on a per hectare basis. The observed values were within the range of values obtained from the interview. 182 Results based on Malaysian calorie expenditure figures. Table 8.2. shows that clearing and preparation for cropping were categorized as heavy work, while planting and harvesting were considered moderate. Travel without burden was regarded as light work, whereas travel with burden could be classified as moderate or heavy work, depending on the type and amount of burden being carried. Harvesting cucumber and bitter solanum was categorized as moderate work. Each harvest only required 1-1.5 hours, but because cucumber was harvested up to eleven times and bitter solanum was harvested up to seven times before the plants died, the total hours spent and the total energy expended for the two harvests were quite high (239 hours and 18.9 Meal). Bean harvest required more labor and longer hours than the other two harvests, but the work was completed in one day. The total energy expended by male and female laborers during the bean harvest was 9.6 Meal. Peeling bean pods was categorized as light work and was done at home by female labor. About 2.1 Meal ha-l was required for this activity. Male laborers only spent 8% of their total energy expenditure (excluding travel) for weeding and fertilizing, while female laborers used up 31% of theirs. Male and female laborers expended 23.6 Meal ha-l and 24.9 Meal ha-l, respectively, for travel during the first year cycle of the talun-kebun rotation. Soil hoeing accounted for approximately 45% of the total energy expended during the second year cropping stage excluding that for travel. Planting cassava required an energy expenditure value of 17.1 Meal ha-l, which was about half of the energy required for harvesting. Travel without burden was categorized as light work, while travel while carrying the cassava tubers was categorized as heavy work. Approximately 14.9 Meal was expended for travel. Figure 8.4. summarizes the human energy inputs to crop production for the two year cropping stage of the talun-kebun in the study sites. The values represent 183 the energy expenditure values of both male and female labor. Approximately 330 Meal ha-l of net energy (including travel) was expended during the first year cycle. The energy expenditure during the second year of the cycle was about 33% ofthe energy expended during the first year cycle (107.5 Meal ha-l). Therefore, a total of 437.5 Meal ha-l of net energy was expended during the two year cropping of the talun-kebun rotation cycle. Results based on Uhl's calorie expenditure figures. Work activities in Table 8.3. are based on Uhl's classification. Soil hoeing, cutting and burning were categorized as very demanding work and accounted for 45% of the total energy expended (excluded travel) by male labors during the first year cropping stage. Most of the other activities during clearing and preparation for cropping were considered as demanding job, but collecting ash and fencing were considered intermediate. Planting, fertilizing, weeding, and harvesting were grouped as intermediate work. Female labor spent 65%, while male labor only spent 9% of their total energy expenditure (excluded travel) for these activities. Travel without burden was categorized as light work and travel with burden was considered intermediate. Male labor spent 15.4 Meal ha-l and female labor spent 15.3 Meal ha- 1 in travel. Approximately 179 Meal ha- 1 of energy was spent for soil hoeing, cassava planting, and harvesting, which was almost twice the result calculated using the Malaysian values (92.6 Meal ha-1). Figure 8.5. illustrates the human energy expenditure in crop production during the two year cycle of the talun-kebun rotation in the study sites using Uhls (1980) expenditure figures. A total of 660.7 Meal ha- 1 of net energy (included travel) was expended during the two years cropping cycle, consisting of 471.6 Meal ha- 1 during the first year cycle and 189.1 Meal ha-l during the second year of the cycle. 184 Approximately 56% was expended by male labor and 15% was spent by female labor during the first year cycle, and 29% was expended by male labor during the second year of the cycle. The energy expenditure during the clearing and preparation for cropping accounted for 46% of the total value, of which 26% was spent for hoeing, cutting and burning. Planting, fertilizing, and weeding required 7% and harvesting required 6% of the total energy expenditure. Uhl (1980) reported a labor input of 606.5 Meal ha-l for two-year yuca (cassava) cropping in slash and burn agriculture at San Carlos, Venezuela. This represented the gross energy value. If the energy for basal metabolism was excluded from the calculation, the net energy value in his study was 421.4 Meal ha-l . Therefore, the net energy expended during the two year cropping of a talun-kebun in Soreang was 1.5 times higher than that spent over the two year cropping period of slash and burn agriculture in San Carlos. Moreover, about 41% of the energy expenditure in Uhl's study was spent in processing the yuca roots and only 59% was spent in field activities, while 99.8% of the energy expenditure in my study was spent in field activities. No processing activity except peeling the hyacinth beans was required after harvests. Clearing and preparation for cropping in the talun-kebun required a considerable amount of energy (46% of the total energy expenditure), while these activities only accounted for 10% of the total energy spent in shifting cultivation at San Carlos. 1 Table 8.2. The time spent, the relative difficulty, and the estimated energy expenditure in field activities during the two-year cropping of a talun-kebun cycle (Based on the Malaysian calorie expenditure values) Activity Time spent (hrs ha-l) Male labor Female labor Relative difficulty Energy expenditure (Meal ha-l) Male labor Female labor FIRST YEAR; Cutting undergrowth Sc small branches Litter raking Hoeing Cutting & burning Collecting ash Cutting & setting bamboo poles Fencing Clearing the remain-ing slash & litter Loosening the soil Sc preparing seedbeds Planting hyacinth beans First fertilization for the beans Mounding/ridging Planting cucumber First fertilization for cucumber Second fertilization for the bean Second fertilization for cucumber Transplanting bitter solanum Heaping the soil Urea fertilization First weeding Cucumber harvest (11 times) Bitter solanum harvest (7 times) Last weeding Bean harvest Cutting up and binding the bamboo poles for firewood Peeling the bean Travel SECOND YEAR: Hoeing Cassava planting Cassava harvest Travel 45 45 450 95 25 140 80 175 200 5 5 140 10 12.5 7.5 25 49 20 50 45 22.5 50 30 70 254.8 35 35 T O T A L 355 145 285 157 2,993.3 140 14 5 14 10 25 15 17.5 120 110 129 120 84 30 348.3 heavy heavy heavy heavy heavy heavy heavy heavy heavy moderate moderate heavy moderate moderate moderate moderate moderate heavy moderate moderate moderate moderate moderate moderate heavy light light, moderate, or heavy heavy heavy heavy light or heavy 1,251.8 5.3 5.3 53.1 11.2 3.0 16.5 9.4 20.7 23.6 0.5 0.5 16.5 1.0 L3 0.8 2.5 5.8 2.0 5.0 4.5 2.3 5.0 3.0 8.3 23.6 41.9 17.1 33.6 143 338.2 3.3 3.3 13.0 1.1 0.4 LI 0.8 2.0 1.2 1.6 9.5 8.7 10.2 9.5 6.6 2.1 24.9 99.3 Table 8.3. The time spent, the relative difficulty, and the estimated energy expenditure in field activities during the two-year cropping of a talun-kebun cycle (Based on Uhl's calorie expenditure values) Activity Time spent (hrs ha-l) Male labor Femaie labor FIRST Y E A R : Cutting undergrowth St small branches 45 35 litter raking 45 35 Hoeing 450 • Cutting St burning 95 -Collecting ash 25 -Cutting & setting bamboo poles 140 -Fencing 80 -Clearing the remain-ing slash St litter 175 140 Loosening the soil St preparing seedbeds 200 -Planting hyacinth beans 5 14 First fertilization for the beans 5 5 Mounding/ridging 140 -Planting cucumber - 14 First fertilization for cucumber 10 10 Second fertilization for the beans 12.5 25 Second fertilization for cucumber 7.5 15 Transplanting bitter solanum 25 -Heaping the soil 49 17. Urea fertilization 20 First weeding 50 120 Cucumber harvest (11 times) 45 110 Bitter solanum harvest (7 times) 22.5 129 Last weeding 50 120 Bean harvest 30 84 Cutting up and binding the bamboo poles for firewood 70 -Peeling the beans - 30 Travel 254.8 348. S E C O N D Y E A R : Hoeing 355 Cassava planting 145 Cassava harvest 285 Travel 157 T O T A L 2,993.3 1,251.8 Relative Energy expenditure (Meal ha-l) difficulty Male labor Female labor demanding 7.9 4.4 demanding 7.9 4.4 very demanding 131.4 -very demanding 27.7 intermediate 2.9 -demanding 24.5 intermediate 9.4 -demanding 30.6 17.8 demanding 35.0 _ intermediate 0.6 L2 intermediate 0.6 0.4 demanding 24.5 -intermediate - L2 intermediate 1.2 0.9 intermediate 1.5 2.1 intermediate 0.9 1.3 intermediate 2.9 _ demanding 8.6 2.2 intermediate 2.3 intermediate 5.9 10.2 intermediate 5.3 9.3 intermediate 2.6 11.0 intermediate 5.9 10.2 intermediate 3.5 7.1 demanding 12.3 . light - 1.3 light or intermediate 15.4 15.3 very demanding 103.7 -demanding 25.4 demanding 49.9 light or intermediate 10.1 560.4 100.3 187 r r r Total Energy Inputs 437.5 Meal clearing 8 4 . 4 Meal 19.3 % preparation for cropping 83.2 Meal 19 % planting, fertilizing, mounding and heaping 39.1 Meal 9 % weeding 2 9.0 Meal 6.6 % harvesting 35.3 Meal 8 % cutting and binding bamboo poles 8.3 Meal 1 .9% peeling beans 2.1 Meal 0.5 % travel during the first year cropping stage 46.5 Meal 11 % hoeing in preparation for cassava planting 41. 9 Meal 9.6 % cassava planting 17.1 Meal 3 .9 % cassava harvest 33.6 Meal 7.7 % travel during the second year (cassava) cropping stage 14.9 Meal 3. 4 % Figure 8 .4 The human energy inputs for each activity during the two year cropping cycle of a hectare talun-kebun (based on the Malaysian calorie expenditure figures) 188 clearing 186.6 Meal 2 8 . 3 % r r Total Energy Inputs 6 60.7 Meal preparation for cropping I 17.3 Meal 17.7 % planting , fertil izing, mounding and heaping 5 2 . 4 Meal 7.9 % weeding 32.2 Meal 4.9 % harvesting the first year crops 3 8 . 8 Meal 5.9 % cutting and binding bamboo poles 12.3 Meal 1 . 9 % peeling beans 1.3 Meal 0. 2 % travel during the f irst year cropping stage 30.7 Meal 4 .6 % hoeing in preparation for cassava planting 103. 7 Meal 15.7 % cassava planting 25.4 Meal 3.8 % cassava harvest 49.9 Meal 7.6 % travel during the second year (cassava) cropping stage 10.1 Meal 1.5 % Figure 8.5 The human energy inputs for each activity during the two year cropping cycle of a hectare talun-kebun (based on Uhl's calorie expenditure figures) 189 Comparison between the two results. Figure 8.4 and 8.5. show that clearing and preparation for cropping accounts for 46% of the total energy expenditure when calculated based on Uhl's figures, but only accounts for 38% of the total energy expenditure when calculated based on the Malaysian figures. On the other hand, travel during the first year cropping constitutes 11% of the total energy expenditure when calculated based on the Malaysian figures, but only accounts for 5% of the total energy expenditure based on Uhl's figures. This activity during the second year of cropping comprises 3% vs 2% of the total energy expenditure, using the Malaysian and Uhls data, respectively. Uhl's calorie expenditure figures seemed to be more appropriate than the Malaysian values, because the former clearly reflected the relative difficulty of each activity. For example, Uhl's calorie expenditure value for intermediate work doubled the value for light work, and the calorie expended for demanding and very demanding work was 1.5 and 2.5 times that for intermediate work, respectively, whereas the Malaysian calorie expenditure figures only showed a slight difference between each category (see section 8.5.5.A). Table 8.2 and 8.3 show that most of the activities done by male labor are categorized as demanding to very demanding or heavy work. The total energy expended by male labor during the two year cropping cycle calculated based on Uhl's values was 1.65 times higher than that calculated based on the Malaysian values. Since most of the activities done by female labor were categorized as light and intermediate or moderate work, there was only a slight difference in energy expenditure calculated using the two different energy data sets (i.e. 100.4 Meal ha-l based on Uhl's values and 99.2 Meal ha-l based on the Malaysian values). 190 B. E N E R G Y OUTPUTS Energy outputs from the system consisted of food energy from crop harvests, the energy content of cassava leaves that were used as fodder, fuel energy from bamboo branches that were collected during the clearing and from immature bamboo poles that were cut up after the bean harvest, and the energy stored in bamboo culms that were cut during the clearing. According to the farmers, approximately 7.5-8 t ha-l of cucumber, 3.5-41 ha-l of bitter solanum, and 10-12 t ha-l of hyacinth bean could be harvested during the first year of cropping. Cassava production ranged from 10-15 t ha-l. The mass of cassava leaves used for fodder was estimated by applying the ratio of foliage biomass to tuber biomass measured on the study sites to tuber production values given by the farmers. The estimated mass of foliage was 1.8-2.7 t ha-l. Farmers did not have any record of the mass of bamboo branches and bamboo culms that were used for fuelwood or building material; therefore, I used the biomass data from my study sites to estimate the production of these harvested materials. Table 8.4. summarizes the energy outputs of the two year cropping stages of a talun-kebun in the study sites. 191 Table 8.4. Energy outputs in a two year cropping stage of the talun-kebun system. Items Production Energy content (kg ha-l) (Meal ha-l)* Food: Cucumber 7,600 . 912 Bitter solanum 3,620 1,134.9 Hyacinth beans 11,450 9,732.5 Cassava 12,225 13,386.4 T o t a l 25,165,8 Fodder: Cassava leaves 2,200 1,397,2 T O T A L 26,563 * Based on the Directorate of Nutrition, Department of Health, the Republic of Indonesia: Cucumber, 1 kg= 120 kcal (moisture contents 94%, edible portions 100%) Bitter solanum, 1 kg=330 kcal (moisture contents 90%, edible portions 95%) Hyacinth bean, 1 kg= 1,250 kcal (moisture contents 67%, edible portions 68%) Cassava tubers, 1 kgs 1,460 kcal (moisture contents 62.5%, edible portions 75%) Cassava leaves, 1 kgs 730 kcal (moisture contents 77%, edible portions 87%) Rappaport (1971) estimated a total energy output of 12,989.3 Meal ha-l yr-l for taro-yam gardens cultivated by the Tsembaga tribe in the highlands of New Guinea. The total energy output of food crops over a two-year period obtained in this study was 25,165.8 Meal ha-l, consisting of 11,779.4 Meal ha-l from the first year crops and 13,386.4 Meal ha-l from the second year crop (cassava). The annual outputs in the present study were comparable to Rappaport's rinding. Uhl (1980) reported a total energy output of 8,406.3 Meal ha-l over two consecutive years of cassava planting, in which 5,247.7 Meal ha-l was obtained during the first year and 3,158.7 Meal ha-l in the second year. Rappaport's finding, 192 energy outputs in comparison to monocropping (Uhl's study). Cassava leaves that were used for fodder increased the energy output during the second year of cropping by 1,397.2 Meal ha-l. Therefore, the total energy output from cassava was 14,783.6 Meal ha-l. Bamboo also contributed to the output of the talun-kebun system at the end of the six year rotation. The mature bamboo culmsc that were harvested at the end of the rotation cycle (after four year fallow period) contained 102,795 Meal ha-l, while the energy output of bamboo branches and immature bamboo poles that were used for fuel wood was 47,227.5 Meal ha-l. Therefore, a total energy output of 150,022.5 Meal ha-l was obtained from four years of bamboo growth (or about 37,500 Meal ha-l yr-l) within a six year talun-kebun rotation cycle. Thus, the total energy output over a six year talun-kebun rotation cycle was 176,585.5 Meal ha-l, and the average annual value was 29,430.9 Meal ha-l. E. ENERGY EFFICIENCY Energy efficiency is expressed as the ratio of energy outputs to energy inputs. Table 8.5. summarizes the ratio of various energy outputs to total energy inputs in the study sites. c Based on Atje (1980, unpublished data): Bamboo branches, 1 kg dry mass has an energy content of 3,600 kcal Immature bamboo culms for poles, 1 kg dry mass has an enrgy content of3,375 kcal Mature bamboo culms, 1 kg dry mass has an energy content of3,850 kcal Table 8.5. Energy efficiency ratios of various inputs and outputs for the two year cropping stage of a talun-kebun cycle. Energy Input Energy Output • Efficiency ratio (kcal/ha) (kcal/ha) (output/input) A, Based on Malaysian values Food: First year: 330 11,779.4 35.7 Second year: 107.5 13,386.4 124.5 Total: 437.5 25,165.8 57.5 Food + fodder: First year: 330 11,779.4 35.7 Second year: 107.5 14,783.6 137.5 Total: 437.5 26,563.0 60.7 B. Based on Uhl's values. Food: First year: 471.6 11,779.4 25.0 Second year: 189.1 13,386.4 70.8 Total: 660.7 25,165.8 38.1 Food & fodder: First year. 471.6 11,779.4 25.0 Second year: 189.1 14,783.6 78.2 Total: 660.7 26,563.0 ' 40.2 194 The ratio of food energy harvested to total energy input during the two years of cropping was 38:1 (based on Uhl's figures). This ratio is high in comparison to the output/input ratio of 17:1 for shifting cultivation by the Tsembaga tribe in the highlands of New Guinea (Rappaport, 1971) and 13.9:1 for slash and burn agriculture in San Carlos, the upper Rio Negro, Venezuela (Uhl, 1980). However, the latter is based on gross energy expenditure. If we use the net energy expenditure to calculate energy efficiency in San Carlos study, the ratio increases to 20:1, but this is still lower than the energy efficiency ratio of food production in the talun-kebun. Assuming an average annual energy output (including bamboo) of 29,430.9 Meal ha-l t the average energy efficiency value over the six year cycle was 44.6:1. Thus, the talun-kebun system can be considered as an energy efficient system. 6. ECONOMICS OF THE TALUN KEBUN PRODUCTS The economic returns from the talun-kebun come from selling food products of the first and second year crops and bamboo culms at the end of the rotation. The costs of labor, fertilizer, pesticides and maintenance during the first year of cropping are the major economic costs involved over a talun-kebun rotation. A.COSTS OF PRODUCING THE FIRST AND SECOND YEAR CROPS Data from the interviews showed that labor costs during the harvest and clearing of the mature talun and the planting, maintenance and harvest of food crops constituted 36% of the total costs during the first year cropping stage. Since most activities in the talun-kebun were scheduled during the off-peak season for ricefield work, the opportunity cost of family labor was excluded from the 195 calculation of net returns. However, if the opportunity cost of family labor is counted, the labor costs increase by 40%. The costs of hired labor during the first year of cropping ranged from Rpd 180,000 to Rp 293,506 per ha with an average of Rp 231,616 per ha. The total costs during the first year of cropping ranged from Rp 512,857 to Rp 775,000 per ha with an average of Rp 634,674 per ha Table 8.6). Approximately 41% of the total cost was spent on fertilizer inputs to crop production, 36% on hired labor, and 23% on maintenance (weeding, loosening and heaping up the soil, and the costs of pesticides). The high proportion for fertilizer cost rendered the system sensitive to both loss of soil fertility and increased costs of fertilizer. There was no cost for seeds and seedlings because these materials were obtained from the previous harvest or from neighbors free of charge. Since all ofthe work during cassava cropping was done by family labor, there was practically no labor cost during this period. If the labor cost of family labor is included, however, the total labor cost ranged from Rp 160,000 to Rp 190,000 per ha. Farmers usually obtained cassava stem cuttings from their other fields or from the neighbors fields, and therefore did not have to purchase them. Cassava cropping thus only required human energy input; it did not need any capital input for labor, nor for fertilizers or pesticides, which are not applied during this period. B . R E T U R N S F R O M T H E H A R V E S T P R O D U C T S Although the combination of hyacinth beans, cucumber, and bitter solanum is the most common crop mixture used in the area, data from the interviews showed that only 25% of the interviewees employed this pattern during the period of the study. About 50% of them also grew other crops, such as bitter solanum, chili pepper, and maize in addition to the basic mixture; 10% only grew hyacinth beans and cucumber; and the rest (15%) mixed hyacinth beans and cucumber with other d One USA dollar = Rp 1,045.00 at the time ofthe study 196 crops instead of with bitter solanum. Gross returns from the harvested products of the first year crops ranged from Rp 1,735,714 to Rp 3,129,464 per ha. Hyacinth beans contributed 47% and cucumber contributed 38% to the gross returns, while bitter solanum and other crops only accounted for 15%. The harvest products were mostly sold, only 5-8% being used for the farmers' own consumption. After harvesting hyacinth beans, the farmers usually cut up the used bamboo poles and sold them as firewood. This generated an income of approximately Rp 371,043 per ha. Cassava tubers gave an average gross return of Rp 450,000 per ha. Approximately 90% of the cassava tubers were sold. In addition to the harvest products of the first and second year crops, bamboo culms from the clearcutting were sold and contributed significantly to the total gross returns. The returns from mature bamboo culms ranged from Rp 1,714,286 to Rp 2,455,357 per ha with an average return of Rp 2,121,970 per ha. Neither capital nor labor input was required and no production was obtained during the four year bamboo talun stage; therefore, this period resulted in a zero return. Returns from the talun-kebun were obtained only at the end of the mature talun stage and during the two year cropping stage. Harvested products from the first year crops constituted 47% of the total gross returns over the six-year cycle (Rp 5,551,557 per ha). Cassava tubers and bamboo culms contributed 8% and 38%, respectively, and the rest (7%) was from selling the used bamboo poles. Table 8.7. summarizes the average economic costs and benefits in talun-kebun production in Soreang, West Java. 197 Table 8.6.The Average Costs and Benefits over a six-year Talun- kebun Production cycle in Soreang, West Java (data given on a per ha basis over the whole cycle) Costs: labor costs for clearing & planting Rp 228,516.00 fertilizer (mcluding trans-portation) Rp 258,809.00 pesticides, weeding & other maintenance Rp 147,249.00 Total Rp 634,574.00 Returns: hyacinth beans Rp 1,231,288.00 cucumber Rp 985,624.00 bitter solanum Rp 145,470.00 other crops Rp 246,164.00 cassava tubers Rp 450,000.00 bamboo culms Rp 2,121,970.00 used bamboo poles (firewood) Rp 371,043.00 Total Rp 5,551,557.00 Benefit-Cost Ratio: 8.75 Productivity of labore (gross returns/ man hours) 1387.9 Net return: Rp 4,916,983.00* * Assuming that a family has six fields, this would constitute an average annual return. Approximately 89% of the total gross return was considered as net return if the opportunity cost of family labor was excluded from the calculation. If this cost was included, the net return decreased to 84% of the total gross return. e Female adult is equivalent to 0.8 male adult (Lagemann, 1981; results of the interviews in this study) 198 from each was Rp 975,260 ha"1. This amount was slightly lower than Rp 1,020,000 ha-l annual return of lowland ricefields in the area, but comparable to Rp 974,300 ha*1 annual return of permanent talun which was dominated by fruit trees (Hadikusumah, 1982). However, if a family owns six fields, the average annual return from the entire holding would be Rp 4,916,983 ha*1. 7. PEST AND CONSERVATION ASPECTS Although pests are a common problem in the talun-kebun, there was no evidence of any severe pest outbreak. Insect damage is found even at the talun stage, but the amount of damage to production is generally low. Culm borer, for example, is common in bamboo; however, the farmers considered the damage was negligible in comparison to the total number of healthy culms. Data from the interviews revealed that the farmers employed mixed cropping to spread the risk of crop losses due to insect attack. They also believed that mixed cropping reduced pest incidence in comparison to monocropping by interfering with the movement of pests between host plants. Altieri & Liebman (1986) suggested that a reduced incidence of insect pest in mixed cropping might result from decreased colonization and reproduction of pests, chemical repellency, masking and/or feeding inhibition by nonhost plants, camouflage, and interference with population development and survival. Root (1973) suggested two possible mechanisms to explain high resistance to pest attack in mixed cropping: (1) the natural enemy hypothesis and (2) the resource concentration hypothesis. The natural enemy hypothesis predicts that a greater abundance and diversity of natural enemies of pest insects would be found in mixed cropping than in mono cropping, keeping the pests under control. The resource concentration hypothesis suggests that insect herbivore populations could be influenced directly by the concentration or spatial dispersion of their host plants. The reduced density of host species or interference from nonhost plants reduces 199 pest problems in mixed cropping. Insect pests were common during the first year cropping stage when the plants were still young; therefore, ignoring pest incidence during this period can lead to a complete harvest failure. As a result, pest control measures were applied. Traditionally, farmers have employed direct pest control by manual removal of insect pests from crop plants, or by other mechanical methods, such as burning herbivorous ant nests, etc. In recent years, however, farmers have adopted the use of pesticides for talun-kebun, particularly during the first year of cropping. Such chemical pest control has been applied in ricefields and upland permanent cropfields in the area for many years. Data from the interviews showed that approximately 50% of the interviewees currently use pesticides; 25% used mechanical pest control; 12.5% noticed the occurrence of insect pests but did nothing about it, while the rest (12.5%) believed that there were no pests in their fields. It seems that farmers are becoming increasingly dependent on the use of pesticides to control insect pests. This may raise new problems, such as the emergence of insect pest resistance to pesticides and the destruction of beneficial insects, but there appears to be little choice. If most farmers continue to use pesticides, those who do not could be in a difficult position because natural mechanisms of regulation have been destroyed by the pesticides used by other farmers, and they may face unacceptable pest losses (Brown & Marten, 1986). Besides insect pests, weeds are common in the talun-kebun. Weeds can be considered as pests, but they can also serve useful functions. Some weeds are used as food, fodder, medicine, or fuel biomass. Weeds usually occur during the food cropping stages and are eliminated from the later stage of bamboo talun by the shading effect of the bamboo canopy and the accumulation of surface litter. Although intercropping was said to suppress weed growth by providing an 200 early ground cover, weeds were still found in between the rows of crop plants. Hand weeding was done during the early growth of the first year crops to eliminate interspecies competition, and prior to harvesting the hyacinth beans to clear paths between poles. Farmers never use herbicides to control weeds in the talun-kebun, because they considered weeds to be a temporary problem that only occurs during the early establishment of the first and second year crops, and because they have found that hand weeding alone provides a satisfactory level of weed control. Most talun-kebun fields in the study area were located on slopes; therefore, there is some risk of erosion during the cropping stages, particularly during the period between hoeing and the establishment of an effective cover by the crops. To avoid this problem, contour terracing is done so that soil eroded from an uphill field is retained in the next field downhill. Mounding and ridging the soil in each row, and orienting the rows and drainage ditches along the contour were anti-erosion practices commonly employed by the farmers. Data from the interviews showed that the farmers' knowledge about erosion control measures was based on tradition. 8. F U T U R E T R E N D S The talun-kebun system is known to have been practised at least since the early 1900's, but its objectives and techniques have changed over this period in accordance with the changing needs of the Javanese society. For example, talun-kebun is now being practised for commercial purposes rather than to fulfill the subsistence needs ofthe family. Data from the interviews showed that at least 90% of the harvested products were sold and only 10% were used for the farmers' own consumption, including the portion to be given away to relatives and neighbors. The use of chemical fertilizers in addition to ash and compost has become common as farmers have attempted to increase production. The trend towards market-oriented rather than subsistence-oriented talun-kebun has forced farmers to 201 be more individualistic. For example, in the old days the field was opened to everyone and villagers could collect firewood freely from their neighbor's fields, but nowadays fields are fenced to prevent trespassing and only family members are allowed to collect firewood. It can be said that the social function of the talun-kebun has diminished with the increase in its economic importance. Although some farmers have already adopted monoculture and have replaced bamboo with perennial cash crops, the majority of farmers in the study area are still practising the traditional cropping pattern. Data from the interviews revealed that most farmers did not want to take risks. Because talun-kebun is usually practised on sloping land and because such lands are frequently far from any water resource, bamboo is considered as the most appropriate perennial crop plant. Moreover, the farmers have already learned about the susceptibility of monoculture to pest outbreaks, for example, the devastation of orange orchards in nearby regions by the CPVD virus (Village statistics, 1978). It seems that for the next few years, talun-kebun will generally be practised in the traditional manner, because the farmers see it as the most sustainable system for sloping or marginal land. However, bamboo in the study area is mainly used for building materials, such as for walls, roof-laths, rafters and ceilings. As more and more houses are now being built from bricks or from wood derived from trees, the importance of bamboo as a major bmlding material will probably decrease. This situation will undoubtedly influence the long-term practice of bamboo talun-kebun, with significant implications for the long-term maintenance of soil fertility, and energy and economic efficiency. 202 8.6. S U M M A R Y A N D C O N C L U S I O N S The motivation to use the talun-kebun land use system, the labor requirements during the cycle, the economic costs and benefits, and the pest and conservation aspects of the system were assessed on the basis of interviews with farmers. In addition, field data on the amount of work and the labor used during various stages of the talun-kebun in my experimental sites were quantified. The energy efficiency of the system was calculated by using the ratio of energy outputs to energy inputs. Two sets of human calorie expenditure values per body weight were used: Malaysian values and Uhl's values. Energy outputs were quantified based on the energy content of harvested products. The following conclusions can be drawn from the results of this chapter: 1. Most of the work during the clearing and preparation for cropping was done by male labor. Women participated more during the planting, fertilizing, weeding and harvesting. All activities during the cassava cropping were done by male labor. 2. Non-family labor was hired mainly during the clearing and the preparation for cropping (96% of the total labor used). 3. Men (male labor) spent 1,500-2,000 hrs ha-l and women (female labor) spent 700-1,000 hrs ha- 1 during the clearing and the first year of cropping. Approximately 650-950 hrs ha-l were spent by male labor during the second year of cropping. 4. A total of either 437.5 Meal ha- 1 or 660.7 Meal ha- 1 of net energy (depending on the calorie expenditure values used in the calculation) was expended during the two year cropping period. Approximately 70-75% of the total energy was expended during the first year cropping. 5. The total energy output from a six year talun-kebun rotation was 176,585.5 Meal ha- 1 : 14% as food energy, 27% as fuel wood, 0.8% in fodder, and 203 the rest (58.2%) in the mature bamboo culms. 6. The energy efficiency ratio of two year food crop production was 38:1, while the average energy efficiency ratio over a six year talun-kebun rotation cycle was 45:1. 7. A six year talun-kebun rotation gave a net return of Rp 4.9 million per ha or Rp 816,667 per ha annually with an economic benefit/cost ratio of 8.8 and a productivity of labor of Rp 1387.9/hour. 8. Pest control measures were only taken for the first year crops. 204 Chapter 9 B I O G E O C H E M I C A L A S S E S S M E N T O F T H E SUSTAINABILITY O F P R O D U C T I O N IN T H E BAMBOO TALUN-KEBUN S Y S T E M 9.1. I N T R O D U C T I O N The objective of this chapter is to synthesize information on the biogeochemistry of the talun-kebun system in order to explain the sustainability of production under the traditional management practice, and to analyze the effects on production of alterations in the traditional practice (eg. an extension of the cropping period and/or a shortening of the fallow period). In the first section of the chapter the input and output balance of a complete rotation cycle will be presented, followed by a calculation of the net difference between inputs and outputs for each major compartment in each stage of the rotation, based on the data provided in Figures 7.7-7.13. These data, together with estimates of uptake per tonne of production for various crop and non-crop species grown during the two year cropping stage, by weeds, and by bamboo in each of the fallow stages, are then used to assess the implications of changing the cropping patterns for the site nutrient budget, and for plant production. Because several untested assumptions must be made in order to make this assessment, the results and hence the interpretations of this chapter should be considered as speculative. 9.2. O V E R A L L I N P U T A N D O U T P U T B A L A N C E O F F I V E M A J O R N U T R I E N T S (N, P, K, Ca, Mg) IN T H E TALUN-KEBUN S Y S T E M . Table 9.1 summarizes the overall input-output balance of five major nutrients (N, P, K, Ca, and Mg) over a six year talun-kebun rotation cycle. The data in this table were presented in more detail in Chapter 7. Table 9.1. The input-output balance of N, P, K, Ca, and Mg in a six year talun-kebun rotation cycle. 205 Table 9.1. The input-output balance of N, P, K, Ca, and Mg in a six year talun-kebun rotation cycle. Statements about the reliability of the various estimates are given in Table 7.11. Nutrient inputs Precipitation (6 yrs) Biological fixation 1st yr cropping 2nd yr cropping four years of fallow N 47 15 60 P K 9.6 Ca 12.0 Mg 3.6 Fertilization 188.5 22.4 101.5 28.9 51.1 Weathering (6 yrs) - ? ? ? ? Total Input 310.5 22.4 111.1 40.9 54.7 Nutrient output Losses in the burn 84.5 1.0 4.5 3.4 1.9 Leaching losses 1st yr cropping 2nd yr cropping four years of fallow 0.3 0.2 0.16 0.3 0.2 0.32 2.8 3.7 3.0 7.7 6.8 9.9 6.1 5.9 5.3 Harvest removal 1st yr cropping 2nd yr cropping end of fallow 292.5 74.4 71.6 15.6 6.3 45.0 106.8 78.2 240.8 18.3 16.6 55.8 24.6 13.4 134.0 Removal of dead branches 10.2 4.3 16.5 3.8 3.0 Total output 533.86 73.02 456.3 122.3 194.2 Net difference -223.4 -50.6 -345.2 -81.4 -139.5 The average annual input-output balance -37.2 -8.4 -57.5 -13.6 -23.3 Table 9.1 shows that harvest removals were clearly the major pathway of nutrient loss, accounting for 84, 97, 97, 77, and 90% of the total N, P, K, Ca, and Mg output from the six year talun-kebun rotation cycle, respectively. Soil leaching losses were small for N, P, and K (less than 1%, 1.2%, and 2.2% of the total loss, respectively), but significant for Ca (about 20%) and Mg (about 9%). Although lysimeter-based estimates of quantitative leaching losses are subject to considerable 2 0 6 error, the data suggest that leaching losses were only important for the two divalent cations. Losses in the burning of litter and slash were only significant for nitrogen (16% of total losses), burn-related loss of P being similar to the leaching losses, while losses of K, Ca, and Mg were less than the leaching losses. About 223.4 kg of N ha-l, or an average of 37.2 kg of N ha-lyr-1, was lost from the system over the six year rotation. The difference between input and output to and from the talun-kebun ecosystem reflected changes in the soil N compartment; mineral soil N to 25 cm depth declined by 160 kg ha-l over the three year period after clearcutting (two years of cropping and one year of fallow; data in Chapter 7). However, the negative balance for the nitrogen budget may have been partly caused by an underestimate of the rate of nitrogen fixation by Albizia and hyacinth bean, for which literature values were used. If the rate of the nitrogen fixation by Albizia is changed to 50 kg/ha/year instead of 15 kg/ha/year, the net difference shifts from -234.5 to -13.4 kg/ha over the six year rotation, or an average annual input-output balance of only -2.2 kg/ha/year. Considering the importance of inputs from biological nitrogen fixation to the input-output balance of the system, an indepth study on the role of N-fixing species and the rate of nitroegn fixation should be conducted. There may also have been some loss to denitrification, but this was not measured. From the data, it is not possible to say whether or not the present system of management is resulting in a decline in site nitrogen. The net differences in the input-output balance of P, K, Ca, and Mg were presumed to be compensated for by the release from soil minerals by weathering, but this was not measured or estimated because of the lack of data. The net loss of K > Mg > Ca > P would presumably lead eventually to a depletion of these elements in the surface soil. There are no data to suggest a current problem of declining availability of these elements on this site, although the use of an NPK fertilizer does suggest concern over the availability of P and K, and the estimated net loss of 207 57.5 kg ha" 1yr" 1 of K is certainly noteworthy. However, the periodic deposition of volcanic ash in this region may replenish the supply of weatherable minerals containing these elements sufficiently frequently that this rate of loss may not be of significance for long-term productivity on these sites. 9.3. COMPAJRTMENTAL ANALYSIS OF THE INPUTS AND OUTPUTS OF FIVE MAJOR NUTRIENTS (N, P, K, Ca, and Mg) IN THE DIFFERENT STAGES OF THE TALUN-KEBUN SYSTEM The objective of this analysis is to account for the net difference between the measured or estimated inputs and outputs for each stage on the basis of empirical data on nutrient dynamics in the vegetation, forest floor, and mineral soil presented in Chapter 7. Tables 9.2-9.6 summarize the inputs and outputs of N, P, K, Ca, and Mg into and out of the vegetation, forest floor, and mineral soil of various stages of the talun-kebun. The net difference between inputs and outputs in each compartment indicates whether the compartment is gaining or losing nutrients at that stage. Fine root turnover represents an important input of nutrients to the forest floor and mineral soil (Vogt et al, 1986), but was not measured in this study. Fine root and rhizome turnover were estimated based on the rhizome and root biomass data in Chapter 4, and the discussion of litterfall and fine root turnover in Chapter 6 and in Appendix 4. Lack of data on fine root production and turnover in the fallow stage is a significant weakness of this study. Data on the vertical distribution of fine roots was used to aportion fine root inputs to the forest floor and mineral soil, and this was incorporated into the calculation of nutrient uptake. Harvest residue (that was composted in the field) was treated as an output from the vegetation, but was then returned as an input to the mineral soil (there was no forest floor at the time of the composting). Aboveground litter decomposition 208 was estimated at 5% annually (Dali, 1983). 9.3.1. THE MATURE TALUN A net difference of +45.3 kg N ha-lyi-1 was observed between inputs and outputs for the vegetation. This reflected the annual increment of N in bamboo biomass during the fallow stage and the accumulation of N in weed biomass at the end of the mature talun. A positive net balance was observed for the other nutrients (P, K, Ca, and Mg). Uptake was an input to the vegetation, but it was considered as an output from the forest floor and mineral soil. Uptake was estimated to be 5.1 kg N ha-lyr-1 from the forest floor and 67.3 kg N ha-lyr-1 from the mineral soil (0-25 cm). The rest of the N uptake (29.5 kg ha-lyr-1) was assumed to be from the lower mineral soil (below 25 cm soil depth). There was an observed increase in forest floor nitrogen from the early to mature fallow of 69.4 kg N ha-l. However, the estimated increase in forest floor N over the 3 year fallow period based on the N budget was 89.9 kg N ha-l with an estimated budget increase in the mature talun year of 29.9 kg ha-l. The difference of 20.5 kg N ha-l between the estimate obtained from the N budget and the measured forest floor N change over the 3 years may reflect unmeasured losses by denitrification, or errors in various other parameters. A similar pattern was found for K, with the estimated increase based on the K budget being higher than that actually observed in the forest floor. The observed increase in forest floor potassium from early to mature fallow was 51 kg ha-l, whereas the estimated increase based on the K budget was 25.6 kg ha-lyr-1, and 76.8 kg ha-l over the 3 years for which the fallow was studied. A possible explanation for the 25.8 kg ha-l difference between the estimate from K budget and the measured forest floor was leaching losses from the forest floor which were not measured. There was a difference of 1.6 kg ha-l yr-l of forest floor P between the budget estimate and the measured value, which may reflect leaching losses. The differences in values for 209 forest floor Ca and Mg between the budget estimate and the empirical field data were very small. Inputs to the mineral soil during this stage consisted of fine root and rhizome turnover, litter decomposition and mineralization which then leached from the forest floor, and biological fixation by Albizia. All of these contributed 41.0 kg N ha-1 yr-l to the mineral soil. There was a net difference of -26.3 kg N ha-lyr-1 between the inputs to and outputs from the upper 25 cm of mineral soil, which lost 13.2, 68.8, 24.4, and 28.3 kg ha-l yr- l of P, K, Ca, and Mg, respectively. However, the losses of P, K, Ca, and Mg may have been balanced by weathering inputs which were not considered in the analysis due to the lack of data. 9.3.2. CLEARING, BURNING, AND HOEING Harvesting removed nutrients from the vegetation. Plant uptake was reduced to uptake by the surviving belowground component, but most live fine roots were then killed by soil hoeing. Part of the nutrients in the forest floor were lost during and after the burn (in the smoke, flyash, and leached by the first rain). Ash from the burn was mixed with animal manure and used as fertilizer during the first year of cropping. 9.3.3. THE CROPPING STAGES The estimate of vegetation uptake during the first year of cropping was based on measurements of foliar leaching, litterfall and plant harvest . There was an accumulation of nutrients in the regrowth of bamboo and in weeds during the second year of cropping. Aboveground litterfall was balanced by decomposition because litter was completely decomposed during the cropping period. Weed and harvest residues were returned as compost to the mineral soil. All fine roots that were killed by soil hoeing were assumed to decompose during the 210 first year of cropping. Rhizome inputs consisted of the decomposition of dead mother rhizome. Except for N and Ca, nutrient inputs to the mineral soil during the first year of cropping were higher than the outputs. The negative net difference for Ca may be compensated for by inputs from weathering. The residue from harvesting the hyacinth bean at the end of the first year cropping was considered as compost input to mineral soil in the second year of cropping. Rhizome inputs in the second year of cropping were estimated from the decomposition of the dead mother rhizomes as well as the dead new rhizomes, NO fertilizer was applied during this stage. There was a withdrawal of nutrients from the mineral soil during this stage which was related to the build up of nutrients in bamboo biomass. Potassium uptake during the cassava cropping seemed to be responsible for the decline of K in the mineral soil. 9.3.4. THE EARLY FALLOW STAGE The early fallow stage was characterized by the accumulation of nutrients in the vegetation and the build up of the forest floor. Weeds from the second year of cropping died and were treated as input to the mineral soil during the early fallow stage (categorized as compost in the analysis). Nutrient uptake was assumed to be 100% from the mineral soil because the forest floor in this stage consisted of intact, unfragmented litterfall. Based on the vertical distribution of fineroot biomass, it was assumed that 71% of the roots was in the 0-25 cm of soil depth, while 29% was below the 25 cm depth. Accordingly, 71% of the nutrient uptake was estimated to be from the upper mineral soil (0-25 cm), while 29% was from the lower mineral soil (below 25 cm depth). There was a net change of-24.4, -21.06, -78.3, -32.5, and -36.9 kg ha-l yr-l of N, P, K, Ca, and Mg in the mineral soil during the early fallow stage based on the estimated budget. 9.3.5. THE SUSTAINABILITY OF THE TALUN-KEBUN SYSTEM 211 The nutrient budget calculations indicate, in agreement with "current wisdom", that the fallow stage is the key to the long-term sustainability of the system. Nutrients are accumulated in the forest floor and in bamboo biomass during this stage, while losses from the system only occur through leaching. The budget showed nutrient decline in the surface 25 cm of the mineral soil during the fallow stages, but for N this decline may have been balanced by N fixation and for P, K, Ca, and Mg by weathering inputs . Moreover, the small but possibly significant contributions of nutrients to the mineral soil in the form of fine root inputs by other tree species on the site (Albizia and fruit trees) was not considered in the budget calculation. Harvest removal was the major pathway of nutrient loss in the talun-kebun rotation, but additional losses (most of the nitrogen and a proportion of the other nutrients) occurred during and after the burn. Commercial fertilizer (NPK) and animal dung were applied in addition to ash as inputs to the first year cropping stage. Eliminating this use of fertilizer would undoubtedly lead to a depletion of soil fertility and affect the long-term sustainability of the system (see discussions in section 9.5). The area is subject to periodic volcanic eruptions (approximately once every 50 years). The deposition of volcanic ash during an eruption will contribute a supply of weatherable minerals containing P, K, Ca, and Mg, which can replace part of the losses of these nutrients. The nutrient budget calculation also indicated that shortening the fallow stage might lead to a depletion of nutrients in the mineral soil because of the lower nutrient accumulation in the forest floor and lower contribution by fine root turnover. Table 9 .2. Estimates of various transfers of N into and out of the vegetation, forest 212 floor. and upper mineral s o i l i n various stages of the talun-kebun cycle. Transfer Nitrogen (kg N n a ^ y r - 1 ) Rotation Stage Mature Clearing F i r s t year Second year Early talun S hoeing of cropping of cropping fallow Inputs to vegetation Precipitation 0. 0. 0. 0. 0. Biological fixation 0. 0. 32.0 0. 0. Plant uptake 101.9 51.5 418.8 195.3 137.6 Total inputs 101.9 51.5 450.8 195.3 137.6 Outputs from tbe vegetation Foliar leaching 0.5 0. 1.1 0.6 Aboveground l i t t e r f a l l 45.2 0. 42.0 48.5 28.2 Belowground l i t t e r f a l l 10.9 10.9 0. 1.8 9.4 Removal of dead branches 2.8 - - - 2.0 Total harvest 0. 114.7 407.7 74.4 0. Total outputs 58.8 114.7 450.8 137.9 40.2 Met change +43.1 -63.2 0. +57.4 +97.4 Inputs to forest floor Foliar leaching 0.5 0. 1.0 1.0 0.6 Aboveground l i t t e r f a l l 45.2 0. 42.0 48.5 28.2 Belowground l i t t e r f a l l 3.2 0. 0. 0. 0. Total inputs 48.9 0. 43.0 49.5 28.8 Outputs from tbe forest floor Plant uptake 5.1 0. 0. 0. 0. Lit t e r decomposition /mineralization 13.9 0. 42.0 48.5 1.4 Burning 0. 84.5 0. 0. 0. Total outputs 19.0 84.5 42.0 48.5 1.4 Net change +29.9 -84.5 +1.0 +1.0 +27.4 2 1 3 Inputs to upper mineral s o i l (25 cm depth) F e r t i l i z e r 0. 0. 188. .5 0. 0. Compost 0. 0. 51. .6 63. 6 34. 2 Fine root £ rhizome inputs 12.1 83.6 83. .6 8. .9 22. ,7 L i t t e r decomposition / m i n e r a l i z a t i o n * 13.9 0. 42. .0 48. .5 1. ,4 B i o l o g i c a l f i x a t i o n * * (by Albizia) 15.0 15.0 15. .0 15. .0 15. .0 Weathering 0. 0. 0. 0. 0. To t a l inputs 41.0 98.6 380. ,7 136. .0 73. .3 Outputs from the upper mineral s o i l (25 cm depth) Plant uptake 67.3 51.5 418. .8 195. .3 97, .7 Leaching l o s s e s 0.04 0.3 0. .3 0. .2 0. .04 To t a l outputs 67.34 51.8 418. ,8 195. .5 97. ,74 Net change -26.34 +46.8 -38. .1 -59. .5 -24. .44 * leached from the f o r e s t f l o o r t o the mineral s o i l ** v i a exudation and root death of N - f i x i n g species. Albizia roots are assumed to be l o c a t e d i n the mineral s o i l . 214 Table 9.3. Estimates of various transfers of P into and out of the vegetation, forest floor. and upper mineral s o i l i n various stages of the talun-kabun cycle. Transfer Phosphorus (kg P h a - l y r _ l ) Rotation Stage Mature Clearing Fi r s t year Second year Early talun £ hoeing of cropping of cropping fallow Inputs to vegetation Precipitation 0. 0. 0. 0. 0. Plant uptake 27.5 13.9 23.7 23.4 39.7 Total inputs 27.5 13.9 23.7 23.4 39.7 Outputs from the vegetation Foliar leaching 0.8 0. 0.8 0.9 0.8 Aboveground l i t t e r f a l l 7.6 0. 1.5 1.1 5.3 Belowground l i t t e r f a l l 2.8 0. 0. 0.4 2.6 Removal of dead branches 1.1 - - - 1.0 Total harvest 0. 54.4 21.4 8.5 0. Total outputs 11.2 54.4 23.7 10.9 9.7 Net change +16.2 -40.7 0. +12.5 +30.0 Inputs to forest floor Foliar leaching 0.8 0. 0.8 0.9 0.8 Aboveground l i t t e r f a l l 7.6 0. 1.5 1.1 5.3 Belowground l i t t e r f a l l 0.8 0. 0. 0. 0. Total inputs 9.2 0. 2.3 2.0 6.1 Outputs from the forest floor Plant uptake 1.4 0. 0. 0. 0. Lit t e r decomposition £ mineralization 2.3 0. 1.5 1.1 0.3 Burning 0. 1.0 0. 0. 0. Total outputs 3.7 1.0 1.5 1.1 0.3 Net change +5.5 -1.0 +0.8 +0.9 +5.9 215 Inputs to upper mineral s o i l (25 cm depth) F e r t i l i z e r 0. 0. 22.5 0. 0. Compost 0. 0. 3.9 1.7 2.6 Fine root £ rhizome inputs 3.2 21.2 21.2 2.3 4.4 Lit t e r decomposition/ 2.3 0. 1.5 1.1 0.3 mineralization* Weathering ? ? ? ? ? Total inputs 5.5 21.2 46.1 5.1 7.3 Outputs from the upper mineral s o i l (25 cm depth) Plant uptake 18.2 13.9 23.7 25.7 28.2 Leaching losses 0.04 0.3 0.3 0.15 0.16 Total outputs 18.24 14.2 23.3 23.15 28.36 Net change -13.24 +7.0 +22.8 -18.05 -21.06 * leached from the forest floor to the mineral s o i l 216 Table 9.4. Estimates of various transfers of K into and out of the vegetation, forest floor, and upper mineral s o i l i n various stages of the talun-kebun cycle. Transfer Potassium (kg K ha'^yr" h Rotation Stage Mature Clearing Fi r s t year Second year Early talun £ hoeing of cropping of cropping fallow Inputs to vegetation Precipitation 1.6 1.6 1.6 1.6 1.6 Plant uptake 138.9 62.6 164.3 159.2 162.4 Total inputs 138.9 64.2 165.9 160.8 163.0 Outputs from the vegetation Foliar leaching 7.0 0. 0.9 2.1 7.6 Aboveground l i t t e r f a l l 31.9 0. 4.8 10.4 19.4 Belowground l i t t e r f a l l 11.2 0. 0. 1.5 9.9 Removal of dead branches i 5.1 - - - 3.1 Total harvest 0. 276.3 158.6 79.6 0. Total outputs 55.2 276.3 164.3 93.6 40.0 Net change +83.7 -212.1 +1.6 +67.2 +123.0 Inputs to forest floor Foliar leaching 7.0 0. 0.9 2.1 7.6 Aboveground l i t t e r f a l l 31.9 0. 4.8 10.4 19.4 Belowground l i t t e r f a l l 3.2 0. 0. 0. 0. Total inputs 42.1 0. 5.7 12.5 27.0 Outputs from the forest floor Plant uptake 6.9 0. 0. 0. 0. Lit t e r decomposition 9.6 0. 4.8 10.4 1.0 Burning 0. 4.5 0. 0. 0. Total outputs 16.5 4.5 4.8 10.4 1.0 Net change +25.6 -4.5 +0.9 +2.1 +26.0 217 Inputs to upper mineral s o i l (25 cm depth) F e r t i l i z e r 0. 0. 101.5 0. 0. Compost 0. 0. 35. 6 15.2 20.9 Fine root £ rhizome inputs 13.7 84.4 84.4 10.0 17.9 Lit t e r decomposition/ 9.6 0. 4.8 10.4 1.0 mineralization* Weathering ? 7 ? ? ? Total inputs 23.3 84.4 226.3 35.6 38.8 Outputs from the upper mineral s o i l (25 cm depth) Plant uptake 91.7 62.6 164.3 159.2 115.3 Leaching losses 0.4 2.8 2.8 3.7 1.8 Total outputs 92.1 65.4 167.1 162.9 117.1 Net change -68.8 +22.0 +59.2 -129.3 -78.3 * leached from the forest floor to the mineral s o i l 218 Table 9.5. Estimates of various transfers of Ca into and out of the vegetation, forest floor. and upper mineral s o i l i n various stages of the talun-kebun cycle * Transfer Calcium (kg Ca h a _ 1 y r _ 1 ) Rotation Stage Mature Clearing Fi r s t year Second year Early talun £ hoeing of cropping of cropping fallow Inputs to vegetation Precipitation 2.0 2.0 2.0 2.0 2.0 Plant uptake 41.2 16.2 90.7 48.6 50.6 Total inputs 43.2 18.2 92.7 50.6 52.6 Outputs from the vegetation Foliar leaching -0.5 0. 0.1 1.0 -0.5 Aboveground l i t t e r f a l l 9.9 0. 34.1 10.8 9.5 Belowground l i t t e r f a l l 2.7 0. 0. 0.5 2.3 Removal.of dead branches 1.0 - - - 0.9 Total harvest 0. 105.1 56.5 17.1 0. Total outputs 13.1 105.1 90.7 29.4 12.2 Net change +29.7 -97.7 +2.0 +11.2 +40.4 Inputs to forest floor Foliar leaching -0.5 0. 0.1 1.0 -0.5 Aboveground l i t t e r f a l l 9.9 0. 34.1 10.8 7.5 Belowground l i t t e r f a l l 0.8 0. 0. 0. 0. Total inputs 10.2 0. 34.2 11.8 7.0 Outputs from the forest floor Plant uptake 2.1 0. 0. 0. 0. Lit t e r decomposition 1.3 0. 34.1 10.8 0.5 Burning 0. 3.4 0. 0. 0. Total outputs 3.4 3.4 34.1 10.8 0.5 Net change +6.8 -3.4 +0.1 +1.0 +6.5 219 Inputs to upper mineral s o i l (25 cm depth) F e r t i l i z e r 0. 0. 28.9 0. 0. Compost 0. 0. 5.5 32.8 2.4 Fine root £ rhizome inputs 3.5 20.2 20.2 2.7 4.4 Litt e r decomposition/ 1.3 0. 34.1 10.8 0.5 mineralization* Weathering ? ? ? ? ? Total inputs 4.8 20.2 88.7 46.3 7.3 Outputs from the upper mineral s o i l (25 cm depth) Plant uptake 27.2 16.2 90.7 48.6 35.9 Leaching losses 2.0 7.7 7.7 6.8 3.9 Total outputs 29.2 23.9 98.4 56.4 39.8 Net change -24.4 -3.7 -9.7 -10.1 -32.5 *leaohed from the forest floor to the mineral s o i l 220 Table 9.6. Estimates of various transfers of Mg into and out of the vegetation, forest floor. and upper mineral s o i l i n various stages of the talun-kebun cycle • Transfer Magnesium (kg Mg na~^ -yr" 1) Rotation Stage Mature Clearing Fi r s t year Second year Early talun & hoeing of cropping of cropping fallow Inputs to vegetation Precipitation 0.6 0.6 0.6 0.6 0.6 Plant uptake 52.0 22.9 57.9 49.7 69.1 Total inputs 52.6 23.5 58.5 50.3 69.7 Outputs from the vegetation Foliar leaching 3.3 0. 0.2 0.8 3.5 Aboveground l i t t e r f a l l 7.2 0. 4.5 6.1 6.6 Belowground l i t t e r f a l l 3.4 0. 0. 0.7 1.4 Removal of dead branches 0.9 - - - 0.6 Total harvest 0. 121.7 53.2 13.4 0. Total outputs 14.8 121.7 57.9 21.0 12.1 Net change +37.8 -98.2 +0.6 +29.3 +57.6 Inputs to forest floor Foliar leaching 3.3 0. 0.2 0.8 3.5 Aboveground l i t t e r f a l l 7.2 0. 4.5 6.1 6.6 Belowground l i t t e r f a l l 0.9 0. 0. 0. 0. Total inputs 11.4 0. 4.7 6.9 10.1 Outputs from the forest floor Plant uptake 2.6 0. 0. 0. 0. Litte r decomposition 2.3 0. 4.5 6.1 0.3 Burning 0. 1.9 0. 0. 0. Total outputs 4.9 1.9 4.5 6.1 0.3 Net change +6.5 -1.9 +0.2 +0.8 +9.8 221 Inputs to upper mineral s o i l (25 cm depth) F e r t i l i z e r 0. 0. 53.6 0. 0. Compost 0. 0. 12.5 16.1 8.3 Fine root £ rhizome inputs 4.9 24.6 24.6 3.9 6.3 Li t t e r decomposition/ 2.3 0. 4.5 6.1 0.3 mineralization* Weathering ? 1 7 ? ? Total inputs 7.2 24.6 95.2 26.1 14.9 Outputs from the upper mineral s o i l (25 cm depth) Plant uptake 34.3 22.9 57.9 49.7 49.1 Leaching losses 1.2 6.1 6.1 5.9 2.7 Total outputs 35.5 29.0 64.0 54.9 51.8 Net change -28.3 -4.4 +31.2 -28.8 -36.9 * leached from the forest floor to the mineral s o i l 222 9.4. NUTRIENT UPTAKE-PRODUCTION RELATIONSHIPS The amount of nutrient uptake per tonne of biomass production was determined for each species in order to estimate the uptake requirements of plants. Table 9.7 summarizes nutrient uptake-production relationships for each species at various stages of the talun-kebun cycle. Nutrient uptake values were based on the data used for compartmental analysis (Section 9.3). It is recognized that nutrient uptake by plants can vary from "luxury consumption" to growth-limiting severe deficiency. Unless appropriate fertilization experiments are conducted, it is difficult to conclude whether or not a particular level of uptake constitutes adequate nutrition, nutrient deficiency, or luxury uptake. The fact that N P K fertilizer is used by the farmers presumably reflects a potential limitation of plant growth by these three elements on the study site. The reported and observed decline in food crop yields in the absence of fertilization supports the contention that at least the food crops are nutrient limited in this system. On this basis, the assumption is made that the calculated uptake/tonne values are causally related to production, at least for nitrogen. The shortcomings of this assumption are recognized, but it is necessary in order to make the following preliminary and speculative assessment. Additional data from factorial fertilizer experiments or controlled plant nutrition studies would be necessary to test this assumption and to extend the analysis to the other nutrient 223 Table 9.7. Nutrient uptake-production relationship for various species grown at various stages of the talun-kebun cycle (the number within brackets represents percentage distribution of uptake at that particular stage). Species Biomass Uptake (kg ha-l) Uptake/production (kg/tonne) in various stage (t ha-l) N F K Ca Mg N P K Ca Mg Fi r s t year cropping Cucumber 0. 25 14.9 2.5 15.0 2.9 2.4 59. .6 9. .6 60 11. .6 9. .6 (4) (11) (9) (3) (4) Bitter solanum 0. 50 22.3 1.7 23.3 4.0 3.5 44. .6 7. .0 46.7 8. .0 7. .0 (5) (7) (14) (4) (6) Hyacinth bean 7. 30 330.3 15.1 93.3 79.0 39.4 45. .2 2. .1 13.0 10. .8 5. .4 (79) (64) (58) (87) (68) Bamboo grass-shoot 0. 51 5.2 1.0 4.6 1.4 1.8 10. .4 2. ,0 9.2 2. .8 3. ,6 (1) (4) (2) (2) (3) Weeds 1. 80 46.1 3.4 28.1 3.4 10.8 25. .6 1. .9 15.6 1. .9 6. .0 (11) (14) (17) (4) (19) Second year cropping Cassava 6. 50 123.7 10.5 92.2 28.0 20.3 19. .0 1. .7 14.4 4. .3 3. .1 (63) (45) (58) (58) (41) Bamboo 6. 50 37.4 10.1 46.1 18.2 21.1 5. .8 1. .6 7.1 2. .8 3. .2 (19) (44) (29) (37) (42) Weeds 1. 50 34.2 2.6 20.9 2.4 8.3 22. .8 1. .7 13.9 1. .6 5. .5 (18)- (11) (13) (5) (17) F i r s t year fallow Bamboo 19. 10 116.0 38.3 149.9 49.1 65.0 6. .1 2. .0 7.8 2. .6 3. .4 (84) (96) (92) (97) (94) Weeds 1. 10 21.6 1.4 12.5 1.5 4.1 19. .6 1. .3 11.4 1. .4 3. .7 (16) (4) (8) (3) (6) Second year fallow Bamboo 17.10 98.4 27.9 139.5 42.5 51.8 5.8 1.6 8.2 2.3 2.9 (89) (97) (94) (98) (95) Weeds 0.80 12.6 0.9 8.6 1.0 3.0 15.8 1.1 10.8 1.3 3.8 (11) (3) (6) (2) (5) 224 Third year fallow Bamboo 17.00 99.6 26.6 141.7 38.9 49.4 5.9 1.6 8.3 2.3 2.9 (94) (98) (97) (98) (96) Weeds 0.50 6.0 0.6 5.1 0.7 1.8 12.0 1.2 10.2 1.4 3.6 (6) (2) (3) (2) (4) Mature talun Bamboo 16.90 98.7 27.3 137.0 40.9 51.3 5.8 1.6 8.1 2.4 3.0 (98) (98) (99) (99) (99) Weeds 0.20 1.7 0.2 1.9 0.3 0.7 8.5 1.0 9.5 1.5 3.5 (2) (2) (1) (1) (1) Annual crops had a higher nutrient uptake requirement (assuming that kg uptake/tonne of production is a measure of uptake requirement) than bamboo. The highest nutrient uptake/tonne of production for bamboo occurred during the "grass stage", during which the ratio of foliage to "grass stem" was high. This value declined during the cassava cropping, which was probably due to interspefic competition between bamboo and cassava and the lower production of bamboo foliage during this stage. The nutrient uptake requirement of bamboo increased again during the early fallow stage for all nutrients except for Ca. Bamboo still showed a high uptake of K at the end of the mature talun, but the uptake of other nutrients declined or remained stable. The continuous decline of nutrient uptake by weed species over various stages of the talun-kebun cycle showed that only those species with low nutrient uptake requirements survived as undergrowth in the developing bamboo field. Data on nitrogen uptake per tonne of biomass production are used in the following section as an example to estimate the availability of nitrogen on the site, and how changes in nitrogen under a different management regime might affect plant production. 225 9.5. IMPLICATIONS OF ALTERATIONS OF THE TRADITIONAL CROPPING PRACTICES FOR CROP PRODUCTION: A TABULAR ANALYSIS There is a variety of approaches that could be used in trying to make biogeochemical analysis of the long term sustainability of the talun-kebun system under various different management practices. One can make a simple inventory of nutrients in plant biomass and the soil at various stages of the rotation cycle, and infer from this the consequences of different intensities and frequencies of harvest on soil nutrient inventories and crop production. However, such a static inventory only shows conditions at one or a few moments in time, and it fails to account for the temporal dynamics of nutrients: the various inputs and outputs to and from the system, and the patterns of, within system circulation (Kimmins, 1987). Documentation of these temporal dynamics in the various stages of the rotation cycle provides the data required for a tabular budget analysis, which can result in a more realistic evaluation of the imports of cropping on plant growth and yield than the simple inventory approach. Incorporation of a large amount of detail in the tabular budget renders this approach cumbersome, so a detailed budget analysis is better undertaken using a computer simulation model of nutrient cycling and plant growth. In this section, a relatively simple tabular analysis is made, in anticipation of using the data in a computer simulation model subsequently. If we assume that plant growth is nutrient-limited, nutrient uptake estimates in various stages of the talun-kebun cycle can be used to represent both the amount of nutrients in mineral soil that were available to the plants, and the amount of uptake required to support a given rate of growth. This availability of soil nutrients is influenced by management practices. The combination of data on nutrient uptake requirements and on the effects of management on soil nutrient availability can then be used to prepare an initial and speculative assessment of the consequences for production of altering the traditional cropping practices. This was 226 the approach used in preparing a tabular analysis of the changes in uptake and production that are predicted to occur if the period of cropping is extended. It is recognized that this is a less satisfactory approach than the use of a detailed simulation model, but in the absence of an appropriate model, it was the most detailed level of analysis that could be undertaken. Two scenarios were chosen. The first one was to assess the effect of extending the period of mixed cropping on N uptake and biomass production, and the second scenario dealt with the effect of extending the cassava cropping on N uptake and biomass production. Nitrogen was selected for the purpose of this analysis because hyacinth bean (a nitrogen fixing species) was the main crop planted during the mixed cropping, and because the assumption that nitrogen availability was limiting growth was the only assumption which could be supported by the available data (however, the use of P and K fertilizer by the farmers suggests that these two nutrients may also be potentially limiting). 9.5.1 .EXTENDED THE PERIOD OF MDXED CROPPING In this analysis, the period of mixed cropping was extended to two years, followed by one year of cassava cropping and three years of bamboo fallow (in contrast to the traditional one-one-four year sequence). Data on the N uptake during the first year cropping stage of the traditional talun-kebun practice were used as the input data for the first year of cropping. The uptake requirement for the second year of cropping was determined based on the following assumptions: 1. The amount of N available for uptake in the second year was equal to that available in the first year cropping stage, less the total amount of N in the fertilizer used in the first year. It is assumed that fertilizers are not used in this second year because of the lack of ash and manure, and that 227 the farmers would be reluctant to buy synthetic fertilizers. This assumption implies that there is no change in the ability of the site to provide N, which may not be true. It is therefore believed that this assumption is a conservative one. 2. There was no change in the percentage distribution of uptake by different species. It is assumed that all species will behave in the same way in response to declining availability. Plant production was based on the uptake requirement per tonne of production in the first year cropping, and the estimated amount of N available for uptake. This assumption implies no change in resource allocation between above and belowground biomass as N availability changes. Evidence (Kimmins, 1987) suggests that in many species, there is a switch from the aboveground to belowground (root) production, and thus of "harvest index" (the % of total production that is harvestable), as N availability declines. Thus, the estimated effect of declining soil fertility or crop yield is considered to be conservative. The sum of the uptake of the two years of mixed cropping was compared with the uptake of one year of mixed cropping and used to calculate the reduction in the amount of N available for cassava cropping. N uptake during cassava cropping was calculated based on the assumption that there was no change in the percentage distribution of uptake by species between the first and second year of cropping.. The sum of the uptake of the two years of mixed cropping and the uptake of one year of cassava cropping was compared with the uptake of the traditional two year of mixed cropping and used to calculate the reduction in the amount of N available for uptake during the early bamboo fallow. N uptake in the early fallow stage was calculated based on the assumption 228 that there was no change in the percentage distribution of uptake by species between the traditional system and the alternative scenario. The total uptake of the three years of cropping and the one year fallow was summed and compared to that of the two year cropping plus one year fallow of the traditional practice. The result was used to estimate the reduction in the amount of N available for uptake during the subsequent years of the fallow stage. Applying a six year talun-kebun cycle, the fallow stage in this scenario was thus shortened to three years instead of the four years usually practised by the farmers. This would result in a lower total bamboo aboveground litterfall production which in turn reduced the amount of N returned to the mineral soil and the amount of ash nutrients available for use as fertilizer in the subsequent six year cycle. The extension of this analysis to a subsequent rotation showed that there would be a continued reduction in production for some time, unless additional fertilizers were added. 229 Table 9.8. Tabular analysis of the implications of an alternative cropping sequence for crop and bamboo production, based on the estimated amount of N available for plant uptake and biomass production in various stages of the scenario. This scenario involves two years of mixed species cropping, one year of cassava cropping, and three years of bamboo fallow. FIRST ROTATION: The f i r s t year of cropping: Total N uptake (excluding N fixation) 418.8 Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) cucumber 4 14.9 59.6 0.25 solanum 5 22.3 44.6 0.50 h. bean 79 330.3 45.2 7.30 bamboo 1 5 2 10.4 0.51 weeds 11 46.1 25.6 1.80 T o t a l 100 418.8 10.36 The second year of cropping (mixed species): Total N uptake = 418.8 - 188.5 (fertilizer) = 230.3 Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) cucumber 4 9.2 59.6 0.15 solanum 5 11.5 44.6 0.26 h. bean 79 181.7 45.2 4.02 bamboo 1 2.3 10.4 0.23 weeds 11 25.3 25.6 0.99 T o t a l 100 230.0 5.65 Total uptake over two years of mixed cropping = 418.8 + 230.0 = 648.8 (a) Total uptake of one year cropping i n the traditional practice = 418.8 (b) A difference of 230 resulted in a 55% reduction in the amount of N available for cassava cropping in comparison to the traditional system. The total food crop production over the two years was 12.5 t/ha 230 The third year of cropping (cassava): Total N uptake = the amount of N available for cassava cropping in the traditional practice - 55% reduction = 195.3 - 107.4 = 87.9 Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) cassava 63 55.4 19 2.9 bamboo 18 15.8 5.8 2.7 weeds 19 16.7 22.8 0.7 T o t a l 100 87.9 6.3 Total uptake of three year cropping = 648.8 + 87.9 = 736.7 (c) Total uptake of two year cropping in the traditional practice = 418.8 + 195.3 = 614.1 (d) A difference of 122.6 resulted i n a 20% reduction in the amount of N available for bamboo uptake during the fallow stage The total production of cassava in the traditional system was 6.5 t/ha, 2.2 times the production i n this scenario The early fallow stage: Total uptake = the amount of N available for bamboo uptake in the early fallow stage of the traditional practice - 20% reduction = 137.6 - 27.5 = 110.1 Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) bamboo 84 92.5 6.1 13.5 weeds 16 17.6 19.6 0.9 T o t a l 100 110.1 14.4 Total uptake of three year cropping + one year bamboo fallow = 846.8 kg/ha Total uptake of two year cropping + one year bamboo fallow in traditional practice = 751.7 kg/ha A difference of 95.1 kg/ha resulted in a 13% reduction in the amount of N available for bamboo uptake during the second year of fallow The total production of bamboo in this scenario was 13.5 t/ha in comparison to 19.1 t/ha in the traditional system 231 The second year of fallow: Total uptake = the amount of N available for bamboo uptake i n the second year fallow stage of the traditional practice - 13% reduction = 111 - 14.4 =96.6 kg/ha Species % distr. Uptake Uptake/prod. Prod. (kg/ha) (kg/tonne) (t/ha) bamboo 89 86 5.8 14.8 weeds 11 10.6 15.8 0.7 T o t a l 100 96.6 15.5 Total uptake of three year cropping + two year fallow = 943.4 kg/ha Total uptake of two year cropping + two year fallow i n traditional practice = 862.7 kg/ha A difference of 80.7 kg/ha resulted in a 9% reduction of the amount of N available for uptake during the third year of fallow The total production of bamboo in the second year of fallow was 14.8 t/ha in comparison to 17.1 t/ha in the traditional practice The third year of fallow: Total uptake = the amount of N available for bamboo uptake i n the third year fallow stage of the traditional practice - 9% reduction = (105.6 - 9.5) kg/ha =96.1 kg/ha Species % distr. Uptake Uptake/prod. Prod. (kg/ha) (kg/tonne) (t/ha) bamboo 97 93.2 5.9 15.8 weeds 3 2.9 12.0 0.2 T o t a l 100 96.1 16.0 Total uptake of a six year rotation cycle in this scenario (3 year cropping + 3 year fallow) = (943.4 + 96.1) kg/ha = 1039.5 kg/ha Total uptake in a six year talun-kebun rotation cycle of the traditional practice (2 year cropping + 4 year fallow) = 1070.2 kg/ha The total uptake in this scenario was 3% lower than the total uptake under the traditional practice. Therefore, the amount of N available for uptake i n the subsequent rotation increased by 3%. However, the biomass production of 4 year bamboo fallow in traditional practice was much higher than that of three year fallow i n this scenario (70.1 t/ha i n comparison to 44.1 t/ha). Thus, shortening the fallow period by one year resulted i n a 37% reduction of biomass yield. This i s probably a conservative estimate of the true reduction. 232 Assuming that the production of aboveground l i t t e r changes in proportion to changes in biomass production, the amount of forest floor accumulated by the end of the rotation would also decline by 37%. As a result, the amount of ash after the bum was reduced. Considering the proportion of ash : manure mixture to be constant, there would be a reduction i n the amount of this f e r t i l i z e r mixture applied to the s o i l during the f i r s t year cropping of the subsequent rotation. Also the reduction i n aboveground l i t t e r f a l l would reduce the amount of N released from the forest floor by mineralization of l i t t e r . Because of lack of data, this aspect of the scenario cannot be developed, but again, the estimated reduction i n production i s undoubtedly conservative. SECOND ROTATION The f i r s t year of cropping: Thoretically, there was a 3% increase in the amount of N available for uptake by the f i r s t year crops, but this may not be true considering the reduced N input from bamboo l i t t e r f a l l . Total N i n the f e r t i l i z e r s = (60.2 + 88) kg/ha = 148.2 kg/ha (based on the conservative assumption that the proportion of ash:manure mixture i s constant, there i s also a reduction of 37% of the total N i n ash-manure mixture) Total uptake (excluding N fixation) = 385.4 kg/ha Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) cucumber 4 13.9 59.6 0.23 solanum 5 19.3 44.6 0.43 h.bean 79 304.5 45.2 6.74 bamboo 1 3.9 10.4 0.38 weeds 11 42.4 25.6 1.66 T o t a l 100 385.4 9.44 (compared to 10.36 in the f i r s t rotation) The second year of cropping (mixed species): Total N uptake = 385.4 - 148.2 = 237.2 kg/ha Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) cucumber 4 8.5 59.6 0.14 solanum 5 11.9 44.6 0.27 h.bean 79 187.4 45.2 4.15 bamboo 1 2.4 10.4 0.23 weeds 11 26.1 25.6 1.02 T o t a l 100 in the f i r s t rotation) 237.2 5.81 (compared to 5.7 2 3 3 Total uptake of two year mixed cropping = 385.4 + 237.2 = 622.6 kg/ha Total uptake of one year mixed cropping in the traditional practice = 418.8 kg/ha A difference of 203.8 kg/ha resulted i n a 49% reduction in the amount of N available for cassava cropping The total food crop production was 12 t/ha, 4% lower than that of the f i r s t rotation The third year of cropping (cassava): Total uptake = 195.3 - 49% reduction =99.6 kg/ha Species % distr uptake uptake/prod prod. (kg/ha) (kg/tonne) (t/ha) cassava 63 62.8 19.0 3.3 bamboo 18 17.9 5.8 3.1 weeds 19 18.9 22.8 0.8 T o t a 1 100 99.6 7.2 Total uptake of three year cropping = 622.6 + 99.6 = 722.2 kg/ha Total uptake of two year cropping i n traditional practice = 418.8 + 195.3 = 614.1 kg/ha A difference of 108.1 kg/ha resulted i n an 18% reduction i n the amount of total N available for uptake during the f i r s t year fallow The total production of cassava was 3.3 t/ha, which was only 51% of the cassava production i n a traditional practice. This amount, however, was 14% higher than the production during the f i r s t rotation of this modified cropping scenario, because of the reduced production and nutrient removal in mixed crops The early fallow stage: Total uptake = 137.6 - 18% reduction = 112.8 kg/ha Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) bamboo 84 94.8 6.1 15.5 weeds 16 18.0 19.6 0.9 T o t a l 100 112.8 16.4 The total production of bamboo i n this stage was 15.5 t/ha (81% of the production following the traditional practice). This amount was 15% higher than that of the f i r s t rotation. Total uptake of the three year cropping and one year fallow = 835 kg/ha in comparison to 751.7 kg/ha i n the traditional practice A difference of 83.3 kg/ha resulted i n an 11% reduction in the amount of N available for uptake during the subsequent stage 234 The second year of fallow Total uptake = 111 - 11% reduction =98.8 kg/ha Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) bamboo 89 87.9 5.8 15.2 weeds 11 10.9 15.8 0.7 T o t a l 100 98.8 15.9 Total uptake of three year cropping + two year fallow = 933.8 kg/ha in comparison to 862.7 kg/ha i n the traditional practice A difference of 71.1 kg/ha resulted in an 8% reduction of the amount of N available for uptake i n the third year of fallow The total production of bamboo in the second year of fallow was 15.2 kg/ha (89% of the production i n the traditional practice). This number, however, was 3% higher than the production of the second year fallow i n the f i r s t rotation The third year fallow: Total uptake = 105.6 - 3% reduction =102.4 kg/ha Species % distr. uptake uptake/prod. prod. (kg/ha) (kg/tonne) (t/ha) bamboo 97 99.3 5.9 16.8 weeds 3 3.1 12.0 0.3 T o t a l 100 102.4 17.1 Total uptake of a six year rotation cycle (3 year cropping + 3 year fallow) = 1036.2 kg/ha (3% lower than that of the traditional rotation). The biomass production of bamboo during the fallow period of this rotation was 47.5 t/ha (68% the production of four year bamboo fallow i n the traditional practice). Thus, the bamboo production in the second rotation was 8% higher than that of the f i r s t rotation, but i t was s t i l l lower than the bamboo production during the fallow period of the traditional practice 235 SUMMARY Biomass production over a six year rotation in the traditional practice and i n the extended cropping practice: Species Traditional Extended mixed cropping practice Rotation 1 Rotation Mixed crop 8.1 12.0 12.5 Cassava 6.5 2.9 3.3 Bamboo 70.1 44.1 47.5 Weeds 5.7 4.6 5.4 T o t a l 90.8 64.1 68.2 The total biomass production in six year cycle of the f i r s t and second rotation under the extended mixed cropping scenario accounted for 71 and 75 % of the total production in the traditional practice, respectively The above results showed that extending the mixed cropping period increased the production by 1.5 times, but decreased the cassava production by 51-55% and the production of bamboo biomass by 32-37%. This result i s unlikely to happen because the shorter fallow reduced the amount of ash for f e r t i l i z e r and theamount of organic matter returned to the s o i l . The increase y i e l d in the second rotation was merely counter-intuitive. It i s clear that the effect of reduced ash and reduced organic matter and nitrogen from the fallow would reduce the yield and not increase i t . Therefore, a simulation model seemed to be a more appropriate approach to handle the complexity of the system and to explain the uncertainty raised in this study. 236 9.5.2. EXTENDING THE PERIOD OF CASSAVA CROPPING This scenario consisted of one year of mixed cropping, two years of cassava cropping, and three years of bamboo fallow. Data on cassava production and uptake during the subsequent cropping were derived from biomass data in Chapter 6 and from the nutrient concentration data in Appendix 3. Cassava production in the third year cropping dropped by 45% and the total N accumulated in the biomass dropped from 74.4 kg/ha to 37 kg/ha. Assuming that the decrease in litterfall is in accordance to the decrease in biomass, litterfall production in the second cassava cropping was 0.77 t/ha and the total N content was 24 kg/ha. Applying the same method as in the first scenario, the result of this assessment is summarized in Table 9.9. 237 Table 9 .9 . The effect of extending the cassava cropping on the amount of tota l N available for plant uptake and biomass production in various stages of the cycle FIRST ROTATION: The f i r s t year of cropping: Total N uptake 418.8 kg/ha Total mixed crop production 8.1 t/ha The second year of cropping: Total N uptake 195.3 kg/ha Total cassava production 6.5 t/ha The third year of cropping (extra cassava cropping): Total N uptake 133.9 kg/ha Total cassava production 4.4 t/ha Total uptake for the three year cropping was 748.0 kg/ha, which was 22% higher than following a two year cropping period as in the traditional practice. Therefore, the amount of total N available during the f i r s t year fallow was reduced by 22%. The f i r s t year of fallow: Total N uptake 107.3 kg/ha Total bamboo production 14.8 t/ha There was a 33% reduction i n bamboo biomass produced during this stage. Total uptake for three year cropping + one year fallow was 855.3 kg/ha, 14% higher than following the traditional pattern. Thus, the amount of total N available for uptake during the second year fallow was reduced by 14% The second year of fallow: Total N uptake 95.5 kg/ha Total bamboo production 14.7 t/ha There was a 14% reduction in bamboo biomass produced during this stage. Total uptake of three year cropping + two year fallow was 950.8 kg/ha. Therefore, the amount of total N available for uptake during the subsequent fallow was reduced by 10%. The third year of fallow: Total N uptake 95.0 kg/ha Total bamboo production 15.6 t/ha Total uptake of a six year rotation cycle i n this scenario was 1045.8 kg/ha, which was 2% less than that following the traditional pattern. This increased the amount of N available for uptake during the f i r s t year cropping of the subsequent rotation by 2%. The production of bamboo i n three year fallow, however, was only 64% of that of four year fallow in the traditional practice. As a result, the amount of l i t t e r poroduced and the ash after the burn also declined at the same proportion. 238 SECOND ROTATION The f i r s t year cropping: Total N i n f e r t i l i z e r 1 5 2 . 3 kg/ha (a 36% reduction in the total ash-mixture) Total N uptake 2 3 4 . 9 + 1 5 2 . 3 = 3 8 7 . 2 kg/ha Total crop production 7 . 5 t/ha There was a 7% reduction i n the total mixed crop production during the f i r s t year cropping of the subsequent rotation The total uptake was 6% less than following the traditional pattern; therefore, the amount of N available for uptake i n the cassava cropping increased by the same proportion. The second year cropping: Total N uptake 2 0 7 . 0 kg/ha Total cassava production 6 . 9 t/ha This value was 6% higher than that during the f i r s t rotation Total N uptake of the two year cropping stage was 5 9 4 . 2 kg/ha in comparison to 614.1 kg/ha following in the f i r s t rotation. Therefore, the amount of N uptake for the second cassava cropping increased by 3%. The third year cropping (second cassava cropping): Total N uptake 1 3 7 . 9 kg/ha Total cassava production 4 . 6 t/ha The production value was 5% higher than that of the f i r s t rotation Total uptake of the three year cropping stage was 7 3 2 . 1 kg/ha, 118 kg/ha higher than that following traditional pattern. Therefore, the total N available for uptake during the f i r s t year fallow decreased by 19%. The f i r s t year fallow: Total N uptake 111.5 kg/ha Total bamboo production 1 5 . 4 t/ha The production value was 4% higher than the value during the f i r s t rotation, but i t was s t i l l 19% less than that following the traditional pattern The total uptake of three year cropping + one year fallow was 8 4 3 . 6 kg/ha (12% higher than following traditional pattern), which resulted in a 12% reduction for the subsequent fallow The second year of fallow: Total N uptake 9 7 . 7 kg/ha Total bamboo production 1 5 . 0 t/ha This value was 2% higher than that of the f i r s t rotation, but 12% lower than that following the traditional practice There was a 9% reduction in the amount of total N available for uptake during the subsequent fallow 239 The third year fallow: Total N uptake 96.1 kg/ha Total bamboo production 15.8 t/ha The production value was 1% higher than that of the f i r s t rotation, but i t was s t i l l 9% lower than that following the traditional practice Total uptake of the second rotation was 1037.4 kg/ha in comparison to 1070.2 kg/ha following the traditional practice. Therefore, there would be a 3% increase in the amount of N available for uptake during the subsequent rotation SUMMARY Biomass production over a six year rotation i n the two rotations of the extended cassava cropping and in the traditional practice was as follow: Species Traditional Extended cassava cropping practice Rotation 1 Rotation 2 (t/ha) (t/ha) (t/ha) Mixed crop 8.1 8.1 7.5 Cassava 6.5 10.9 11.5 Bamboo 70.1 45,1 46.2 (fallow stage) The total biomass production in six year cycle of the f i r s t and second rotation under the extended cassava cropping accounted for 76 and 79% of the total production in the traditional practice, respectively. Although this analysis showed an increase cassava production during the second rotation, i t i s very unlikely to happen in reality, because of other growth factors which were not counted here. Moreover, the reduction i n bamboo l i t t e r production reduced the amount of nutrients returned to the s o i l , and, therefore, reduced the ava i l a b i l i t y of nutrients for the subsequent cropping. The assumption that was made here was very simple and based on the direct relationship between uptake and production. Considering the less amount of nutrients taken by mixed crops in the second rotation, i t was assumed that the difference was used by cassava, which then led to the increasing production. This proved the complexity of the system and that a simple tabular analysis alone was not appropriate enough in order to explain the sustainability of the system. There i s a definite need to include computer simulation models in the future research. 240 9.6. DISCUSSION AND CONCLUSIONS The reduction in biomass production as the result of extending the period of cropping in the two scenarios indicated that in the absence of additional fertilizer inputs, the traditional pattern was more suitable than either of the modified patterns. For a given length of management cycle, the sustainability of the traditional system was determined by the length of the cropping period, because food crops have a high uptake requirement. The scenario showed that the use of fertilizer was critical for maintaining the sustainability of production. Extended cropping without applying additional inputs to the system would lead to a depletion of the soil nutrients. Long-term sustainability under altered cropping practice might be achieved if losses in harvest removals could be compensated by sufficient fertilizer inputs to the system. Although this scenario only dealt with nitrogen, the reduced biomass production of bamboo and the shorter fallow would cause a reduction in the forest floor accumulation. As a result, the amount of ash after the clearing and burning a mature talun would decrease. This would probably influence the availability of phosphorus and other elements (K, Ca, and Mg) in subsequent rotation. There is clearly a need for field experiments on the effects of various fertilization levels on selected crops before we start altering the existing traditional pattern of the talun-kebun rotation cycle. Moreover, considering the complexity of the system and the need for repeated analysis, computer models will be a more efficient tool for such an analysis than the static inventory or tabular approach. Models such as FORCYTE-11.4 (Kimmins & Scoullar, 1989) could be used for this purpose. This model is able to simulate the growth of trees, bamboo, shrubs and herbs, the growth-determining effects of nutrient availability and light, and all the major management practices involved in the talun-kebun rotation cycle. 241 C H A P T E R 10 S U M M A R Y A N D C O N C L U S I O N S A series of questions regarding the sustainability of the talun-kebun system were formulated in the Introduction of this thesis. The first parts of the study were designed to examine biomass production, accumulation and removals in various stages of the talun-kebun rotation cycle,in order to answer the first subsidiary question of key question # 1: What is the temporal pattern of crop productivity, and what changes in biomass occur during a complete cycle of the bamboo talun-kebun1? The biomass data were then used to calculate nutrient content in each biomass component in each species to give an inventory of nutrients in plants. The latter was important for assessing changes in the site nutrient capital and future site productivity under alternative cropping strategies. Nutrient dynamics in plant were studied by the measurement of litterfall and forest floor mass and nutrient concentrations in each of the talun-kebun stages. Litterfall increased with increasing crop age and bamboo age, but the rapid decomposition of leaf litter of the food crops prevented the accumulation of an ectorganic layer during the cropping stages. The outputs and inputs of nutrients to and from the system were studied by measuring and/or estimating nutrients in precipitation, soil leaching, fertilizer, biological fixation, compost, and losses in the burn. An inventory of nutrients in the mineral soil was made to evaluate changes in site nutrient capitals related to cropping. The total N in the surface 25 cm of mineral soil was 24 times higher than the N held in the aboveground bamboo and about 60 times higher than the N content of the forest floor in the mature talun. Clearing and hoeing increased the amount of available P in the surface 25 cm of the mineral soil. Cassava cropping decreased the content of exchangeable K in the soil. The nutrient data were synthesized into several flow charts (Figures 7.9-7.13) which illustrated the distribution and transfer of nutrients in various stages of the 242 talun-kebun cycle (Subsidiary question # 2: How are the site nutrient resources redistributed during the different stages ofthe cycle?). An input-output balance was used to estimate the gain or loss of nutrients over a complete talun-kebun rotation cycle. The results indicated harvest removals as the major pathway for nutrient losses from the system. A nutrient budget analysis was done to estimate transfer of nutrients in each compartment. Lack of data on fine root production and turnover in the fallow stage of the system is a significant weakness of this study, because fine root turnover represents an important input of nutrients to the forest floor and mineral soil. The fallow stage functioned as the recovery stage of the talun-kebun, during which nutrients was accumulated in the system. The negative balance of mineral nutrients in the surface 25 cm of the mineral soil was thought to be compensated for by weathering inputs which were not estimated because ofthe lack of data. The use of fertilizer during the first year cropping stage was critical in order to obtain a high production. A tabular analysis was done as an initial and speculative assessment of the consequences of extending the period of mixed cropping for nitrogen uptake and biomass production. This assessment was an attempt to answer the subsidiary question # 3: How are the production, biomass and nutrient patterns of each stage changed by departure from the traditional practice, and what is the implication of these changes for future productivity? Extending the period of mixed cropping without increasing the use of fertilizer would lead to a depletion of the site nutrients. A decrease in biomass production would be followed by a decrease in litter production, which in turn would reduce the amount of nutrients returned to the soil. Therefore, the traditional practice could be changed, but higher fertilization levels would be required to ensure sustained production. Besides biophysical knowledge of the talun-kebun system, a socio-cultural 243 understanding of the existing systems from the farmers' point of view would be required before one could make any changes in the traditional practice. Some of the management aspects of the talun-kebun practice were studied. The results of interviewing the farmers about their motivation to continue practising the traditional talun-kebun rotation showed that the farmers had adequate knowledge about the traditional practice, but they lack knowledge of new innovations. Lack of capital to undertake changes was another reason given. The interview was related to subsidiary question # 3 of key question # 2: What factors influence the farmers' decision in selecting their cropping pattern, and what measures are taken to maintain crop production? Work in the talun-kebun fields only took place during the clearing, preparation for cropping, cropping, and harvesting. Therefore, it was considered as secondary to work in ricefields. Most work except that during the clearing and the preparation for cropping was done by family labor. The talun-kebun proved to be an energy efficient cropping system. The ratio of net energy input/output of food crop production in the talun-kebun was 38:1 in comparison to 20:1 obtained by Uhl (1980) in the slash and burn agriculture in Venezuela. This answered the subsidiary question # 1 of the key question # 2: What is the energy efficiency of the bamboo talun-kebun in comparison with traditional shifting cultivation? The sale of the products from the talun-kebun were not considered as the main source of income for the owners, because they usually also own ricefield and engaged in other revenue-producing work. However, the costs of production in the talun-kebun were also limited to labor costs during the cropping stages. No labor work was required during the fallow stages. Therefore, the income from crop and bamboo sales were much higher than the cost of production (Subsidiary question #2 of key question #2: What is the economic yield of the bamboo talun-kebun in 244 comparison with alternative cash crop systems, and what are the economic consequences of changing the traditional pattern?). In contrast to the shifting cultivation practice, observations in the field showed that erosion problems were negligible in the talun-kebun. The farmers already practised contour terracing to prevent erosion. Some farmers have already changed from the traditional system to monoculture cropping and have replaced bamboo with cash crops, but the majority of farmers are still practising the traditional cropping pattern. The interviews showed that most farmers did not want to take risks. Bamboo is considered as the most appropriate perennial crop plant for talun-kebun, because it is usually practised on sloping land which is frequently located far from any water resource. Data on nutrient accumulation in the forest floor during the fallow stage indicated the build up of soil humus and supported the farmers saying "without bamboo the land dies". Although this study tried to cover many aspects of the talun-kebun, the limited resources and manpower that were available at the time of the study resulted in several shortcomings. For example, the lack of data on litter decomposition and fine root turnover which are very important in estimating transfer of nutrients from plant to the mineral soil, is a significant shortcoming of the budget analysis. The short period of the lysimeter and precipitation study resulted in a low-intermediate confidence level of data. The exclusion of Albizia and other trees from the biogeochemistry of the system may have caused discrepancy in nutrient budget. The lack of weathering input data, and of empirical evidence on nutrient limitation on growth are other weaknesses that suggest the need for further work. The present study provides a basis for the following recommendations concerning future studies on nutrient cycling and management aspects of tropical 245 agroforestry systems: 1. Study on litter decomposition should be conducted for both annual and perennial crops in various stages of the rotation cycle (over a number of years for bamboo and other trees). 2. Fine root distribution and turnover rate should be examined in each of the talun-kebun stage. 3. The longevity and decomposition of bamboo rhizomes should be studied. 4. Fertilizer experiments with different fertilization levels should be conducted for various species mixtures. 5. Dominant tree species should be included in the biomass and nutrient studies. 6. Precipitation and lysimeter studies should be conducted in a longer time frame (at least 2 years). 7. If the site is dominated by nitrogen-fixing species, reliable data on nitrogen fixation and denitrification rates should be sought or laboratory experiments should be conducted on these two aspects. 8. Weathering inputs should be considered in the analysis. 9. The effect of fire on the burn spots (under the piles) on crop production should be analyzed. 10. The economic analysis should be broadened to include the role of the system in relation to other sources of income in the household and village economy. 2 4 6 REFERENCES Altieri,M.A. and M.Liebman. 1986. Insect, Weed, and Plant Disease Management in Multiple Cropping Systems, pp 183-218. In Francis.C.A. 1986. Multiple Cropping Systems, MacMillan Publ.Co, New York. Andrews,D.J. and A.H.Kassam. 1976. The importance of multiple cropping in increasing gworldfood supplies, pp 1-10. In R.I.Papendick, P.A.Sanchez, and G.B.Triplett (eds), Multiple cropping. Special Publication 27, American Society of Agronomy, Madison. Baskerville.G.L. 1982. Use of logarithmic regression in the estimation of plant biomass. Can.J.For.Res.2:49-53. Beckett,P.H.T. and R.Webster. 1971. Soil Variability: A Review. Soils and Fertilizers 34(1):1-15. BPS, 1985. Statistik Indonesia. Biro Pusat Statistik, Jakarta, Indonesia. Bene,J.G., H.W.Beall, and A.Cote. 1978. Trees, food and people: Land management in the tropics. International Development Agency, Ottawa. Berish.C.W. 1982. Root biomass and surface area in three successional tropical forests. Can. J.For.Res. 12(3): 699-704. Booth,A. and R.M. 1984. Labor Absorption in Agriculture: Theoretical Analysis and Empirical Investigation. Oxford University Press, Delhi. 327 pp. Brasher.B.R., D.P.Franzmeier, V.Valassis, and S.E.Davidson. 1966. Soil clods for bulk density and water retention measurements. Soil Sci 101(2). Bray.J.R. and E.Gorham. 1964. Litter Production in Forests of the World. Adv.Ecol.Res. 2:101-158. Bremner,J.M. 1965. Nitrogen availability indexes, pp 1324-1345. In C.A.Black (ed.), Methods of soil analysis. Brown.B. 1982. Productivity and herbivory in high and low diversity tropical successional ecosystems in Costa Rica. PhD dissertation, University of Florida, Gainesville. Bruijnzel.L.A. 1982. Hydrological and biogeochemical aspects of man-made forests in south-central Java, Indonesia. Thesis:Final Report,Vol.9, Nuffic Serayu Valley Project,ITC/GUA/VU/l, the Hague, Netherlands. Burgos,C.F. 1980. Soil Related Intercropping Practices in Cassava Production, pp 71-85. In Weber,E.J., J.C.Toro, and M.Graham (eds), Cassava Cultural Practices. Proc.of a workshop held in Salvador, Bahia, Brazil. Caroll,D. 1970. Rock weathering. Plenum Press, New York. 200 pp. Chang,J. 1968. Climate and Agriculture. Aldine Publ.Co. Chicago, Illinois. 247 Christanty.L., J.Iskandar, O.S.Abdoellah, and G.G.Marten. 1986. Traditional Agroforestry in West Java: The Pekarangan (Homegarden) and Kebun-Talun (Annual-Perennial Rotation) Cropping Systems, pp.132-158. In G.G.Marten (ed), Traditional Agriculture in South East Asia: A Human Ecology Perspective. Westview Press, Boulder. Cunningham,R.K 1963. The effect of clearing a tropical forest soil. Journal of Soil . Science 14:334-345. Dali.J. 1980. The effect of fertilization on bamboo growth in Borisalu, South Sulawesi, Indonesia. Dept.of Silviculture, Forest Research Institute, Bogor, Indonesia (in Indonesian). 19 pp. Dennis,E.A. 1977. Nodulation and Nitrogen Fixation in Legumes in Ghana.pp 217-232. In Ayanaba,A. and P.J.Dart (eds.), Biological Nitrogen Fixation in Farming Systems of the Tropics. John Willey & Sons, Chichester. Direktorat Gizi Departemen Kesehatan RI. 1981. Food composition tables. Penerbit Bhatara Karya Aksara, Jakarta (in Indonesian). Ewel.J.J. 1976. Litter fall and leaf decomposition in a tropical forest succession in Eastern Guatemala. Journal of Ecology 64:816-829. Ferreira,M.D.G.M. 1984. An Analysis of the Future Productivity of Eucalyptus grandis plantations in the"Cerrado" Region in Brazil: A Nutrient Cycling Approach. PhD thesis, University of British Columbia. Vancouver. 230 pp. Finnegan.E.J. 1981. The use of agrisilviculture as a resource conservation and rural community development method in the Tropical Wet Forest in Colombia. PhD thesis, Dept.of Natural Resources,Cornell University, Ithaca, New York. Fox.D.J. and K.E.Guire. 1976. Documentation of Midas. Statistical Res.Lab.Univ. of Michigan. 203 pp. Hadikusumah,H.Y. 1982. The role of homegarden, talun and ricefield in supporting the daily needs of the rural people in Legokkole hamlet, Sadu village, West Java, Indonesia. M.S. thesis, Dept.of Biology, Padjadjaran Univ, Bandung, Indonesia (in Indonesian). Hidayat,A. 1978. Soil Chemical Analysis. Japan International Cooperation Agency (JICA) Indonesia-Japan Joint Food Crop Research Program, Bogor, Indonesia. 105 pp. Howeler, R.H. 1980. Soil Related Cultural Practices for Cassava, pp 59-69. In Weber.E.J., J.C.Toro, and M.Graham (eds), Cassava Cultural Practices. Proc.of a workshop held in Salvador, Bahia, Brazil. Howeler,R.H. and L.F.Cadavid. 1983. Accumulation and distribution of dry matter and nutrients during a 12-month growth cycle of cassava. Field Crops Research^: 123-139. Jordan.C.F. 1985. Nutrient Cycling in Tropical Forest Ecosystems. Principles and their Application in Management and Conservation. John Wiley & Sons. 248 Chichester. Jordan,C.F. 1987. Shifting Cultivation, pp 9-23. In Jordan.C.F. (ed.), Amazonian Rain Forests Ecosystem Disturbance and Recovery. Ecological Studies 60, Springer-Verlag, New York. Jordan,C.F., F.Golley, J.Hall, and J.Hall. 1980. Nutrient scavenging of rainfall by the canopy of an Amazonian Rain Forest. Biotropica 12(l):61-66. Kass,D.C.L. 1978. Polyculture cropping systems:Review and analysis. Cornell International Agriculture Bulletin No.32, Ithaca, New York. Kelly,J.L. 1975. The role of crop yields in shifting cultivation.PhD dissertation, University of Kansas, Lawrence. Kenworthy, J.B. 1971. Water and nutrient cycling in a tropical rain forest. In J.R.Flenly (ed.) The water relations of Malesian forests, pp 49-59. Transactions of the First Aberdeen-Hull Symposium on Malesian Ecology. Institute for Southeast Asian Biology, University of Aberdeen, Aberdeen. Kimmins,J.P. 1987. Forest Ecology. MacMillan Publ.Co. New York. 531 pp. Koh.B.H. 1978. The flow of energy in a Orang Asli community (A case study of a Temuan Community at sungai Lui, Ulu Langat, Selangor). Dept.of anthropology and Sociology, University of Malaya, Kualalumpur. Kondas,S. 1981. Bamboo biology, culm potential and problems of cultivation, pp 184-190. In Higuchi,T. (ed.), Bamboo Production and Utilization. Proc.of the Congress Group 5.3A, Production and Utilization of Bamboo and Related Species, XVIIIUFRO World Congress, Kyoto, Japan. Krumlik,G.J. 1979. Comparative study of nutrient cycling in the subalpine Mountain Hemlock Zone of british Columbia. PhD thesis.University of British Columbia, Vancouver. 195 pp. Lagemann.J. 1977. Traditional African Farming Systems in Eastern Nigeria. Weltforum Verlag, Munchen. 269 pp. Luck, P.E. 1965. Dolichos lablab: a valuable grazing crop. Qd Agric J.(91): 308-309. Lundgren.B. 1978. Soil conditions and nutrient cycling under natural and plantation forests in Tanzanian Highlands. Report in Forest Ecology and Forest Soils 31, Swedish University of agricultural Sciences. Madgwick,H.A.I. 1983. Above-ground weight of forest plots- comparison of seven methods of estimation. N.Z. J.For.Sci. 13(1):100-107. Manokaran,N. 1980. The nutrient contents of precipitation, throughfall, and stemflow in a lowland tropical rain forest in peninsular Malaysia. The Malaysian Forester 43:266-289. McCaulley,D. 1982. Soil Erosion and land-use patterns among upland farmers in the Cimanuk watershed of West Java, Indonesia. East West Center Environment and Policy Institute. Working Paper, Honolulu. 62pp. 249 Marten.G.G., and D.M. Saltman, 1986. The Human Ecology Perspective, pp. 20-53. In G.G.Marten (ed.), Traditional Agriculture in South East Asia: A Human Ecology Perspective. Westview Press, Boulder. McClure,F.A. 1966. The Bamboos: A Fresh Perspective. Harvard Univ.Press, Cambridge. 347 pp. Mountford.M.D. and R.G.H. Bunce. 1973. Regression sampling with allometrically related variables, with particular reference to production studies. Forestry 46:203-212. Norman.M.T.J. 1978. Energy inputs and outputs of subsistence cropping systems in the tropics. Agroecosystems 4:355-366. Norman.M.J.T. 1979. Annual Cropping Systems in the Tropics. An Introduction. Univ.Press of Florida, Gainesville. Nye,P.H. and D.J.Greenland. 1960. The Soil under Shifting Cultivation. Commonwealth Bureau of Soils, Harpenden, England. Nye,P.H. and D.J.Greenland. 1964. Changes in the soil after clearing a tropical forest. Plant and Soil 21:101-112. Okoli.P.S.O. and G.F.Wilson. 1986. Response of cassava (Manihot esculenta Crantz) to shade under field conditions. Field Crops Research 14(4):349-360. Popenoe,H.L. 1960. The influence of shifting cultivation on natural soil constituents in Central America. PhD thesis, University of Florida, gainesville. Rappaport,R.A. 1971. The flow of energy in an agricultural society. Scientific American 224(3):117-132. Root,R.B. 1973. Organization and a plant-arthropod association in simple and diverse habitats: The fauna of collards (Brassica oleracea), Ecol.Monogr. 43:95-124. Sabhasri.S. 1978. Effects of Forest Fallow Cultivation on Forest Production and Soil, pp 160-174. In P.Kundstadter, E.C.Chapman, and S.Sabhasri (eds.), Farmers in the forest./ University of Hawaii, Honolulu. Sanchez.P.A. 1976. Properties and Management of Soil in the Tropics. John Wiley & Sons, New York. 618 pp. Santantonio,D., R.K.Hermann and W.S.Overton. 1977. Root biomass studies in forest ecosystems. Pedobiologia, 17(5):1-31. Satoo,D. and H.A.I.Madgwick. 1982. Forest Biomass. Martinus Nijhoff, The Haque. 151 pp. Scaaffhausen.R.v. 1963. Dolichos lablab or Hyacinth Bean: Its uses for Food and Soil Improvement. Economic Botany : 146-153. 250 Seth.S.K., O.N.Kaul, and A.C.Gupta. 1963. Some observations on nutrient cycle and return of nutrients in plantations at new forest. Indian Forester :90-99. Sharma.Y.M.L. 1980. Bamboos in the Asia Pacific Region. pp99-120. In Lessard,G. and A.Choinard (eds.), Bamboo Research in Asia. Proc.of a workshop held in Singapore. IDRC, Ottawa. Sirie, M.S. 1981. The Talun Structure: A Research in Sadu Village. Minor M.S. Thesis. Dept. of Biology, Padjadjaran University, Bandung, Indonesia, (in Indonesian). Soemarwoto.O. 1980. Interrelations among population, resources, environment and development in the ESCAP Region with special reference to Indonesia. Ecology and Development Publication No.7, Institute of Ecology, Padjadjaran Univ., Bandung, Indonesia. Soemarwoto,0., L.Christanty, Henky, Y.H.Herri, J.Iskandar, Hadyana, and Priyono. 1985. The Talun-kebun: A Man-made Forest Fitted to Family Needs. Food and Nutrition Bull. 7(3):48-51. Stark,N.M. and CFJordan. 1978. Nutrient retention by the root mat of an Amazonian rain forest. Ecology 59:434-437. Steiner,K.O. 1984. Intercropping in Tropical Smallholder Agriculture with special reference to west Africa (2nd ed.) Deutsche Gesellscaht fur Technische Zusammenarbeit (GTZ) Gmb H.Eschborn. Swift,M.J., O.W.Heal, and J.M.Anderson. 1979. Decomposition in terrestrial ecosystems (Studies in Ecology,Vol 5). University of California Press, Berkeley. Terra,G.J.A. 1953. The distribution of mixed gardening in Java. Landbouw 25:163-223. Terra,G.J.A. 1958. Farm systems in Southeast Asia. Netherland Journal of Agricultural Sciences 6:157-182. Toky,O.P. and P.S.Ramakrishnan. 1981. Run-off and infiltration losses related to shifting agriculture (Jhun) in Northeastern India. Environmental Conservation 8(4):313-321. Trenbath.B.R. 1974. Biomass productivity of mixture. Adv.Agron. 26:177-210. Trenbath.B.R. 1981. Light use efficiency of crops and the potential for improvement through intercropping. In International Workshop on Intercropping. International Crops Research Institute for the Semi-Arid Tropics, Patanchertu, India, pp. 141-154. Tritton,L.M. and J.W.Hornbeck. 1982. Biomass equations for major tree species of the northeast. USDA Forest Service Northeastern For.Exp.Sta., Gen.Tech.Rep. NE-69. 46 pp. Tsutsumi.T., K.Yoda, P.Sahunalu, P.Dhanmanonda, and B.Prachaiyo. 1983. Forest: Felling, burning and regeneration, pp 13-61. In KKyuma and C.Pairintra 251 (eds.), Shifting cultivation. An experiment at Nam Phrom, northeast Thailand, and its implications for upland farming in the Monsoon Tropics. Faculty of Agriculture, Kyoto University, Kyoto. Uchimura.E. 1980. Bamboo Cultivation. ppl51-60. In Lessard.G. and A.Choinard (eds.), Bamboo Research in Asia. Proc.of a workshop held in Singapore. IDRC, Ottawa. Ueda,K. 1960. Studies on the Physiology of Bamboo: With a special reference to practical application. Resource Bureau Reference Data No.34. Resource Bureau, Science and Technics Agency, Prime Minister's office, Tokyo, Japan. 137 pp. Uhl,C. 1980. Studies of Forest, Agricultural, and Successional Environment in the Upper Rio Negro Region of the Amazon Basin. PhD dissertation, Michigan State University. van Bemmelen.R.W. 1949. The Geology of Indonesia. Vol IA: General Geology of Indonesia and Adjacent Archipelago. Gov't Printing Office. The Hague, pp 25-28. Vitousek.P.M. and R.L.Sanford.Jr. 1986. Nutrient cycling in moist tropical forest. Ann.Rev.Ecol.Syst. 17: 137-167. Vogt,K.A., C.C.Grier, and D.J.Vogt. 1986. Production, turnover, and nutrient dynamics of above-and below-ground detritus of world forests. Advances in Ecol.Res. 15:303-366. Wang.T.T. 1981. The Aboveground Biomass and Net Production of a Moso-bamboo (Phyllostachys pubescens Mazel ex H . de Leh.) Stand in Taiwan, pp. 125-130 . In Higuchi.T. (ed.), Bamboo Production and Utilization. Proc.of the Congress Group 5.3A, Production and Utilization of Bamboo and Related Species, XVII IUFRO World Congress, Kyoto, Japan. Webster,C.C. and P.N.Wilson. 1980. Agriculture in the Tropics. Longman, London. Widjaya.E.A. 1980. Indonesia, Country Reports, pp 63-68. In Lessard.G. and A.Choinard (eds.), Bamboo Research in Asia. Proc.of a workshop held in Singapore. IDRC, Ottawa. Wiersum.KF. 1979. Introduction to Principles of Forest Hydrology and Erosion: with a special reference to Indonesia. Inst.of Ecology. Padjadjaran University, Bandung, Indonesia. 76 pp. Willey.R.W. 1979a. Intercropping-Its importance and research needs. I.Competition and yield advantages. Field Crop Abstracts 32(1): 1-10. Willey,R.W. 1979b. Intercropping-Its importance and research needs. II.Agronomy and research approaches. Field Crop Abstracts 32(2):73-85. Yadav,J.S.P. 1963. Site and Soil Characteristics of Bamboo Forests. Indian Forester :177-193. 252 Yoshida.S., D.A.Forno, J.H.Cock, and K.A.Gomez. 1976. Laboratory manual for Physiological Studies of Rice. The international Rice Institute. Los Banos, Laguna, Philippines, pp 17-33. Zinke,P.J., S.Sabhasri, and P.Kundstadter. 1978. pp 134-159. In P.Kundstadter, E.C.Chapman, and S.Sabhasri (eds.), Farmers in the forest. University of Hawaii, Honolulu. 2 5 3 APPENDIX 1. Botanical terminology for Bamboo (McClure, 1966; Ueda, 1960) Culm Monopodial Rhizome Sympodial ;(L.culmus, s t a l k , stem). A segmented a e r i a l a x i s t h a t emerges from a rhizome, and forms a p a r t of gramineous p l a n t . :a stem of a s i n g l e and contiuous a x i s . The rhizome which has one bud at each nodedevelops monopodially each year. Some of these buds grow i n t o sprouts which appear on the ground at a c e r t a i n d i s t a n c e , and these newly grown culms take a s i n g l e - c u l m f ormation. : (Gr. rhizoma, a mass of roots) . An i n d i v i d u a l component bransh of the subterranean system of segmented axes t h a t c o n s t i t u t e the r o o t s t o c k of a bamboo p l a n t . :the apex of the rhizome which has nodes, but no buds, p r o t r u d e s out of the ground and grow i n t o a culm. In the f o l l o w i n g year, the bud on the b a s a l p a r t of the culm develops i n t o a sho r t thizome, which p r o t r u d e s out of the ground t o make a secondary culm, thus forming a clump culm. APPENDIX 2. Estimates of fine root turnover Assumptions: 1. Final year root l i t t e r f a l l = 2.2 t h a - 1 (the difference between forest floor mass and the total estimated l i t t e r f a l l over 4 years of fallow). See Chapter 6. 2. Turnover rate = 1 year The ratio of leaf l i t t e r f a l l to fine root •= 3.8/2.2 = 1.73 The ratio i s applied to leaf l i t t e r f a l l to estimate root l i t t e r f a l l i n other years: Year Leaf L i t t e r f a l l Root l i t t e r f a l l (t ha - 1) (t ha - 1) 3 2.2 1.27 4 2.9 1.68 5 3.2 1.85 6 3.8 2.2 Total in 4 years 7.0 3. Turnover rate = 2 years The ratio of leaf l i t t e r f a l l to fine root l i t t e r f a l l = 3.8/1.1 = 3.45 The ratio i s applied to leaf l i t t e r f a l l to estimate root l i t t e r f a l l i n other years Year Leaf l i t t e r f a l l Root l i t t e r f a l l (t ha - 1) (t ha - 1) 3 2.2 0.64 4 2.9 0.84 5 3.2 0.93 6 3.8 1.1 Total i n 4 years 3.51 APPENDIX 3 Soil organic matter concentration (%) in various s o i l depths under different stages of the talun-kebun cycle. Numbers within brackets represent standard deviation. Stages Organic matter concentration (%) Soil depth (cm) 0 - 5 5 - 25 25 - 45 45 - 75 Mature talun 7.15a (0.31) Clearing & hoeing 7.76b (0.35) End of l e t yr 7.04° cropping (0.62) End of 2 n d yr 6.23d cropping (0.49) End of 1 s t yr 6.64cd fallow (0.57) 5.04a 3.03a 1.39a (0.29) (0.58) (0.08) 6.12b 3.06a 1.40a (0.59) (0.31) (0.16) 5.46= 3.08a 1.40a (0.65) (0.27) (0.13) 5.02a 3.04a 1.36a (0.38) (0.31) (0.16)) 5.26 a o d 3.08a 1.36a (0.56) (0.43) (0.11) One-way analysis of variance was performed separately for each s o i l depth. Mean followed by the same letter do not d i f f e r significantly at P <0.05. APPENDIX 3 Nutrient concentrations i n various components of plant species and i n forest f l o o r 257 T a b l e A4.1. The l e a n c o n c e n t r a t i o n (X) o f N, P, K, Ca, and Mg i n r o o t , s t e a , and f o l i a g e o f c u c u i b e r grown d u r i n g t h e f i r s t y e a r c r o p p i n g s t a g e o f a talun-kebut c y c l e . N u s b e r s w i t h i n b r a c k e t s r e p r e s e n t s t a n d a r d d e v i a t i o n . Age Root Stem F o l i a g e ( d a y s ) N P K Ca Mg N P K Ca Hg N P K Ca Mg 22 1.91 0.46 7.02 0.15 0.45 (0.13) (0.03) (0.17) (0.02) (0.06) 26 1.99 0.51 6.50 0.32 0.50 (0.15) (0.07) (0.30) (0.04) (0.08) 32 1.95 0.59 5.42 0.48 0.49 (0.13) (0.06) (0.20) (0.03) (0.06) 38 2.18 0.43 5.28 0.49 0.46 (0.24) (0.06) (0.32) (0.06) (0.05) 45 1.61 0.65 4.34 0.59 0.67 (0.14) (0.09) (0.12) (0.08) (0.10) 64 1.41 0.15 3.54 0.28 0.34 (0.09) (0.04) (0.31) (0.03) (0.03) 2.18 0.38 4.30 1.40 0.86 (0.16) (0.03) (0.16) (0.10) (0.11) 2.66 0.46 5.93 1.09 0.94 (0.12) (0.05) (0.50) (0.14) (0.11) 2.49 0.52 7.48 1.02 0.82 (0.23) (0.03) (0.36) (0.10) (0.09) 1.81 0.52 6.27 0.75 0.55 (0.13) (0.08) (0.25) (0.10) (0.06) 1.08 0.17 3.50 0.74 0.45 (0.10) (0.02) (0.22) (0.11) (0.05) 0.87 0.09 1.76 0.47 0.36 (0.04) (0.02) (0.22) (0.08) (0.02) 4.14 0.48 3.15 2.47 1.30 (0.12) (0.02) (0.35) (0.22) (0.12) 4.61 0.79 3.25 2.85 1.63 (0.26) (0.09) (0.20) (0.17) (0.11) 4.45 0.56 2.37 2.78 1.37 (0.29) (0.11) (0.09) (0.24) (0.11) 4.86 0.46 2.78 2.21 1.16 (0.26) (0.09) (0.11) (0.18) (0.13) 2.99 0.20 1.68 1.41 0.84 (0.28) (0.04) (0.16) (0.16) (0.09) 2.21 0.14 1.12 1.06 0.65 (0.16) (0.02) (0.14) (0.10) (0.07) Table A4.2. The aean concentration (X) of N, P, K, Ca, and Hg i n flower and f r u i t of cucuiber grown during the f i r s t year cropping stage of a talun-kebun c y c l e . Nuabers within brackets represent standard d e v i a t i o n . Age Flower F r u i t (days) N P K Ca Hg N P K Ca Hg 22 26 32 38 4.57 0.88 4.73 0.76 0.76 (0.20) (0.14) (0.22) (0.07) (0.03) 45 2.05 0.43 4.03 0.40 0.49 (0.11) (0.05) (0.11) (0.09) (0.10) 4.29 0.77 5.18 0.48 0.57 (0.11) (0.14) (0.36) (0.08) (0.06) 4.05 0.73 4.27 0.56 0.55 (0.13) (0.07) (0.20) (0.12) (0.12) 60 2.04 0.38 4.02 0.39 0.50 4.05 0.74 4.35 0.59 0.57 (0.10) (0.05) (0.11) (0.05) (0.03) (0.06) (0.04) (0.04) (0.05) (0.01) 259 Table A3.3. The sean co n c e n t r a t i o n (I) of N, P, K, Ca, and Mg i n root, s t e a , and f o l i a g e of b i t t e r solanua grown during the f i r s t year cropping stage of a talan-kebun c y c l e . Nuibers within brackets represent standard d e v i a t i o n . Age Root S t e i F o l i a g e (days) N P K Ca fig N P K Ca Mg N P K Ca Hg 70 0.76 0.04 1.40 0.05 0.15 1.10 0.05 3.23 0.45 0.34 6.38 0.55 3.84 0.84 0.72 (0.05) (0.01) (0.18) (0.01) (0.02) (0.16) (0.01) (0.16) (0.08) (0.06) (0.15) (0.13) (0.14) (0.15) (0.08) 86 1.00 0.07 2.33 0.05 0.21 1.19 0.10 4.82 0.53 0.50 6.38 0.34 3.40 1.08 0.70 (0.04) (0.02) (0.05) (0.01) (0.02) (0.05) (0.02) (0.17) (0.07) (0.07) (0.11) (0.02) (0.25) (0.07) (0.09) 100 1.05 0.08 2.17 0.05 0.49 1.40 0.07 4.20 0.64 0.54 4.92 0.30 2.62 1.56 0.89 (0.05) (0.02) (0.05) (0.00) (0.06) (0.07) (0.01) (0.24) (0.07) (0.05) (0.12) (0.04) (0.15) (0.15) (0.11) 130 1.01 0.05 1.26 0.14 0.46 1.76 0.05 3.64 0.74 0.57 4.05 0.28 1.80 1.56 0.57 (0.05) (0.00) (0.05) (0.02) (0.05) (0.13) (0.01) (0.09) (0.09) (0.06) (0.28) (0.01) (0.16) (0.11) (0.10) 160 0.79 0.05 0.92 0.11 0.67 0.81 0.03 1.63 0.66 0.30 2.35 0.20 1.59 0.60 0.30 (0.02) (0.01) (0.09) (0.02) (0.10) (0.09) (0.00) (0.08) (0.11) (0.06) (0.06) (0.02) (0.07) (0.07) (0.05) 2 6 0 Table A4.4. The lean concentration (X) of N, P, K, Ca, and Hg in flower and f r u i t of bitter solanun grown during the f i r s t year cropping stage of a taluv-ltebui) cycle. Numbers within brackets repre-sent standard deviation. Age Flower Fruit (days) — N P K Ca Hg N ; P K Ca Hg 70 3.27 (0.04) 0.33 (0.03) 2.27 (0.08) 0.64 (0.02) 0.45 (0.05) 2.53 (0.16) 0.18 (0.02) 2.46 (0.06) 0.27 (0.02) 0.27 (0.04) 86 S.S3 (0.11) 0.60 (0.02) 3.56 (0.10) 0.80 (0.07) 0.69 (0.10) 4.51 (0.13) 0.43 (0.02) 3.69 (0.12) 0.41 (0.08) 0.51 (0.05) 100 5.02 (0.09) 0.48 (0.01) 3.33 (0.13) 1.00 (0.11) 0.72 (0.07) 4.03 (0.06) 0.34 (0.03) 3.83 (0.13) 0.41 (0.04) 0.45 (0.05) 130 6.01 (0.07) 0.44 (0.03) 3.00 (0.18) 0.66 (0.07) 0.71 (0.07) 4.22 (0.05) 0.34 (0.14) 3.54 (0.10) 0.39 (0.05) 0.50 (0.05) 160 — ~ ~ 3.18 (0.09) 0.14 (0.03) 2.77 (0.09) 0.33 (0.05) 0.30 (0.05) 261 Table A4.S. The sean concentration (X) of N, P, K, Ca, and Hg in root, stea, and foliage of hyacinth bean grown during the f i r s t year cropping stage of a talun-kebun cycle. Age Root Stei Foliage (days) N P K Ca Hg N P K Ca Hg N P K Ca Hg 45 2.43 0.16 1.18 0.47 0.34 2.58 0.26 1.10 0.73 0.34 6.39 0.20 1.51 1.94 0.65 (0.14) (0.02) (0.04) (0.07) (0.04) (0.08) (0.03) (0.06) (0.05) (0.06) (0.13) (0.02) (0.07) (0.11) (0.05) 62 2.86 0.15 1.79 0.71 0.27 3.42 0.25 2.44 0.55 0.33 4.67 0.20 1.63 1.31 0.49 (0.09) (0.01) (0.07) (0.04) (0.02) (0.06) (0.04) (0.06) (0.05) (0.05) (0.06) (0.02) (0.03) (0.05) (0.03) 84 2.30 0.06 1.43 0.73 0.34 3.58 0.18 1.74 0.53 0.45 4.76 0.18 1.27 1.48 0.56 (0.15) (0.01) (0.05) (0.03) (0.05) (0.07) (0.02) (0.07) (0.04) (0.05) (0.07) (0.02) (0.03) (0.06) (0.05) 125 2.74 0.10 1.35 0.80 0.62 4.33 0.06 1.14 0.95 0.61 4.31 0.20 1.05 2.29 0.43 (0.10) (0.01) (0.05) (0.07) (0.05) (0.10) (0.01) (0.03) (0.08) (0.05) (0.05) (0.02) (0.05) (0.08) (0.05) 160 2.61 0.07 0.73 1.17 0.59 2.56 0.07 0.68 0.92 0.78 3.54 0.09 0.52 3.90 0.29 (0.04) (0.01) (0.05) (0.06) (0.03) (0.10) (0.01) (0.09) (0.04) (0.09) (0.07) (0.01) (0.06) (0.14) (0.04) 262 Table A4.6. The lean concentration (X) of N, P, K, Ca, and Hg i n flower and f r u i t of hyacinth bean grown during the f i r s t year cropping stage of a talun-kebun c y c l e . Nuabers within brackets represent standard d e v i a t i o n . Age Flower F r u i t (days) N P K Ca Hg N P K Ca Hg 45 62 84 6.48 2.03 2.03 0.47 0.33 (0.06) (0.10) (0.05) (0.05) (0.04) 125 5.43 2.42 2.42 0.63 0.34 4.98 0.35 1.94 0.27 0.48 (0.10) (0.07) (0.10) (0.04) (0.04) (0.07) (0.14) (0.09) (0.05) (0.06) 160 3.51 0.93 0.93 0.07 0.35 5.09 0.22 1.44 0.24 0.37 (0.06) (0.08) (0.07) (0.01) (0.06) (0.07) (0.02) (0.06) (0.05) (0.02) 263 Table A4.9. The lean c o n c e n t r a t i o n (Z) of N, P, K, Ca, and Hg i n the aboveground banboo component over the age sequence of a talun-kebun. Nuibers wi t h i n brackets represent stnadard d e v i a t i o n . Age C u l l F o l i a g e Branch d o n t h s ) N P K Ca Hg N P K Ca Hg N P K Ca Hg 16 0.18 0.16 0.70 0.36 0.40 2.04 0.28 1.13 O.SO 0.57 0.52 0.24 0.86 0.23 0.25 (0.02) (0.03) (0.05) (0.05) (0.04) (0.06) (0.04) (0.07) (0.06) (0.03) (0.12) (0.04) (0.04) (0.05) (0.04) 24 0.18 0.13 0.65 0.33 0.36 2.02 0.25 1.01 0.48 0.54 0.44 0.20 0.80 0.20 0.23 (0.01) (0.03) (0.07) (0.04) (0.04) (0.13) (0.04) (0.09) (0.06) (0.06) (0.09) (0.02) (0.05) (0.03) (0.03) 36 0.17 0.14 0.60 0.29 0.32 1.29 0.24 1.00 0.43 0.50 0.43 0.20 0.70 0.17 0.20 (0.02) (0.03) (0.06) (0.03) (0.05) (0.04) (0.04) (0.14) (0.04) (0.07) (0.12) (0.04) (0.04) (0.02) (0.03) 72 0.16 0.11 0.56 0.25 0.28 1.22 0.20 0.95 0.26 0.32 0.31 0.12 0.65 0.11 0.16 (0.01) (0.02) (0.05) (0.05) (0.03) (0.07) (0.02) (0.07) (0.02) (0.03) (0.08) (0.03) (0.04) (0.02) (0.02) 264 Table A4.10. The lean c o n c e n t r a t i o n (Z) of N, P, K, Ca, and Hg in the belowground batboo component over the age sequence of a talun-kebun. Numbers with i n brackets represent standard d e v i a t i o n . Age Rhizoie Coarse root Fine root (fionths) N P K Ca Hg N P K Ca Hg N P K Ca Hg 16 0.52 0.10 0.66 0.20 0.29 0.44 0.11 0.62 0.19 0.23 0.45 0.12 0.59 0.18 0.20 (0.05) (0.01) (0.03) (0.03) (0.02) (0.0B) (0.03) (0.06) (0.04) (0.05) (0.04) (0.02) (0.05) (0.03) (0.01) 24 0.44 0.12 0.62 0.18 0.25 0.53 0.13 0.58 0.16 0.20 0.50 0.14 0.55 0.15 0.18 (0.05) (0.01) (0.03) (0.03) (0.04) (0.05) (0.03) (0.05) (0.03) (0.03) (0.05) (0.03) (0.04) (0.03) (0.03) 36 0.43 0.15 0.58 0.14 0.23 0.36 0.12 0.55 0.14 0.17 0.52 0.13 0.50 0.12 0.15 (0.04) (0.01) (0.02) (0.03) (0.04) (0.06) (0.02) (0.04) (0.02) (0.03) (0.03) (0.02) (0.04) (0.03) (0.02) 72 0.31 0.11 0.50 0.13 0.19 0.35 0.11 0.48 0.12 0.14 0.42 0.11 0.42 0.10 0.12 (0.02) (0.03) (0.04) (0.02) (0.02) (0.02) (0.03) (0.04) (0.02) (0.03) (0.04) (0.02) (0.03) (0.01) (0.02) 265 Table A 4 . l l . The lean concentration (X) of N, P, K, Ca, and fig in weeds at various stages of the talan-kebut cycle. Numbers within brackets representstandard deviation. Stage N P K Ca Hg 1"* yr cropping 2.56 (0.05) 0.19 (0.01) 1.53 (0.07) 0.19 (0.01) 0.60 (0.01) 2 n d yr cropping 2.2B (0.05) 0.17 (0.02) 1.39 (0.05) 0.16 (0.02) 0.55 (0.05) Early fallow 1.96 (0.07) 0.13 (0.01) 1.14 (0.04) 0.14 (0.03) 0.37 (0.04) Nature taint 0.83 (0.04) 0.09 (0.01) 0.96 (0.03) 0.12 (0.01) 0.36 (0.03) r 266 APPENDIX 5. The mean concentration (%) of N, P, K, Ca, and Hg i n the forest floor/ectorganic layer at various stages of the talun-kebun cycle. Numbers within brackets represent standard deviation. Stage N P K Ca Mg 1 s t yr cropping 2.66 0.12 0.33 1.87 0.42 (0.61) (0.02) (0.11) (0.29) (0.25) 2 n d yr cropping 1.95 0.07 0.38 0.52 0.34 (0.16) (0.02) (0.12) (0.04) (0.11) Early fallow 1.39 0.13 0.70 0.27 0.29 (0.28) (0.04) (0.19) (0.07) (0.05) Mature talun Surface l i t t e r 0.91 0.09 0.66 0.23 0.21 (0.11) (0.03) (0.14) (0.07) (0.02) Fragmented l i t t e r 0.87 0.10 0.44 0.19 0.17 (0.12) (0.02) (0.10) (0.05) (0.00) Humus 0.39 0.05 0.44 0.18 0.15 (0.03) (0.02) (0.09) (0.05) (0.03) 267 APPENDII 5.1. GENERAL QUESTIONNAIRES To be valid as a respondent, the person mist have a talai-iebat f ie ld . Then ask whether or not he is presently engaged in the talan-keban activities (clearing, cropping). This answer wi l l be used to provide a l i s t of farters eligible for detailed interviews. Nate of Head of Household: Hailet/Village: Case nutber: Date: Enumerator: 1. The household Note: Al l questions are directed to the head of the household. List a l l tethers of the household starting with the head of the household, his wife and children followed by relatives. A. List of household tetbers living with the head of the household. No. Relationship to Sex Age household head H or F No. of years Schooling Occupation Hain Part t i te Note 268 B. Household leobers away froi hoae No. Relationship Sex Age Occupation Education Support to the household 2. The size of land holding, status, and land use Land use The size and status of cultivated land (ha) Total Owned Sharecropping Rent Hotegarden lalan-kebat Ricefield Others 3. The talm-kebu A. Cropping pattern and rotation length 1. Since when do you own and cultivate your talui field? 2. List the naae of a l l lajor species that grew in the talun at the t i te you bought or inherited i t . 3. Have you cleared your talut since you own i t ? If yes, what species did you plant after the clearance? Did you plant i t with the saie species as before or have you changed thei with other species? Why? 269 4. Have you considered changing the species composition in your talaa-kebaaf If yes, what species do you want to grow and why? S. How often do you dearcut your talas? Why? Do other faraers have the sate pattern? 6. Is the rotation length of a talun-kebun cycle now similar to that 20 or 30 years ago? If i t was different, why and what was the rotation length back then? 7. What factors influence the cropping pattern and rotation length of the talttt-kebui in the area? 8. If you have 1 hectare of talui/ baaboo f ie ld , wil l you clearcut a l l of i t or do you plan to do partial clearing? Why and what is the size of the clearing? 9. When does the clearing take place? (aonth/season) 10. Please l i s t a l l activities related to clearing and cropping (nuabered in sequence with the duration of each activity) . 11. How did you learn about the talun-kebun rotation pattern? a. tradition b. froa the neighbor c. own idea d suggestion froa the village leaders or agriculture extension service e. others (explain) 12. Is the cropping pattern in the talun-kebun now different froa that 20 years ago? If yes, what factors influence the changes? 270 13. Do you consider the cropping practice now as satisfactory? Do you plant to continue doing i t or do you plan to change i t? Why? 14. Why do you choose hyacinth bean as the lain crop after clearing? Could you plant hyacinth bean as the second year crop? Why? 15. Where do you get the crop seeds/ seedlings? a. buy b. froi neighbor c. others: 16. Why do you prefer Mixed-cropping to aono-cropping? What are the benefits of lixed-cropping? 17. Have you tried to extend the period of lixed cropping? Why? Have you seen anyone doing i t? What happened to the production in the second year? What are the reasons for i t? IB. Is cassava always used as the second year crop? Why do you grow cassava as sole crop and not tilted with other crops? 19. What are the reasons for having several years of fallow after the cassava cropping? Why the cropping ceased after two years? 20. Is there any particular reason why baiboo is chosen as the lain component of the talaa-kebat systei? Explain. 21. Have you seen other farters replace baiboo with other tree crops? Why and what species are used to replace baiboo? What is your perception toward this? Do you intend to follow this trend or do you s t i l l prefer baiboo to other tree crops? Why? B. LABOR USE 1. Mho usually do the work in the Ul at-kebat"! Faaily aeaber working Percentage aiount Type of work Note in the Ula»-kebai of work Father Mother Son Daughter Others: 2. Do you usually hire non fa i i ly labor to do the work in the tilnt-kebas? If the answer is yes, ask the following Work activity Month Class of labor No. of No.of No.of Uage Heal Note •an= 1 person days hours per value woaan= 2 hired per day day children 5 3 Start with the aonth when clearing takes place. In which aonth do you norially hire aost labor? Did you eaploy non fa i i ly labor 20 years ago? For which job? Do you eaploy aore or less labor now? More? Hhy? " Less? ~ Hhy? - - - - - - - - - -List a l l the work activities done during a coeplete Ulat-kebat rotation cycle, who do the work, and the labor tiae needed for each activity. 272 C. MARKETING 1. Do you usually sel l the harvested products froi the tahn-kebaa or do you use them for your own consumption? Which products are for sale and which are for own consumption? 2. If you sell the products, where do you usually go? a. local market b. district market c. sold to broker d. others: 3. Uhat is the distance of the distance from the field to the market? How do you transport the products? D. HAINTENANCE 1. How often do you fert i l ize your talut-keba* field? Uhat is your opinion on soil f e r t i l i ty in the talas-kebat sites? Do you have lower or higher production than 20 years ago? Uhat factors influence production in the talat-kebat? 2. List the types of fert i l izer used in the talat-kebat? Uhen do you use them and what species required such fert i l izer? 3. lalan-keban is usually practised on a sloping land, how do you prevent erosion during the cropping stage? Have you received advice from the extension agent? Uhere did you learn about i t? ; 4. Have you experienced any severe pest problem in your talat-kebat? How do you overcome pest problem in your talus-kebai? Do you use pesticides? Uhat kind of pesticides are commonly used in the taluB-kebwl 5. Uhat are the major problems you face in practising and maintaining taJae? Uhat kind of assistance do you wish to receive from the government in order to increase production? APPENOII 5.2. DETAILED QUESTIONNAIRES Name of Head of Household: Hamlet/Village: Case number: Date: Enumerator: Croppino activities & Labor use: Hixed cropping Note: the farmers usually ovn several fields representing different taltfieban stages 1. What is the size of your crop field? 2. What did you grow in your field this year? 3. Did you get a better yield than the last cropping? What factors do you think influenced the crop yields? 4. Please l i s t the amount of yield per crop species. Species Production (kg) 5. Uhat kind of fert i l izers did you use? Do you use more or less fert i l izer nov than 10 or 20 years ago? Why? Uhat happened to your crop yield i f you did not use fert i l izer? Where did you get your ferti l izers? 6. Did you have any insect problem? Did you experience any significant damage on production because of the insect problem? Uhat are your criteria for significant damage? How did you control pest/insect in your aixed cropping field? Did you use pesticides? If yes, where did you get the pesticides? 7. Please conplete the following table on labor use (faaily or non-faaily), tiae allocation, and wages paid: Activities Labor use Tiae Wage Faaily Hired Allocation (including aeal) H F C H F C H F C H= aale F- feaale C= children The average labor tiae= S hours per day 8. When you hired non-faaily labor, did you choose local laborers (those who lived nearby the field) or did you choose people froa your own haalet? Why? " " " " 8. How auch tiae you and your faaily spent for travel froa your house to the field and back to the house? Cassava cropping: 1. Did you use fert i l izer in your cassava field? Why? . „ ™ " „ _ " „ " 2. Did you hire any non-faaily labor during the cassava cropping? What are the reasons that aost faraers did not hire non-faaily labor? 3. Please conplete the following: Activities Labor use* Tiie allocation Wage equivalent H F C H F C i f using hired labor t Assuiing there is no hired labor 4. Has there any pest problei during the cassava cropping? If yes, how did you control i t? 5. Did you have a good harvest this year? Uhat was the tain factor that influenced cassava yield? Haintenance costs and sales of the products: 1. Did you spend any loney on crop seeds and cassava stalks? If not, where did you get the»? 2. How auch did you spend on ferti l izers? Please l i s t the price of each ferti l izer including aniial aanure. 3. How euch did you spend on pesticides? 4. Please coeplete the following (for vegetables and cassava): Crop species Products Price/kg Incote Own consumption Sale froi sales 5. Where did you sell the harvest products? Did you spend any money to transport the products? Why? 6. Now much did you receive from selling the bamboo culms after clearcutting the field? To whom did you sell the bamboo culms? Why? 7. Did you sel l the used bamboo poles after harvesting the bean? How much money did you receive from the sales? Future trends; 1. Do you consider your talat-kebia as a profitable system? Why? 2. Do you intend to continue practising the bamboo talan-keban rotation or do you plan to change i t into a more intensive cropping system? Why? 3. Some farmers already switch from bamboo talan-kebat rotation to continuous cropping. Do you think of i t as a wise decision? Why? 4. What are the constraints of replacing bamboo with other species? 5. Do you think that the current talan-kebat practice s t i l l needs to be improved? If yes, which aspects and how? Publications Christanty, L. 1981. An ecosystem analysis of West Javanese Home-gardens. Working Paper, EAPI, East West Center, Honolulu, Hawaii. Christanty,L. 1981. Biophysical and Environmental Constraints of Tree Plantations in Hawaii. Working Paper EAPI, East West center, Honolulu, Hawaii. Christanty,L. and J.Iskandar. 1985. Development of Decision Making and Management Skills in Traditional Agroforestry: Examples in West Java. In: Rao,Y., N.Vergara, and G.W.Lovelace. Community Forestry: Socio-economic Aspects. FAO.RAPA. Bangkok. Christanty.L., Hadyana, and H.Y.Hadikusumah. 1985. The Home-garden Potentials in Contributing to the Diet and Income of People in Transmigration Areas of Lampung and in Transmigrants Origin Villages in Java. Paper presented at the 2nd SUAN-EAPI Regional Symposium on Agroforestry Research in Rural Resource Management and development. Baguio City, Philippines. Christanty,L. 1985. Home-gardens in Tropical Asia. Paper presented at the First International Symposium on tropical Homegarden. Bandung, Indonesia. Christanty,L.,O.S.Abdoellah,G.G.Marten, and J.Iskandar. 1986. Traditional Agroforestry in West Java: The Pekarangan (Homegarden) and Kebun-talun (Annual-Perennial Rotation) Cropping Systems, pp.129-58. In G.G.Marten (ed.)., Traditional Agriculture in Southeast Asia: A Human Ecology Perspective. Westview Press, Boulder, Colorado. Christanty,L. 1986. Shifting Cultivation and Tropical Soils: Patterns, Problems, and Possible Improvement, pp.226-239. In G.G.Marten (ed)., Traditional Agriculture In Southeast Asia: A Human Ecology Perspective. Westview Press, Boulder, Colorado. Christanty.L. and J.Iskandar. —. Community Participation in Rural Development: Some Insights from Yogyakarta, Indonesia, pp 189-203. In Rao.Y.S., M.W.Hoskins, N.T.Vergara, and C.P.Castro (eds.), Community Forestry: Lessons from Case Studies in Asia and the Pacific region. FAO.RAPA and EAPI.EWC. Bangkok and Honolulu. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098259/manifest

Comment

Related Items