UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An Analysis of the future productivity of Eucalyptus grandis plantations in the "Cerrado" region in Brazil… Ferreira, Maria Das Graças Moreira 1984

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

Item Metadata

Download

Media
831-UBC_1985_A1 F47.pdf [ 11.4MB ]
Metadata
JSON: 831-1.0096642.json
JSON-LD: 831-1.0096642-ld.json
RDF/XML (Pretty): 831-1.0096642-rdf.xml
RDF/JSON: 831-1.0096642-rdf.json
Turtle: 831-1.0096642-turtle.txt
N-Triples: 831-1.0096642-rdf-ntriples.txt
Original Record: 831-1.0096642-source.json
Full Text
831-1.0096642-fulltext.txt
Citation
831-1.0096642.ris

Full Text

AN ANALYSIS OF THE FUTURE PRODUCTIVITY OF Eucalyptus grandls PLANTATIONS IN THE "CERRADO" REGION IN BRAZIL: A NUTRIENT CYCLING APPROACH by MARIA DAS GRACAS M. FERREIRA For. Eng., Universldade Federal de Vicosa, B r a z i l , 1973 M.Sc, Universldade Federal de Vicosa, B r a z i l , 1977 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES Department of Forestry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1984 (Maria das Gracas M. F e r r e i r a , 1984 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements fo r an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis f o r s c h o l a r l y purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publi c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date ZLk^Mf /3 /ft*/ ABSTRACT Extensive areas of Eucalyptus plantations are being established i n the "cerrado" region of B r a z i l f o r both pulp and charcoal production. These plantations are being managed i n t e n s i v e l y using short rotations and high stand density. Because the s o i l s of the area are very i n f e r -t i l e , there i s concern over the s u s t a i n a b i l i t y of yields i n the absence of repeated f e r t i l i z a t i o n . Therefore, the present study was undertaken to evaluate the e f f e c t s of intensive management of these plantations on the P and N status of the "cerrado" s o i l s and the implications thereof for future productivity. Biomass and P and N content of each tree component (including the root system) were determined over age sequences of Ej_ grandis planta-tions growing on good and poor "cerrado" s i t e s i n Minas Gerais State, B r a z i l . The dynamics of P and N were studied by analysing l i t t e r f a l l , f o r e s t f l o o r accumulation, decomposition of branches and leaves, and i n t e r n a l c y c l i n g due to heartwood formation and the shedding of branches and f o l i a g e . The data that were col l e c t e d permitted both a t r a d i t i o n a l s t a t i c inventory assessment and a dynamic assessment of the effects of management on the s i t e nutrient c a p i t a l . However, because these two approaches are inadequate for the development of long-term sustained y i e l d strategies f or the management of these plantations, the thesis work included the development of conceptual models of two important aspects of eucalypt management i n the "cerrado": coppice regeneration and phosphorus dynamics i n the mineral s o i l . Studies of coppice sprouting were conducted to provide guidance for the development of the coppice model. i i i By using information on both accumulation and dynamics of biomass and nutrients over age sequences of E. grandis growing on two d i f f e r e n t "cerrado" s i t e s , the following predictions were obtained. Based on the s t a t i c approach, the maximum y i e l d over an i n i t i a l 18 year period was obtained f o r a 54 month r o t a t i o n , on both s i t e s . However the use of t h i s method to assess the effects of intensive management on s i t e nutrient status would give misleading r e s u l t s because there i s no feed-back between changes i n nutrient a v a i l a b i l i t y and y i e l d . Therefore, a dynamic approach to the assessment was also used. This showed that over an 18 year period harvesting only stems on the good s i t e i s l i k e l y to cause a reduction i n productivity of 48% while whole tree harvesting i s l i k e l y to r e s u l t i n a reduction of 61%. On the poor s i t e , there i s l i k e l y to be only a 12% decrease due to whole tree harvesting, while harvesting stems only i s u n l i k e l y to r e s u l t i n any decrease i n y i e l d . If whole tree harvesting i s adopted on the good s i t e , f e r t i l i z e r quantities required to maintain the same l e v e l of productivity i n the second r o t a t i o n would be approximately 24 kg/ha of P while the amount required using stem harvesting only would be approximately 15 kg/ha. The amount of P required to produce one unit of biomass on the poor s i t e was less than that required on the good s i t e . In addition, the tree root systems on the poor s i t e contain a greater proportion of t o t a l tree P as compared to those on the good s i t e . This reduces the proportion of the t o t a l tree P that i s removed at harvesting and gives an advantage to sprout growth on the poor s i t e i n subsequent r o t a t i o n s . Thus, intensive biomass harvesting on the poor s i t e i s expected to have r e l a t i v e l y less impact on future s i t e p r o d u c t i v i t y compared to that on i v the good s i t e . However, the good s i t e can be expected to y i e l d a greater biomass than the poor s i t e . The data on accumulation and dynamics of biomass and P and N proved to be h e l p f u l i n understanding the effects of intensive forest management on s i t e nutrient status. However, the evaluation of long-term effects i s rather complex because of the i n t e r a c t i o n of several factors when d i f f e r e n t management conditions are considered. Therefore, computer simulation models are required to evaluate the long-term e f f e c t s of tree harvesting on s i t e nutrient status. Because the growth strategies of a tree growing from seed are d i f f e r e n t from that of a tree developing from a stump, and the dynamics of P i n mineral s o i l (the most l i m i t i n g nutrient for eucalypt growth i n the "cerrado" region) d i f f e r from that of N, coppice and phosphorus models were developed that could be implemented i n the future i n ecosystems models used to evaluate long-term e f f e c t s of intensive management. V TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS v LIST OF TABLES x LIST OF FIGURES x i i i LIST OF APPENDICES xv ACKNOWLEDGEMENTS xvi CHAPTER 1. GENERAL INTRODUCTION 1 1.1 The "Cerrado" Region 3 1.2 Eucalypt Plantations i n B r a z i l 7 1.2.1 Site Preparation and Tending Operations . . . . 8 1.3 T r a d i t i o n a l Approaches to the Evaluation of the Risk of S i t e Nutrient Depletion under Intensive Forest Management 11 1.4 Simulation Approaches to the Evaluation of Nutrient Cycling 13 1.5 Objectives and Design of the Thesis 15 CHAPTER 2. ACCUMULATION OF BIOMASS AND NUTRIENTS IN AGE SEQUENCES OF E. grandis PLANTATIONS GROWING ON GOOD AND POOR "CERRADO" SITES 18 2.1 Introduction 18 2.2 L i t e r a t u r e Review 19 2.2.1 Biomass 19 a. Regression equations to estimate biomass . 19 b. Biomass studies for eucalypts 22 2.2.2 P and N i n L i v i n g Biomass 26 2.3 Description of the Study Area 28 v i Page 2.4 F i e l d Methods 29 2.4.1 Biomass Sampling 29 2.4.2 S o i l Sampling 32 2.5 Laboratory Methods 32 2.5.1 Plant Analysis 32 2.5.2 S o i l Analysis 33 2.6 Data Analysis 33 2.7 Results and Discussion 35 2.7.1 Biomass 35 a. Regression equations to estimate biomass . 35 b. Estimated biomass 41 2.7.2 P and N i n l i v i n g biomass 49 2.8 Summary and Conclusions 59 CHAPTER 3. DYNAMICS OF NUTRIENTS (INTERNAL CYCLING, LITTERFALL AND LITTER DECOMPOSITION) IN AGE SEQUENCES OF E. grandis PLANTATIONS GROWING ON GOOD AND POOR "CERRADO" SITES 64 3.1 Introduction 64 3.2 L i t e r a t u r e Review 65 3.2.1 Biomass, P and N i n L i t t e r f a l l 65 3.2.2 Biomass, P and N i n Forest Floor 69 3.2.3 L i t t e r Decomposition 71 3.2.4 Internal Cycling 75 3.3 Description of the Study Area 77 3.4 F i e l d Method 77 3.4.1 L i t t e r f a l l Sampling 77 3.4.2 Forest Floor Sampling 78 v i i Page 3.4.3 L i t t e r Decomposition 78 3.5 Plant Analysis 80 3.6 Data Analysis 80 3.7 Results and Discussion 81 3.7.1 Biomass, P and N i n L i t t e r f a l l 81 3.7.2 Biomass, P and N i n Forest Floor 86 3.7.3 L i t t e r Decomposition 91 3.7.4 Internal Cycling 97 3.8 Summary and Conclusions 104 CHAPTER 4. FACTORS DETERMINING COPPICE REGENERATION AND GROWTH, AND A SIMULATION OF COPPICE GROWTH 110 4.1 Introduction 110 4.2 Review of Factors Determining Coppice Regeneration . . I l l 4.2.1 Phys i o l o g i c a l Aspects Related to Sprouting and Sprout Growth . . . . . . I l l a. Carbohydrates 112 b. Water and nutrients 113 4.2.2 Root Dynamics a f t e r Coppicing 114 4.2.3 Productivity of Coppice 115 a. Stump mortality 115 b. Root:shoot r a t i o and n u t r i t i o n a l aspects 118 c. Number of sprouts per stump 121 d. Stump diameter 122 4.3 P i l o t Studies of Biomass and Nutrient Accumulation i n Sprouts of Ej|_ grandis i n the Greenhouse and i n the F i e l d 123 4.3.1 Introduction 123 v i i i Page 4.3.2 Materials and Methods 125 a. Greenhouse experiment 125 b. F i e l d experiment 127 4.3.3 Results and Discussion 129 a. Greenhouse experiment 129 b. F i e l d experiment 135 4.4 Simulation of Coppice Growth 140 4.4.1 Data Requirements 144 4.5 Summary and Conclusions 148 CHAPTER 5. FACTORS DETERMINING THE DYNAMICS OF P IN THE SOIL SYSTEM, AND A SIMULATION OF P CYCLING IN THE ECOSYSTEM 151 5.1 Introduction 151 5.2 Phosphorus Status of "Cerrado" S o i l s , and Implications f o r Eucalypt Plantations 152 5.3 Evaluation of P A v a i l a b i l i t y to Plants 153 5.3.1 Adsorption Isotherms 154 a. Desorption 157 5.4 E f f e c t s of Residual P on Plant Growth 158 5.5 E f f e c t s of Root Growth on P Uptake 158 5.6 Simulation of P i n the S o i l System 159 5.6.1 Data Requirements 160 5.7 Summary and Conclusions 167 CHAPTER 6. EVALUATION OF THE TRADITIONAL METHODS OF PREDICTING THE LONG-TERM CONSEQUENCES OF INTENSIVE MANAGEMENT OF EUCALYPTS 169 6.1 Introduction 169 6.2 E f f e c t s of Intensive Management of E. grandis planta-tions on P and N Removal: A T r a d i t i o n a l Evaluation . . 171 ix Page 6.2.1 S t a t i c Inventory Assessment 171 a. Proportion of P and N i n the biomass removed at harvest 171 b. Amount of P and N removed per tonne of biomass 174 c E f f e c t s of harvesting on s o i l nutrient status 176 6.2.2 Assessment Based on Both Inventory and Nutrient Dynamics Data 182 6.3 Conclusions 190 REFERENCES 194 APPENDICES 209 X LIST OF TABLES Table Page 1 Aboveground l i v i n g biomass (t/ha) and nutrient content (kg/ha) of Eucalyptus spp. under natural conditions .23 2 Aboveground l i v i n g biomass (t/ha) and nutrient content (kg/ha) of Eucalyptus plantations 25 3 Stand c h a r a c t e r i s t i c s of the s i t e s studied 36 4 Equations f or biomass (g/tree) of each tree com-ponent of 15. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l . 38 v 5 Biomass accumulation (kg/ha) of each tree component over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 42 6 Percentage of basal disc area represented by heartwood, over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 48 7 Equations for phosphorus (g/tree) of each tree component of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l . 50 8 Equations f or nitrogen (g/tree) of each tree component of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l . 51 9 Phosphorus content (kg/ha) of each tree component over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 52 10 Nitrogen content (kg/ha) of each tree component over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 53 11 Forest f l o o r and l i t t e r f a l l biomass and P and N content of native eucalypt forests (studies reported up to 1977) 66 x i Table Page 12 L i t t e r f a l l biomass (t/ha/yr) and P and N content (Kg/ha/yr) of native and planted eucalypts (studies reported since 1978) 67 13 Forest f l o o r biomass (t/ha) and P and N content (kg/ha) of native and planted eucalypt species . . . 70 14 L i t t e r decomposition (% loss of o r i g i n a l weight) of eucalypt s 73 15 Monthly mean temperature and p r e c i p i t a t i o n during the study period (1981-1983) i n Bom Despacho and Carbonita, Minas Gerais, B r a z i l 79 16 Biomass (kg/ha) and nutrient content (g/ha) of seasonal l i t t e r f a l l over an age sequence of E_. grandis plantations growing i n the "cerrado" region, Bom Despacho, Minas Gerais, B r a z i l 82 17 Biomass (kg/ha) and nutrient content (g/ha) of seasonal l i t t e r f a l l over an age sequence of E_. grandis plantations growing i n the "cerrado" region, Carbonita, Minas Gerais, B r a z i l 83 18 Biomass (kg/ha) and nutrient content (g/ha) of forest f l o o r over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 87 19 Internal c y c l i n g of P i n stands of 15. grandis older than 48 months growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 98 20 Internal c y c l i n g of N i n stands of E_. grandis older than 48 months growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 99 21 Stand c h a r a c t e r i s t i c s at the end of seedling and subsequent coppice rotations of Ej_ grandis i n Kenya . 116 22 Pro d u c t i c i t y (m^/ha) at the end of the seedling and subsequent coppice rotations of Eucalyptus spp. -CAF - B r a z i l 117 23 Productivity for early stages of f i r s t and second rotations of Ej_ alba i n the "cerrado" region . . . . 119 0 x i i Table Page 24 Biomass (g), P and N (mg) of coppicing E. grandis growing on two d i f f e r e n t s o i l s i n the greenhouse . . 130 25 Total root:shoot r a t i o of 9.5 months old seedlings and sprouts up to 4.5 months old of Ej_ grandis growing on two d i f f e r e n t s o i l s i n the greenhouse . . 134 26 Total aboveground biomass (g/stump) and stem: foliage r a t i o of sprouts and rooted cuttings of E. grandis growing on a poor "cerrado" s i t e , Itamarandiba, Minas Gerais, B r a z i l 136 27 A summary of P and N that would be removed (g/ha) at harvest for d i f f e r e n t ages and levels of u t i l i -z ation of E_^ grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l . 173 28 A summary of biomass (t/ha) and P and N (kg/ha) removal over 18 years, at three d i f f e r e n t levels of u t i l i z a t i o n and r o t a t i o n length i n E_^ grandis plantations (managed by coppice) growing i n the "cerrado" region, Minas Gerais, B r a z i l 178 29 Tree growth P budget (kg/ha) over two 54 month rotations, at two le v e l s of u t i l i z a t i o n i n E. grandis plantations (managed by coppice) growing i n the "cerrado" region, Minas Gerais, B r a z i l . . . 183 30 Summary of the predicted biomass y i e l d (t/ha) of E. grandis plantations growing i n the "cerrado" region, over an 18 year period, for a 54 month ro t a t i o n length, at two le v e l s of u t i l i z a t i o n , produced by the s t a t i c and the dynamic approaches . 188 LIST OF FIGURES Figure 1 Map of the "cerrado" d i s t r i b u t i o n i n B r a z i l 2 Map of loc a t i o n of the study area i n Minas Gerais State 3 Mean annual increment of Ej_ grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 4 Biomass (t/ha) of stem, t o t a l aboveground and t o t a l roots over an age sequence of E ^ grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 5 Phosphorus content (kg/ha) of stem, t o t a l aboveground and t o t a l roots over an age sequence of E_^ grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 6 Nitrogen content (kg/ha) of stem, t o t a l aboveground and t o t a l roots over an age sequence of E. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 7 Biomass (t/ha) of l i t t e r f a l l and forest f l o o r over an age sequence of Ej_ grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 8 Percentage change of P and N quantities i n decom-posing l i t t e r for d i f f e r e n t stand ages of E. grandis growing i n Bom Despacho (good s i t e ) , Minas Gerais, B r a z i l 9 Percentage change (average from d i f f e r e n t stand ages) of weight, P and N quantities i n foliage and branch l i t t e r of E_^ grandis growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l 10 Flowchart summarizing the factors represented i n the coppice model 11 Flowchart of P dynamics i n the ecosystem as represented i n the P model x i i i Page 4 30 37 43 54 55 89 95 96 143 161 x i v Figure Page 12 Graphic representation of P adsorption-desorption-re-adsorption processes 163 13 Biomass (t/ha) and P and N content (kg/ha) over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l : (a) Bom Despacho - good s i t e and, (b) Carbonita -poor s i t e 172 X V LIST OF APPENDICES Appendix Page 1 S o i l data at two d i f f e r e n t depths f o r the stands studied 209 2 Diameter d i s t r i b u t i o n per plot for the stands studied . 211 3 Phosphorus and nitrogen concentrations of each component of the sampled trees 214 4 Phosphorus and nitrogen concentrations of each component of l i t t e r f a l l , forest f l o o r and l i t t e r i n the decomposition bags 217 5 Results of the test of s i g n i f i c a n c e of the differences i n means of l i t t e r f a l l and decomposing l i t t e r 221 6 Percentage change i n the nutrient content of decomposing l i t t e r 224 7 Results of the test' of s i g n i f i c a n c e of the differences i n means for biomass, and P and N content and concen-trations of coppicing 15. grandis growing i n the greenhouse 226 8 Adsorption isotherms f o r the two s i t e s studied . . . . 229 ACKNOWLEDGEMENTS The author wishes'to express gratitude to Dr. J.P. Kimmins, Faculty of Forestry, University of B r i t i s h Columbia (UBC), for his assistance with the preparation of t h i s t h e s i s . I am also g r a t e f u l for his encouragement and friendship during my stay i n Canada. Thanks are also due to the committee members for t h e i r guidance and i n t e r e s t : Drs. N.F. Barros, M.C. F e l l e r , A. Kozak, L.M. Lavkulich, and G.F. Weetman. The guidance and support given by Dr. N.F. Barros, Department of S o i l Science, Universidade Federal de Vicosa (UFV), and the support of G.C. Rezende, Companhia Agric o l a e F l o r e s t a l Santa Barbara (CAF), Minas Gerais, i n the f i e l d and laboratory work i n B r a z i l , are g r a t e f u l l y acknowledged. I am g r a t e f u l to the s t a f f of UFV, CAF and UBC involved i n c o l l e c t i n g and processing samples, and i n the support for the development of the t h e s i s . The understanding and support of my friends and family are appreciated. The f i n a n c i a l support was given by the Canadian International Development Agency (CIDA) and the Universidade Federal de Vicosa. The help of Dr. V.J. Nordin and Mr. A.F. Shirran (CIDA), and of Dr. Arno Brune, coordinator of the CIDA/UFV agreement i n B r a z i l , i s acknowledged. The research was supported by the contract IBDF/UFV/SIF. This thesis i s dedicated to the author's family for their under-standing. 1 CHAPTER 1 GENERAL INTRODUCTION B r a z i l i s a country with a great v a r i e t y of c l i m a t i c and edaphic conditions. It covers a l a t i t u d i n a l range from about 5°N to 33°S and an a l t i t u d e range from sea l e v e l to 1600 m. The mean annual p r e c i p i t a t i o n ranges from 253 to 3700 mm ( G o l f a r i , 1978). This wide var i e t y of conditions, together with a favorable temperature regime (most of the country has a mean annual temperature higher than 18°C), an abundant supply of land and cheap labour, and a high demand f o r forest products, creates conditions suitable for the development of an intensive forest plantation program. Pinus and Eucalyptus are the most common genera planted due to t h e i r r e l a t i v e l y fast growth under a wide var i e t y of e c o l o g i c a l conditions. Forest plantations in B r a z i l began i n 1920 when eucalypts were planted for r a i l r o a d t i e production. After 1940, conifers were also planted i n the south. However, i n t e r e s t i n establishing plantations on a large scale developed a f t e r 1966 when government l e g i s l a t i o n i n s t i -tuted tax concessions to encourage tree plantations (including f r u i t t r e e s ) . There was an increase i n the annual area approved for planta-tions (with government f i s c a l incentive payments) from 35,000 ha i n 1967 to 436,000 ha i n 1980, bringing the t o t a l area of such plantations to approximately 4.2 m i l l i o n ha. Most of these plantations are a source of wood for pulp and paper (1.38 m i l l i o n ha) and of charcoal for the i r o n -work industry (1.36 m i l l i o n ha) (Barrichelo and B r i t o , 1982). 2 Despite the great increase i n forest plantations over the l a s t decade, the planted area i s far from being s u f f i c i e n t to meet the demand. Projections of forest plantation needs for the year 2000 i n d i -cate that 16.3 m i l l i o n ha w i l l be necessary to supply the demand based on an assumed average productivity of approximately 17 m^/ha/yr. How-ever, i f p r o d u c t i v i t y could be increased by an average of 2% a year, the area needed to supply t h i s demand could be reduced to 10.9 m i l l i o n ha (Brandao and L u p a t t e l l i , 1982). Because the increase i n area of plantations i s not s u f f i c i e n t to s a t i s f y the r a p i d l y growing demand, e x i s t i n g plantations are being managed more i n t e n s i v e l y . Techniques such as s i t e preparation, planting/coppice regeneration, f e r t i l i z a t i o n , and weeding are r e g u l a r l y used. In the State of Minas Gerais, where eucalypts have been planted mainly for charcoal production, there i s a trend towards a reduction i n spacing from 3x2 m to 3x1.5 m or even 2x1 m, as well as a trend towards a shortening of the r o t a t i o n . Forest plantations were i n i t i a l l y established on the f e r t i l e s o i l s of the southern and southeastern region of B r a z i l where most of the population i s concentrated. However, the a v a i l a b i l i t y of f e r t i l e land i n t h i s region was limited so plantation establishment spread to the "cerrado"! region. Because "cerrado" s o i l s are n a t u r a l l y very i n f e r t i l e , a major aspect of forest management must be s i t e nutrient management. There i s Central region of B r a z i l covered by a vegetation named "cerrado" (vegetation consisting of crooked trees, less than 15 m high, and a continuous grass l a y e r ) . 3 growing concern that the productivity of future rotations w i l l be reduced unless s i g n i f i c a n t f e r t i l i z e r applications are made and there i s a growing need f o r a biogeochemical evaluation of the s u s t a i n a b i l i t y of yie l d s i n these plantations. However, i n spite of i t s much lower f e r t i l i t y , the "cerrado" region offers several advantages including low land and labour costs, easy mechanization due to topographic conditions and good a c c e s s i b i l i t y (Goedert, 1980). 1.1 The "Cerrado" Region The study was conducted i n plantations of E. grandis that have been established on two s i t e types i n the "cerrado" region of southeast B r a z i l f o r charcoal production. According to Goedert (1980), t h i s region covers 180 m i l l i o n ha (22% of t o t a l B r a z i l i a n area). The la t i t u d e ranges from about 3°N to about 24°S (Figure 1). The "cerrado" region has a great v a r i a t i o n i n climate due to i t s wide d i s t r i b u t i o n . I t includes the following types of climate based on Koppen's c l a s s i f i c a -t i o n (Camargo, 1963): Am (warm and moist, with winter dry period); Aw (warm and moist, with pronounced dry winter, representing the climate of the c e n t r a l "cerrado"); Cwa ( t r o p i c a l a l t i t u d i n a l , with mild summer and dry winter); Cfa (su b t r o p i c a l , with warm summer but no dry period); Cfb (subtropical, with mild summer and no dry period). The mean annual p r e c i p i t a t i o n i n the "cerrado" region ranges from 500 to 2300 mm with a dry season varying from 2 to 9 months and the mean annual temperature ranging from 19 to 26°C (Azevedo and Caser, 1980). The topography i s generally gently undulating with two or more erosion surfaces (Sanchez, 1976). However, there i s a great hetero-4 Figure 1. Map of the "cerrado" distribution in Brazil. Source: Camargo et a l . (1976) 5 geneity of geological structure represented by 22 d i s t i n c t sub-regions (Azevedo and Caser, 1980). The predominant rocks (granite, gabbro, gneiss, s c h i s t , p h y l l i t e and q u a r t z i t e (precambrian) are partly exposed and p a r t l y covered by younger paleozoic and mesozoic sedimentary materials (AIA, 1963 ( c i t e d by Lopes, 1975)). There i s also a great d i v e r s i t y i n s o i l types i n the "cerrado" region. However, the predominant s o i l s are yellow red l a t o s o l s (41% of the area) (Freitas and S i l v e i r a , 1977). Latosols are highly weathered s o i l s , with depths greater than 3 m, where s i l i c a and cations such as Na, K, Mg and Ca have been leached and the predominant residual components are Fe and A l oxides, k a o l i n i t e and quartz. The horizon sequence i s A,B,C with very small v a r i a t i o n i n morphological features between horizons (Fr e i t a s and S i l v e i r a , 1977). The water retention capacity of these s o i l s i s very low (less than 1.5 mm H20/cm s o i l (Ranzani, 1963)). Very stable sand-size aggregates frequently form due to the high cl a y content and iron and aluminum oxide coatings (Sanchez, 1976). More information on s o i l P i s given i n Chapter 5. The physiognomy of the B r a z i l i a n "cerrado" i s very heterogeneous, varying from almost pure grass with low crooked trees (campo sujo) to a vegetation type c a l l e d "cerradao" with a very high density of t a l l e r (less than 15 m) and less crooked trees. The standing volume varies from 10 m^/ha (poorest "cerrado") up to 80 m^/ha i n "cerradao". Several tree species can develop i n a l l gradations of "cerrado" but with d i f f e r e n t height growth ( G o l f a r i , 1975; Veloso, 1966). As the trees have a xeromorphic appearance, "cerrado" vegetation was considered a c l i m a t i c climax f o r several years. However, several 6 studies that were started a f t e r 1942 by Rawitscher, F e r r i and others demonstrated that leaf-shedding of a l l trees does not occur at the same time and stomata remain open even during the dry season, implying that climate does not exert an overpowering influence; several trees have deep roots and the s o i l i s s t i l l moist i f deeper than 2 m ( F e r r i , 1977). I t was also observed that "cerrado" vegetation occurs over a wide range of c l i m a t i c conditions and that dense forest can occur within the same range (Azevedo and Caser, 1980; Camargo, 1963; Reis, 1971). These studies confirmed the theory that climate i s not the main conditioning fa c t o r of "cerrado" occurrence. A theory of edaphic climax started to develop based on observations that "cerrado" does not occur on f e r t i l e s o i l s . Arens (1958) (cited by Goodland, 1971), proposed the theory of oligotrophic-scleromorphism. More detailed studies of s o i l chemistry and i t s c o r r e l a t i o n with vegetation were developed a f t e r 1970. Goodland (1979) studied a gradient of "cerrado" vegetation from which he found a strong c o r r e l a -t i o n between basal area/ha and N, P and K, P being the most strongly correlated. There was an increase of a l l nutrients i n the gradient from "campo sujo" to "cerradao". Aluminum levels were found to be very high throughout the gradient (average of 75 ppm), but there was a decrease of A l saturation with an increase i n vegetation density. Because of the high aluminum l e v e l s , the scleromorphism observed i n "cerrado" vegeta-t i o n has been p a r t i a l l y a t t r i b u t e d to the excess of A l (Goodland, 1971, 1979). Although the d i s t r i b u t i o n of d i f f e r e n t types of "cerrado" vegeta-ti o n i n the Central Plateau of B r a z i l i s considered to be p r i m a r i l y due 7 to edaphic f a c t o r s , repeated f i r e i s also considered to play an important role i n t h e i r development. Furthermore, the c l i m a t i c factors are s t i l l considered to be one of the factors determining t h i s vegeta-t i o n (Alvim and S i l v a , 1980; Goodland, 1971; Heringer, 1971). 1.2 Eucalypt Plantations i n B r a z i l Eucalypt plantations i n B r a z i l began i n 1920 when eucalypts were planted for r a i l r o a d t i e production. After 1940, conifers were also planted i n the south of B r a z i l . However, the i n t e r e s t i n establishing plantations on a large scale was shown a f t e r 1966 when governmental l e g i s l a t i o n allowed a w r i t e - o f f a proportion of t h e i r income tax against the cost of establishing tree plantations. Most of these plantations have been established i n the south-eastern region of B r a z i l and, more recently, i n the "cerrado" region. The e a r l i e r eucalypt plantations were established on the basis of the growth of eucalypt species introduced i n Sao Paulo i n the early 1900's. It was assumed that these species would behave i n a s i m i l a r manner i n d i f f e r e n t regions, and plantations were subsequently established i n other areas mostly using seeds c o l l e c t e d from the early B r a z i l i a n plantations. However, many of the species and provenances used i n i t i a l l y proved to be unsuitable for these other regions: low p r o d u c t i v i t y , v a r i a b i l i t y i n growth, poor coppicing a b i l i t y and high s u s c e p t i b i l i t y to diseases were common. The s t e a d i l y increasing wood demand encouraged the testing of new species and d i f f e r e n t seed sources to control these problems. 8 Most of the research related to forest establishment began i n the l a t e 1960's when several provenance t r i a l s were established. An ecolo-g i c a l zoning for r e f o r e s t a t i o n was undertaken for d i f f e r e n t regions. The most adapted species and provenances were indicated based on planta-tions and t r i a l s already established i n the region or under si m i l a r con-d i t i o n s elsewhere; new species and provenances were selected f o r testing based on c l i m a t i c analogies between the region of natural occurrence and l o c a l conditions ( G o l f a r i , 1967, 1971, 1975; G o l f a r i and Caser, 1977; G o l f a r i et a l . , 1978). Several seed orchards were established for the . most widely planted species, decreasing the dependency on imported seeds, and rooted cuttings are now being widely used to speed the s e l e c t i o n process for eucalypts. 1.2.1 S i t e Preparation and Tending Operations Eucalypt plantations i n B r a z i l have a very high productivity p o t e n t i a l . For this reason a high technology l e v e l of management can be adopted. Site preparation i n the "cerrado" region i n v o l ves the use of large bulldozers or a heavy chain drawn by two tractors to push the native vegetation i n t o windrows for burning. Ploughing and discing are then used to improve s o i l physical conditions to encourage better i n i t i a l root growth. I r r i g a t i o n i s applied when planting i s done during the dry season. Within the f i r s t 10 days a f t e r planting, the plants receive 1-3 i r r i g a t i o n s as required. Discing i s used to control compet-ing vegetation between tree rows over the f i r s t 2 or 3 years. The number of discing operations depends on crown closure and weed growth; up to 3 times where spacing i s 3x2 m. If spacing i s 2x1 m, weeding 9 might not be required. The system of weeding by discing has the advantage of incorporating organic matter into the s o i l (Simoes et a l . , 1981; Rezende, pers. comm.^). Ant control i s very important for eucalypt establishment. The most important l e a f - c u t t i n g ants belong to the genus Atta and Acromirmex. Because of t h e i r importance, ants are c o n t r o l l e d before ploughing and a f t e r planting. Plantations are generally monitored for ants 2-3 times over a 5-year r o t a t i o n period (Moraes, 1982). E a r l i e r plantations were established at 3x2 m spacing. However, with increasing wood demand there was a great i n t e r e s t i n increasing stocking density to produce more biomass per unit area per unit time. Very extensive experiments were established over the l a s t decade to t r y to obtain as much information as possible related to the e f f e c t s of spacing on productivity, establishment, harvesting cost, and wood qu a l i t y , for the most commonly planted eucalypt species (Rezende et a l . , 1982a; Pereira et a l . , 1982). Up to 50 d i f f e r e n t spacing treatments are being tested by varying distance between rows, distance between plants, and the pattern of plant d i s t r i b u t i o n i n the same row, with the area per plant ranging from 1.75 to 7 m^ . Based on r e s u l t s obtained from these experiments, recent plantations have been established using spacing as narrow as 2x1 m. Besides the increase i n volume, i t was observed that no weeding i s required due to e a r l i e r crown closure. Gustavo C. Rezende, Companhia Agrfcola e F l o r e s t a l Santa Barbara, Belo Horizonte, Minas Gerais, B r a z i l . 10 E a r l i e r plantations were established on more f e r t i l e s o i l s where f e r t i l i z a t i o n was not required. However, because "cerrado" s o i l s are very poor i n nutrients, the f i r s t plantations without f e r t i l i z a t i o n attained very low productivity. Hence, e a r l y plantations i n the "cerrado" region r o u t i n e l y received f e r t i l i z a t i o n at planting time. Each plant receives approximately 100-150 g of a mixture containing N, P and K (e.g. 10-28-6). A trace of B and Zn was also included i n the mixture. Recently, extensive experiments have been undertaken to i d e n t i f y more pr e c i s e l y the f e r t i l i z e r requirements on d i f f e r e n t s i t e s (e.g. Barros et a l . , 1981; Rezende et a l . , 1982b; Rocha et a l . , 1982a,b,d). The re s u l t s of these experiments have shown that f e r t i l i z a -t i o n i s c r u c i a l i n the "cerrado" region. For example, Rezende et a l . (1982b) reported volumes of 8.4 and 2.2 m3/ha for 24-month-old E.  grandis planted without f e r t i l i z a t i o n i n Bom Despacho (good) and Carbonita (poor) "cerrado" s o i l s , r e s p e c t i v e l y . By adding 150 g of NPK + BZn per plant, the volume increased to 45.7 and 49.7 nrVha, respec-t i v e l y . When, i n the same experiment, 2 tonnes/ha of rock phosphate was applied, i n addition, the volume increased to 86.8 and 92.3 m-Vha, resp e c t i v e l y . F e r t i l i z a t i o n of subsequent rotations managed under the coppice system i s not commonly adopted. However, some experiments have been undertaken, showing p o s i t i v e e f f e c t s of f e r t i l i z e r on sprout growth (for more d e t a i l s see Chapter 4). Because of the great capacity of eucalypts to regenerate by coppicing, the plantations are managed by coppice over 2 or 3 ro t a t i o n s . The r o t a t i o n age varies from about 5 to 10 years. The sprouts can be thinned to 2-3 per stump at age 10-12 months, depending on the f i n a l 11 product to be obtained. If stump mortality i s high i n plantations that were established at wide spacing (e.g. 3x2 m), new rows of seedlings are planted between the rows of stumps a f t e r the s o i l has been s c a r i f i e d by d i s c i n g . The coppice sprouts are cut back and the stumps allowed to resprout at the time of planting, to prevent the sprouts from shading the new seedlings. By age 24 months, the seedlings and sprouts usually both a t t a i n the same s i z e . Where narrower spacing (e.g. 2x1 m) i s adopted, the plantation of new seedlings i s not required. When produc-t i v i t y of the stand i s very low due to stump mortality and/or loss of vigour of rootstocks, an entire new seedling plantation i s established. The resprouting of old stumps i s prevented by debarking and the old stumps are completely covered by s o i l when the s i t e i s disced. New seedlings are planted between stumps i n the o r i g i n a l rows to permit the use of machines between rows (Simoes e_t a l . , 1981). (For more d e t a i l s on the physiology and productivity of coppice, see Chapter 4). 1.3 H i s t o r i c a l Approaches to the Evaluation of the Risk of Site  Nutrient Depletion under Intensive Forest Management Because of the increasing i n t e n s i t y of forest management, there has long been a concern about the effects of biomass removal on s i t e p r o d u c t i v i t y as a consequence of nutrient depletion. Early studies showed adverse e f f e c t s of l i t t e r raking on forest productivity i n Germany (Kreutzer, 1972). Rennie (1955) emphasized the importance of studies of the amount of nutrients i n d i f f e r e n t components of the forest ecosystem i n order to better analyse the s u s t a i n a b i l i t y of a s i t e for continuous growth of forest over a r o t a t i o n . Numerous studies have been 12 done on t h i s topic since Rennie raised the issue i n 1955, but the question of the n u t r i t i o n a l e f f e c t s of t o t a l or whole tree harvesting on s i t e nutrient depletion remains c o n t r o v e r s i a l (Freedman, 1981; Morrison and Foster, 1979). Modern examples of declines i n forest y i e l d i n successive rotations have been reported (e.g. for Pinus radiata i n South A u s t r a l i a (Bednall, 1968; Woods, 1980)). The most commonly used method for the determination of the e f f e c t s of intensive management on nutrient status has been the s t a t i c approach. This inventory approach involves the estimation of nutrient content of biomass components, l i t t e r and s o i l . A more dynamic nutrient budget approach includes the balance between natural inputs and outputs to and from the ecosystem (van Hook et a l . , 1982). Good examples of nutrient budget studies for eucalypts are: E. obliqua ( A t t i w i l l , 1968, 1979, 1980; A t t i w i l l et a l . , 1978) and E. obliqua/E. dives ( F e l l e r , 1980, 1984). However, understanding the s u s t a i n a b i l i t y of the s i t e over several subsequent rotations i s more complex. Kimmins (1977) has discussed several aspects that should be analysed i n order to evaluate future p r o d u c t i v i t y of an ecosystem, such as a v a i l a b i l i t y of nutrients to plants, r a t i o of losses to replenishment, nutrient requirements of subsequent crops, changes i n nutrient demand with age, magnitude of other losses (such as by burning, erosion or s o i l leaching), and the frequency of harvesting. An ecologically-based computer simulation model has been suggested by Kimmins and Scoullar (1984b) as the best way to handle the extensive 13 knowledge required to predict the consequences of forest management a c t i v i t i e s over a very long time f o r s i t e nutrient status and productivity. 1.4 Simulation Approaches to the Evaluation of Nutrient Cycling Models can be b u i l t with d i f f e r e n t scope and levels of r e s o l u t i o n . For example, the model developed by L a y z e l l and LaRue (1982) summarizes the measured transfers of C and N from leaves to developing' f r u i t of soybean, and excludes any potential transfers between the f r u i t and other plant parts. A model that includes within-tree retranslocation of l i m i t i n g nutrients as tissues age and senesce ( i n t e r n a l cycling) was developed by Fagerstrom and Lohm (1977) for Scots pine. These two models dealt only with within-plant processes and are unsuitable in scope to deal with s i t e budget questions. Ingestad et a l . (1981) have developed a computer model that describes f o l i a r biomass production, accumulation and turnover for a co n i f e r stand. This model can be used to examine f e r t i l i z a t i o n e f f e c t s on development and production of the f o l i a r biomass and forest f l o o r nitrogen content of coniferous stands. However, they noted that the representation of s o i l nitrogen dynamics i n t h i s model i s very simple and that the nitrogen d r i v i n g function s t i l l requires f i e l d t e s t i n g . The model FORTNITE (Aber and M e l i l l o , 1982), a combination of a forest growth simulator and a forest f l o o r decomposition model, predicts the e f f e c t s of harvest i n t e n s i t y , species s e l e c t i o n , r o t a t i o n length, and f e r t i l i z a t i o n on s i t e nutrient status. The model was designed and works 14 for multi-age, multi-species f o r e s t s , but can be applied to single species monocultures. FORTNITE i s only a nitrogen model and cannot be used with other nutrients without modification. Another model, FORCYTE, i s an ecologically-based computer simu-l a t i o n designed to examine the long-term consequences of intensive forest management on s i t e nutrient c a p i t a l , biomass production and the economic performance and energy e f f i c i e n c y of a l t e r n a t i v e management scenarios (Kimmins and Scoullar, 1983). The objective of the authors was to develop a model that could be applied to even age forest anywhere i n the world as a management t o o l . Models are developed i n order to represent natural phenomena, and the accuracy with which they do so depends i n part on the knowledge of the phenomena at the time of model development. Therefore, models are subjected to evolution as knowledge increases. None of the models currently a v a i l a b l e are suitable for an analysis of the s u s t a i n a b i l i t y of eucalypts plantations because of the lack of representation of coppicing a b i l i t y . In addition, these models are developed based on concepts of N that i s considered the most l i m i t i n g factor for plant growth i n a temperate region. However, P i s the l i m i t i n g nutrient i n the "cerrado" region where eucalypts have been recently planted. Before any model can be used to evaluate the long-term consequences of Inten-sive forest management on productivity of eucalypts i n the "cerrado" region, modifications should be made to include the a b i l i t y to simulate coppicing and the dynamics of P i n the ecosystem. 15 1.5 Objectives and Design of the Thesis The present study was undertaken to investigate ecosystem P and N inventories and P and N cy c l i n g i n Eucalyptus grandis plantations i n the "cerrado" region of B r a z i l , i n order to evaluate the effects of i n t e n -sive management on s i t e P and N status. It was recognized from the out-set that t h i s approach would be inadequate for the development of long-term sustained y i e l d management strategies f or t h i s area. Consequently, the thesis work included the development of conceptual models of two important aspects of eucalypt management i n the "cerrado": coppice regeneration and phosphorus dynamics i n the mineral s o i l . Studies of coppice sprouting were conducted to provide guidance for the development of the coppice model. The major objectives of the thesis are: 1. to quantify the biomass and P and N inventories of age sequences of Ej_ grandis plantations on two "cerrado" s i t e s that vary i n productivity; 2. to quantify the biomass and P and N dynamics i n l i t t e r f a l l and l i t t e r decomposition and to evaluate i n t e r v a l c y c l i n g due to the shedding of branches and f o l i a g e and due to heartwood formation over the age sequences on the two s i t e types; 3. to develop conceptual models of coppice regeneration and of phosphorus c y c l i n g ( e s p e c i a l l y sorption-desorption processes i n the mineral s o i l ) , and 4. to compare predictions concerning future biomass y i e l d s using the s t a t i c inventory approach and a dynamic budget approach. Because of the complexity of the data to be presented here the thesis i s subdivided into s i x chapters. 16 Chapter 1 - "General introduction". This chapter includes i n f o r -mation on ec o l o g i c a l conditions of the "cerrado" region and on the h i s t o r y and management of eucalypt plantations i n B r a z i l . The importance of nutrient c y c l i n g i n eucalypt plantations i n B r a z i l , and the need for a simulation approach to evaluate the e f f e c t s of intensive management on nutrients are also discussed. Chapter 2 - "Accumulation of biomass and nutrients i n age sequences of E. grandis plantations growing on good and poor "cerrado" s i t e s " . This chapter constitutes the c l a s s i c a l inventory approach to the evaluation of the effects of shortening r o t a t i o n and t o t a l or whole tree harvesting on s i t e nutrient status: an approach which analyses the d i s t r i b u t i o n of nutrients i n d i f f e r e n t biomass components. Chapter 3 - "Dynamics of nutrients ( i n t e r n a l c y c l i n g , l i t t e r f a l l and l i t t e r decomposition) i n age sequences of E. grandis plantations growing on good and poor "cerrado" s i t e s " . This chapter describes the dynamics of P and N i n E. grandis plantations i n the "cerrado" region by analysing l i t t e r f a l l , forest f l o o r accumulation, decomposition of branches and f o l i a g e , and i n t e r n a l c y c l i n g due to heartwood formation and the shedding of fo l i a g e and branches. Chapter 4 - "Factors determining coppice regeneration and growth, and a simulation of coppice growth". A review of factors determining coppice regeneration and growth, and the r e s u l t s of two experiments on sprout growth (greenhouse and f i e l d ) , are used i n the development of a conceptual model of coppice sprouting and growth. 17 Chapter 5 - "Factors determining the dynamics of P i n the s o i l system, and a simulation of P c y c l i n g i n the ecosystem". Based on a review of P a v a i l a b i l i t y to plants, a conceptual model of P dynamics i n the mineral s o i l i s developed. Chapter 6 - "Evaluation of the t r a d i t i o n a l methods of predicting the long-term consequences of intensive management of plantations i n B r a z i l " . 18 CHAPTER 2 ACCUMULATION OF BIOMASS AND NUTRIENTS IN AGE SEQUENCES OF E. grandis PLANTATIONS GROWING ON GOOD AND POOR "CERRADO" SITES 2.1 Introduction The "cerrado" region of B r a z i l that i s being extensively used f o r eucalypt plantations has a low l e v e l of s o i l f e r t i l i t y , and the planta-tions there are being managed on very" short rotations. This raises important questions as to the s u s t a i n a b i l i t y of y i e l d i n the plantations without much c o s t l y f e r t i l i z a t i o n . At the s t a r t of the thesis study i n 1981 there were i n s u f f i c i e n t data i n the l i t e r a t u r e to permit a meaning-f u l analysis of the p o t e n t i a l f o r declines i n future y i e l d due to s i t e nutrient impoverishment. For t h i s reason, the study reported i n t h i s thesis was undertaken to obtain the information on the accumulation and dynamics of biomass and nutrients that i s needed for an evaluation of the p o t e n t i a l e f f e c t s of intensive management on the future productivity of the s i t e s studied. Most of the studies of nutrient c y c l i n g i n eucalypt forests have been conducted i n A u s t r a l i a . Before 1977, most of the research was con-cerned with forest f l o o r and l i t t e r f a l l biomass and i t s nutrient content (as reviewed by Bevege, 1978). Most studies of biomass and nutrient content of l i v i n g tree components have been developed only recently. The most complete studies, which included l i v i n g biomass, l i t t e r f a l l , f o r e s t f l o o r and decomposition studies, were conducted on E. signata (Westman and Rogers, 1977a,b; Rogers and Westman, 1977) and E. obliqua 19 ( A t t i w i l l , 1968, 1979, 1980; A t t i w i l l et_ a l . , 1978; Lee and C o r r e l l , 1978; F e l l e r , 1980). Despite the importance of understanding the patterns of accumula-t i o n of biomass and nutrients over the l i f e of a stand, few studies have been conducted over an age sequence of eucalypts. Notable exceptions include studies of E. obliqua ( A t t i w i l l , 1979, 1980); E. grandis (Bellote et^a^., 1980; Bradstock, 1981) and E. globulus (Cromer et a l . , 1976; Cromer and Williams, 1982). Also, the effects of intensive management (such as short r o t a t i o n and total-tree-harvesting) on s i t e nutrient budgets have not been studied u n t i l recently (see Hingston et a l . , 1979; Crane and Raison, 1980; Wise and Pitman, 1981; Raison et a l . , 1982; Richards and Charley, 1983). Because of the lack of appropriate B r a z i l i a n data on the accumula-t i o n and dynamics of biomass and nutrients over age sequences of eucalypt stands, a study was undertaken to document the biomass and P and N content of E. grandis plantations of various ages. This was c a r r i e d out for two d i f f e r e n t s i t e s that vary i n productivity i n the "cerrado" region of the State of Minas Gerais. 2.2 L i t e r a t u r e Review 2.2.1 Biomass a. Regression equations to estimate biomass Because of the increasing i n t e r e s t i n t o t a l or whole tree harvesting, studies of biomass including crown components and roots 20 have become very important. The d i r e c t determination of the biomass of an en t i r e stand i s very time consuming and stand biomass i s normally estimated using regression equations based on the destructive sampling of a few trees. Most such work has only considered aboveground biomass. In r e l a t i v e l y few cases has the biomass of the root system been evaluated (Satoo and Madgwick, 1982). The regression equations used to i estimate the biomass of aboveground tree components are mostly based on the logarithmic r e l a t i o n s h i p between component weight and stem diameter; sometimes t o t a l height and i t s combination with diameter are also used (see the review by T r i t t o n and Hornbeck, 1982). However, diameter and t o t a l height are not considered to be s u f f i c i e n t to estimate the biomass of crown components. According to the review presented by Satoo and Madgwick (1982), variables such as crown length, sapwood basal area and diameter at the base of the l i v e crown should be included to improve the estimates of crown biomass. Regression estimates of root biomass are not commom. For t h i s reason these authors give no s p e c i a l recommenda-tio n concerning predictive variables for roots. Santantonio et a l . (1977) reported a logarithmic equation for the root biomass of three trees based on stem diameter, and a regression of l a t e r a l root fresh weight on root-end diameter. As already mentioned, the biomass equations used by most researchers have mainly been logarithmic because they reduce the variance associated with successive increase i n tree s i z e . According to B a s k e r v i l l e (1972), t h i s type of r e l a t i o n s h i p gives biased estimates unless a correction i s made, and c o r r e c t i o n methods have been developed (Satoo and Madgwick, 1982). However, Madgwick (1983) found that the 21 bias and v a r i a b i l i t y of estimates were greater when these correction factors were used, as compared with logarithmic regressions without the correction f a c t o r s . The question as to whether or not biomass estimates obtained using logarithmic equations should be corrected thus remains unresolved. In many cases, equations obtained to estimate forest biomass have been based on data which are s i t e - and stand-age s p e c i f i c . The s i m i l a r i t y of equations f o r d i f f e r e n t s i t e s and ages can be tested (Kozak, 1970) to see i f generally applicable regression can be used. T r i t t o n and Hornbeck (1982) and Crow (1983) observed that regression equations obtained from d i f f e r e n t stands of red maple (Acer rubrum) gave s i m i l a r tree biomass estimates. Bradstock (1981) obtained a commom equation to predict stem biomass of planted grandis over an age sequence ranging from 2 to 27 years old, each stand age being located i n a d i f f e r e n t s i t e with variable c h a r a c t e r i s t i c s . On the other hand, Koerper and Richardson (1980) reported s i m i l a r regression equations for good and intermediate s i t e s for a l l biomass components, but d i f f e r e n t equations for poor s i t e s . They found that the use of a general equation for a l l s i t e s gave the least error for stemwood and stembark. Satoo and Madgwick (1982) presented r e s u l t s of several studies that showed that variables other than tree si z e are required to permit a general a p p l i c a -b i l i t y of a single regression equation to estimate crown components over a range of s i t e . 22 b. Biomass studies for eucalypts Several biomass studies have been completed over the l a s t decade f o r eucalypt species i n natural conditions. Table 1 presents the t o t a l biomass and N and P content obtained from these studies. The stands studied are generally older than 27 years and the data were obtained for only one s i t e per study. In the case of natural f o r e s t s , a change i n s i t e w i l l sometimes imply a change of species composition. Consequent-l y , comparison of the productivity of the same species between s i t e s i s d i f f i c u l t . For example, Hingston ejt a l . (1979) studied E. d i v e r s i c o l o r forests i n two d i f f e r e n t s i t e s . On a red earth s i t e t h i s species was predominant, while on a yellow podzolic s o i l s i t e i t was associated with E. c a l l o p h y l l a . Despite the variations i n species composition, these studies are useful In order to compare the productivity of eucalypts i n general between d i f f e r e n t ecosystems. As can be seen from Table 1 there i s great v a r i a b i l i t y i n productivity i n the ecosystems studied so f a r . The aboveground biomass of eucalypt species occurring i n those ecosystems ranges from 104 t/ha for E. radiata/E. dalrympleana forest (Turner, 1980) to 883 t/ha f o r E. s i e b e r i forest (Ashton, 1976). Studies of natural eucalypt forests over an age sequence have only been undertaken f o r E_^ obliqua ( A t t i w i l l , 1979); age-sequence studies help to understand the dynamics of the ecosystem better than studies of a single stand age. Root studies are very d i f f i c u l t to carry out. For t h i s reason t h i s aspect of tree biomass has been less i n t e n s i v e l y studied despite i t s importance for plant growth. F e l l e r (1980) reported that root biomass accounted for about 10% of t o t a l biomass i n his study; however, TABLE 1. Aboveground living biomass (t/ha) and nutrient content (kg/ha) of Eucalyptus spp. under natural conditions* Biomass Source E. obliqua - 44 yr old 242 21.4 -55 yr old 298 26.7 -66 yr old 371 33.7 -E. obliqua/E. dives 2 418 (10.9) 26.0 (34.6) 485 (13 •2) E. 2 regnans* 664 (9.5) 38.0 (31.6) 412 (31 .6) E. regnans 793 15.8 -E. sieberi 883 13.1 -E. signata/E. umbra^ 118 (43.5) 10.3 (49.3) 301 (37 •5) E. diversicolor 225 18.0 189 E. diversicolor/E. callophylla 284 26.0 333 E. marginata/E. callophylla 262 13.0 321 E. radiata/E. dalrympleana 104 51.4 301 A t t i w i l l (1979, 1980) Feller (1980) Ashton (1976) Westman and Rogers (1977a, 1977b) Hingston et a l . (1979) Turner (1980) * Excluding understory and other tree species. 2 Including root system without taproot. 3 Including root system with taproot. 4 Numbers within brackets represent percentage of total biomass, P and N in the root system. 24 the taproot was not included (Table 1). On the other hand, Westman and Rogers (1977a) found that root biomass of E^ _ signata/E. umbra comprised an average of 43.5% of t o t a l l i v i n g biomass of these two species when a l l roots down to 5 mm diameter were considered. Ashton (1976) sampled the biomass of fine roots i n the upper 12.5 cm of s o i l and forest f l o o r of Ej_ regnans/E. s i e b e r i forest and found 0.4% of the t o t a l tree biomass i n t h i s biomass compartment. Planted eucalypts have been studied recently (Table 2). Singh and Sharma (1976) presented biomass of above- and belowground components for planted E. t e r e t i c o r n i s aged 5-9 years. The t o t a l biomass was s t i l l increasing by age 9 years, while the maximum fol i a g e biomass (16097 kg/ha) was attained at age 8 years. Root biomass constituted 13% (average for 5 ages) of t o t a l tree biomass. Cromer et a l . (1976) and Cromer and Williams (1982) reported on the biomass of planted E. globulus at age 2, 4, 6, and 9.5 years under four d i f f e r e n t f e r t i l i z a t i o n regimes. S i g n i f i c a n t differences between treatments were observed i n the accumulation of biomass. For example, foliage biomass of the u n f e r t i l i z e d plot was s t i l l increasing l i n e a r l y at age 9.5 years, while i n the most heavily f e r t i l i z e d plot maximum leaf biomass had been approached by age 4 years and leaf biomass had s t a b i l i z e d by age 6 years. Total stand biomass was s t i l l increasing at 9.5 years In each of the plots due to continued increase i n stem biomass. Studies over an age sequence of E. grandis were reported by Bellote et a l . (1980) i n Sao Paulo, B r a z i l , and by Bradstock (1981) i n New South Wales, A u s t r a l i a . Unfortunately, the data for d i f f e r e n t stand ages were obtained from d i f f e r e n t stands, which may have been growing on si t e s of d i f f e r e n t TABLE 2. Aboveground l i v i n g biomass (t/ha) and nutrient content (kg/ha) of Eucalyptus plantations Age (yr) Stand density (trees/ha) Biomass* P I N I Source E. grandis 1 1500 2 12.1 (66.5) 7.5 (85.3) 112 (91.2) Bellote et a l . (1980) 2 39.4 (31.4) 11.4 (58.8) 199 (68.9) 3 64.3 (23.5) 19.5 (45.1) 294 (53.0) 4 109.9 (U.6) 27.3 (17.5) 256 (36.7) 5 149.1 (9.6) 24.5 (26.9) 306 (32.6) 6 289.6 (8.8) 47.9 (22.3) 709 (26.9) 7 224.2 (5.2) 31.8 (14.6) 429 (21.8) E. grandis 2 996 18.3 (53.8) Bradstock (1981) 5 961 53.2 (31.2) 6 810 27.5 (28.7) 11 762 84.2 (14.8) 12 830 196.7 (9.9) 15 1219 164.7 (9.2) 16 756 189.4 (12.5) 27 790 394.0 (6.8) 25.1 (45.9) 435 (36.1) Turner and Lambert (1983) E. grandis 2.5 5333 85.7 (15.6) 28.0 (38.2) 276 (60.7) Pogglanl et a l . (1983) E. nitens 4 6470 81.8 (27.0) 23.2 (59.9) 332 (63.9) Madgwick et_ a l . (1981) E. fastlgata 4 7250 61.8 (38.2) 19.9 (69.8) 319 (73.4) " " " E. globulus 2 A3 1.05 (63) 0.53 (81) 10.1 (89) Cromer eit a l . (1976) D 8.55 (53) 4.87 (72) 53.1 (84) 4 A 6.30 (35) 1.60 (66) 26.3 (76) D • 30.30 (35) 9.26 (53) 92.1 (73) 6 A 3.52 42.9 Cromer & Williams (1982) D 14.16 125.3 9.5 A 4.87 70.6 D 15.28 153.2 Numbers within brackets represent the percentage of total biomass in crown components. Stand density used to transform the average biomass per tree given by the author into biomass per ha. ^Different levels of f e r t i l i z a t i o n - (A) u n f e r t i l i z e d plot and (D) heaviest f e r t i l i z a t i o n . U l 26 q u a l i t y . Consequently, the observed v a r i a t i o n between d i f f e r e n t stand ages may be at least p a r t l y due to variations i n s i t e c h a r a c t e r i s t i c s and stand density. For example, Be l l o t e et^ al^ (1980) observed smaller t o t a l biomass at age 7 years as compared with 6 years. Similar age sequence data anomalies were observed by Bradstock (1981). The proportion of aboveground biomass allocated to crown com-ponents i s presented i n Table 2. A comparison of the two studies of E. grandis at age 6 years reveals a wide v a r i a t i o n i n the r e l a t i v e magni-tude of the crowns; the stand studied by Bradstock (1981) had a much larger proportion of biomass i n crown components than that studied by B e l l o t e et^ a l . (1980). This difference might be due to s i t e d i f f e r -ences. Cromer et a l . (1976) observed that u n f e r t i l i z e d plots at age 2 years had a greater proportion of biomass i n crown components than the plot which was f e r t i l i z e d . However, by age 4 years no difference was observed between treatments. Problably there was l i t t l e r esidual effect of the f e r t i l i z e r by t h i s age and the carrying capacity of the s i t e had become more a function of the natural s i t e f e r t i l i t y . 2.2.2 P and N i n L i v i n g Biomass Studies of the d i s t r i b u t i o n of nutrients within tree components are basic to a determination of the e f f e c t s of intensive management on s i t e nutrient c a p i t a l and hence on future s i t e p r o d u c t i v i t y . The nutrient content of each tree component i s usually obtained by m u l t i p l y -ing the estimated biomass by the average nutrient concentration i n that given component ( F e l l e r , 1980; Westman and Rogers, 1977b). The nutrient 27 content can also be calculated by using regression equations based on e a s i l y measured vari a b l e s . A t t i w i l l (1980) determined the nutrient content of each tree component by using the allometric r e l a t i o n s h i p between nutrient content and tree diameter. Several studies have already been undertaken for native eucalypts i n A u s t r a l i a (Table 1). The reported P and N content of aboveground l i v i n g biomass ranges from 13.0 to 38.0 kg/ha and 249 to 485 kg/ha, res p e c t i v e l y . Nutrient concentrations i n aboveground tree components usually decrease i n the following order: leaves > stembark > branches > stemwood ( F e l l e r , 1980; Hingston e_t a l . , 1979; Madgwick et a l . , 1981). When T r i t t o n and Hornbeck (1982) reviewed biomass equations, they included those that did not take into account foliage biomass, consider-ing that t h i s component makes a small contribution to t o t a l tree bio-mass. However, fo l i a g e i s very important when nutrients are being analysed because of the high nutrient concentration i n leaves. For example, A t t i w i l l (1980) observed that 18% of the P i n the aboveground biomass of native E. obliqua at age 51 years was located i n the leaves. The effect of age on nutrient d i s t r i b u t i o n within tree components i s s t i l l not well documented. Most of the studies of eucalypts c o n s t i -tute a description of the nutrient budget at a single stand age. How-ever, i t i s very important to know the dynamics of accumulation of nutrients over an age sequence. This w i l l permit a better understanding of the e f f e c t s of shortening r o t a t i o n on nutrient c y c l i n g . Nutrient d i s t r i b u t i o n within tree components over an age sequence of plantations was studied f o r E_^ grandis by Bellote e_t al^. (1980) and for E^ globulus by Cromer et a l . (1976) and Cromer and Williams (1982). The r e s u l t s 28 presented i n Table 2 show that more than 25% of the t o t a l P and N of an E. grandis stand less than 6 years old i s located i n crown components. This proportion increases s i g n i f i c a n t l y at younger ages; at age 4 the crown accounted f o r more than 50% f o r E_^ nitens and E^ _ f a s t i g a t a (Madgwick et a l . , 1981) and for E. globulus (Cromer et^ al_. , 1976). The root system can account f o r up to 49.3% of the P and up to 37.5% of the N i n the t o t a l tree (Table 1). These r e s u l t s show the importance of studies of root systems f or the analysis of nutrient dynamics i n the ecosystem. The effect of s i t e on the nutrient content of eucalypts has not yet been well studied. As pointed out e a r l i e r , when native forests are studied i n d i f f e r e n t s i t e s there may be a v a r i a t i o n i n species composi-t i o n . The only study for eucalypt species that allows analysis of s i t e e f f e c t i s that presented by Cromer et^ a l . (1976) and Cromer and Williams (1982) for E. obliqua under four d i f f e r e n t f e r t i l i z e r l e v e l s . At age 2 years, when the f e r t i l i z e r e f f e c t was s t i l l strong, plants of the heaviest f e r t i l i z e r plot had 5.2 and 9.2 times more N and P, respective-l y , as compared with u n f e r t i l i z e d p l o t s . A s l i g h t l y greater proportion of both nutrients was allocated to crown components i n the f e r t i l i z e d plot as compared to the u n f e r t i l i z e d plot due to a greater proportion of biomass i n the crown i n the f e r t i l i z e d p l o t s . 2.3 Description of the Study Area The plantations are located at "Companhia A g r i c o l a e F l o r e s t a l Santa Barbara" (CAF), at Bom Despacho and Carbonita, Minas Gerais 29 (Figure 2). Mean annual temperatures for the period 1981-1983 were 21.5°C and 20.4°C, and mean annual p r e c i p i t a t i o n were 1731 mm and 1351 mm f o r Bom Despacho and Carbonita, r e s p e c t i v e l y . Monthly temperature and p r e c i p i t a t i o n are given i n Table 15 (page 79). The Bom Despacho plantation i s considered to be a good "cerrado" s i t e , compared with the Carbonita plantation which i s considered to be a poor s i t e ("good" and "poor" are used as r e l a t i v e terms within the "cerrado" region). The two s i t e s were d i f f e r e n t i a t e d into good and poor based on the productivity of natural vegetation and e x i s t i n g eucalypts plantations. The natural vegetation at Carbonita i s smaller and more scattered than that of Bom Despacho. The productivity of 24-month-old E_. grandis plantation, as reported by Rezende et a l . (1982b), i s 8.4 m^  f o r Bom Despacho and 2.2 m^  f o r Carbonita. At each s i t e type, plantations of d i f f e r e n t ages with 3.0x1.5 m or 3.0x2.0 m spacing were studied (1667 or 2220 trees/ha). At planting time, each tree received 150 g of 10-28-6 + micro (N, P, K, B, Zn) f e r t i l i z e r . This i s approximately equivalent to the following dosage (kg/ha) f o r the 3x2 m spacing stands: N - 25; P - 30.5; K - 12.5 and a trace of B and Zn. 2.4 F i e l d Method 2.4.1 Biomass Sampling I n i t i a l biomass sampling was conducted i n June 1981 f o r stands 15, 26, 38, 51 and 62 months old for the good s i t e , and 21, 32, 43 and 56 months old i n the poor s i t e . Additional trees were f e l l e d i n May 1982 i n the youngest and oldest stands of each s i t e . At t h i s time these two Figure 2. Map of location of the study areas i n Minas Gerais State. Belo Horizonte i s the State c a p i t a l . Source: G o l f a r i (1975) o 31 stands were 26 and 73 months old at Bom Despacho and 32 and 67 months old at Carbonita. Three plots of 600 m^  (15x40 m) each were measured for each stand age at each s i t e type. Diameter at breast height was taken f o r a l l trees i n each p l o t . Average t o t a l height for each d i a -meter class was obtained from inventory information. Three trees were f e l l e d i n each stand for biomass evaluation and nutrient a n a l y s i s . A t o t a l of 21 trees were f e l l e d on the good s i t e and 18 trees on the poor s i t e . Sampled trees had diameters ranging from 2.5 to 17.4 cm on the good s i t e and 2.9 to 12.7 cm on the poor s i t e . Total height and diameter at breast height were measured f o r each tree sampled. Sampled trees were subdivided into the following components: stemwood, stembark, f o l i a g e , branches, and roots. Stems sections less than 3 cm diameter were considered as branches as t h i s i s the size l i m i t used f o r charcoal production. Total fresh weight of leaves and branches was measured i n the f i e l d . A homogenized subsample of each was taken fo r dry weight determination. The stem (wood + bark) was subdivided into two parts (base and top) which were weighed i n the f i e l d . A disc was cut at the base, middle and top of the stem. The following measure-ments were taken for each di s c : thickness of the disk, circumference inside and outside of bark, heartwood diameter, and fresh and dry weight of bark and of wood, separately. Where heartwood was already developed, i t was separated from sapwood f o r chemical analysis. Root biomass was determined by hand digging the root system of at le a s t one tree i n each stand age i n order to have v a r i a t i o n i n tree size for regression analysis (12 trees on the good s i t e and 14 trees on the 32 poor s i t e ) . One quarter of the horizontal s o i l area occupied by each tree was excavated to a depth of 40 cm to sample l a t e r a l roots. This depth was based on previous studies of rooting depth i n the same area (Rezende, pers. comm.). Taproots were excavated to a depth of 1.0 to 1.5 m. Late r a l roots were subdivided into three diameter c l a s s e s , as follows: smaller than 0.3 cm; 0.3 to 1.0 cm and greater than 1 cm. For both l a t e r a l roots and the taproot, t o t a l fresh weight was obtained i n the f i e l d and a sample was taken f o r dry weighing. 2.4.2 S o i l Sampling A composite s o i l sample of four subsamples was co l l e c t e d per plot at depths of 0-10 cm and 10-30 cm. This gave three bulked samples for each depth per stand age. 2.5 Laboratory Methods 2.5.1 Plant Analysis Dry weight of a l l samples was evaluated a f t e r constant weight was obtained i n an oven at 80°C within one week a f t e r being c o l l e c t e d . Samples of plant components were ground to 2 mm immediately a f t e r being dr i e d . The samples were analysed chemically according to the following methods: nitrogen was determined by the Kjeldahl method; and phosphorus was determined by the ascorbic acid method a f t e r digestion with n i t r i c acid and perc h l o r i c aci d (Braga and Defelipo, 1974). The analyses were done at the Federal University of Vicosa, B r a z i l . 33 2.5.2 S o i l Analysis Chemical analysis was done at the Federal University of Vicosa, B r a z i l for pH, t o t a l and available P and exchangeable K, Ca, Mg and A l . A proportion of 1:2.5 s o i l r d i s t i l l e d water was used f o r pH evaluation. T o t a l phosphorus was determined c o l o r i m e t r i c a l l y after digestion with concentrated sulphuric acid. The available nutrients were analysed as follows. The Mehlich extractor ( d i l u t e hydrochloric acid and sulphuric acid) was used f or phosphorus and potassium. Phosphorus was determined c o l o r i m e t r i c a l l y using the molybdenum blue method. A flame photometer was used f or potassium determination. Calcium, magnesium and aluminum were extracted by IN potassium chloride. Calcium and Magnesium were determined by atomic absorption spectrophotometry (Black, 1965). Data for p l o t t i n g phosphate sorption isotherms were obtained by e q u i l i b r a t i n g 2.5 g of s o i l i n 25 ml of 0.01 C a C l 2 containing 0, 30 and 60 ppm of P. The tubes were shaken for 120 min. and after f i l t r a -t i o n the s o l u t i o n was c o l o r i m e t r i c a l l y analysed f o r P. A l l P not detected i n s o l u t i o n was considered sorbed. 2.6 Data Analysis Dry weight and nutrient content (P and N) were calculated f o r stemwood, stembark, f o l i a g e , branches, and each root size class for each sample tree. The biomass (on a dry weight basis) of the basal stem section was obtained by using the fresh-dry weight r e l a t i o n s h i p of the basal and middle d i s c s . The mean nutrient concentration of the basal and middle discs was m u l t i p l i e d by the biomass of the basal stem section to obtain nutrient content of the basal stemwood and bark. S i m i l a r l y , 34 data for the middle and top discs were used to obtain biomass and nutrient content of the top stem section. Taproots were analysed together with l a t e r a l roots greater than 1 cm diameter. Biomass and nutrient content equations (g/tree) were obtained f o r each tree component using a stepwise regression technique. Independent variables considered were diameter at breast height (cm), t o t a l height (m) and age (months), and various combinations and transformations thereof. The independent variable age was included as i t i s e a s i l y obtained for plantations. Also, i t has been demonstrated that stand age a f f e c t s the regression constants for crown components (Tadaki, 1966; cit e d by Satoo and Madgwick, 1982). At f i r s t , logarithmic equations were t r i e d f o r every component as they are the most commonly used type of equation (Satoo and Madgwick, 1982; T r i t t o n and Hornbeck, 1982). Equations with a non-significant intercept were forced through the o r i g i n . After the regressions were f i t t e d for each s i t e they were tested i n order to find out whether or not they could be combined to reduce the number of equations. The trees from each sample plot were subdivided into 1.2 cm diameter classes and biomass and nutrient content were estimated for the tree of mean diameter i n each cla s s for each stand age. These values were then m u l t i p l i e d by the number of trees i n each c l a s s . The r e s u l t s are presented on a per hectare basis over an age sequence f o r the two s i t e types. 35 2.7 Results and Discussion Table 3 gives stand density and average tree diameter and height and Figure 3 gives MAI f o r the s i t e s studied. The volume was obtained by mult i p l y i n g the biomass on hectare base by wood density.^ Because the data for each age were obtained i n d i f f e r e n t stands, the r e s u l t s presented here r e f l e c t some unavoidable v a r i a t i o n i n s i t e character-i s t i c s within each of the two major s i t e types. There was also v a r i -a t i o n i n spacing. Stands older than 51 months on the good s i t e and 43 months on the poor s i t e were planted at 3.0x2.0 m spacing. Younger stands were planted at 3.0x1.5 m spacing. The r e s u l t s of s o i l analysis of each stand studied are given i n Appendix 1 and, diameter d i s t r i b u t i o n of each plot studied i s given i n Appendix 2. 2.7.1 Biomass a. Regression equations to estimate l i v i n g biomass Table 4 gives the biomass equations of each tree component for the two s i t e s studied. Logarithmic equations s i g n i f i c a n t at 5% p r o b a b i l i t y l e v e l were obtained for the biomass of stembark and stemwood on both s i t e types, and for medium roots on the good s i t e . The best independent variables obtained for those equations were mostly logD or logD^H, as has been found i n most other biomass studies (as reviewed by T r i t t o n and Hornbeck, 1982). No correction was made for bias due to use of Companhia A g r i c o l a e F l o r e s t a l Santa Barbara, Belo Horizonte, Minas Gerais, B r a z i l (Internal r e p o r t ) . TABLE 3. Stand c h a r a c t e r i s t i c s of the si t e s studied Age I n i t i a l Stand Average Average-^ spacing density diameter height (months) (m) (trees/ha) (cm) (m) (a) Bom despacho (good s i t e ) 15 3.0 x 1.5 2117 4.0 6.6 26 3.0 x 1.5 2133 7.2 8.9 38 3.0 x 1.5 2078 8.9 2 12.9 51 3.0 x 2.0 1561 12.9 16.4 62 3.0 x 2.0 1528 12.2 16.7 73 3.0 x 2.0 1417 12.8 18.0 Carbonita (poor s i t e ) 21 3.0 x 1.5 2111 4.5 6.9 32 3.0 x 1.5 2111 7.1 8.1 43 3.0 x 2.0 1533 8.3 8.2 56 3.0 x 2.0 1622 8.5 10.2 67 3.0 x 2.0 1611 9.0 10.9 ^Height data were taken from e x i s t i n g inventory of d i f f e r e n t plots within the same stands used i n this study. This accounts for the lack o agreement between height and diameter data f or months 32 and 43 on the poor s i t e . ^The lack of diameter increase at t h i s age i s thought to r e f l e c t s i t e v a r i a b i l i t y . Such data inconsistencies are a consequence of using d i f f e r e n t stands to make up the age sequence, rather than following the development of one stand over a number of years. 37 Age (months) Figure 3 - Mean annual increment of grandis plantations growing on two di f f e r e n t "cerrado" s o i l s i n Minas Gerais - B r a z i l TABLE 4. Equations f o r biomass ( g / t r e e ) of each t r e e component of 12. g r a n d i s p l a n t a t i o n s growing on two d i f f e r e n t " c e r r a d o " s o i l s I n Minas G e r a i s , B r a z i l 2 No. of Component Equations R Sy.x o b s e r v a t i o n s (a) Bom Despacho (good s i t e ) Stembark l o g Y - 0.67068 + 0.92456 l o g D2H 0.99 1 0.07 1 21 Stemwood l o g Y - 1.04834 + 1.02902 l o g D2H 0.99 1 0.05 1 21 Branches Y - 1617 .53 + 80.27794 D 2 - 44.21629 Dli 0.91 880.21 21 F o l i a g e (age <38) Y - 67.050D 2 - 0.47512 H 2Age 0.82 850.89 12 (age >51) Y - 45.458D 2 - 0.44152 D 2Age 0.93 698.99 9 Fine r o o t s Y - -1559.04 + 896.43834 l o g H Age 0.83 158.94 12 Medium roots l o g Y - -2.53333 + 2.14969 l o g II Age - 0.00082669 H Age 0.91 1 0.15 1 12 Large roots Y 31.88751 Dll 0.92 1045.70 11 C a r b o n i t a (poor s i t e ) Stembark l o g Y 1.05905 + 2.51491 l o g D 0.96 1 0.08 1 18 Stemwood l o g Y - 1.10581 + 1.01205 l o g D2H 0.98 1 0.07 1 18 Branches Y - 880.35 + 46.01188 D Z - 0.33741 H 2Age 0.95 313.56 18 F o l i a g e (age <43) Y 577.01 + 34.92379 D 2 0.84 376.09 12 (age >56) Y - 6803.6 + 0.71449 D 2Age - 3284.75139 l o g D 2 0.996 99.41 6 Fine roots Y - 774.96 + 2.70617 H Age 0.73 437.74 14 Medium roots Y - -2952.97 + 2651.43551 l o g Age 0.44 541.79 14 Large r o o t s Y - 1139.41 + 0.71249 D 2Age 0.85 984.28 14 D • Diameter at breast height (cm); II = T o t a l height (m); Age - Months. 1 R 2 and Sy.x are based on l o g a r i t h m i c y-valuea. A l l equations are s i g n i f i c a n t a t P <0.05. Co oo 39 logarithmic r e l a t i o n s h i p s . Various d i f f e r e n t independent variables proved best for the other biomass components. Foliage biomass data for the poor s i t e showed two very d i s t i n c t groups, so equations were obtained f o r each group separately. This d i f f e r e n t i a t i o n into two groups probably occurred due to v a r i a t i o n i n s o i l conditions because stands were located i n d i f f e r e n t s i t e s . The data for the two oldest ages (>51 months) were obtained from the same stand i n two subsequent years. The three good s i t e stands older than 38 months were planted at a wider spacing and also were located i n the same area and consequently i n the same type of s o i l . At t h i s age the stands would have reached f u l l canopy closure. Shinozaki et a l . (1964b) ( c i t e d by Koerper and Richardson, 1980) suggest that shade-intolerant species present greater v a r i a b i l i t y i n the r e l a t i o n s h i p between crown component weight and diameter as compared to shade-tolerant species. E. grandis i s a very l i g h t demanding species (Jacobs, 1955), and the foliage bio-mass and tree size r e l a t i o n s h i p w i l l consequently change a f t e r canopy closure. For example, a tree with small diameter w i l l have greater f o l i a g e biomass before canopy closure as compared to a tree of the same size at older ages when l i g h t competition i s high. Tadaki (1966) ( c i t e d by Satoo and Madgwick, 1982) demonstrated the ef f e c t s of stand age on the regression constants for crown components. A highly s i g n i f i c a n t regression equation (R 2 = 0.94) f o r fol i a g e biomass based on D^H was obtained by Poggiani et a l . (1983) for 2.5-year-old E. grandis growing i n the Bom Despacho region (the good s i t e of the present study). This shows that when an even-aged stand i s considered, a good r e l a t i o n s h i p can be obtained between dry weight of fol i a g e and tree size as opposed 40 to a poorer r e l a t i o n s h i p obtained i n the present study with stands of d i f f e r e n t ages, without s t r a t i f i c a t i o n . The independent variable age i s not commonly considered for b i o -mass equations because the biomass studies are usually developed f o r stands of a single age and also because of the d i f f i c u l t i e s i n obtaining "this information f o r uneven-aged natural stands. However, for planta-t i o n s , t h i s variable i s e a s i l y obtained and may contribute greatly to improve biomass estimation. For t h i s reason, age was included i n the present study and i t proved to be very important as an estimator of the biomass of crown and root system components. Other variables, such as length of l i v e crown and sapwood basal area, have been used i n f o l i a g e biomass equations (Satoo and Madgwick, 1982) and should be included i n future studies of eucalypts i n order to try to Improve the estimation of those components. The test of p a r a l l e l l i s m (Kozak, 1970) revealed that the data for the two s i t e s could be combined to give a single equation for stembark, stemwood, branches and large roots, independent of s i t e . The following common equations were obtained: Stembark: LogY = 0.95642 + 2.63401 logD (R2=0.97, Sy.x=0.089); Stemwood: LogY = 1.07757 + 1.02073 logD 2H (R2=0.99, Sy.x=0.058); Branches: Y = 653.93787 - 0.15115 H 2Age + 41.27635 D 2 (R2=0.88, Sy.x=0.88); and Large Roots: Y = 1626.433 + 0.55703 D2Age (R2=0.83, Sy.x=0.83). As pointed out e a r l i e r , regression equations to estimate stem biomass can e a s i l y be applied to d i f f e r e n t s i t e s due to the consistent r e l a t i o n s h i p between stem weight and tree s i z e . The common equation 41 obtained for stemwood biomass gave estimates that d i f f e r e d very l i t t l e from the s i t e s p e c i f i c equations, while the common equations for stem-bark, branches and large roots gave underestimates of biomass for some ages and overestimates for other ages, on both s i t e s . Because i t i s important for the present study to have accurate biomass estimates, s i t e s p e c i f i c equations were used f o r these biomass components. b. Estimated biomass The estimated biomass f o r each tree component over an age sequence i s presented i n Table 5, and the biomass of stem, t o t a l aboveground and t o t a l roots i s presented i n Figure 4 f o r both s i t e s . Rapid early growth occurred on both s i t e s . By age 15 and 21 months on the good and poor s i t e s , r e s p e c t i v e l y , the stem (wood + bark) production was 27 5 and 268 kg/ha/month. This r e f l e c t s the effect of f e r t i l i z e r applied at the time of planting. In the second year of growth the effect of s i t e quality became evident, with stem increment values of 725 and 421 kg/ha/month at ages 26 and 32 months for the good and poor s i t e s , r e s p e c t i v e l y . The biomass of stembark, stemwood, and large roots s t a b i l i z e d at age 51 months on the good s i t e , while on the poor s i t e they were s t i l l increasing at age 67 months. At the oldest age studied, the stemwood biomass was about 2.5 times greater on the good s i t e as compared to the poor s i t e . However, biomass production at the oldest age on the good s i t e was only twice as great as that on the poor s i t e when t o t a l biomass was considered, due to the higher root production on the poor s i t e . Foliage biomass increased up to age 38 months (8546 kg/ha) on the good s i t e and to age 32 months (4353 kg/ha) on the poor s i t e . A TABLE 5. Biomass accumulation (kg/ha) of each tree component over an age sequence of E. grandis plantations growing on two different "cerrado" soils in Minas Gerais, Brazil Age Total Fine Medium Large Total Total 1 1 2 3 4 1 (months) Bark Wood Branches Foliage aboveground roots roots roots roots tree (a) Bom Despacho (good site) 15 824 3307 3744 1742 9617 (45.6) 471 (3.7) 6 102 1853 2426 (20.I) 5 12043 26 3299 15548 6700 5796 31343 (33.3) 1184 (4.9) 503 4533 6220 (16.6) 37563 38 5007 24716 9631 8546 47900 (32.4) 1501 (5.7) 951 5798 8250 (14.7) 56150 51 10549 58253 7941 5516 82259 (14.1) 1632 (3.4) 1723 10014 13369 (14.0) 95628 62 10522 58525 8764 4439 82250 (13.8) 1692 (2.6) 1794 9674 13160 (13.8) 95410 73 10956 61842 9135 3328 85261 (12.7) 1669 (2.0) 1689 9633 12991 (13.2) 98252 i Carbonita (poor site) 21 1163 4464 3185 2778 11590 (32.6) 2464 (1.13) 1167 3074 6705 (36.7) 18295 32 2817 10663 4468 4353 22301 (27.5) 3113 (1.4) 2191 4452 9756 (30.4) 32057 43 2670 9894 3493 3683 19740 (25.0) 2644 (1.4) 2113 4202 8959 (31.2) 28699 56 4519 19095 3840 1015 28469 (11.6) 3763 (0.27) 2729 6802 13294 (31.8) 41763 67 5308 22947 3477 2356 34088 (14.5) 4417 (0.53) 3043 8552 16012 (32.0) 50100 Including stems smaller than 3 cm diameter. Diameter less than 0.3 cm. Diameter between 0.3 and 1.0 cm. Diameter greater than 1.0 cm (including taproot). Percentage of the total biomass allocated to crown components and root system, respectively. Foliage/fine roots ratio. 43 100 n Legend A Total Aboveground O Stem_ _ _ _ • Total Roots (a) 26 38 Age (months) r 51 73 (0 at C O E o m 50 -1 4 0 -Figure 4. 21 32 43 Age (Months) Biomass (t/ha) of stem, total aboveground and total roots over an age sequence of E. grandis plantations growing on two different "cerrado" soils in Minas Gerais, Brazil: (a) Bom Despacho - good site and, (b) Carbonita - poor site. subsequent decrease occurred on both s i t e s . This decrease was probably a r e s u l t of an imbalance between leaf mortality and leaf production due to increasing s o i l nutrient depletion as the effects of the i n i t i a l f e r t i l i z a t i o n ceased (see Appendix 1 f o r data on available P). The r a t i o f o l i a g e / f i n e roots (Table 5) decreased at about the same ages, which r e f l e c t s d e c l i n i n g i n P a v a i l a b i l i t y . In a experiment with E.  grandis and E. saligna growing adjacent to the present study s i t e s , Barros et a l . (1981) noted that the growth rate i n height was much higher on f e r t i l i z e d plots than on u n f e r t i l i z e d c o n t r o l plots during the f i r s t 2 years, but that they were s i m i l a r a f t e r 4.5 years. However, Cromer and Williams (1982) observed no decline i n f o l i a g e biomass for E. obliqua under d i f f e r e n t f e r t i l i z e r l e v e l s . In t h e i r study the stemwood increment for f e r t i l i z e d E. obliqua was s t i l l greater than for unfer-t i l i z e d trees at age 9.5 years, so presumably there was s t i l l some effect of the i n i t i a l f e r t i l i z a t i o n . Quantities of f e r t i l i z e r applied and s o i l c h a r a c t e r i s t i c s should be taken into account in order to better in t e r p r e t these r e s u l t s . As discussed i n Chapter 5, s o i l s from the "cerrado" region are l a t o s o l s with a very high P f i x a t i o n capacity. This a f f e c t s the a v a i l a b i l i t y of P (which i s the major l i m i t i n g nutrient for plant growth i n these s o i l s ) , and i s probably the reason f o r the s h o r t - l i v e d e f f e c t of the i n i t i a l f e r t i l i z a t i o n . The maximum estimated f o l i a g e biomass for age 38 months on the good s i t e (8546 kg/ha) was high compared to r e s u l t s of other studies for the same species. Poggiani et a l . (1983) reported a biomass of 7618 kg/ha for the same species, on the same s i t e , at age 30 months, with a density of 5333 trees/ha, the t o t a l aboveground tree biomass being 85714 kg/ha. Bellote et a l . (1980) reported a maximum fol i a g e biomass for 45 E_. grandis of 7284 kg/ha at age 36 months (this biomass estimate was obtained by multiplying the mean of four trees given by the authors by a density of 1500 trees/ha). Considering that t o t a l aboveground biomass i n t h e i r study (Table 2) was greater than that obtained on the good s i t e of the present study i t i s expected that f o l i a g e biomass should also have been greater than i n this study. Bradstock (1981) reported a maxi-mum foliage biomass of 6150 kg/ha for planted E. grandis at age 27 years i n a stand with 790 trees/ha. Foliage biomass as well as t o t a l above-ground biomass was s t i l l increasing up to t h i s age. However, the value obtained by Singh and Sharma (1976) f o r 8-year-old E ^ t e r e t i c o r n i s (16097 kg/ha) i s much higher than those reported for E. grandis. As already discussed, the re l a t i o n s h i p between predictive tree size variables and fol i a g e biomass changes with age. For t h i s reason, equations f o r fo l i a g e biomass estimates should be developed for each stand age or other pr e d i c t i v e variables should be included i n the equation i n order to improve the r e s u l t s . The biomass of f i n e roots s t a b i l i z e d at age 38 months on the good s i t e , while i t was s t i l l increasing on the poor s i t e at age 67 months. The maximum f i n e root biomass achieved on the poor s i t e (4417 kg/ha) was 2.6 times greater than that of the good s i t e (1692 kg/ha). Keyes and Grie r (1981) noted that fin e root production and turnover i n 40-year-old Pseudotsuga menziesii growing on a poor s i t e was about 2.5 times greater than i n the same species growing on a good s i t e . The r a t i o of above-ground biomass between the good and poor s i t e i n their study was 1.9, while the r a t i o for the present study was 2.5. This suggests that the difference between s i t e s i s much greater i n the present study and i t can 46 be hypothesized that the difference i n fine root biomass w i l l also be greater by the time f i n e root biomass s t a b i l i z e s on the poor s i t e of my study area. No studies of the root systems of eucalypts growing on d i f f e r e n t s i t e s had been undertaken p r i o r to this study. The d i s t r i b u t i o n of t o t a l biomass between tree components varied with s i t e . The crown accounted for about the same proportion of the t o t a l biomass on both s i t e s . The proportion of aboveground biomass i n crown components observed i n t h i s study was higher than those obtained by Bellote et a l . (1980) (Table 2). The proportion of stembark was about the same on both s i t e s (approximately 11% at the oldest age studied). However, stemwood was only 45.8% of t o t a l tree biomass at age 67 months on the poor s i t e , while i t was about 60.0% on the good s i t e at ages older than 51 months. The main reason for t h i s d i s s i m i l a r i t y between the two s i t e s i s probably the increased a l l o c a t i o n of photo-synthates to the root system on the poor s i t e ( c . f . Keyes and Grier, 1981). The proportion of t o t a l tree biomass allocated to the root system on the poor s i t e was 29.4-36.7%, while on the good s i t e the values were 13.2-14.0% for ages older than 51 months and 14.7-20.1% for younger stands (Table 5 and Figure 4). The a l l o c a t i o n of biomass to the root system on both s i t e s i s between the range of 14-52% given by Santantonio et a l . (1977) for coniferous and deciduous fore s t s . A value of 43.5% was observed f o r E. signata/E. umbra by Westman and Rogers (1977a). F e l l e r (1980) reported that l a t e r a l roots of E. obliqua/E. dives c o n s t i -o tute about 10% of t o t a l biomass. Considering that the taproots of sampled trees i n the present study constituted an average of 57% of the 47 t o t a l roots on the good s i t e , and 35% on the poor s i t e , the values observed by F e l l e r , corrected to give estimates for t o t a l root biomass, would be s l i g h t l y greater than values observed for the good s i t e i n the present study. Also, the low values observed on the good s i t e are i n agreement with the value of 13% reported by Duvigneaud and Denayer-De Smet (1967) f o r deciduous mixed fore s t s , and by Singh and Sharma (1976) for t e r e t i c o r n i s plantations aged 5-9 years. D i s t r i b u t i o n of l a t e r a l roots according to s o i l depth was not measured i n the present study. The t o t a l l a t e r a l root sampling depth of 40 cm was selected based on preliminary r e s u l t s obtained from 5 trees of the same species, aged 1 to 5 years, growing on the good s i t e . More than 68% of the t o t a l roots of these trees were located i n the top 20 cm of s o i l (Rezende, pers. comm.). This value would have been higher on the poor s i t e , where most of the f i n e roots are concentrated i n a thick root mat i n the upper 10 cm of the s o i l . The taproot was excavated u n t i l the diameter of the root was about 1 cm. Only 5 out of 26 trees sampled had a taproot with diameter greater than 1 cm at a depth below 100 cm. In two trees, both on the poor s i t e , the taproot had a diameter of 5.7 and 3.5 cm at a depth of 150 and 127 cm, r e s p e c t i v e l y . The average proportion of taproot biomass below 40 cm depth was 16.5 and 19.8% i n sampled trees on the good and poor s i t e , r e s p e c t i v e l y . Based on data c o l l e c t e d i n the f i r s t year of sampling, only 8.2% of t o t a l taproot biomass was located below 60 cm depth. These r e s u l t s show that i t i s not worthwhile to excavate tap-roots to a depth greater than 100 cm for Ej_ grandis. 48 The d i s t r i b u t i o n of stemwood between sapwood and heartwood was not measured at the time of sampling. However, some measurements were made on photographs of the basal stem di scs. The d i f f e r e n t i a t i o n i n co l o r between sapwood and heartwood was quite sharp on the good s i t e which permitted a reasonably precise measurement. However, i t was less evident on the poor s i t e and i n t e r p r e t a t i o n of the res u l t s presented i n Table 6 should be made with caution. Table 6. Percentage of basal disc area represented by heartwood, over an age sequence of E_^_ grandis plantations growing on two di f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l . a. Bom Despacho (good s i t e ) Age (months) 26 38 51 62 73 % heartwood 8.9 19.2 39.8 39.6 36.1 b. Carbonita (poor s i t e ) Age (months) 32 43 56 67 % heartwood 7.2 20.5 29.5 44.2 It can be observed that by age 38 months on the good s i t e and 43 months on the poor s i t e , about 20% of stemwood i n the basal disc was made up of heartwood. The apparent decline i n the percentage of heart-wood formed at ages older than 62 months on the good s i t e might be due to lack of prec i s i o n of the measurement method or due a v a r i a b i l i t y i n heartwood formation between trees sampled. Because t h i s process i s very important i n terms of i n t e r n a l c y c l i n g of nutrients, more de t a i l e d studies to quantify v a r i a t i o n i n the timing of heartwood formation should be developed. 49 2.7.2 P and N i n L i v i n g Biomass Equations for the P and N content of each tree component selected by the stepwise regression method are presented i n Tables 7 and 8. P content of f i n e roots on the good s i t e and of f o l i a g e on the poor s i t e at younger ages showed d i s t i n c t l y lower values than for the two oldest ages. For t h i s reason they were s t r a t i f i e d . No s i g n i f i c a n t (p < 0.5) equations for P content were obtained for fin e roots for the two oldest stands on the good s i t e , or for medium roots on the poor s i t e , and no equation for N content was obtained for f i n e roots on the good s i t e . Also, equations for P and N content of medium roots on the good s i t e , and stembark on the poor s i t e and for N of stemwood on the good s i t e , gave negative estimates for the youngest age. Whenever equations could not be obtained to estimate P and N content, an average concentration f o r the component was obtained and m u l t i p l i e d by estimated biomass per hectare for each age. Tables 9 and 10 show the estimated P and N content for each tree component over an age sequence of plantations on the two s i t e s . P and N content of stem, t o t a l aboveground and t o t a l roots are presented i n Figures 5 and 6. On both s i t e s , the accumulation of P and N i n most tree components r e f l e c t e d the pattern of accumulation of biomass. P content of stemwood s t a b i l i z e d a f t e r age 51 months while N content was s t i l l increasing up to age 73 months on the good s i t e . Both P and N content was s t i l l increasing up to age 67 months on the poor s i t e . P and N accumulation i n stembark was s t i l l occurring on both s i t e s , except for N on the good TABLE 7. Equations f o r phosphorus ( g / t r e e ) of each t r e e component of E. gr a n d i s p l a n t a t i o n s growing on two d i f f e r e n t "cerrado" s o i l s i n Mlnaa G e r a i s , B r a z i l 2 No. of Component Equations R Sy.x o b s e r v a t i o n s (a) Bom Despacho (good s i t e ) Stembark Stemwood Branches F o l i a g e F i n e roots (age <51) (age >62) Medium roots Large roots Y - 0.00046639 D 2Age Y - 0.026803 D 2 Y - 0.053382 D 2 - 0.015875 DH - 0.001102 D2H Y - 1.4358 + 0.03925 D 2 - 0.00022696 H 2Age Y - 0.07066 + 0.00373 D 2 No v a r i a b l e s Y - -0.62145 + 0.22091 l o g H 2Age Y = 0.083516 D 0.88 0.91 0.90 0.70 0.95 0.62 0.45 1.15 0.91 O.34 1.34 0.04 0.10 0.48 16 18 21 21 8 12 11 (b) Carbonlta (poor s i t e ) Stembark Stemwood Branches F o l i a g e (age ^43) (age >56) Fine roots Medium roots Large roots Y - -0.16623 + 0.01113 D 2 Y'- 0.0014411 D2H - 0.00007203 H 2Age Y - 0.01266 D 2 - 0.00061663 HAge Y - 0.65097 + 0.06229 D 2 - 0.03452 Dil Y - 0.00072759 D 2Age - 0.0277 DH Y - 0.25044 + 0.00324 DH No v a r i a b l e s Y - 0.0050248 D 2 0.97 0.95 0.92 0.90 0.83 0.44 0.56 0.13 0.13 0.13 0.26 0.57 0.17 0.16 13 17 18 12 6 13 13 D - Diameter at breast height (cm); H ° T o t a l height (m); Age = Months-A l l equations are s i g n i f i c a n t at P <0.05. o TABLE 8. Equations f o r n i t r o g e n ( g / t r e c ) of each tree component of E. grandis p l a n t a t i o n s growing on two d i f f e r e n t "cerrado" s o i l s I n Minus G e r a i s , B r a z i l 2 No. of Component Equations R Sy.x ob s e r v a t i o n s (a) Bom Despacho (good s i t e ) Stembark Y - 0.0088593 D2II 0.98 2.69 16 Stemwood Y - 0.011981 D 2Age - 0.052115 D Age 0.97 10.36 18 Branches Y - 5.75427 + 0.41785 D 2 - 0.24075 Dll 0.95 3.08 21 F o l i a g e Y - 17.698 D - 5.0286 II - 0.063901 11 Age 0.80 18.92 21 Fine roots No v a r i a b l e s Medium roots Y - -11.9036 + 5.87417 l o g 11 Age 0.74 1.39 12 Large roots Y 1.7375 D 0.63 7.22 11 Carbonita (poor s i t e ) Stembark Y > -2.24397 + 0.13884 D 2 0.95 1.57 13 Stemwood Y - 0.014435 D21I - 0.00068614 ll 2Age 0.99 1.02 17 Branches Y - 0.23262 D 2 - 0.0013565 ll 2Age 0.90 2.38 18 F o l i a g e Y - 15.947 D - 7.9962 II - 0.0034755 ll 2 A g e 0.87 8.01 18 Fine roots Y - 5.46581 + 0.00114 ll 2Age 0.77 2.31 14 Medium roots Y - 0.10922 Age 0.53 1.67 14 Large roots Y - 0.0022978 D 2Age 0.83 2.84 13 D - Diameter at breast height (cm); II - T o t a l height (m); Age - Months. A l l equations are s i g n i f i c a n t at P <0.05. TABLE 9. Phosphorus content (kg/ha) of each tree component over an age sequence of E. grandis plantations growing on two different "cerrado" soils in Minas Gerais, Brazil Age Total Fine Medium Large Total Total (months) Bark Wood Branches Foliage aboveground* roots roots roots roots* tree (a) Bom Despacho (good site) 15 0.40 0.97 0.72 4.13 6.22 (66.9) 0.28 0.04 0.71 1.03 (14.2) 7.25 26 1.47 3.18 2.82 6.69 14.16 (58.5) 0.59 0.23 1.28 2.10 (12.9) 16.26 38 3.16 4.79 4.64 8.35 20.94 (55.0) 0.81 0.32 1.55 2.68 (11.4) 23.62 51 5.72 6.44 3.30 6.87 22.33 (40.1) 1.01 0.45 1.56 3.02 (U.9) 25.35 62 7.10 6.58 3.73 6.45 23.86 (38.6) 0.48 0.45 1.55 2.48 (9.4) 26.43 73 8.37 6.59 3.76 5.63 24.35 (35.3) 0.37 0.44 1.48 2.29 (8.6) 26.64 >) Carbonlta (poor site) 21 0.25 0.32 0.38 1.83 2.78 (54.2) 0.75 0.32 0.23 1.30 (31.9) 4.08 32 0.66 0.80 0.80 3.12 5.38 (55.0) 0.89 0.41 0.45 1.75 (24.5) 7.13 43 0.64 0.71 0.68 2.83 4.86 (57.2) 0.68 0.33 0.27 1.28 (20.9) 6.14 56 1.12 1.28 1.00 1.00 4.40 (31.7) 0.88 0.40 0.62 1.90 (30.2) 6.30 67 1.30 1.44 1.06 2.29 6.09 (41.0) 0.94 0.44 0.71 2.09 (25.6) 8.18 Numbers within brackets are percentage of total tree biomass allocated to crown components and root system, respectively. TABLE 10. Nitrogen content (kg/ha) of each tree component over an age sequence of IE. grandis plantations growing on two different "cerrado" s o i l s i n Minas Gerais, B r a z i l Age Total Fine Medium Large Total Total (months) Bark Wood Branches Foliage aboveground roots roots roots roots tree (a) Bom Despacho (good s i t e ) 15 2.27 5.19 13.23 66.68 87.37 (75.5) 3.20 0.53 14.73 18.46 (17.4) 105.83 26 10.20 33.16 27.68 143.47 214.51 (56.6) 8.06 4.37 26.51 38.94 (15.4) 253.45 38 16.00 44.54 42.67 181.61 284.82 (67.5) 10.21 5.16 32.13 47.50 (14.3) 332.32 51 36.56 97.14 33.84 123.22 290.76 (45.9) 11.10 8.06 32.54 51.70 (15.1) 342.46 62 36.66 120.81 38.33 114.15 309.95 (42.1) 11.51 8.51 32.20 52.22 (14.4) 362.17 73 37.46 147.54 39.47 98.57 323.03 (36.9) 11.36 8.47 30.87 50.70 (13.6) 373.74 (b) Carbonita (poor s i t e ) 21 2.98 6.22 7.46 26.29 42.95 (52.8) 14.00 4. 84 2.16 21.00 (32.8) 63.95 32 7.90 14.30 14.77 61.00 97.97 (59.0) 16.68 7. 38 6.60 30.66 (23.8) 128.63 43 7.78 13.50 12.43 53.41 87.12 (56.8) 13.60 7. 20 7.92 28.72 (24.8) 115.84 56 13.65 26.58 15.61 52.29 108.13 (44.0) 20.02 9. 92 15.98 45.92 (29.8) 154.05 67 15.96 32.80 14.99 46.45 110.20 (36.7) 23.72 11. 79 21.66 57.17 (34.2) 167.37 ^Numbers within brackets are percentage of total tree biomass allocated to crown components and root system, respectively. 54 30-, 25-Legend A Total Aboveground O Stem_ _ _ _ • Total Roots (a) 26 38 Age (Months) _ 15 C O — Ui — 10 to O .c a «j o (b) 32 43 Age (Months) 67 Figure 5. Phosphorus content (kg/ha) of stem, t o t a l aboveground and t o t a l roots over an age sequence of grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l : (a) Bom Despacho - good s i t e , and (b) Carbonita - poor s i t e . 55 Age (Months) Figure 6. Nitrogen content (kg/ha) of stem, total aboveground and total roots over an age sequence of grandis plantations growing on two different "cerrado" soils in Minas Gerais, Brazil: (a) Bom Despacho - good site, and (b) Carbonita - poor si t e . 56 s i t e where i t s t a b i l i z e d a f t e r 51 months. The maximum P and N accumula-t i o n i n branches and leaves occurred at age 38 months on the good s i t e . On the poor s i t e , P and N content of foliage reached i t s maximum at age 32 months, corresponding to the maximum biomass. The same species studied by Bellote et a l . (1980) showed continuous increase i n P and N content i n stems up to age 6 years, whereas a maximum f o r f o l i a g e was observed at ages 3 and 6 years, r e s p e c t i v e l y . However, the 6-year-old stand of t h e i r study seems to have been located i n a better s i t e as compared to t h e i r other stand ages because i t s t o t a l biomass was much greater than that of the 5- and 7-year-old stands (Table 2). The f o l i a r data for the present study are i n agreement with the peak at age 3 years observed by B e l l o t e et^ al^. (1980). The maximum P and N content of the aboveground components was 24.4 and 323.0 kg/ha, respectively, for the good s i t e , and 6.1 and 110.2 kg/ha, res p e c t i v e l y , f o r the poor s i t e at the oldest age studied. These values are lower than those (31.8 and 429 kg/ha at age 7 years) observed by Bellote e£ a l . (1980) f o r the same species (Table 2). Nutrient content data for the poor s i t e i n the present study are also lower than values observed for mature native eucalypts (Table 1). However, they are s i m i l a r to those observed for planted and f e r t i l i z e d E. globulus at young ages (Table 2). At the oldest age studied, both P and N content of stemwood on the good s i t e was about 4.5 times greater than on the poor s i t e . Consider-ing that the difference i n biomass was only 2.5 times, these r e s u l t s show that the amount of stemwood biomass per unit P or N i s much greater 57 on the poor s i t e . When t o t a l tree biomass i s considered, the r a t i o be-tween the good and the poor s i t e s was only 2.0, 3.3 and 2.2 for biomass and P and N content, r e s p e c t i v e l y , as a re s u l t of the greater biomass and nutrient accumulation i n the root system on the poor s i t e . Accumulation of P and N i n the belowground components also followed the pattern of biomass accumulation, except for P content i n fin e roots on the good s i t e . Biomass of fine roots had s t a b i l i z e d by age 51 months, while P content reached i t s maximum at age 51 months and decreased thereafter. The d i s t r i b u t i o n of t o t a l P and N between tree components was di f f e r e n t from that of biomass because of high nutrient concentrations i n components with low biomass. Based on the average concentration of a l l trees sampled on each s i t e , the following order of magnitude was observed: P - f o l i a g e > stembark ^ f i n e roots > branches > medium roots > large roots > stemwood; and N - f o l i a g e > fine roots > medium roots > branches > stembark > large roots > stemwood. P and N concen-t r a t i o n for each tree components of a l l trees studied are presented i n Appendix 3. On the good s i t e , the biomass of crown components (Table 5) was only 12.7-14.1% of t o t a l biomass after age 51 months, while i t repre-sented 35.3-40.1% of t o t a l P and 36.9-45.9% of t o t a l N (Tables 9 and 10) because leaves always have the greatest nutrient concentration. Due to the high concentration of P i n the bark (second highest), 23-31% of t o t a l P was located i n th i s component aft e r age 51 months on the good s i t e , while bark contained only 10% of t o t a l N. Fine roots had r e l a -t i v e l y high P and N concentrations, but due to the very low biomass of 58 f i n e roots on the good s i t e , t o t a l roots contained only 8.6-11.9% of t o t a l P and 3.0% of t o t a l N at ages greater than 51 months (these values are low compared to those presented i n Table 1). Stemwood had the lowest P and N concentration, and consequently i t represented only about 25% of t o t a l P and 28.3-39.4% of t o t a l N on the good s i t e . On the poor s i t e , a l l o c a t i o n of P and N to d i f f e r e n t components was d i f f e r e n t from that observed on the good s i t e due to differences i n biomass a l l o c a t i o n . At ages older than 56 months, roots contained 25.6-30.1% of t o t a l P and 29.8-34.2% of t o t a l N, which i s very high compared to values of less than 12% f o r both nutrients on the good s i t e . The proportion of P i n the root system on the poor s i t e was s l i g h t l y lower than values of 31.6-49.3% reported f o r mature forests of three d i f f e r e n t eucalypt species, while those for N agree with values of 31.6-37.5% for two of those eucalypt species (Table 1). Because of a very low concen-t r a t i o n of nutrients i n the stemwood and a greater proportion of nutrients i n the root system, stemwood contains only 17.3-20.3% of t o t a l P and N. The proportion of nutrients i n crown components (31.7-41.0% of t o t a l P and 36.7-44.0% of t o t a l N) i s only s l i g h t l y greater than that for the good s i t e . Cromer et a l . (1976) also observed that the propor-t i o n of P and N allocated to crown components was s i m i l a r when f e r t i l i z e d plants were compared to u n f e r t i l i z e d ones at age 4 years. Accumulation of P and N per unit biomass of stem (wood + bark) and per unit biomass of t o t a l tree decreases with age for both s i t e s . When the t o t a l tree i s considered the requirement i s higher due to high concentration of nutrient i n other components, es p e c i a l l y i n leaves. 59 The P and N requirement for stemwood growth decreased sharply between ages 38 and 51 months on the good s i t e , and between age 21 and 32 months on the poor s i t e . By t h i s age, the effect of f e r t i l i z e r i s thought to have been greatly reduced and growth i s more dependent on native f e r t i l i t y . The r a t i o f o l i a g e / f i n e roots (Table 5) decreased at about the same ages, which also r e f l e c t s d e c l i n i n g In P a v a i l a b i l i t y . Also, heartwood formation had begun at these ages, which reduced the rate of accumulation of P and N i n the stemwood and led to a small reduction i n the proportion of t o t a l tree P contained i n stemwood. 2.8 Summary and Conclusions Accumulation of biomass and nutrients was studied over ages sequences of grandis plantations located at Companhia A g r i c o l a e F l o r e s t a l Santa Barbara, growing on good (Bom Despacho) and poor (Carbonita) "cerrado" s i t e s . Biomass and nutrient content of each l i v i n g tree component (including the root system) were determined as a basis for better understanding of eucalypt strategies as related to nutrients on such poor s o i l s . This study provides information to evaluate the e f f e c t of intensive management of eucalypts on s i t e nutrient status, to be presented i n Chapter 6. From the data presented i n t h i s chapter, c e r t a i n conclusions follows. 1. Equations f o r estimating biomass of E. grandis should be s i t e s p e c i f i c i f high pr e c i s i o n i n estimation i s required. 2. Rapid early growth occurred on both s i t e s . The stem (wood + bark) production was 275 kg/ha/month at age 15 months on the good s i t e and, 268 kg/ha/month at age 21 months on the poor s i t e . This 60 r e f l e c t s the effect of f e r t i l i z e r applied at the time of planting. The e f f e c t of s i t e q u ality became evident l a t e r with stem i n c r e -ment of 725 and 421 kg/ha/yr at ages 26 and 32 months for the good and poor s i t e s , r e s p e c t i v e l y . 3. The biomass of stembark, stemwood, and large roots s t a b i l i z e d at age 51 months on the good s i t e , while on the poor s i t e they were s t i l l increasing at age 67 months. At the oldest age studied, the stemwood biomass was about 2.5 times greater on the good s i t e as compared to the poor s i t e . 4. Foliage biomass increased up to 38 months (8546 kg/ha) on the good s i t e , and to age 32 months (4353 kg/ha) on the poor s i t e . A subsequent decrease occurred on both s i t e s . The maximum f o l i a g e biomass on the good s i t e was considered high compared to r e s u l t s of others studies f o r the same species. 5. The biomass of fine roots s t a b i l i z e d at age 38 months on the good s i t e while i t was s t i l l increasing on the poor s i t e at age 67 months. The maximum fine roots biomass achieved on the poor s i t e (4417 kg/ha) was 2.6 times greater than that of the good s i t e (1692 kg/ha). 6. The d i s t r i b u t i o n of t o t a l biomass between tree components varied with s i t e . The crown accounted for about the same proportion of the t o t a l biomass on both s i t e s . The stemwood represented only 45.8% of t o t a l tree biomass at age 67 months on the poor s i t e , while i t was about 60.0% on the good s i t e at ages older than 51 months. The proportion of stembark was about the same on both s i t e s (11% at the oldest age studied). The proportion of t o t a l 61 tree biomass allocated to the root system on the poor s i t e was 29.4-36.7%, while on the good s i t e the values were 13.2-14.0% f or ages older than 51 months. 7. S i g n i f i c a n t heartwood formation did not start u n t i l about age 43 months on the poor s i t e and 51 months on the good s i t e . 8. P content of stemwood s t a b i l i z e d a f t e r age 51 months (about 6.6 kg/ha), while N content was s t i l l increasing up to age 73 months (147.5 kg/ha) on the good s i t e . Both P and N were s t i l l increasing up to age 67 months on the poor s i t e (1.4 kg/ha of P and 32.8 kg/ha of N). 9. At the oldest age studied, both P and N content of stemwood on the good s i t e was about 4.5 times greater than on the poor s i t e , while comparable value f o r biomass was 2.5. These r e s u l t s show that the amount of stemwood biomass per unit P or N i s much greater on the poor s i t e . 10. P and N accumulation i n stembark was s t i l l occurring on both s i t e s , except for N on the good s i t e where i t s t a b i l i z e d a f t e r 51 months. 11. The maximum P and N accumulation i n branches and leaves occurred at age 38 months on the good s i t e . On the poor s i t e , the maximum P and N content was obtained at age 32 months. 12. The maximum P and N content of the aboveground components was 24.4 and 323.0 kg/ha, r e s p e c t i v e l y , f o r the good s i t e and 6.1 and 110.2 kg/ha, respectively, for the poor s i t e , at the oldest age studied. 13. The maximum P and N content of the belowground components was 3.0 and 52.2kg/ha, respectively, for the good s i t e , and 2.1 and 57.2 kg/ha, r e s p e c t i v e l y , f o r the poor s i t e . 62 14. The d i s t r i b u t i o n of t o t a l P and N between tree components was d i f f e r e n t from that of biomass because of high nutrient concentra-tions i n components with low biomass. On both s i t e s , the biomass of crown components was only 11.6-14.5% of t o t a l biomass a f t e r age 51 months, while i t represented 31.7-41.0% of t o t a l P and 36.7-45.9% of t o t a l N. Stemwood contained only 25% of t o t a l P and 28.3-39.4% of t o t a l N on the good s i t e , and 17.3-20.3% of t o t a l P and N on the poor s i t e . Roots contained about 20% of t o t a l P and 15% of t o t a l N on the good s i t e and 20-34% of t o t a l P and N on the poor s i t e . The present study provided a basis for the following recommenda-tions concerning sampling for biomass and nutrients i n plantations. (a) More intensive f o l i a g e sampling i s required i n studies of this type. Foliage regression equations for fast growing trees should probably be r e s t r i c t e d to stands varying i n age by no more than 2 or 3 years due to change i n canopy c h a r a c t e r i s t i c s with age. A l t e r n a t i v e l y , separate regressions should be developed for pre-canopy closure stands and post-canopy closure stands. However, i f sampling were more intensive and other predictive crown variables (e.g. crown width, crown length) were measured, s t r a t i f i c a t i o n might not be required. (b) Pre d i c t i v e equations for roots, e s p e c i a l l y for f i n e and medium roots, were also inadequate. More intensive sampling i s required for these two classes of roots unless better pr e d i c t i v e variables can be found. 63 Site s p e c i f i c equations of biomass should be used i f more precise estimates are required, e s p e c i a l l y f or crown and root components. Taproots should not be sampled to a depth greater than 100 cm for E. grandis growing i n the "cerrado" region, because the proportion of roots below t h i s depth appears to be i n s i g n i f i c a n t . A d d i tional observations should be made for other s i t e types to determine i f t h i s statement can be generalized. Future biomass and nutrient studies of plantations with the objec-t i v e of comparing the effects of d i f f e r e n t levels of s i t e nutrient status on biomass and nutrient accumulation should preferably be done on a single stand under d i f f e r e n t levels of f e r t i l i z a t i o n , rather than on d i f f e r e n t s i t e s i n order to avoid the e f f e c t s of changes i n non-nutritional factors and differences i n h i s t o r y of plantation establishment. 64 CHAPTER 3 DYNAMICS OF NUTRIENTS (INTERNAL CYCLING, LITTERFALL AND LITTER DECOMPOSITION) IN AGE SEQUENCES OF E ^ grandis PLANTATIONS GROWING ON GOOD AND POOR •,CERRADO,, SITES 3.1 Introduction Internal c y c l i n g of nutrients, amount and rate of loss of nutrients from trees by l i t t e r f a l l , and the subsequent release of nutrients through l i t t e r decomposition are very important, e s p e c i a l l y for s i t e s with low nutrient status. A t t i w i l l (1980) stressed the importance of the contribution of the sapwood-to-heartwood conversion process to i n t e r n a l c y c l i n g of P i n eucalypts. The withdrawal of nutrients from plant parts before they are shed as l i t t e r also contributes s i g n i f i c a n t l y to i n t e r n a l c y c l i n g of nutrients ( A t t i w i l l £t a l . , 1978; O'Connell et^ a l . , 1978; O'Connell and Menage", 1982). Because of the poor nutrient status of the "cerrado" s o i l s , the return of nutrients through l i t t e r i s very important. As reported i n Chapter 2, E_^ grandis planted i n the "cerrado" region forms a very thick mat of f i n e roots i n the upper layer of s o i l . Nutrients released from the forest f l o o r w i l l be promptly absorbed by these roots. This chapter describes the dynamics of P and N i n E_^_ grandis plantations on the "cerrado" region by analysing l i t t e r f a l l , f o r e s t f l o o r accumulation, decomposition of branches and fol i a g e and i n t e r n a l 65 c y c l i n g due to heartwood formation and the shedding of plant foliage and branches. 3.2 L i t e r a t u r e Review 3.2.1 Biomass, P and N i n L i t t e r f a l l Several studies have been conducted on native eucalypts i n A u s t r a l i a . These studies include both the overstory and understory. Eucalypt l i t t e r f a l l studies developed before 1977 were reviewed by Bevege (1978) and are presented i n Table 11. Results of recent studies on native and planted eucalypts are presented i n Table 12. These r e s u l t s show that l i t t e r f a l l biomass i s extremely v a r i a b l e . This v a r i a -b i l i t y might be due to s i t e conditions, age of the stand and/or species c h a r a c t e r i s t i c s . As can be observed from Table 12, the t o t a l biomass of l i t t e r f a l l of mature E. obliqua (dry s c l e r o p h y l l f o r e s t ) i s small com-pared to E. d i v e r s i c o l o r (wet s c l e r o p h y l l f o r e s t ) , which has a substan-t i a l v a r i a t i o n i n l i t t e r f a l l biomass as a function of age, e s p e c i a l l y at younger ages. As mentioned by O'Connell and Menage (1982), t h i s v a r i a -t i o n i s mainly due to an increase i n the biomass of branches, bark and f r u i t s , as leaf biomass only varies from 3.1 to 3.8 t/ha/yr a f t e r age 6 years. Despite the increase i n proportion of other components, the leaves s t i l l contribute more than 50% of aboveground l i t t e r biomass. At older ages t h i s v a r i a t i o n i s smaller as observed for E_^ p i l u l a r i s (Table 11). Seasonal v a r i a t i o n i n l i t t e r f a l l has also been studied by several authors. In general, most l i t t e r f a l l occurs during the summer (Cozzo and Riveros, 1971; Ashton, 1975; A t t i w i l l et a l . , 1978; F e l l e r , 1979; TABLE 11. Forest floor and l i t t e r f a l l biomass and P and N content of native eucalypt forests (studies reported up to 1977) Species Forest floor L i t t e r f a l l Source Biomass (t/ha) P N (kg/ha) Biomass (t/ha/yr) P N (kg/ha/yr) E. obliqua 18.0 5.0 - 3.6 1.0 - A t t i w i l l (1972) E. regnans — — — 6.6 7.8 1.3 1.9 43.5 57.6 Ashton (1975) E. fastigata 27.0 18.1 19.3 10.7 216 109 — — — Meakins (1966) •• E. robertsoni 9.2 5.2 55 - - - M II E. delegatensis 17.8 14.3 125 - - - ! • E. marginata 8.8 - - 2.4 0.5 10.3 Hatch (1955) E. saligna 12.A 4.0 92 10.0 7.5 77.0 Richards and Charley (1977) E. olminalis 4.5 2.0 28 3.9 2.5 33.5 I I II E. signata 27.0 6.8 195 6.4 2.1 39.6 Rogers and Westman (1977) E. pilularis - - - 6.5 1.4 44.0 Webb et a l . (1969) E. pilularis* - - - 7.5 1.1 21.2 Florence (1961) E. pilularis^ - - - 7.0 1.0 23.6 • I M E. pilularis-^ - - - 7.2 1.0 21.2 E. pilularis^ 15.0 1.9 87 - - - Richards and Charley (pers. comm.) E. pilularis^ 15.3 1.3 105 •• I I •» •• Pole stand; mature regrowth; overmature virgin stand; 25 year old pole stand; 39 year old pole stand. Source: Adapted from Bevege (1978). TABLE 12. L i t t e r f a l l biomass (t/ha/yr) and P and N content (kg/ha/yr) of native and planted eucalypts (studies reported since 1978) Species (age-yr) Biomass*" Source E. obliqua/E. baxteri 2 2.33 (81.5) 0.33 9.8 Lee and Correll (1978) E. obliqua (mature)2 3.56 (46.1) 1.03 - A t t i w i l l et a l . (1978) E. obliqua (37) 6.56 1.40 31.1 Feller (1979) E. regnans (37) 9.81 2.20 34.8 E. diversicolor (2)3 1.13 (88.5) 0.48 (79.2) 8.6 (83.7) O'Connell and Menage (1982) E. diversicolor (6) 3.70 (83.8) 0.99 (37.4) 32.5 (28.0) if #i •• E. diversicolor (9) 4.46 (69.5) 0.85 (52.9) 33.8 (36.4) •I at ft E. diversicolor (40) 7.15 (53.1) 1.40 (18.6) 49.3 (33.1) •• M E. diversicolor (mature) 9.45 (37.0) 1.89 (21.2) 58.1 (18.9) •I •• It E. obliqua (70-80) 5.18 (51.5) 1.47 (54.4) 30.2 (66.4) Baker (1983) E. obliqua (80-90) 3.88 (66.8) 0.94 (71.2) 24.1 (80.2) •• •« E. sieberi (60) 5.37 (37.7) 1.08 (52.7) 21.4 (68.2) it •• E. regnans (19)^ 6.86 (49.7) 2.01 (43.3) 46.4 (44.9) M •• E. saligna (6)^ 5.506 3.00 51.0 Poggiani (1976) cited by Carpanezzi (1980) E. saligna/E. grandis (5)^ 7.57 (61.9) 2.89 (79.2) 39.8 (81.9) Carpanezzi (1980) E. globulus (12) 7 1.946 0.80 31.0 Venkataramanan et a l . (1983) E. grandis (27) 8 7.49 (56.4) 3.15 (45.7) 51.6 (70.3) Turner and Lambert (1983) Numbers within brackets are percentage of biomass and nutrient in leaves; Dry sclerophyll forest; Wet sclerophyll forest; Plantation in Victoria, Australia; Plantation in Sao Paulo, Brazil ('Cerrado' region); Leaves only; Plantation in N i l g i r i s , India; Plantation in N.S.W., Australia. 68 Carpanezzi, 1980; O'Connell and Menage, 1982; Baker, 1983). Cozzo and Riveros (1971) found that 63% of l i t t e r f a l l i n an 8-year-old E.  camaldulensis plantation i n B r a z i l occurred i n the summer season, a period that corresponds to the highest monthly p r e c i p i t a t i o n . A t t i w i l l et a l . (1978) mentioned that part of the seasonal v a r i a t i o n observed for E. obliqua was explained by temperature; a s i m i l a r c o r r e l a t i o n was obtained by Lee and C o r r e l l (1978) for l e a f y l i t t e r f a l l of E. obliqua/E.  baxteri f o r e s t . In contrast, Venkataramanan et a l . (1983) observed maximum leaf l i t t e r f a l l during the winter (lower temperature and p r e c i p i t a t i o n and higher number of sunshine hours). Twig and branch l i t t e r f a l l i s highly variable because i t i s affected by storms, wind or other mechanical e f f e c t s (Ashton, 1975; A t t i w i l l et a l . , 1978; O'Connell and Menage, 1982). The amount of P and N returned through l i t t e r f a l l i s presented i n Table 11 and i n Table 12. L i t t e r f a l l from E. obliqua/E. baxteri forest (dry s c l e r o p h y l l f o rest) contained the lowest amount of P (0.33 kg/ha/yr) and N (9.8 kg/ha/yr); i n addition to the low amount of l i t t e r f a l l , nutrient concentrations were amongst the lowest reported (0.014 and 0.42% for P and N r e s p e c t i v e l y ) . The maximum l i t t e r f a l l nutrient content reported was 7.5 and 77 kg/ha/yr for P and N, r e s p e c t i v e l y , for E. saligna due mainly to the large l i t t e r f a l l biomass (10 t/ha/yr) produced by the species (Table 11). Lower concentrations of P and N were observed for the same species planted i n the "cerrado" region i n B r a z i l (Table 12). 69 3.2.2 Biomass, P and N In Forest Floor A review of e a r l i e r studies of forest f l o o r accumulation i n eucalypt stands i s presented i n Table 11, and more recent studies are summarized i n Table 13. The reported forest f l o o r biomass ranges from 4.5 t/ha for E^ _ v i m i n a l i s to 47.5 t/ha for E^ _ regnans. As was the case for l i t t e r f a l l , forest f l o o r values vary according to s i t e type, stand age and species c h a r a c t e r i s t i c s . For example, i n Table 13 one can see that dry s c l e r o p h y l l forest produces much less forest f l o o r than wet s c l e r o p h y l l f o r e s t . V a r i a t i o n i n forest f l o o r biomass can be observed between d i f f e r e n t s o i l types for E. marginata/E. c a l l o p h y l l a f o r e s t , and the amount of forest f l o o r i s affected by the age of the stand. O'Connell and Menage (1982) pointed out that increasing forest f l o o r biomass with age i s mainly due to an increasing proportion of branches, f r u i t s and bark that are more slowly decomposed than f o l i a g e . Comparisons between ecosystems of the amount of forest f l o o r should be done c a r e f u l l y . Forest f l o o r biomass depends on disturbances that have occurred on the s i t e such as f i r e , logging, insect d e f o l i a t i o n (Rogers and Westman, 1977; Lee and C o r r e l l , 1978) and on stand c u l t u r a l practices (e.g. s o i l c u l t i v a t i o n between tree rows to control weeds) that would increase the proportion of branches and logs i n the forest f l o o r (Rogers and Westman, 1977). Also, l a r g e r branches are sometimes excluded during the determination of forest f l o o r biomass. For example, Pressland (1982) sampled only branches < 2.5 cm diameter and obtained forest f l o o r biomass values of only 2.5-3.7 t/ha i n d i f f e r e n t s i t e s of a E. laevopinea f o r e s t . TABLE 13. Forest floor biomass (t/ha) and P and N content (kg/ha) of native*- and planted eucalypt species Species (age-yr) Biomass P N Source E_^  obliqua/E. dives 2 13.1 3.0 132 Feller (1980) E. obliqua (40) 1 26.4 10.0 114 Feller (1978) E. regnans (40)^ 44.0 7.0 75 ti •• E. regnansJ 47.5 11.0 394 Feller (1980) E. marginata/E. callophylla*»2 Yellow sand 9.2 1.4 35 O'Connell et_ a l . (1978) Grey sandy gravel 10.7 2.0 47 it it it Grey sandy gravel 15.0 2.4 59 tt tt tt Reddish gravel 16.9 4.2 83 tt it I I Reddish gravel 18.4 3.3 85 ti ti it Yellow gravel 13.0 1.6 42 tt ft it Yellow gravel 13.2 2.0 . 51 tt i i tt E. marginata/E. callophylla*»2 11.1 2.0 51 Hingston et^ a l . (1981) E. diversicolor/E. callophylla 19.5 5.0 105 Hingston et_ a l . (1979) E. diversicolor (36)^ 27.3 7.0 224 tt it •• E. radiata/E. dalrympleana 19.8 10.6 124 Turner (1980) E. grandis (27)4 17.4 3.3 47 Turner and Lambert (1983) E. citriodora (24) 5 20.0 8.9 212 Haag et a l . (1978) Includes understory and other tree species; 2Dry sclerophyll forest; "Hfet sclerophyll forest; Plantation in N.S.W. Australia; ^Plantation in 'Cerrado* area. Sao Paulo. Brazil. 71 Forest f l o o r nutrient content (Tables 11 and 13) i s extremely variable due to v a r i a t i o n i n species c h a r a c t e r i s t i c s , s i t e and/or stand age. The minimum P content reported was found for E. p i l u l a r i s (1.3 kg/ha) due to a very low P concentration (0.009%), while the maximum was for E. f a s t i g a t a (19.3 kg/ha), which had one of the highest reported f o r e s t f l o o r P concentrations (0.071%). The minimum N content (35 kg/ha) was observed f o r Ej_ marginata/E. c a l l o p h y l l a forest on a yellow sand s i t e while the minimum N concentration (0.17%) was found for a stand of E. regnans (Table 5). On the other hand, a very high forest f l o o r N content (394 kg/ha) was observed f o r another stand of Ej_ regnans due to a high N concentration (0.83%). Studies of forest f l o o r do not usually report the composition of the forest f l o o r by l i t t e r types. V a r i a t i o n i n the proportion of d i f f e r e n t categories of l i t t e r may explain this great v a r i a b i l i t y i n forest f l o o r nutrient concentration. 3.2.3 L i t t e r Decomposition Studies of accumulation of forest f l o o r i n conjunction with i n f o r -mation on l i t t e r f a l l are very h e l p f u l i n understanding the release of nutrients to the s o i l . A decay rate ( i . e . a turnover rate) obtained by using the r e l a t i o n s h i p between these two types of data i s presented for several species by Bevege (1978). E. saligna and E. v i m i n a l i s had the highest decomposition rate (k values of 0.81 and 0.85, r e s p e c t i v e l y ) with a h a l f l i f e of 0.8 years. The lowest rate was observed for E.  p i p e r i t a (0.10) with a half l i f e of 6.8 years. However, Birk and Simpson (1980) emphasize that t h i s method of estimating decomposition was developed for steady state conditions, and i n most cases eucalypt 72 forest i n A u s t r a l i a does not reach steady state due to w i l d f i r e s or prescribed burning. Another method of studying decomposition involves placing bags containing a known amount of l i t t e r on the forest f l o o r and studying the l i t t e r weight loss over time. Table 14 summarizes published studies of thi s type for eucalypt. The low rate of weight loss obtained by A t t i w i l l (1968) as compared to McColl (1966) might be due to d i f f e r e n t proportions of branch to leaves. Also, they studied d i f f e r e n t species i n a d i f f e r e n t climate. Ashton (1975) reported that decomposition of leaves of E. regnans i s four times greater than for branches over the f i r s t three months of decomposition. Studies developed by Woods and Raison (1983) and O'Connell and Menage (1983) showed that loss of weight of fresh leaf l i t t e r over the i n i t i a l 6 to 7.5 months period i s 24-30%; af t e r 12 to 14 months the loss i s 33-39%. Similar values for green leaves were 36-59% and 51-88%. These re s u l t s showed that green leaves decompose faster than fresh leaf l i t t e r . According to Woods and Raison (1983), this difference i n decay rates was correlated with i n i t i a l N and P concentrations (higher for green leaves). The effect of season on l i t t e r decomposition was studied by Ashton (1975). He observed that decomposition of leaves i s s l i g h t l y faster i n autumn and winter (rainy season). Woods and Raison (1983) observed that the i n i t i a l fast decomposition rate during the spring was reduced during the summer and increased again i n the f a l l and winter. No seasonal effect was observed i n the second year of decomposition. The authors considered that changes i n leaf c h a r a c t e r i s t i c s and moisture and temperature were responsible for these v a r i a t i o n s . TABLE 14. L i t t e r decomposition (% loss of o r i g i n a l weight) of eucalypts Species Period of exposure (months) Component Loss of weight (%) Source E_. maculata E. obliqua 15. regnans* IS. globulus (planted) E^ . delegatansis E. pauciflora IS. calophylla^ IS. marginata^ 18 24 5; 12 7.5; 14; 23.5 7.5; 14; 23.5 6; 12; 18 6; 12; 18 6; 12; 18 Leaves, capsules, twigs Leaves, branchwood Green branchwood Green leaves Leaves Bark Twig Fresh leaf l i t t e r Fresh leaf l i t t e r Green leaves Fresh leaf l i t t e r Green leaves Fresh leaf l i t t e r Fresh leaf l i t t e r Green leaves 30-39 20 19 54 12 ' 4 3 24; 82 30; 34; 45 59 26; 33; 44 43; 58; 88 24; 35; 42 29; 39; 45 36; 51, 69 McColl (1966) A t t i w i l l (1968) •• II Ashton (1975) II II II i ) Upadhyay (1982) Woods and Raison (1983) I I II II Woods and Raison (1983) O'Connell and Menage (1983) O'Connell and Menage (1983) Average of a 23 yr old and a mature stand. Average of three s i t e types. Average of three burning intensity. 74 The effect of age on decomposition rate was studied by Ashton (1975). The only difference observed was for leaves. The loss of weight of leaves over 3 months i n a mature stand was 13.9% and i n the 27 year old stand i t was 10.1%; t h i s difference was attributed to d i f f e r -ences i n microclimatic conditions. O'Connell and Menage (1983) studied the e f f e c t s of d i f f e r e n t s o i l types and burning i n t e n s i t i e s on decomposition. Their r e s u l t s showed no major difference i n decomposition rate of fresh leaf l i t t e r for Banksia  grandis and Ej_ c a l l o p h y l l a over a period of 18 months, while decomposi-t i o n was s i g n i f i c a n t l y higher for E_^_ marginata on an i n t e n s i v e l y burned s i t e as compared to a l i g h t l y burned or unburned s i t e . A s i g n i f i c a n t difference was found between three s o i l types i n the decomposition rates of green leaves of E. marginata; the r i c h e s t s i t e showing the f a s t e s t decomposition. Most of these studies reported loss of l i t t e r biomass. Loss of nutrients from decomposing l i t t e r over a period of time was studied by A t t i w i l l (1978), Upadhyay (1982) and O'Connell and Menage" (1983). A t t i w i l l (1968) mentioned that there was no s i g n i f i c a n t change i n P concentration i n decomposing l i t t e r and green branchwood of Ej_ obliqua over a period of 24 months, the loss of P being proportional to the loss of weight. In the case of green leaves there was a substantial increase i n concentration over the f i r s t 12 months followed by a decline to i n i t i a l l e v e l s . Upadhyay (1982) observed seasonal variations i n nutrient concentration of JU_ globulus leaf l i t t e r i n decomposition. There was an increase i n P and N concentrations for 7 months, followed by a decrease u n t i l the end of the 12 month period. O'Connell and 7 5 Menage (1983) observed a continuous increase i n P and N concentration i n fresh leaf l i t t e r f o r 18 months, which resulted i n a net gain i n the quantity of P i n the l i t t e r , and a small decrease i n the quantity of N. When green leaves were analysed, the release of P and N was higher. The percentage loss of P i n the l i t t e r followed that of biomass, while the loss of N was less than that of biomass. Because studies of decomposition of eucalypts l i t t e r are scarce, and because extensive plantations have been established on poor s o i l s such as those of the "cerrado" region i n B r a z i l , more studies are required to understand the release of nutrient from decomposing l i t t e r to the s o i l . 3.2.4 Internal Cycling Internal c y c l i n g i s considered very important, e s p e c i a l l y f or s i t e s with very low nutrient content. A high percentage of some nutrients such as P and N may be withdrawn from senescing foliage p r i o r to l i t t e r f a l l and from sapwood as i t turns into heartwood ( A t t i w i l l , 1980; Rogers and Westman, 1977; O'Connell et a l . , 1978). In eucalypts, i n t e r n a l c y c l i n g i s p a r t i c u l a r l y important for phosphorus, e s p e c i a l l y considering that they have been planted i n P d e f i c i e n t s o i l s such as those from the "cerrado" region. The process of sapwood-to-heartwood conversion i s the major con-t r i b u t o r to i n t e r n a l c y c l i n g of nutrients i n eucalypts. A t t i w i l l (1980) observed a withdrawal of 70% of P during the sapwood-to-heartwood conversion based on the amount of P per ha, and of 87.50% based on difference i n concentration between sapwood and heartwood. Lambert 76 (unpublished) studied i n t e r n a l c y c l i n g i n 41 d i f f e r e n t eucalypt species i n mature stands. She found 34 species with sapwood-heartwood with-drawal higher than 80% (based on difference i n concentration between sapwood and heartwood), the highest value being 97% for E. s a l i g n a . Values of 53 to 96% (average of 81%) were observed for d i f f e r e n t provenances of mature E_. grandis. According to A t t i w i l l (1980), 46% of the P required for new growth of E. obliqua i s supplied by biochemical ( i n t e r n a l ) c y c l i n g , whereas only 36% i s supplied by biogeochemical eyeling. Crane and Raison (1980) emphasize the e f f e c t of shortening rota-t i o n on increasing nutrient removal due to greater sapwood removal with shorter r o t a t i o n . They observed that E_^ delegatensis (7 years of sap-wood) requires an uptake of 0.73 kg P/ha/yr to produce stemwood at age 18 years, while i t requires only 0.44 kg P/ha/yr at age 57 years, the difference being accounted for by i n t e r n a l c y c l i n g . Internal c y c l i n g of N f o r eucalypts i s much lower than that observed for P. Withdrawal of N during sapwood-to-heartwood conversion ranges from 28.6 to 58.1% for 12 d i f f e r e n t eucalypt species (Turner, 1980; Turner and Lambert, 1983; Lambert, unpublished). The actual amount of nutrients i n l i t t e r f a l l depends on the a b i l i t y of a species to withdraw nutrients from plant parts before they are shed as l i t t e r . Usually there i s a withdrawal of more than 50% of P and N when comparing mature with senescent leaves (Ashton, 1975, 1976; A t t i w i l l et a l . , 1978; O'Connell et a l . , 1978; O'Connell and Menage', 1982). A study of P withdrawal i n E. d i v e r s i c o l o r showed an increase i n the amount withdrawn from 47.9 % at age 2 to 65.4% f o r mature stands, 77 based on concentration differences between mature and senescent leaves; the v a r i a t i o n of N withdrawal with age was i n s i g n i f i c a n t , being 45.5% at age 2 years and 51% i n mature forest (O'Connell and Menage, 1982). This suggests that i n mature stands, more P f o r new growth i s supplied by i n t e r n a l c y c l i n g i n the leaves than by uptake, and consequently l i t t e r -f a l l nutrient concentration i s decreased. O'Connell and Menage (1982) pointed out that for d i v e r s i c o l o r the increase i n l i t t e r f a l l a f t e r 6 years i s mainly due to the shedding of branches, bark and f r u i t s . A high degree of r e c y c l i n g of P before branches and bark are shed has also been observed (Ashton, 1975; A t t i w i l l e_t a l . , 1978; Madgwick et a l . , 1981). This contributes to the decrease i n the percentage of P i n t o t a l l i t t e r f a l l of older stands. 3.3 Description of the Study Area The same area described i n Chapter 2 was used. 3.4 F i e l d Methods 3.4.1 L i t t e r f a l l Sampling Aboveground l i t t e r f a l l was studied i n four of the stands (aged between 19 and 62 months old) at each of the two s i t e types. Three l i t t e r traps of 1 m2 per plot (9 per age) were placed as follows: the f i r s t one between trees along the tree l i n e , the second one between the tree l i n e s , and the t h i r d one at the base of one tree. This arrangement was employed to account for any v a r i a t i o n that might have been due to crown p o s i t i o n . These traps had 10 cm high wooden sides, with nylon f i b r e screen mesh on the bottom, supported 50 cm above the s o i l on 78 wooden legs. L i t t e r f a l l c o l l e c t i o n started i n May, 1981 f o r Bom Despacho and i n June, 1981 for Carbonita. The l i t t e r from each trap was co l l e c t e d every three months for one year; fresh weight of leaves and branches was determined. When f r u i t or bark were present they were com-bined with branches because i t occurred only occasionally i n very small proportions. A f t e r fresh weighing, a l l samples of a given component from the twelve traps of the same age were bulked for measurement of dry weight and f o r chemical a n a l y s i s . 3.4.2 Forest Floor Sampling Four samples of forest f l o o r per plot (3 plots and therefore 12 samples per stand age) were taken by removing a l l organic material from 50x50 cm quadrats at the time of biomass sampling. Two samples were taken along tree lines and two between the lines to account for v a r i a -t i o n due to c u l t i v a t i o n e f f e c t s and crown p o s i t i o n . The samples of each plot were bulked and sorted into leaves and branches for fresh and dry weight. 3.4.3 L i t t e r Decompositon L i t t e r c o l l e c t e d i n the f i r s t s ix months, mostly from the dry season, from a l l four stand ages at each of the two s i t e types was used to determine decomposition rates. Bags containing 100 g of l i t t e r were placed out i n November and December, 1981, for Bom Despacho and Carbonita, r e s p e c t i v e l y , i n the wet season (Figure 15). Another set of l i t t e r bags using l i t t e r c o l l e c t e d i n the second 6 months of the study (wet season) was placed out on the plots i n May and June, 1982, i n the dry season. 79 TABLE 15. Monthly mean temperature and p r e c i p i t a t i o n during the study period (1981-1983) i n Bom Despacho and Carbonita, Minas Gerais, B r a z i l Bom Despacho Carbonita Month Temperature P r e c i p i t a t i o n Month Temperature P r e c i p i t a t i o n (°C) (mm) (°C) (mm) May 4 19.7 34.0 June 17.2 4 3.9 June 19.4 15.2 July 16.9 11.6 July 18.4 11.3 August 18.2 0.3 19.2 1 60.5 2 17.41 15.8 2 Augus t 20.2 11.2 September 20.2 3.8 September 21.8 30.6 October 21.4 150.6 October 22.1 224.7 November 22.4 199.2 21.4 266.5 21.3 353.6 3 November 23.2 239.4 December 22.9 3 178.8 December 22.6 322.4 January 21.9 308.9 February 22.5 393.6 February 22.5 83.6 22.8 955.4 22.4 571.3 February 23.7 99.6 March 22.4 329.1 March 22.9 262.9 A p r i l 20.3 70.1 A p r i l 21.3 86.5 May 18.9 11.2 22.6 449.0 20.5 410.4 *Mean temperature f or each three month period. 2 Cumulative p r e c i p i t a t i o n for each three month period. F i r s t set of l i t t e r ' bags placed i n the f i e l d (wet season). ^Second set of l i t t e r bags i placed i n the f i e l d (dry season). 80 Leaves and branches were weighed separately and placed together i n l i t t e r bags to keep conditions f or decomposition as natural as possible. The proportion of branches and leaves used was based on t h e i r proportion i n the i n i t i a l l i t t e r c o l l e c t i o n s . The samples were placed i n 30 cm square, 5 cm high nylon mesh bags. A metal frame treated with a n t i -corrosion paint was used to maintain the bags i n a f l a t p o s i t i o n . One edge of the bag was l e f t open to permit small animals to enter and leave the bags. The bags were placed f l u s h with the s o i l surface a f t e r the e x i s t i n g forest f l o o r was removed, but without disturbing the s o i l beneath. S u f f i c i e n t bags were l a i d out to permit one bag to be c o l l e c t -ed from each of the plots every s i x months for two years for dry weight and chemical analysis of leaves and branches separately. This sampling schedule was not e n t i r e l y successful as noted l a t e r . 3 .5 Plant Analysis Samples of a l l organic materials and f o r e s t f l o o r were ground to pass through a 2 mm mesh sieve immediately a f t e r being c o l l e c t e d , while l i t t e r and decomposition samples were stored at a i r temperature i n paper bags a f t e r t h e i r dry weight was measured. The samples were analysed chemically as described i n Chapter 2. 3.6 Data Analysis Analysis of variance was performed on data for decomposition rates and for biomass of l i t t e r f a l l . B a r t l e t ' s test was used to test the homogeneity of variance, and Duncan's test was used to evaluate d i f f e r -ences between treatment means. 81 3.7 Results and Discussion 3.7.1 Biomass, P and N i n L i t t e r f a l l The raw data of this experiment had two major defects for s t a t i s -t i c a l a n a l y s i s . These were non-normality and lack of homogeneity. For these reasons a logarithmic transformation was used. This l a r g e l y removed the non-normality and improved (although i t did not e n t i r e l y removed) the lack of homogeneity of variance. Tables 16 and 17 present the biomass and P and N content of seasonal l i t t e r f a l l for each stand age studied on the two d i f f e r e n t s i t e s . P and N concentrations are presented i n Appendix 4. Branch and leaf l i t t e r f a l l showed a s i g n i f i c a n t i n t e r a c t i o n between age of the stand and season on both s i t e s , probably because the seasonal i n t e r v a l (3 months) i s long enough to r e f l e c t the changes i n growth of such fast-growing species. The r e s u l t s of the test of the means are presented i n Appendix 5. There was no d i s t i n c t seasonal v a r i a t i o n i n branch l i t t e r f a l l on either s i t e , but some trends can be observed. For example, the least branch l i t t e r f a l l usually occurred between May and August except for the 6-month-old stand on the good s i t e and the 42-month-old on the poor s i t e (Tables 16 and 17). On the poor s i t e , there was a very high standard deviation, probably because the branches were thicker and t h e i r shedding was more random as compared to the good s i t e . The lack of seasonality observed for branch l i t t e r f a l l i s i n agreement with other studies already developed f o r native eucalypts (Ashton, 1975; A t t i w i l l et a l . , 82 TABLE 16. Biomass (kg/ha) and nutrient content (g/ha) of seasonal l i t t e r f a l l over an age sequence of E_. grandis plantations growing i n the "cerrado" region, Bom Despacho (good s i t e ) , Minas Gerais, B r a z i l 1 „ 2 Branches Foliage Age Season (months) Biomass P N Biomass 1 51 (51) 3 4 137 879 (318) 295 5387 2 129 (313) 8 286 459 (100) 212 3996 3 130 (174) 7 276 1165 (448) 317 7867 4 444 (458) 63 1303 1614 (392) 500 11186 Total 754 82 2002 4117 1324 28436 1 385 (325) 18 650 1055 (146) 287 7029 2 464 (300) 19 660 188 (65) 73 1591 3 392 (471) 29 662 1273 (207) 428 10182 4 968 (611) 65 1806 1599 (365) 495 7954 Total 2209 131 3778 4115 1283 26756 1 397 (230) 47 740 799 (68) 199 7812 2 540 (306) 34 816 470 (39) 164 3505 3 749 (554) 28 1131 1557 (160) 483 8164 4 839 (527) 119 1938 1092 (97) 284 5529 Total 2525 228 4625 3918 1130 25010 1 530 (372) 40 1177 611 (152) 221 4340 2 424 (279) 60 1317 556 (51) 257 5038 3 826- (497) 62 2055 1552 (92) 501 10342 4 386 (240) 44 755 756 (62) 244 5713 Total 2166 206 5304 3475 1223 25433 Age of the plantation at the commencement of l i t t e r sampling. L i t t e r traps set i n the f i e l d on May 5, 1981 and collected every 3 months: (1) May-July; (2) August-October; (3) November-January; (4) February-April. 'Numbers within brackets represent standard deviations. 83 TABLE 17. Biomass (kg/ha) and nutrient content (g/ha) of seasonal l i t t e r f a l l over an age sequence of E_. grandis plantations growing in the "cerrado" region, Carbonita (poor s i t e ) , Minas Gerais, B r a z i l 1 „ 2 Branches Foliage Age Season _ (months) Biomass P N Biomass P N 1 0 0 0 458 (269) 72 2036 2 6 ( I D 3 1 15 512 (288) 133 3683 3 63 (145) 4 167 2765 (1630) 965 22358 4 35 (76) 2 77 680 (282) 194 4231 Total 104 7 259 4416 1364 32308 1 31 (70) 3 61 408 (127) 106 2393 2 159 (296) 9 282 814 (243) 242 5932 3 344 (385) 34 611 2960 (700) 771 17883 4 234 (214) 15 437 353 (85) 109 2793 Total 768 61 1391 4535 1228 29001 1 105 (129) 11 234 508 (188) 126 2484 2 133 (82) 13 308 573 (206) 208 4122 3 202 (198) 17 270 1819 (545) 409 9860 4 50 (61) 5 66 255 (84) 76 1493 Total 491 46 878 3155 819 17959 1 176 (93) 27 468 323 (62) 135 2957 2 444 (300) 30 789 718 (123) 251 6189 3 297 (280) 38 845 1291 (158) 522 12048 4 495 (367) 25 1144 409 (36) 171 3817 Total 1412 120 3246 2742 1079 25011 Age of the plantation at the commencement of l i t t e r sampling. L i t t e r traps set i n the f i e l d on June 5, 1981 and collected every 3 months: (1) June-August; (2) September-November; (3) December-February; (4) March-May. Numbers within brackets represent standard deviations. / 84 1978; O'Connell and Menage, 1982), e s p e c i a l l y due to the effect of mechanical fac t o r s , such as wind, forcing the shedding of branches. Seasonal v a r i a t i o n of leaf l i t t e r f a l l was more evident than for branches on both s i t e s , and the standard deviation was lower than for branches (Tables 16 and 17). This lower standard deviation of leaf l i t t e r f a l l i n comparison to branches might be explained by the higher c o r r e l a t i o n of leaf l i t t e r f a l l with environmental factors as observed by Cozzo and Riveros (1971), A t t i w i l l et_ a l . (1978) and Lee and C o r r e l l (1978). On the good s i t e , the lowest leaf l i t t e r f a l l was observed during the period of May-October, while the highest values occurred i n November-April. A much stronger seasonal v a r i a t i o n was observed for the poor s i t e (Appendix 5) with more than 50% of leaf l i t t e r f a l l occurring i n December-February for stands younger than 42 months, and 31% for the 55 months old stand. The lowest leaf l i t t e r f a l l (Tables 16 and 17) coincides with the lowest temperature and p r e c i p i t a t i o n (Table 15) on both s i t e s . However, when the highest leaf l i t t e r f a l l i s analysed, i t seems that there i s a better c o r r e l a t i o n with temperature than p r e c i p i t a t i o n . On the good s i t e , temperatures were s i m i l a r l y high for both periods November-January and February-April, and leaf l i t t e r f a l l was also s i m i l a r l y high. P r e c i -p i t a t i o n was very high (955.4 mm) i n November-January, but much lower i n February-April (449.0) when leaf l i t t e r f a l l was s t i l l high. On the poor s i t e , the difference i n p r e c i p i t a t i o n between December-February and March-May was much smaller than on the good s i t e . On the other hand, the difference i n temperature between December-February and March-May, was higher than that observed on the good s i t e which might explain the 85 predominance of leaf l i t t e r f a l l during December-February. A t t i w i l l et  a l . (1978) and Lee and C o r r e l l (1978) also reported that seasonal leaf l i t t e r f a l l was p a r t i a l l y c o r r e l a t e d with temperature. Annual l i t t e r f a l l biomass, P and N content data are presented i n Tables 16 and 17. On both s i t e s , there was a s i g n i f i c a n t increase i n branch l i t t e r f a l l with age. On the good s i t e branch l i t t e r f a l l values were s i m i l a r from ages 36 to 60 months, and s i g n i f i c a n t l y higher than that at age 25 months. On the poor s i t e , branch l i t t e r f a l l increased s i g n i f i c a n t l y at age 56 months. The reason for the delay i n branch l i t t e r f a l l on the poor s i t e might be due to the fact that branch diameter was much greater than on the good s i t e . Leaf l i t t e r f a l l was s i g n i f i c a n t l y smaller at age 60 months on the good s i t e , and at ages 43 and 56 months on the poor s i t e as compared to younger ages. Live leaf biomass had also decreased at these ages (Table 5). The maximum annual t o t a l l i t t e r f a l l (branches + f o l i a g e ) was observed at ages 36 and 49 months on the good s i t e (6325 and 6443 kg/ha/yr, r e s p e c t i v e l y ) , and at age 31 months on the poor s i t e (5304 kg/ha/yr). These values are lower than those observed f o r E. saligna and E. saligna/E. grandis plantations elsewhere i n B r a z i l , but greater than the annual l i t t e r f a l l observed for E^ d i v e r s i c o l o r f o r e s t at age 6 years (Table 12). Analysis of variance of the l i t t e r f a l l nutrient content data could not be done as a l l the samples of a given season for each age were bulked for chemical a n a l y s i s . However, i t can e a s i l y be seen that P and N content of both branches and f o l i a g e followed the same pattern as that of biomass (Tables 16 and 17). This c o r r e l a t i o n between seasonal 86 l i t t e r f a l l and seasonal nutrient content was also observed by F e l l e r (1979) and by A t t i w i l l (1980). The t o t a l annual P and N content of l i t t e r f a l l (branches + f o l i a g e ) averaged about 1.4 and 30.3 kg/ha/yr for a l l stand ages studied on the good s i t e . On the poor s i t e , a decline of both t o t a l annual P and N content with age was observed, the maximum being, r e s p e c t i v e l y , 1.4 and 32.6 kg/ha/yr at 21 months. The stand aged-42 months on the poor s i t e had very low quantities of P (0.9 kg/ha/yr) and N (18.8 kg/ha/yr) i n l i t t e r f a l l content because of very low l i t t e r f a l l nutrient concentrations during the season of heavy l i t t e r f a l l (Appendix 4). This stand also had very low t o t a l l i v e f o l i a r biomass (Table 5). The t o t a l annual P and N quantities i n l i t t e r f a l l per stand age were lower than those observed f o r other plantations i n B r a z i l . However, they are within the range observed for eucalypts i n A u s t r a l i a (Tables 11 and 12). Leaf l i t t e r f a l l represents more than 82 and 88% of the aboveground l i t t e r transfers of P and N, on the good and poor s i t e s , r e s p e c t i v e l y . The great v a r i a b i l i t y observed for branch l i t t e r f a l l thus has a propor-t i o n a t e l y smaller effect i n P and N transfers through l i t t e r than would be suggested by i t s biomass. 3.7.2 Biomass, P and N i n Forest f l o o r The biomass and nutrient content of forest f l o o r materials are presented i n Table 18, and P and N concentrations are i n Appendix 4. The maximum accumulation of branches i n the forest f l o o r on the good s i t e (12232 kg/ha) at age 62 months seems to be very high considering that the same stand had only 6895 kg/ha of branch biomass 11 months TABLE 18. Biomass (kg/ha) and nutrient content (g/ha) of forest f l o o r over an age sequence of E. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l ^ Branches Foliage Total (months) ' Biomass P N Biomass P N Biomass P N (a) Bom Despacho (good s i t e ) 15 0 0 0 1854 597 26 106 13 369 3674 1061 38 2660 173 5440 5258 940 51 7867 506 15214 9165 2809 62 12232 1435 30696 7927 3071 73 6895 563 10969 7131 2640 Carbonita (poor s i t e ) 21 42 12 358 2169 487 32 325 55 2625 4513 1301 43 366 31 1448 4621 1211 56 354 30 2572 6047 2041 67 3270 249 10110 5369 1774 16796 1854(100.0) 1 597(100.0) 16796(100.0) 27217 3780 (97.2) 1074 (98.8) 27586 (98.7) 35887 7918 (66.4) 1113 (84.5) 41327 (86.8) 69055 17032 (53.8) 3315 (84.7) 84269 (81.9) 58982 20159 (39.3) 4506 (68.2) 89678 (65.8) 62803 14026 (50.8) 3203 (82.4) 73772 (85.1) 16123 2211 (98.1) 499 (97.6) 16481 (97.8) 40235 4838 (93.3) 1356 (95.9) 42860 (93.9) 41900 4987 (92.7) 1242 (97.5) 43348 (96.7) 62245 6401 (94.5) 2071 (98.6) 64817 (96.0) 57871 8639 (62.1) 2023 (87.7) 67981 (85.1) Percentage of biomass and nutrients in leaves. 88 l a t e r . This might have been due to sampling error due to the great v a r i a b i l i t y of branch d i s t r i b u t i o n i n the forest f l o o r . The t o t a l f o r est f l o o r biomass at age 62 months (20159 kg/ha) corresponds to that of a plantation of E_^  c i t r i o d o r a at age 24 years (Table 13) when accumu-l a t i o n of branches might have reached steady state. Also, i f the proportion of leaves i n the forest f l o o r i s analysed (Table 18), one can see that at age 62 months leaf contribution dropped from 53.8 to 39.3% and increased the year a f t e r to 50.8%. If the proportion of leaves i s to be maintained at 51%, the amount of branches i n the forest f l o o r at age 62 months should not be greater than 7600 kg/ha, a figure that f i t s the o v e r a l l pattern of the branch data. On the poor s i t e , there was a sudden increase i n branch biomass i n the forest f l o o r at age 67 months. This increase was s i m i l a r to that observed on the good s i t e at age 38 months. The increase i n forest f l o o r branch biomass coincides with the increase i n l i t t e r f a l l branch biomass (Figure 7), which suggests that most of the branches star t to f a l l around ages 36 and 56 months on the good and poor s i t e s respective-l y . One of the reasons for this difference i s that branches on the poor s i t e were thicker than on the good s i t e and therefore took longer to f a l l a f t e r dying. The biomass of fo l i a g e i n the forest f l o o r increased up to 51 and 56 months, on the good and poor s i t e s , r e spectively, then decreased thereafter. This decrease corresponds to the decrease observed i n leaf l i t t e r f a l l (Figure 7) and l i v e leaf biomass (Table 5). The proportion of foliage i n the forest f l o o r remained higher than 94.5% u n t i l 28 and 56 months on the good and poor s i t e , r e s p e c t i v e l y . 89 CD sz CO CO CD E o CO. X ) . - i 9 8 -7-6 5-4-3-2-1-0-Legend O Forest Floor m Litterfall 15 26 38 51 62 73 Age (Months) CD 14 10 « 8 -co CO CD E en o in 4-1 2 0 J (b) i 15 26 38 51 62 73 Age (Months) Figure 7. Biomass (t/ha) of l i t t e r f a l l and forest f l o o r over an age sequence of Ej_ grandis plantations growing i n two different "cerrado" s o i l s i n Minas Gerais, B r a z i l : (a) Foliage and (b) branches, on the good s i t e and, (c) foliage and (d) branches, on the poor s i t e . L i t t e r f a l l was collected over a 12 month period, so each l i t t e r f a l l bar represents the annual l i t t e r f a l l since the previous stand age. 90 When most of the branches started to f a l l , the proportion of leaves decreased to less than 66.4% on both s i t e s , which shows that the pattern of accumulation for d i f f e r e n t components i n the forest f l o o r i s s i m i l a r on both s i t e s , the difference being mainly the length of time f o r the process to occur. Accumulation of P and N i n branches i n the forest f l o o r reached i t s maximum at age 62 and 67 months on the good and poor s i t e s , respec-t i v e l y , when biomass was at i t s maximum, while P and N concentration was much higher at age 26 months on the good s i t e and 21 and 32 months on the poor s i t e than at older ages. The maximum accumulation of forest f l o o r P i n fo l i a g e on the good s i t e was observed at age 62 months when P concentration was at i t s maximum (0.038%) (Appendix 4). At th i s age fol i a g e l i t t e r f a l l also had i t s highest P concentration. P concentration at age 38 months was very low (0.018%) as compared to other ages (0.030-0.038%). The maximum N content i n forest f l o o r f o l i a g e on the good s i t e was observed at age 51 months when fo l i a g e biomass reached i t s maximum. On the poor s i t e , maximum P and N content was observed at age 56 months. P concentration did not show a major change with age, while N concentration increased with age. The maximum t o t a l forest f l o o r P content (branches + f o l i a g e ) was 4.51 and 2.07 kg/ha and the maximum t o t a l N content was 89.68 and 67.98 kg/ha f o r the good and poor s i t e s , r e s p e c t i v e l y . These values are within the range of nutrient content observed for forest f l o o r s of other eucalypt forests (Table 13). Interpre t a t i o n of the data on accumulation of biomass and nutrients i n the forest f l o o r should be made with caution as they are affected by 91 disturbances that occurred i n each stand. Usually these plantations are cu l t i v a t e d p e r i o d i c a l l y u n t i l the age of about 24 months. The s o i l between the tree rows i s mechanically disturbed burying any l i t t e r that has accumulated. 3.7.3 L i t t e r Decomposition Due to problems of vandalism to f i e l d c o l l e c t o r s ( e s p e c i a l l y on the poor s i t e ) , and because of the accidental loss of some l i t t e r weight data i n the laboratory, the data presented here are incomplete; for some c o l l e c t i o n s there i s only one datum and for some sampling periods there i s no weight loss data. For th i s reason, the "general least square analysis of variance method" (Greig and B j e r r i n g , 1980) was used to analyse this experiment. Logarithmic transformation of the percentage data was required i n order to remove the non-normality and improve the homogeneity of variance. On the good s i t e , the weight loss data were analysed as a function of stand age (25, 36, 49 and 60 months) and the season i n which bags were placed i n the f i e l d (wet, dry). Only the 18 month sample data were analyzed due to lack of weight data for the 6 and 12 month samples. There was no s i g n i f i c a n t difference i n weight loss of fol i a g e and branch l i t t e r between eit h e r season or stand age, aft e r 18 months. At th i s age, fo l i a g e had l o s t an average of 45.5% of i t s i n i t i a l weight; branches had l o s t 9.5%. At 24 months these values were 59.9 and 15.8%, for leaves and branches, r e s p e c t i v e l y , based on data from the wet season sample. 92 The percentage change of weight of branches was -4.4% (wet) and 9.8% (dry) a f t e r 6 and 12 months, re s p e c t i v e l y . The i n i t i a l increase i n branch weight contrasts with a decrease of 3.1% over 3 months, reported f o r E. regnans (Ashton, 1975). However, the result reported here for 24 months of exposure i s s i m i l a r to that observed by A t t i w i l l (1968), for green branchwood of E. obliqua (Table 14). The percent loss of weight of f o l i a g e was 33.1% (wet) and 33.0% (dry) a f t e r 6 and 12 months, re s p e c t i v e l y . The lack of s i g n i f i c a n t difference between these two periods of decomposition might be due to the fact that the bags were placed i n the f i e l d i n d i f f e r e n t seasons. The bags from the 6 month sample (wet) received 1521 mm of p r e c i p i t a t i o n during this 6 month period, while the bags from the 12 month sample (dry) received a t o t a l of 1904 mm over the 12 month period (only 383 mm during the f i r s t 6 months). These re s u l t s suggest that decomposition of leaves occurs mainly during the wet season, and that moisture rather than temperature i s the major c o n t r o l l i n g f a c t o r . The re s u l t s of leaf l i t t e r decomposition observed i n t h i s experiment are i n agreement with data reported i n Table 14 for other eucalypt species. On the poor s i t e , s t a t i s t i c a l analysis could not be performed for weight loss data because there were too few samples obtained f o r the set of bags placed i n the f i e l d i n June, 1982, for 43- and 56-month-old stands. They were co l l e c t e d a f t e r 12 (5 samples) and 18 (7 samples) months of decomposition. Average percent loss of foliage biomass was 19 and 28%, and of branches was -2.0 and 4.5% at 12 and 18 months, respec-t i v e l y . These values are much lower than those for the good s i t e . This difference i n decomposition rate due to s i t e i s i n agreement with res u l t s reported by O'Connell and Menage'(1983) . 93 The r e s u l t s of the test of means of changes i n P and N content performed for the 18 months of decomposition, for both season samples, are presented i n Appendix 5. There was a s i g n i f i c a n t i n t e r a c t i o n between age of the stand and season (on which the decomposition bags were placed i n the f i e l d ) for the percentage change i n the quantity of P i n branches and the quantity of P and N i n f o l i a g e . The smallest loss of f o l i a g e P (14.6%) occurred i n the 49-month-old stand for samples placed i n the f i e l d i n the wet season. No s i g n i f i c a n t differences were observed between other values of loss of P i n f o l i a g e . The loss of N i n fo l i a g e was smaller f or samples placed i n the f i e l d i n the dry season, on stands older than 36 months. There was a net gain i n P content of branches e s p e c i a l l y for samples placed i n the f i e l d i n the dry season at younger ages which d i f f e r e d s i g n i f i c a n t l y from those located on the 60-month-old stand. No s i g n i f i c a n t i n t e r a c t i o n between age and season was observed for changes i n the quantities of f o l i a g e N. However, the samples placed on the 60-month-old stand were s i g n i f i c a n t l y greater than those for younger stands. No s t a t i s t i c a l analysis was performed for P and N losses on the poor s i t e due to i n s u f f i c i e n t numbers of samples. Changes i n the amount of P and N a f t e r 18 months was, r e s p e c t i v e l y , 16.3 and 2.2% for fo l i a g e ( l o s s ) , and -23.8 and -89.2% for branches (gain). These loss values are low as compared to an average of 33.6 and 15.3% for f o l i a g e , and the gain values are high compared to 0.4 and -6.1% for branches on the good s i t e . 94 Changes i n the quantity of P and N i n decomposing l i t t e r on the good s i t e as a percentage of the i n i t i a l values are presented i n Figure 8 and Appendix 6 for d i f f e r e n t stand ages over a period of 24 months. The changes i n the quantity of P and N i n fol i a g e for the stand studied followed the same pattern over the 24 months of exposure (except for N i n the 49-month-old stand, a f t e r 6 months of exposure). There was no consistent pattern for branch decomposition over time for the stand ages studied, probably because the samples were too small and p r e c i s i o n i n weight determination was low (the amount of branches placed i n decomposition bags was based on the proportion of branches to f o l i a g e i n l i t t e r f a l l over the 6 previous months). Changes i n weight and of P and N over a period of 24 months are presented i n Figure 9 as an average of a l l stands studied. On the good s i t e , l o s s of P from f o l i a g e (Figure 9a) occurred mainly due to loss of weight, as also observed by A t t i w i l l (1968). P concentration i n f o l i a g e increased s i g n i f i c a n t l y over the f i r s t 6 months and then remained constant u n t i l 24 months (Appendix 4). Upadhyay (1982) also observed an increase i n P concentration of leaf l i t t e r during 7 months of decomposition. However, th i s was followed by a decrease u n t i l the end of the 12 month period. On the other hand, N concentration i n the present study increased s i g n i f i c a n t l y u n t i l 18 months (Appendix 4) and, as a consequence, N loss was lower than weight l o s s . A continuous increase i n N concentration over 18 months was also observed by O'Connell and Menage (1983). On the poor s i t e , percentage change i n mass and i n P and N quanti-t i e s (Figure 9c) and concentrations (Appendix 4) of foliage followed the Figure 8. Percentage change of P and N quantities i n decomposing l i t t e r for d i f f e r e n t stand ages of m E. grandis growing i n Bom Despacho, Minas Gerais, B r a z i l : (a) Foliage P, (b) branch P, (c) f o l i a g e N and, (d) branch N. Pos i t i v e change represents percentage l o s s and negative change represents percentage gain of P and N. lOO-i BO Legend A W e i g h t O P h o s p h o r u s • N i t r o g e n (a) « 12 IB Length of Decomposition (Months) 24 a 12 ia Length of Decomposition (Months) » 12 18 Length of Decomposition (Months) 24 a> c 10 •to -30 a o> n c g) -SO U o a. -so (d) 6 12 18 Length of Decomposition (Months) 24 Figure 9. Percentage change (average from different stand ages) of weight, P and N quantities in foliage and branch litter of grandis growing on two different "cerrado" soils in Minas Gerais, Brazil: (a) Foliage and (b) branches, on the good site and, (c) foliage and (d) branches, on the poor site. Positive change represents percentage loss and negative change represents percentage gain of P and N. 97 same pattern observed on the good s i t e . The difference between s i t e s was mainly i n the proportion of loss over a same period of time. The pattern of release of P and N from branches (Figures 9b and 9d) was somewhat s i m i l a r to that observed for f o l i a g e : P loss follows mass loss more c l o s e l y than does N l o s s . The major difference between decomposition of f o l i a g e and of branches was a greater immobilization of both P and N by the branches. Also, branch N concentration (Appendix 4) does not follow the same pattern that was observed f o r f o l i a g e , which might explain the i r r e g u l a r i t i e s i n N loss on the good s i t e (Figure 9b). Interpretation of the r e s u l t s presented for decomposition i n t h i s study, e s p e c i a l l y for the poor s i t e , should be made c a r e f u l l y because the number of samples was small. 3.7.4 Internal Cycling Tables 19 and 20 give the demand and i n t e r n a l c y c l i n g of P and N of E_^ grandis stands older than 48 months. This age l i m i t was selected because i n t e r n a l c y c l i n g for younger stands would be small. In addi-t i o n , harvesting would not be applied before this age because there i s s t i l l s i g n i f i c a n t stem biomass increase. The data on P and N c y c l i n g could not be presented on an annual basis because there i s a s i g n i f i c a n t year-to-year v a r i a t i o n ; the stands have not yet reached steady state conditions. For t h i s reason the r e s u l t s are presented based on accumu-l a t i o n of P and N up to p a r t i c u l a r age. L i t t e r f a l l was only c o l l e c t e d for stands older than 21 months on the poor s i t e and older than 25 months on the good s i t e . Forest f l o o r accumulation up to these ages was assumed to be l i t t e r f a l l for that period. This w i l l have resulted i n an TABLE 19. I n t e r n a l c y c l i n g of P i n stands of E. grand Is o l d e r than 48 months growing on two d i f f e r e n t " c e r r a d o " s o i l s i n Minas U e r a i s , b r a z i l L i t t e r f a l l Age (months) Branches F o l i a g e T o t a l Stemwood L i v e t r e e s * Dead t r e e s ^ T o t a l a. Bom Despacho (good s i t e ) 51 Demand (kg/ha) 0. ,29 14, .76 15.05 7. .22 Returned as l i t t e r (kg/ha) 0.23 3 .67 3.90 A c t u a l content (kg/ha) 6. ,44 Withdrawal (kg/ha) 0.06 11. 09 11.15 0. ,78 Withdrawal (% of demand) (20.7) (75, •1) (74.1) (10, .8) 62 Demand (kg/ha) 0. ,54 19, .62 20.16 10. .00 Returned as l i t t e r (kg/ha) 0. ,45 4, .80 5.25 A c t u a l content (kg/ha) 6. .58 Withdrawal (kg/ha) 0. ,09 14, .82 14.91 3. 42 Withdrawal (% of demand) (16. 7) (75 .5) (74.0) (34, •2) 73 Demand (kg/ha) 0. ,74 23, .93 24.67 10, .57 Returned as l i t t e r (kg/ha) 0, .66 6 .02 6.68 A c t u a l content (kg/ha) 6. .59 Withdrawal (kg/ha) 0. ,08 17 .91 17.99 3, .98 Withdrawal (% of demand) (10, • a) (74 .8) (72.9) (37 •7) C a r b o n i t a (poor s i t e ) 56 Demand (kg/ha) 0. .29 11 .99 12.28 1, .49 Returned as l i t t e r (kg/ha) 0. .13 3 .90 4.03 A c t u a l content (kg/ha) 1, .29 Withdrawal (kg/ha) 0, .16 8 .09 8.25 0, .21 Withdrawal (% of demand) (55, •2) (67 .5) (67.2) (14 •1) 67 Demand (kg/ha) 0, .59 14 .30 14.89 1, .70 Returned as l i t t e r (kg/ha) 0, .25 4 .98 5.23 A c t u a l content (kg/ha) 1 .44 Withdrawal (kg/ha) 0, .34 9 .32 9.66 0, .26 Withdrawal (% of demand) (57 •6) (65 •2) (64.9) (15 •3) ^Except stemwood. 18.91 2.54 43.72 (6.5)3 11.93 (27.3) 19.76 2.63 52.55 18.33 (34.9) 20.05 2.66 57.95 21.97 (37.9) 5.02 0.63 (2.5)3 8.46 (43.6) 6.74 0.82 (2.6)3 9.92 (41.1) 3 C o n t r i b u t i o n (Z) ol i n l c r i u i l c y c l i n g in SIUIIIWOIKI LO t u t ; i l i ill t; rn.i 1 c y c l i n g . TABLE 20. Internal cycling of N in stands of L. grandis older ttian 48 months growing on two different "cerrado" s o i l s In Minas Gerais, brazil L i t t e r f a l l Branches Foliage Total Stemwood Live trees* Dead trees'"1 Total a - Bom Despacho (good site) 51 Demand (kg/ha) 11. .97 266. .69 278. ,66 102. .89 245. .32 34.25 661. , 12 Returned as l i t t e r (kg/ha) 6, .15 82 .41 88. .56 Actual content (kg/ha) 97. .14 Withdrawal (kg/ha) 5. ,82 184. .28 190. .10 5. ,75 (2. •9) 3 195. .85 Withdrawal (Z of demand) (48. .6) (69 •1) (68. 2) (5 .6) (29 .6) 62 Demand (kg/ha) 21. .82 354, .46 376. .28 132. .08 241, .36 36.22 785. .94 Returned as l i t t e r (kg/ha) 10. .77 107 .42 118. .19 Actual content (kg/ha) 120. .81 Withdrawal (kg/ha) 11. .05 247 .04 258, .09 11. .27 (4. •2) 3 269, .36 Withdrawal (X of demand) (50. .6) (69 •7) (68. .6) (8 •5) (34 .3) 73 Demand (kg/ha) 30. .26 432, .30 462. .56 158, .05 226. .20 37.37 884. .18 Returned as l i t t e r (kg/ha) 16. .08 132 .85 148. .93 Actual content (kg/ha) 147. .54 Withdrawal (kg/ha) 14. .IB 299.45 313. .63 10 .51 (3 •2) 3 324. .14 Withdrawal (X of demand) (46 •9) (69 •3) (67, .8) (6 •6) (36 • 7) Carbonita (poor site) 56 Demand (kg/ha) 4. .91 261 .25 266. .16 31. .70 127.45 15.41 440, .72 Returned as l i t t e r (kg/ha) 2 .89 95 .39 98, .28 Actual content (kg/ha) 26, .60 Withdrawal (kg/ha) 2. .02 165, .86 167. .91 5, .10 (2. •9) 3 173 .01 Withdrawal (% of demand) (41. •1) (63 .5) (63, •1) (16, •1) (39, • 3) 67 Demand (kg/ha) 9. .86 311, .43 321. ,29 38. .02 134. .57 16.74 510. .62 Returned as l i t t e r (kg/ha) 6 .13 12U.40 126, .53 Actual content (kg/ha) 32, .80 Withdrawal (kg/ha) 3. .73 191 .03 194 , .76 5. .22 (2, •6) 3 199, .98 Withdrawal (X of demand) (37 .8) (61 •3) (60, .6) (13 •7) (39. • 2) Except stemwood. •102 of t o t a l l i v e t r e e s . 'cont r i hut i on (X) of i i i t e r u . i l c y c l i n g in ulemwoou' to t o t a l iuLeriwil i - y c l i n g . Age (months) 100 underestimate of i n i t i a l l i t t e r f a l l because there was c u l t i v a t i o n i n the younger stands ( l i t t e r was incorporated into the s o i l ) , and also because the l i t t e r was already p a r t i a l l y decomposed. However, these losses are probably very small because leaf l i t t e r f a l l does not occur u n t i l age 8 months. The method used to analyse i n t e r n a l cycling of P and N followed that used by A t t i w i l l (1980). The t o t a l demand f o r P and N to produce the branches and fo l i a g e that were returned as l i t t e r f a l l was obtained by multiplying the biomass of each l i t t e r component, accumulated over a given period, by the average nutrient concentration (see Appendix 3) of the green components for that period. The withdrawal was calculated by subtracting the amount returned as l i t t e r from the demand. The demand for P and N to produce stemwood was calculated based on regression equations for P and N content and assuming that the trees had only sapwood (P and N content of each sampled tree were calculated based on sapwood concentration of basal, middle and top d i s c s ) . Withdrawal of P and N was calculated by subtracting the actual content of stemwood from the demand. Withdrawal of P from branches was s i g n i f i c a n t l y lower, while that of N was s l i g h t l y higher, on the good s i t e as compared to the poor s i t e (Tables 19 and 20). However, the effect of P and N withdrawal from branches on t o t a l withdrawal before l i t t e r f a l l was i n s i g n i f i c a n t because of the low nutrient content of branches. Dead branches attached to the trees were not sampled i n the present study. Bellote et a l . (1980) found that less than 5% of t o t a l biomass of E. grandis consisted of dead 101 branches a f t e r 48 months. Hence, the contribution of i n t e r n a l c y c l i n g from dead branches would also be i n s i g n i f i c a n t as compared to that from leaves or stemwood because of the low nutrient content by branches. Withdrawal of P and N from f o l i a g e showed no s i g n i f i c a n t d i f f e r -ences between the stand ages studied on both s i t e s . O'Connell and Menage (1982) observed an increase i n P withdrawal with increasing age, while no difference was observed for N over a range of 40 years. How-ever, they observed only a small increase i n P withdrawal between ages 6 (53.8%) and 9 (56.0%) years. The average withdrawals of P from f o l i a g e observed i n the present study were 75.1 and 66.4%, and of N were 69.4 and 61.9%, on the good and poor s i t e s , r e s p e c t i v e l y . A t t i w i l l (1980) reported an average of 65.0% of P withdrawal from f o l i a g e of 47-year-old E. obliqua. O'Connell and Menage (1982) observed values of 53.8 and 50.0% for P and N withdrawal from f o l i a g e of 6-year-old E_^ d i v e r s i c o l o r ; values lower than those observed i n the present study. These d i f f e r -ences i n r e s u l t s could be due to v a r i a t i o n i n s i t e and/or species c h a r a c t e r i s t i c s . No other studies of i n t e r n a l c y c l i n g of E. grandis have been undertaken so far to permit a better comparison with the data presented here. Withdrawal of P from sapwood was s t i l l very low (10.8%) at 51 months, but i t increased s i g n i f i c a n t l y a f t e r 62 months (average of 36%) on the good s i t e , even though i t was observed that the proportion of heartwood at the base was s i m i l a r for these three stand ages (Table 6). On the poor s i t e , P withdrawal from sapwood was s t i l l low at 67 months (15.3%). However, on the poor s i t e , when heartwood was completely formed, i t s P concentration (3.2 ppm) was lower than that f or the good 102 s i t e (6.4 ppm). By using the difference i n concentration between sapwood (average of concentration at the base, middle and top discs) and the lowest heartwood concentration, i t was observed that an average of 95% of P was withdrawn on the good s i t e , and 96% on the poor s i t e (values higher than the average of 84% observed by Lambert (Unpublished) f o r mature E. grandis i n A u s t r a l i a ) . These r e s u l t s show that by the time the stemwood on the poor s i t e has the same proportion of heartwood as on the good s i t e , the amount of P withdrawn from sapwood would be s i m i l a r to that on the good s i t e . Withdrawal of N from sapwood was very low on the good s i t e (average of 6.9%), while on the poor s i t e i t was 14.9%. These low values observed f o r N withdrawal show that the process sapwood-to-heartwood conversion does not contribute s i g n i f i c a n t l y to i n t e r n a l c y c l i n g of N f o r the species studied over the range of stand age studied. By the time more heartwood i s formed t h i s process w i l l be important because the difference i n concentration between sapwood and heartwood are 35 and 17% for the good and poor s i t e s , r e s p e c t i v e l y . However, these values are much lower than those observed f o r P with-drawal. Turner and Lambert (1983) also observed a difference i n N concentration of only 37.2% between sapwood and heartwood for the same species at age 27 years. Withdrawal of P and N from bark and roots was not analysed i n the present study because s i g n i f i c a n t bark l i t t e r f a l l did not occur during the study period and root turnover was not determined. The contribution of stemwood P and N i n t e r n a l c y c l i n g to t o t a l i n t e r n a l c y c l i n g was very low (< 6.5%), except for P on the good s i t e at 62 and 73 months (average of 18.4%). A t t i w i l l (1980) reported that 103 i n t e r n a l c y c l i n g of P i n stemwood of 47-year-old E. obliqua represented 31% of t o t a l i n t e r n a l c y c l i n g . This greater contribution of stemwood at older ages occurs due to a s i g n i f i c a n t production of heartwood. The proportion of t o t a l demand supplied by i n t e r n a l c y c l i n g i s also presented i n Tables 19 and 20. On the good s i t e , i n t e r n a l c y c l i n g supplied 27.3% of t o t a l P demand at 51 months. This proportion increased to 34.9% at 62 months and 37.9% at 73 months, pr i m a r i l y due to increased contribution of i n t e r n a l c y c l i n g i n stemwood. The contribution of i n t e r n a l c y c l i n g to t o t a l N demand was 29.6, 34.3 and 36.7% at 51, 62 and 73 months, resp e c t i v e l y . On the poor s i t e , a higher contribution of P and N i n t e r n a l c y c l i n g to t o t a l demand was found: an average of 42.4 for P and 39.3% f o r N f o r ages older than 56 months. This higher contribution of i n t e r n a l c y c l i n g to t o t a l demand on the poor s i t e i s mainly because of a lower P and N requirement to produce one unit of biomass as compared to the good s i t e , as discussed i n Chapter 2. A t t i w i l l (1980) found that 46% of the t o t a l P demand of a 47-year-old E. obliqua i s supplied by i n t e r n a l c y c l i n g . In the present study, i n t e r n a l c y c l i n g supplied 37.9 and 41.0% of t o t a l P demand at the oldest ages studied on the good and poor s i t e s , r e s p e c t i v e l y . These values are lower than A t t i w i l l ' s because of a smaller proportion of heartwood i n young stands and/or due to differences between species. The contribution of l i t t e r f a l l to t o t a l P and N demand was too complex to be determined i n the present study. In A t t i w i l l ' s study ca l c u l a t i o n s were made a f t e r steady state conditions were reached. The 104 amount of l i t t e r being added to the forest f l o o r i n a given period would correspond to the amount being decomposed. In the present study, the l i t t e r f a l l data correspond to the amount accumulated u n t i l the age being studied. L i t t e r i n the forest f l o o r w i l l then be under d i f f e r e n t decom-position stages. Comparisons between l i t t e r f a l l and forest f l o o r data to obtain nutrient release would not be appropriate because, as already discussed i n th i s chapter, the forest f l o o r data were obtained from stands with d i f f e r e n t s i l v i c u l t u r a l h i s t o r y , e s p e c i a l l y related to the age at which the la s t c u l t i v a t i o n was made. 3.8 Summary and Conclusions The dynamics of nutrients i n E_^_ grandis plantations were studied by analysing l i t t e r f a l l , forest f l o o r accumulation, decomposition of branches and leaves and i n t e r n a l c y c l i n g due to heartwood formation and the shedding of branches and f o l i a g e . From the data presented i n this chapter, c e r t a i n conclusions follows. 1. Branch l i t t e r f a l l showed no d i s t i n c t seasonal v a r i a t i o n whereas most leaf l i t t e r f a l l occurred during the summer (wet season and high temperature) and least during the winter (dry season and low temperature). 2. S i g n i f i c a n t branch l i t t e r f a l l occurred only a f t e r 36 and 55 months on the good and poor s i t e s , respectively; the delay i n branch l i t t e r f a l l on the poor s i t e i s probably due to greater branch d i a -meter as compared to the good s i t e . 3. The maximum annual t o t a l aboveground l i t t e r f a l l (branches + fol i a g e ) was observed at 36 and 49 months on the good s i t e (6325 105 and 6443 kg/ha/yr, r e s p e c t i v e l y ) and at 31 months on the poor s i t e (5304 kg/ha/yr). 4. P and N return i n l i t t e r f a l l f o r both branches and f o l i a g e followed the same pattern as that of biomass. 5. The t o t a l annual P and N content of aboveground l i t t e r f a l l (branches + f o l i a g e ) was about 1.4 and 30.5 kg/ha/yr for a l l stand ages studied on the good s i t e . On the poor s i t e , maximum quantities were 1.4 and 32.6 kg/ha/yr for P and N, r e s p e c t i v e l y , at 21 months, which decreased with increasing age. 6. Leaf l i t t e r f a l l accounts for more than 82 and 88% of the above-ground l i t t e r f a l l transfers of P and N, on the good and poor s i t e s , r e s p e c t i v e l y . Thus, branch l i t t e r f a l l has less influence on these l i t t e r f a l l transfers of P and N. 7. There was a sudden increase i n branch biomass i n the forest f l o o r at 38 and 67 months on the good and poor s i t e s , r e s p e c t i v e l y . This increase i n branch forest f l o o r coincided with the increase i n branch l i t t e r f a l l . 8. The biomass of foliage i n the forest f l o o r increased up to age 51 and 56 months, on the good and poor s i t e s , r e s p e c t i v e l y , and decreased thereafter. This decrease corresponded to the decrease observed i n leaf l i t t e r f a l l . 9. The proportion of leaves i n the forest f l o o r remained higher than 95% u n t i l age 28 and 56 months on the good and poor s i t e s , r espectively, and then decreased to less than 66% on both s i t e s due to substantial increases i n branch l i t t e r f a l l at these ages. 106 10. The maximum forest f l o o r accumulation was 20159 and 8639 kg/ha at 62 and 67 months on the good and poor s i t e s , r e s p e c t i v e l y . However, as already discussed, there might have been a sampling error due to the great v a r i a b i l i t y i n branch d i s t r i b u t i o n on the forest f l o o r . The maximum value should probably not be greater than 17000 kg/ha. 11. The quantity of P and N i n the forest f l o o r followed the same pattern of i t s biomass, the maximum t o t a l P content being 4.5 and 2.1 and the maximum t o t a l N content being 89.7 and 68.0 kg/ha on the good and poor s i t e s , r e s p e c t i v e l y . 12. On the good s i t e , there was no s i g n i f i c a n t difference i n loss of decomposing fo l i a g e and branch l i t t e r biomass as a res u l t of either season or stand age, a f t e r 18 months. At this age fol i a g e had l o s t an average of 45.5% of i t s i n i t i a l mass, and branches had l o s t 9.5%. At 24 months these values were 59.9% f o r leaves and 15.8% for branches. 13. On the poor s i t e , the average loss of mass was 28.0 and 4.5% f o r decomposing f o l i a g e and branch l i t t e r , r e s p e c t i v e l y , after 18 months. 15. On the good s i t e , l o s s of P from decomposing f o l i a g e followed that of loss over a period of 24 months; P concentration remained constant throughout this period. On the other hand, N loss was lower than biomass loss due to a continuous increase i n N concen-t r a t i o n u n t i l 18 months. On the poor s i t e i t was observed a si m i l a r pattern i n the percentage change i n the quantities of P and N from decomposing f o l i a g e . 107 16. On the good s i t e , the pattern of loss of P and N i n decomposing branches was not as clear as that f o r foliage throughout the study period whereas, on the poor s i t e , i t was observed the same pattern as that for f o l i a g e . 17. On the good s i t e , i n t e r n a l c y c l i n g contributed 27.3, 34.9 and 37.9% of t o t a l P demand at 51, 62 and 73 months, r e s p e c t i v e l y . The respective values for N were 29.6, 34.3 and 36.7%. 18. On the poor s i t e , a higher contribution of i n t e r n a l c y c l i n g of P (average of 42.4%) and N (average of 39.3%) to t o t a l demand was observed as compared to the good s i t e , on stands older than 56 months• 19. The contribution of stemwood P and N i n t e r n a l c y c l i n g to t o t a l i n t e r n a l c y c l i n g was very low (< 6.5%) except for P on the good s i t e at 62 and 73 months (average of 18.4%). This may r e f l e c t the young age of the stand and the small biomass of heartwood that had been formed. However, where heartwood was formed, P was recycled very e f f i c i e n t l y . P withdrawal based on the difference between sapwood and the lowest heartwood concentration was 95% on the good s i t e and 96% on the poor s i t e . The present study provided a basis for the following recommenda-tions concerning studies of l i t t e r f a l l , forest f l o o r , and of l i t t e r decomposition i n plantations. (a) In natural stands, random sampling i s commonly used for l i t t e r f a l l and forest f l o o r . However, i n planted conditions, the trees are d i s t r i b u t e d systematically, which means that processes such as l i t t e r f a l l or accumulation of forest f l o o r w i l l occur non-randomly. For t h i s reason, systematic sampling should be used for such a r t i f i c i a l conditions. Branch l i t t e r f a l l and forest f l o o r sampling should employ larger sampling units than were used i n t h i s study. For example, branches should be co l l e c t e d from an area equivalent to the area occupied by each tree, which i s easy to determine for plantations. The tree stem would be used as the centre of the p l o t . A l l branches on the forest f l o o r would be c o l l e c t e d from this area at the beginning of the study and further periodic c o l l e c t i o n s would be made from the same area to obtain information on seasonal b r a n c h f a l l . Forest f l o o r leaf l i t t e r should be c o l l e c t e d from an area that corresponds to 1/4, 1/2 or the e n t i r e area occupied by the tree, using the tree stem as one of the corners or centre of the p l o t . This method would avoid sampling v a r i a t i o n due to c o l l e c t i o n between rows and between plants i n the same row and due to the effects of c u l t i v a t i o n between rows. It i s suggested that leaf l i t t e r f a l l also be co l l e c t e d from an area that corresponds to 1/4 of the area occupied by the tree to decrease sample v a r i a b i l i t y . Usually plantations i n B r a z i l are established at a spacing less than 3.0x2.0 m, which means that l i t t e r traps w i l l usually be smaller than 1.0x1.5 m. An experiment should be developed to evaluate more precisely the e f f e c t of season and age on decomposition rates. At f i r s t , t h i s should be conducted for only one s i t e , using a larger number of 109 samples. Based on the r e s u l t s obtained i n th i s experiment, a second one should be established to evaluate s i t e e f f e c t s . ( f ) Because leaf l i t t e r f a l l represented more than 82% of the above-ground l i t t e r transfers of P and N, more e f f o r t s should be made to understand the process of ret r a n s l o c a t i o n of nutrients from f o l i a g e at the time of leaf senescence, and during decomposition. (g) More detai l e d studies of heartwood formation should be done for eucalypt plantations considering the contribution of this process to i n t e r n a l c y c l i n g of nutrients. 110 CHAPTER 4 FACTORS DETERMINING COPPICE REGENERATION AND SUBSEQUENT SPROUT GROWTH, AND A SIMULATION OF COPPICE GROWTH 4.1 Introduction Coppice i s a type of s i l v i c u l t u r a l system i n which regeneration i s obtained by cutting the tree stems near the ground and allowing the development of sprouts from the stumps or the formation of suckers from the roots (Daniel et a l . , 1979; Smith, 1962). This kind of forest management has been practised f o r centuries i n many parts of the world, and plantations of several species of eucalypts are presently being s u c c e s s f u l l y managed under the coppice system i n several countries including B r a z i l , South A f r i c a , India, Portugal and Spain (FAO, 1981). There i s increasing i n t e r e s t i n growing species such as poplar, cotton-wood and other broadleaved species under the coppice system as a source of biomass f o r energy (Anderson et a l . , 1983). A major advantage of the coppice system i s the high i n i t i a l rate of growth of the sprouts as compared to seedlings, due to the presence of an established root system and the stimulation of growth by wound hormones (Daniel et a l . , 1979). This offers the p o s s i b i l i t y of shorter rotations and increased y i e l d s . Considering the differences i n growth strategies between a tree growing from seed and one developing from a stump (sprout), i t was necessary to develop a conceptual model of coppicing that could subse-quently be incorporated into models such as FORCYTE (Kimmins and Scoullar, 1983) to enable them to be used to evaluate the y i e l d I l l consequences of a l t e r n a t i v e eucalypt management s t r a t e g i e s . In order to obtain basic information for the development of the coppice model, the factors determining coppice productivity were reviewed and w i l l be ' presented here. In addition, two experiments were conducted: one i n the greenhouse to analyse root dynamics and the r e l a t i v e contributions of root nutrient reserves to sprout production, and one i n the f i e l d to compare the growth of sprouts with that of seedlings and to quantify the e f f e c t s of stump diameter and number of sprouts per stump on the biomass production of sprouts. 4.2 Review of Factors Determining Coppice Regeneration 4.2.1 P h y s i o l o g i c a l Aspects Related to Sprouting and Sprout Growth Despite the fact that eucalypts are being extensively planted and managed by the coppice system, there i s a lack of research information on the p h y s i o l o g i c a l aspects of sprouting and sprout growth. Eucalypts have a great capacity to regenerate from the cut stump. Jacobs (1955) and Chattaway (1958) pointed out that this regrowth i s mainly due to the presence of lignotubers and root swellings containing high concentrations of dormant buds. Most eucalypts produce lignotubers; however some develop only c a r r o t - l i k e swellings such as i n Ej_ grandis (Jacobs, 1955). The coppice system has the advantage of being based on the parent root system (Daniel et a l . , 1979). The carbohydrate reserves i n the root system may be important at the stage of bud development and/or early sprout growth. However, another important function of the established root system i s the absorption of water and nutrients to sustain the fast sprout growth. 112 a. Carbohydrates The starch reserve i n the lignotubers or root swellings, together with starch reserves i n the root system, i s considered to lead to better sprouting and a fast early growth of the sprouts (Clark and Liming, 1953; Jacobs, 1955; Schier and Zasada, 1973; Steinbeck and Nwoboshi, 1980). A c o r r e l a t i o n between carbohydrate reserves and time required for suckering to occur was observed by Tew (1970) i n cuttings of aspen (P_. tremuloides). Schier and Zasada (1973) showed that seasonal v a r i a t i o n i n t o t a l nonstructural carbohydrates i n the roots was correlated with dry weight of suckers; there was no c o r r e l a t i o n with sucker numbers. However, other studies of the seasonal v a r i a t i o n i n carbohydrate reserves i n eucalypts showed no c o r r e l a t i o n between lev e l s of carbohydrate and sprouting (Cremer, 1973), or si z e and number of sprouts (Wenger, 1953). Also, Blake (1972) and Taylor £t a l . (1982) observed that the starch l e v e l of eucalypt species remained constant a f t e r cutting during the period of the experiment (118 hours and 8 months, r e s p e c t i v e l y ) ; the only v a r i a t i o n they found was related to soluble sugar concentrations that reached a minimum when sprouting started. Recent studies have shown that plant hormones also play a very important r o l e i n sprouting (Bhatnagar and J o s h i , 1973; Blake, 1972, 1981; Taylor et a l . , 1982). Detailed reviews of the e f f e c t s of these substances on sprouting are presented by Blake (1981) and Schier (1981). 113 b. Water and nutrients Demand f o r water and nutrients i s higher f o r sprouts than f o r seedlings. Both Blake (1980) and Reis (1984) observed a higher t r a n s p i r a t i o n rate i n the sprouts of Ej_ camaldulensis as compared to seedlings. The sprouts had a higher stomatal density and length than seedlings d i d . Hence, a high root density a f t e r coppicing w i l l be important to maintain the water absorption rate required for maximum growth. Very few studies have been conducted on n u t r i t i o n a l aspects of coppicing. Blake (1972) found no c o r r e l a t i o n between nutrient reserves i n the root system and sprouting. However, according to Westman and Rogers (1977b), the root system of E_. signata/E. umbra contained 49.3% of the t o t a l P i n the tree biomass, and th i s nutrient reserve i n the root system presumably contributes to early sprout growth. S i m i l a r l y , r e s u l t s presented i n Chapter 2 showed that up to 30% of P and 34% of N are allocated to the root system of E_. grandis, on the poor s i t e . This higher nutrient content l e v e l i n the root system should have some influence on sprouting and/or sprout growth. Hansen and Baker (1979) and Kaul et a l . (1979) observed that the nutrient concentrations found i n sprouts i s s i m i l a r to that of seedlings. However, evaluations based on concentration alone can be deceptive. The biomass of sprouts i s greater than that of seedlings, e s p e c i a l l y at early stages of growth. Therefore greater uptake of nutrients by sprouts i s required to maintain a given concentration. As a conse-quence, coppice has a higher annual nutrient requirement for new growth than does a seedling (Hansen and Baker, 1979). 114 4.2.2 Root Dynamics a f t e r Coppicing Obviously, studies of root systems a f t e r coppicing, e s p e c i a l l y those concerned with root dynamics, are very important for an under-standing of the r e l a t i o n s h i p between water and n u t r i t i o n and the growth of sprouts. Sanyal (1975) reported the formation of a new taproot for Shorea  robusta 3-4 years after coppice. However, according to Jacobs (1955) a new taproot i s r a r e l y developed a f t e r eucalypts coppice, although younger roots of higher order can probably develop. Lee (1979) studied d i f f e r e n t clones of hybrid poplar and observed an increase i n horizontal and v e r t i c a l growth of roots for one clone a f t e r coppicing, as well as a high c o r r e l a t i o n between root dry weight and coppice growth. Lees (1981) studied the roots of sprouts of red maple (Acer rubrum) and observed that only part of the roots from the parent root system were adopted by the sprouts; new roots were formed that grew into the s o i l and into the decaying stump. Steinbeck and Nwoboshi (1980) sampled taproot and l a t e r a l roots greater than 2 mm of 700 rootstocks of sycamore (Platanus o c c i d e n t a l i s ) to evaluate the e f f e c t s of spacing and r o t a t i o n length on root biomass. They found no effect of spacing, while shorter rotations reduced root biomass, e s p e c i a l l y when a 1-year r o t a t i o n was applied. This e f f e c t was explained i n terms of the higher requirement for carbohydrates for production of new shoots, and of the low leaf area produced by a one-year r o t a t i o n . Because f e r t i l i z e r was applied p e r i o d i c a l l y , the authors considered that n u t r i t i o n was not the l i m i t i n g f a c t o r . 115 4.2.3 P r o d u c t i v i t y of Coppice A higher rate of growth of sprouts as compared with seedlings has been reported for several species. For example, Zavitkovski (1982) observed that seedlings of hybrid poplar reached 6.6 m i n height at age 5 years while dominant shoots a f t e r f i r s t coppicing reached the same height at age 3 years. Similar r e s u l t s were reported by Anderson (1979) for the same species. Kaumi (1983) reported that t o t a l height, DBH, and volume for the f i r s t coppice crop of E^ _ grandis were much greater than for the seedling crop (Table 21). On the other hand, Rezende et a l . (1980) found seedlings to be more productive than coppicing for several eucalypts species i n B r a z i l (Table 22). Sharma (1979) observed maximum MAI for Eucalyptus hybrid at ages 5 and 6 for s i t e s I and I I , respec-t i v e l y , i n the f i r s t coppice crop while for seedling crop t h i s maximum was reached at ages 8 and 11, r e s p e c t i v e l y . Luckhoff (1955) observed that the advantage i n terms of height of E. grandis coppice as compared with seedlings l a s t s f o r 10, 11 and 13 years f o r s i t e q u a lity I, I I and I I I , r e s p e c t i v e l y . Daniel et a l . (1979) mentioned that the decreasing advantage of sprouts over seedlings with increasing sprout age r e s u l t s from the fact that the sprouts use only part of the o r i g i n a l root system. a. Stump mortality There i s usually a steady decline i n productivity i n successive coppice crops (Carter, 1974; Rezende e l a l . , 1980; Kaumi, 1983). According to Raitanen (1978), the current annual increment of coppice forest i n eastern Ontario diminishes a f t e r 4-5 years and the area has to 116 TABLE 21. Stand c h a r a c t e r i s t i c s at the end of seedling and subsequent coppice rotations of E_. grandis i n Kenya. Parameter Seedling Coppice crop crop* 1* 2* 3** Stumps/ha 1303 1195 1103 1038 Sprouts/ha*** - 2076 1976 1910 Mean t o t a l height (m) 14.3 17.2 15.1 13.9 DBH over bark (cm) 11.8 12.2 11.1 10.3 Volume/ha to 5 cm top diameter (m 3) 93.78 193.45 132.29 114.78 Source: Adapted from Kaumi (1983). * 5 yr r o t a t i o n ** 7 yr and 8 months r o t a t i o n ***After thinning to 2 sprouts/stump 117 TABLE 22. Productivity (m3/ha) at the end of the seedling and subsequent coppice rotations of Eucalyptus spp. - CAF -B r a z i l . 1 Species Seedling crop Coppice crop Total 1 2 E_. microcorys E_. propinqua 163 147 130 440 E_. grandis E. alba 140 127 112 379 E. saligna E. robusta 123 113 98 334 IS. c i t r i o d o r a 15. maculata E. paniculata 110 120 88 318 E_. t e r e t i c o r n i s E. camaldulensis 106 88 85 279 Average 128 119 103 350 Source: Rezende et a l . , 1980. 1 A l l rotations were 7 years. 118 be replanted every 3 cutting r o t a t i o n s . In South A f r i c a , replanting of eucalypts i s done a f t e r 2-4 crops (Poynton, 1983). In B r a z i l , the rota-t i o n age for eucalypts varies from 5 to 10 years and a f t e r 3-4 rotations the area i s replanted. Poynton (1983) mentioned that the decrease i n productivity i s not due to s o i l depletion or loss of vigour of the stools because he had observed that surviving stools are s t i l l vigorous a f t e r 4 coppice r o t a t i o n s . He considered stump mortality to be the main reason for the decline i n productivity with consecutive coppice crops. A decrease i n productivity i n the f i r s t coppice crop as compared with the seedling crop was observed by Rezende et a l . (1980) f o r several species i n B r a z i l (Table 22). This decrease might be p a r t i a l l y due to high stump mortality. Table 23 shows an estimate of the productivity of the early stages of the f i r s t coppice, with and without stump mortality. b. Root;shoot r a t i o and n u t r i t i o n a l aspects Besides a decrease i n number of stumps that regenerate, some authors have observed a decrease i n the mean stem s i z e with consecutive coppice crops (Table 21). Perala (1979) also observed a decrease i n the mean stem size of aspen i n successive crops. This decrease In s i z e could be a t t r i b u t e d to loss of vigour of the rootstock and/or to nutrient depletion. Results obtained f o r several species have shown that the advantage of coppice as compared with seedlings occurs only during early stages of development. For example, Clark and Liming (1953) observed that height growth of sprouts of blackjack oak (Quercus marilandica) was 5 times 119 TABLE 23. Productivity for early stages of f i r s t and second rotations of E_. alba i n the "cerrado" region. The numbers within brackets are an estimate of coppicing produ c t i v i t y i f there had been no stump morta l i t y . Seedling crop Coppice crop Age (months) 18 30 16 26 Survival (%) 92 88 73 65 Diameter (cm) 5.8 8.7 7.2 12.6 Height (m) 5.7 12.2 7.7 10.6 Volume (m3/ha) 32 94 34 (37.4) 90 (106.2) MAI (m 3/ha/yr) 20.0 34.8 21.2(28.8) 40.9 (48.3) Source: Companhia A g r i c o l a e F l o r e s t a l Santa Barbara, Belo Horizonte, Minas Gerais, B r a z i l ( i n t e r n a l r e p o r t ) . 120 greater i n the f i r s t year than for the following 5 years. Anderson (1979) mentioned a decrease of one-third i n height growth of poplar i n the second year of the f i r s t coppice crop, while no decrease was observed i n the f i r s t year of the second coppice when 1-year-old sprouts were cut back again and allowed to resprout. This reduction i n growth of sprouts with increasing age might be due to root:shoot r e l a t i o n s h i p s . Johnson (1979) reported a higher number of growth flushes per growing season for coppice shoots of oak as compared to i n t a c t seedlings, saplings or larger trees. He attributed t h i s e ffect to a larger root:shoot r a t i o for coppice shoots. The accelerated shoot growth would be expected to continue u n t i l the proper root:shoot r a t i o i s re-established. Besides the e f f e c t s of root:shoot r a t i o on coppice productivity, there may be an effect of n u t r i t i o n . The maximum growth of the coppice crop i s attained at an e a r l i e r age when compared with seedlings (Sharma, 1979). This phenomenon, i n addition to the observed reduction of stem s i z e with consecutive coppice crops, might be explained by the imbalance between high nutrient demand for fast sprout growth and low rate of nutrient replenishment i n the s o i l . According to Daniel et al_. (1979) the coppice system should be applied only on s o i l s with good f e r t i l i t y and moisture i n order to avoid the e f f e c t s of s i t e nutrient depletion i n successive crops. There i s a lack of information about the e f f e c t s of f e r t i l i z a t i o n on sprout growth. However, preliminary r e s u l t s have shown s i g n i f i c a n t e f f e c t s of f e r t i l i z e r when applied to plantations i n the "cerrado" region. B a l l o n i (1978) c i t e d by Pereira and Brandi (1981) recommended 121 app l i c a t i o n of 490 g of N-P-K (6-10-5) per plant, i n a band between rows. Rezende et a l . (1980) reported better response to f e r t i l i z e r applied immediately after harvesting as compared to that applied 6 months a f t e r harvesting. Rocha et a l . (1982c) observed a s i g n i f i c a n t e f fect of rock phosphate on growth of sprouts of E_. robusta over 12 months. The differences between treatment had disappeared by age 24 months. The f e r t i l i z e r was applied before harvesting the trees. Experiments should be developed to analyse the difference i n growth between sprouts and seedlings, with and without f e r t i l i z e r applications i n order to test the nutrient depletion hypothesis. c. Number of sprouts per stump The number of sprouts per stump a f f e c t s the productivity of coppice greatly. Carter (1974) reported that the volume of the f i r s t coppice crop of Ej_ globulus was 20% greater than the seedling crop, while on the other hand, Rezende et a l . (1980) observed that the volume of the f i r s t coppice crop was smaller than that f or seedlings f or several eucalypt species i n B r a z i l (Table 22). One of the reasons for these d i f f e r e n t r e s u l t s i s that i n the f i r s t case there was no thinning of sprouts while i n the second case, there was thinning to 1-4 sprouts per stump. The thinning to one sprout per stump usually increases the mean diameter but decreases t o t a l basal area or volume as compared with no thinning or l i g h t thinning as observed f or E_^ saligna (Luckhoff', 1955; Howland, 1969), E. alba (Rezende et a l . , 1980), and Eucalyptus spp. (Pereira et a l . , 1980a; Paiva et a l . , 1983). One possible explanation 122 for the decrease i n productivity with a decrease i n number of sprouts per stump might be that each sprout i s able to use only a proportion of a given stump; several sprouts are required to u t i l i z e the stump root system e f f e c t i v e l y . d. Stump diameter Several studies have investigated the rel a t i o n s h i p between sprout-ing and sprout vigour, and the diameter of the stump. In general, there i s an increase i n the percentage of stumps that sprout as stump diameter increases up to a range of 10 to 20 cm, and a decrease at greater diameters (Clark and Liming, 1953; Johnson, 197 5; McDonald and Powell, 1983). Kramer and Kozlowski (1960) explained the decrease on sprouting with increasing diameter i n terms of the increase of bark thickness, which probably has a mechanical e f f e c t on sprouting. The number of sprouts per stump and sprout vigour are also a f f e c t -ed by stump diameter. Grunwald and Karschon (1974) observed that the number of sprouts, the height of the t a l l e s t sprout, and the sprout biomass were highly correlated with height and diameter of the trees before harvesting. For Eucalyptus spp. , Pereira et a l . (1980b) observed an increase i n the number of sprouts up to a stump diameter of 17.9 cm, while the height and diameter of sprouts were s t i l l increasing at a stump diameter of 37.5 cm. Reich e_t a l . (1980) observed that height of sprouts increased as stump diameter increased to 10 cm for white oak (Quercus alba) and 18 cm for black oak (Q. marilandica). Mann (Unpublished) observed a maximum number of sprouts and biomass per stump fo r oaks and hickories when the stump diameter reached 20 cm. McDonald 123 and Powell (1983) observed a bimodal d i s t r i b u t i o n for number and height of sprouts, with the maximum at stump diameter of 10 and 70 cm and the minimum at 25 cm. No explanation was given for the increase after 25 cm. Johnson (1979) observed an increase i n oak sprout height with increase i n stump diameter up to 15.2 cm and a decrease a f t e r t h i s diameter. The author explained this decrease i n terms of the reduction i n leaf area as stump size increased, because of the reduction i n number of sprouts per stump with increase i n stump diameter. Clark and Liming (1953) observed that the crown area occupied by sprout clumps increased with diameter up to a c e r t a i n l i m i t , and then decreased at larger diameters. According to Kharitonovich (1937) (c i t e d by Johnson, 1979) the parent root system cannot be mantained i f leaf area i s low. Considering the observed c o r r e l a t i o n between stump diameter and coppice vigour, Sander et a l . (1976) and Johnson (1977) used stump diameter as one of the variables to indicate the p o t e n t i a l of oak to produce vigorous sprouts. However, according to Johnson (1975), the e f f e c t of stump diameter on sprout growth i n t h i s species disappears by age 12. This suggests that the use of stump diameter to predict sprout growth i s e s p e c i a l l y v a l i d when short rotations are being considered. 4.3 P i l o t Studies of Biomass and Nutrient Accumulation i n Sprouts of  E. grandis i n the Greenhouse and i n the F i e l d 4.3.1 Introduction The presence of an established root system i s considered to be the reason for the high i n i t i a l rate of growth of sprouts as compared with seedlings. However, the dynamics of the root system after coppicing and 124 the contribution of its nutrient reserves to sprout growth are not yet well understood. In order to clarify some aspects of the contribution of the estab-lished root system to sprout growth, two experiments were conducted: one in the greenhouse and one in the field. The objectives of the greenhouse experiment were: 1. to determine the P and N budgets of Ej_ grandis sprouts as a means of assessing the relative contributions of root nutrient reserves and of nutrient uptake from soil to sprout production; and 2. to assess the effect of cutting the stem and the subsequent sprouting on root biomass and dynamics. The objectives of the field experiment were: 1. to compare the growth of sprouts with that of seedlings in order to quantify the difference in growth attributable to the influence of the established root system; 2. to quantify the influence of the size of the established root system on sprout growth by examining sprouting of stumps of different diameter assuming that the size of the root system increases with stump diameter; and 3. to identify the effects of the number of sprouts per stump on sprout growth. The results obtained from these experiments, together with infor-mation presented in the literature review, were then used in the development of the conceptual coppice model that could be implemented in any model that evaluates the effects of intensive forest management on site productivity. 125 4.3.2 Materials and Methods a. Greenhouse experiment Plant establishment. Seedlings of E_. grandis were grown for 9.5 months i n two types of s o i l i n the greenhouse at UBC. The "good" s o i l was a bulked sample of Ah and Bm horizon material from a d y s t r i c b r u n i s o l (Canada S o i l Survey Committee, 1978) developed i n w e l l - s t r u c -tured fine-textured g l a c i a l marine materials, taken i n a young Douglas f i r plantation at the UBC Research Forest at Haney. The "poor" s o i l was constituted by mixed horizons from a mini humo-ferric podzol: i c e - r a f t e d t i l l material with some water r e s o r t i n g , taken from an area cleared of mixed conifer f o r e s t , at the UBC campus. Seeds were germinated on A p r i l 4th, 1982 on f i l t e r paper. The resultant seedlings were planted i n 5 L pots on A p r i l 15th, 1982 and watered when required. Additional nutrients (0.6 g of diammonium phos-phate 21-53-0 (N-P-K) per pot) were supplied i n s o l u t i o n to both types of s o i l s when the seedlings were 5 months old following the development of v i s u a l symptoms of P deficiency. An addit i o n a l f e r t i l i z a t i o n was given to the good s o i l one month l a t e r (same amount as before) and 15 days before harvesting at 9.5 months (0.15 g/pot) to maintain the d i f -ference i n nutrient status between the two types of s o i l . The green-house a i r temperature was set at 20°C (average) and a r t i f i c i a l l i g h t was provided to maintain a photoperlod of 12 hours from e a r l y November to la t e March. Windows were made i n the sides of 10 pots of each s o i l type on December 15, 1982, to permit the observation of root growth. Windows 126 were made by cutting out an area of 9x15 cm and replacing i t by a thick but c l e a r p l a s t i c sheet. Light was excluded from this window by a cover except for b r i e f inspections. Sampling. Total seedling height and diameter at the root c o l l a r were measured at 9.5 months. Black and white photographs of the root systems, v i s i b l e through the pot windows, were taken at the same time. The tops of a l l plants were harvested on January 31, 1983, and dry weight of f o l i a g e , branches and stems were measured for each plant separately. At the same time, the root systems of four plants were sampled for dry weight and chemical a n a l y s i s . The remaining stump-root systems were permitted to resprout. Sprouting started between 10 and 15 days a f t e r harvesting the tops. Every 30 days a f t e r sprouting started, four more pots were des t r u c t i v e l y sampled for f o l i a g e , stem and root biomass. Roots were subdivided into two diameter classes: f i n e roots (< 0.5 mm) and large roots (> 0.5 mm). Samples of each biomass component from each pot were analysed for P and N. Photographs of the root systems of sprouted plants were taken every 15 days for 45 days a f t e r harvesting the tops, and every 30 days thereafter u n t i l a l l the plants had been harvested. The photographs were used to provide a q u a l i t a t i v e assessment of the timing of commence-ment of production of new root production following c u t t i n g . An attempt to estimate the p a r t i t i o n i n g of the source of nutrients for new sprout growth between stump-root reserves and s o i l uptake was made by comparing the t o t a l nutrient content of the root system i n the f i r s t stump-root system sample (at the time of harvesting) with the 127 t o t a l nutrient i n the root systems and i n the sprouts at each successive sampling time. As an i l l u s t r a t i o n of the method, the following i s the c a l c u l a t i o n for the f i r s t time step on the good s o i l : Difference i n nutrient content of t o t a l roots between tg and t^: 14.21 - 12.92 = 1.29 mg Difference i n biomass of t o t a l roots between tg and t^: 13180 - 12730 = 450 mg Amount of P l o s t due root mortality between tg and t ^ : 450 x 0.0006 = 0.27 mg Amount of P required to produce sprouts between tg and t-^: 1.35 mg. Estimate of the amount of P i n roots used to produce sprouts between tg and t j : 1.29-0.27 = 1.02. This value i s 75% of the amount required for sprout growth tg - t ^ . b. F i e l d experiment Plant establishment. The study was conducted i n plantations of E.  grandis, at Bom Despacho (good s i t e ) and Itamarandiba (poor s i t e ) , Minas Gerais. Itamarandiba i s located i n the same region of Carbonita and was used because plantations i n Carbonita are not yet being coppiced. Rotation ages on these s i t e s were 5 and 6 years, r e s p e c t i v e l y . The design of the experiment was randomized blocks with 3 r e p l i c a t i o n s . Each block was subdivided into three p l o t s . Each plot represented a d i f f e r e n t number of sprouts per stump (1, 2, or 3) and contained 10 128 stumps of each of the following diameter (cm): 7.5, 12.5, and 17.5. Each stump was numbered i n the f i e l d . Seeds being used for recent plantations are of a d i f f e r e n t provenance from those used i n the establishment of the stand studied i n t h i s experiment. For t h i s reason, cuttings were taken from 3-month-old sprouts (approximately 30 cm high) on each stump and rooted i n a green-house to produce plants of genetic make up i d e n t i c a l to that of the parent stump. After the cuttings had been c o l l e c t e d , a l l remaining sprouts on the experimental stumps were removed and the stumps were allowed to resprout. The r e s u l t i n g rooted cuttings were then planted out on a s i t e s i m i l a r to that of the parent stumps. At the time of planting the rooted cuttings received the same f e r t i l i z a t i o n as the seedlings from which the stumps o r i g i n a l l y developed; the resprouting stumps were not f e r t i l i z e d . Sampling. Due to problems i n getting the experiments established i n the f i e l d , data could not be c o l l e c t e d on the good s i t e . On the poor s i t e , sprouts were cut from stumps on June 1st, 1982, rooted and planted out on December 28, 1982. The dry weight of leaves and stem were measured f o r the second crop of stump sprouts (which had resprouted from the stumps after the i n i t i a l crop of sprouts had been removed to obtain cuttings) and the corresponding rooted cuttings for each plot ( t o t a l of 27 plants) on the following dates: May 19th, August 18th, November 18th, 1983, and February 20th, 1984. Because of lack of adequate stump re-sprouting i n some treatments, the number of r e p l i c a t i o n s was unequal. 129 4.3.3 Results and Discussion a. Greenhouse experiment Table 26 shows the biomass and nutrient data obtained from the greenhouse experiment. The r e s u l t s of the test of the means are presented i n Appendix 7. Analysis of variance showed a s i g n i f i c a n t difference (at the 5% l e v e l of p r o b a b i l i t y ) between the two s o i l types for a l l variables studied except P concentration i n roots and sprouts. Standard deviation was very high for the biomass of sprouts and conse-quently for nutrient content of sprouts. Biomass and P and N content of the tops of the seedlings from which the sprouts originated showed no s i g n i f i c a n t difference between groups of plants to be sampled for sprout growth analysis at d i f f e r e n t post-harvest dates. P content and concentration of f i n e and t o t a l roots decreased s i g n i f i c a n t l y u n t i l age 2.5 months post-harvest and increased a f t e r this age for both s o i l s . Similar r e s u l t was observed for fin e root biomass on the poor s o i l ; however, there was a decline at age 4.5 months that can probably be explained by the fact that the plants i n t h i s s o i l were again showing symptoms of strong nutrient deficiency at t h i s age. No s i g n i f i c a n t difference was observed for N content and concentration of f i n e and t o t a l roots on both s o i l s and, f o r biomass or t o t a l roots on the good s o i l , as a function of age, even though a trend s i m i l a r to that of P content and biomass of f i n e roots can be observed. The dynamics of root biomass shown by these r e s u l t s i s i n agree-ment with the photographs taken through the root-observation windows over the age sequence. From the photographs, i t was observed that roots ceased to grow immediately a f t e r the tops were harvested. Commencement TABLE 24. Bloaass (g), P and N concent (ag) of coppicing E. grandis growing on tvo different soils In th* greenhouse. Age of sprouts (aonthe) Blouse Fine roots Total roots Blouses Sprouts Bloaass Soil P2 ppa ag/pot (s) Good soil 0.0 '1.5 2.5 3.5* «.S 6.67 (1.34)3 7.36 (1.70) 6.56 (1.16) 6.57 (1.46) 4.85 (0.78) 5.22 (2.27) 6.05 (0.68) 2.87 (1.04) 6.95 (2.96) 8.49 (1.27) 35.69 (10.02) 28.55 (6.95) 24.25 (7.83) 31.17 (12.59) 34.44 (4.68) 13.18 (4.23) 12.72 (1.12) 10.91 (2.18) 13.15 (6.05) 13.80 (4.84) 14.21 (3.26) 12.92 (3.02) 6.33 (2.09) 17.59 (8.11) 19.49 (7.50) 66.91 (18.51 55.66 (10.59 53.58 (15.23 78.61 (34.17 77.95 (23.92 0.23 (0.10) 4.69 (0.77) 9.36 (5.34) 17.40 (5.75) 1.35 (0.72) 9.15 (1.72) 23.61 (11.00) 32.17 (8.03) 6.52 (4.07) 58.03 (3.12) 125.07 (51.90) 165.46 (44.59) 8.99 5.08 2.98 6.72 5.08 22.37 12.64 7.44 46.72 12.64 (b) Poor soli ( 0.0 1.5 2.5 3.5* 4.5 6.1' (1.87) 5.00 (0.61) 3.53 (0.81) 5.48 (0.47) 4.35 (0.73) 5.90 (0.83) 4.04 (2.38) 2.41 (0.87) 6.14 (2.10) 6.87 (1.46) 22.53 17.32 14.46 22.32 18.23 (4.71) (9.51) (7.87) (2.22) (1.99) 11.09 (2.22) 10.48 (1.82) 9.31 (3.06) 11.47 (2.26) 9.15 (2.23) 10.79 (1.41) 8.37 (5.05) 6.12 (1.98) 12.41 (2.89) 14.66 (5.24) 42.81 (6.08) 35.86 (19.91) 35.74 (19.50) 46.06 (1.70) 38.00 (5.57) 0.19 (0.14) 1.42 (0.14) 3.97 (2.02) 7.37 (0.99) 0.93 2.69 7.72 12.89 (0.93) (0.46) (2.23) (1.70) 5.60 (5.01) 13.20 (0.88) 40.66 (13.74) 48.94 (4.77) 4.55 4.29 3.76 5.08 4.02 23.59 22.24 19.49 26.33 20.84 1 Soots ensiler then 0.5 sn disaster. 2 Extracted by Melllch aechod bseed on one ssaple per age. 3 Numbers within bracketa represent standard deviation. * Addition of 1.2 g (good soli) and 0.2 g (poor soli) of dlaaaonlua phosphate. o 131 of new growth was observed 3.5 months post-harvest i n 70% of the plants on the good s o i l ; only 10% showed new roots at 2.5 months. On the poor s o i l , 40% of the plants produced new roots at 2.5 months, 40% at 3.5 months and the rest a f t e r t h i s age. There was no new growth of roots up to the t h i r d time step a f t e r coppicing, so the decrease i n root weight during t h i s period can be considered to be a measure of t o t a l root mortality over this period: 3.4 and 14.3% i n the f i r s t and second time step, r e s p e c t i v e l y on the good s o i l . This suggests that root mortality i s increased by coppice i f we consider that 3.4% i s close to root mortality (per 6 weeks) before coppice. Similar values for the poor s o i l were 5.5 and 11.6% f o r the f i r s t and second time step, respectively, showing that i n i t i a l root mortality i s higher than on the good s o i l . The value i n the second time step was lower for the poor than for the good s o i l because new roots were already being formed by this age on the poor s o i l . The f i r s t objective of t h i s study was to evaluate the importance of root nutrient reserves for sprout growth. For each s o i l , the amount of t o t a l net root mortality was calculated for each time step and mu l t i p l i e d by the lowest P concentration observed i n the roots a f t e r coppice (0.06% on the good s o i l and 0.07% on the poor s o i l (Appendix 7)) i n order to estimate the t o t a l P loss due root mortality. I t was assumed that t h i s lowest concentration represented the concentration i n roots immediately p r i o r to death, a f t e r any r e t r a n s l o c a t i o n of nutrients had occurred. In f a c t , the concentration after translocation may well be even lower than t h i s , so that the estimate of the contribution of root mortality to the nutrient supply for sprout growth i s probably conservative. 132 The difference i n nutrient content of t o t a l roots between each time step, minus the estimated losses due to root mortality, was assumed to be nutrient transfer to the growing sprouts. These cal c u l a t i o n s were done only for P as there was no s i g n i f i c a n t difference i n N concentra-t i o n and content i n the root system as a function of age. After new roots started to grow ( t h i r d time step a f t e r coppicing) i t was impossible to evaluate t h i s effect because there was no information on the amount of root m o r t a l i t y . The r e s u l t s showed that, at 1.5 months, P from root reserves could have accounted for 7 5% of the P required for sprout growth on the good s o i l , while the apparent retranslocation of P exceeded the sprout requirements on the poor s o i l by 100%. The v a r i a b i l i t y of sprout growth i n the f i r s t time step was very high on the poor s o i l which might explain t h i s excess of P. Between 1.5 and 2.5 months, the P r e t r a n s l o -cation values were 70% and 81% of the sprout growth requirement for the good and poor s i t e s , r e s p e c t i v e l y . These findings suggest that P i n root reserves are important at very early stages of sprout growth. This dependence on root reserves decreases with time i n favor of s o i l reserves. As can be seen from Tables 26 and 28, at 1.5 months the biomass of shoots on the good s o i l was only 21% greater than on the poor s o i l and the root:shoot r a t i o was the same, while l a t e r on the difference between the two s o i l s was much greater. I t was also observed that the P concentration i n sprouts (Appendix 7) i n the second time step was about half that i n the f i r s t time step on both s o i l s , and that i t remained low for the rest of the experiment. Presumably, t h i s r e f l e c t s the depletion of root reserves 133 and a switch to s o i l uptake. The difference i n sprout growth between the two s o i l s becomes more evident with time because of increasing dependence on s o i l reserves that d i f f e r i n magnitude. P content i n the s o i l showed a decrease i n the second and t h i r d time step when there i s increasing dependence of sprout growth on s o i l n u t r i e n t s . Addition of 1.2 g and 0.2 g of diammonium phosphate per pot on the good and poor s o i l , r e s p e c t i v e l y , was required between ages 2.5 and 3.5 months due to strong symptoms of P deficiency. The addition of t h i s f e r t i l i z e r promoted a s l i g h t increase i n P concentration i n sprouts, which might indicate an increasing importance of s o i l nutrient reserves. The fact that the r i s e i n concentration of P and N i n the sprouts was modest while the increase i n sprout growth was considerable indicates that the sprouts were indeed d e f i c i e n t i n these n u t r i e n t s . As discussed before, the high root:shoot r a t i o of coppice seems to have a great effect on the fast e a r l y growth of sprouts. In the present experiment, the root:shoot r a t i o of the 9.5-month-old seedlings was 0.43 and 0.58, and at 1.5 months after the top was harvested the r a t i o was 54.38 and 54.31 for the good and poor s o i l s , r e s p e c t i v e l y (Table 25). As the r a t i o i s high, no photosynthate has to be u t i l i z e d for new root growth at t h i s age. However, as a consequence of the rapid early sprout growth this r a t i o i s d r a s t i c a l l y reduced and new roots have to be produced. S i g n i f i c a n t growth of new roots started when the t o t a l -root: shoot r a t i o was 3.5 and 5.5 times the o r i g i n a l seedling r a t i o for the good and poor s o i l s , r e s p e c t i v e l y . These r e s u l t s suggest that photosynthate i s used for root growth only after the root:shoot r a t i o begins to approach the r a t i o that existed before coppicing. The data 134 TABLE 25. Total root:shoot r a t i o of 9.5 months old seedlings and sprouts up to 4.5 months old of E_. grandis growing on two d i f f e r e n t s o i l s i n the greenhouse. Age (months) Good s o i l Poor s o i l (a) Seedlings 9.5 0.40 0.53 (b) Sprouts 1.5 54.38 54.31 2.5 2.33 6.55 3.5 1 1.41 2.89 4.5 0.79 1.24 Addition of 1.2 g (good s o i l ) and 0.2 g (poor s o i l ) of diammonium phosphate. 135 also suggest that new roots do not s t a r t to grow u n t i l the dependence of new sprout growth on s o i l nutrients increases. The fact that the root reserves are r e a l l y important at the early stage of growth would suggest that f e r t i l i z e r should not be applied u n t i l the dependence on root reserves has declined s i g n i f i c a n t l y . b. F i e l d Experiment The major objective of this part of the study was to analyse the difference i n i n i t i a l growth between plants regenerated by coppice and by rooted cuttings. Total raw aboveground biomass data and stem:foliage r a t i o of coppice and of corresponding seedlings over an age sequence and for d i f f e r e n t stump diameter and number of sprouts are presented i n Table 26. A l l sprouts were eliminated from the stump when the cuttings were taken. This allowed stumps to resprout and star t growing at the same time as the rooted cuttings. Resprouting and the development of roots i n the cuttings occurred at about one month a f t e r the experiment started; age was then counted from t h i s date. As was the case for the l i t t e r f a l l data, the raw data from t h i s experiment did not meet the basis analysis of variance assumptions. For t h i s reason a logarithmic transformation was used. Because of unequal sample s i z e , the "general least squares analysis of variance method" (Greig and B j e r r i n g , 1980) was used to analyse the data. The analysis of the logarithm of biomass showed that i n i t i a l accumulation of biomass of E^ grandis regenerated by rooted cuttings i s s i g n i f i c a n t l y slower than i f regenerated by coppicing when age of the TABLE 26. Total aboveground biomass (g/stump) and stem:foliage ratio of coppice and rooted cuttings of E_. grand!s growing on a poor "cerrado" s i te , Itamarandi ha, Minas Gerais, B r a z i l . Type of regenerat ion Type of generation Type of regeneration 1 Stump No. of Age Rooted diameter Rooted sprouts Rooted (months) cuttings Coppice (cm) cuttings Coppice per stump cuttings Coppice (a) Aboveground (or above stump) biomass 10.5 (A.5) 2 230 (85) 3a 1966 (1721) 3bc 13.5 (7.5) 432 (264) a 3051 (2362) c 7.5 759 (848)a 1010 (719) b 1 1118 (I372)a 1953 (1664) b 16.5 (10.5) 1063 (590) b 2579 (1418) A c 12.5 1081 (1056)ab 2576 (1628) c 2 936 (1043)a 3142 (2497) be 19.5 (13.5) 2627 (1486) c 3497 (3100) c 17.5 1423 (1619) b 4733 (2369) c 3 1224 (1295)a 3304 (2451) c Stem: foliage ratio 10.5 (4.5) 1 0.67 (0.13) a 1.40 (0.44) b 13.5 (7.5) 0.81 (0.27) a 1.78 (0.77) be 7.5 1.05 (0.62)a 1.80 (0.95) b 1 1.11 (0.64)a 2.02 (1.09) b 16.5 (10.5) 1.24 (0.46) a 2.17 (0.90) c 12.5 1.07 (0.52)a 2.46 (1.52) be 2 0.94 (0.40)a 2.73 (1.90) b 19.5 (13.5) 1.42 (0.63) b 4.20 (1.80) d 17.5 0.99 (0.40)a 2.90 (1.90) c 3 1.06 (0.46)a 2.41 (1.54) b One-way analysis of variance was performed separately for aboveground biomass and stem:foliage r a t i o , as a function of type of regenera-tion, age, stump diameter and number of sprouts per stump. Type of regeneration, age, stump diameter and number of sprouts per stump were arranged in a f a c t o r i a l experiment. Means followed by the same letter for aboveground biomass or stem:foliage ratio for each interaction between type of regeneration and age, stump diameter or number of sprouts do not d i f f e r s i g n i f i c a n t l y at P <0.05. Test of means Is based on logarithmic transformed data. ^ Age was counted from one month after the experiment started (time when sprouting and development of roots In the cuttings occurred). ^ Age of the rooted cuttings after planted In the f i e l d . Numbers within brackets represent standard deviation. ^ The decrease in growth observed at this age is explained by the large standard variation of sprout growth and the small number f — i of samples collected. W 137 rooted cuttings i s counted from the day roots started to develop. The biomass of sprouts at age 10.5 months corresponds to the biomass of planted rooted cuttings at 16.5 months. However, i f the age of the rooted cuttings i s counted a f t e r planting, by age 10.5 months the i r growth i s only s l i g h t l y less than that of the sprouts and the s i g n i f i -cant difference between sprouts and rooted cuttings disappears a f t e r 13.5 months. The sprout biomass s t a b i l i z e d a f t e r age 13.5 months, while the growth of the seedlings continued throughout the sampling period. The r e s u l t s observed for stem:foliage r a t i o were s i m i l a r to that for biomass. Sprouts had a stem:foliage r a t i o greater than that for rooted cuttings planted i n the f i e l d during the i n i t i a l stages of growth. This suggest that the sprouts i n i t i a l l y have a greater e f f i c i e n c y i n the production of stemwood per unit of foliage because of the e f f e c t s of the established root system. The i n i t i a l f a s t growth of the sprouts occurs because t h e i r root system i s already established and most of the net photosynthate i s being u t i l i z e d f o r aboveground biomass accumulation. In the case of the rooted cuttings, the root system had to grow as the shoot grew. A f a s t e r i n i t i a l growth of sprouts as compared to seedlings was reported by Anderson (1979) and Zavitkovski (1982) for hybrid poplar. Kaumi (1983) also observed that mean DBH and t o t a l height of grandis, i n Kenya were 11.8 cm and 14.3 m, r e s p e c t i v e l y , at the end of 5-year rot a t i o n f or the seedling crop, while those values for the f i r s t coppice crop were 12.2 cm and 17.2 m, r e s p e c t i v e l y . The decreasing advantage of sprouts over seedlings with increasing age was also observed by Clark and Liming (1953) and Anderson (1979). Daniel et a l . (1979) explained this decreased advantage based on the 138 fact that sprouts use only part of the o r i g i n a l root system. In addi-t i o n , the root mortality rate increases sharply a f t e r coppicing, as observed i n the greenhouse experiment, and the rate of new root growth might be lower than that required to keep an adequate root:shoot r a t i o to s a t i s f y the fast i n i t i a l sprout growth. In the present study, another two factors should also be consider-ed. When the cuttings were taken from the stumps to be rooted a l l remaining sprouts were removed. The stumps were allowed to resprout so that growth of sprouts would begin at the same time as that of rooted cuttings. In t h i s case, part of the stumps "growth power" had already been used by the sprouts that grew before cuttings were c o l l e c t e d . A second factor i s that at the time the rooted cuttings were planted i n the f i e l d they received 150 g N-P-K (10-28-6) and a trace of B and Zn; the same amount of f e r t i l i z e r received by the o r i g i n a l seedlings that produced the stumps for t h i s experiment. However, the stumps received no f e r t i l i z a t i o n , because that i s the regular procedure adopted i n coppice management i n B r a z i l . According to Barros et a l . (1981), the e f f e c t of f e r t i l i z e r on the growth rate i n height of grandis and E.  s a l l g n a , growing i n the "cerrado" region disappears by age 4.5 years. Consequently, there would be no residual e f f e c t of the f e r t i l i z a t i o n from the f i r s t r o t a t i o n on sprout growth i n subsequent r o t a t i o n s , other than that r e s u l t i n g from the o r i g i n a l f e r t i l i z e r P s t i l l i n the stump and root system and i n the accumulated l i t t e r . If the stumps had received f e r t i l i z a t i o n i n t h i s experiment, the difference i n growth between sprouts and planted rooted cuttings would probably have been much greater and would have lasted longer. 139 The second objective of this study was to analyse the effects of stump diameter and number of sprouts per stump on i n i t i a l biomass accumulation of sprouts. A s i g n i f i c a n t increase i n biomass accumulation and stem:foliage r a t i o occurred as a function of increasing stump diameter (Figure 26). Consequently, the difference between rooted cuttings and sprouts was greater when the sprouts originated from stumps with large diameter. A p o s i t i v e c o r r e l a t i o n between stump diameter and the diameter and height of sprouts has been reported for Eucalytpus spp. for stumps up to 37.5 cm i n diameter (Pereira et a l . , 1980b). The carbohydrate reserves i n the root system are considered to be important for the fast early growth of the sprouts (Clark and Liming, 1953; Jacobs, 1955; Schier and Zasada, 1973; Steinbeck and Nwoboshi, 1980). In addition, the root nutrient reserves are high, e s p e c i a l l y on poor s i t e s , as discussed i n Chapter 2 (Tables 9 and 10). Consequently, the increase i n sprout biomass per stump with increasing stump diameter could be due to the effect of the amount of carbohydrates and nutrient reserves i n the root system. The s i g n i f i c a n t increase of the stem: foliage r a t i o with increase i n stump diameter observed i n the present study provides evidence of the importance of root reserves for sprout growth. Large diameter stumps have more reserves due to larger root systems and consequently the e f f i c i e n c y of production of biomass per unit of f o l i a g e i s increased. In addition, this r e l a t i o n s h i p could be p a r t i a l l y due to the e f f e c t s of a greater root surface area on nutrient absorption. 140 There was an increase i n biomass of sprouts with increasing number of sprouts per stump. The biomass and the stem:foliage r a t i o of sprouts was s i g n f i c a n t l y greater than that for rooted cuttings, independent of number of sprouts per stump. The increase i n biomass with increasing number of sprouts i s i n agreement with r e s u l t s obtained for other eucalypt species planted i n B r a z i l (Rezende et a l . , 1980a; Pereira et a l . , 1980a; Paiva et a l . , 1983). The reason for this effect i s not yet known. One possible explanation i s that each sprout i s able to use only a portion of the resources of the established root system of a given stump. The portion of the root system not being used would die and decompose. 4.4 Simulation of Coppice Growth The present conceptual coppice model could be implemented i n any model that evaluates the e f f e c t s of intensive forest management on future s i t e p r o d u c t i v i t y . However, because i t w i l l be implemented i n FORCYTE (Kimmins and Scoullar, 1983) several concepts presented here w i l l represent some of the philosophies being adopted i n the development of the FORCYTE model. The FORCYTE model or any other model used to analyse the r e l a t i o n s h i p s between s i t e nutrients and y i e l d were developed from a conceptual model of forests that regenerate by means of seed. However, species such as poplars and eucalypts have well developed vegetative reproductive power through coppicing (root or stump coppicing). The dynamics of coppice growth of these species i s somewhat d i f f e r e n t than 141 that of seedling growth, and t h i s must be represented i n the model before i t can be u s e f u l l y applied to the management of these species. F o l i a r - N biomass increment r e l a t i o n s h i p s ( c f . f o l i a r nitrogen productivity concept of Agren, 1983a, 1983b, see also B r i x , 1983) form the d r i v i n g force for the simulation of plant growth i n FORCYTE-11 (Kimmins and Scoular, 1984a). However, the f o l i a r - N increment r e l a t i o n -ship or other nutrient such as P alone cannot explain the very fast e arly growth of coppice sprouts i n which rapid sprout biomass increase occurs before s i g n i f i c a n t leaf area has been developed. Carbohydrate and nutrient reserves i n the root system, as discussed e a r l i e r , play an important r o l e i n t h i s early sprout growth. In addition, the root system i s already established and most of this "growth power" i s used to produce shoot growth. I n i t i a l l y , only a small proportion of the photo-synthate i s allocated to the production of f i n e roots. The large f i n e roof.shoot r a t i o f a c i l i t a t e s nutrient and water absorption and therefore moisture and nutrients may not be l i m i t i n g to the early growth of sprouts. As a consequence, there i s a rapid growth of coppice sprouts as compared with the shoots of seedlings of the same age. Therefore, the biomass of sprouts, i n addition to the biomass of the o r i g i n a l seedling, has to be given i n the model to take into account the d i f f e r -ences i n growth between seedlings and sprouts. Based on the review and preliminary f i e l d r e s u l t s presented i n t h i s chapter, i t appears that stump diameter has an important influence on sprouting and sprout biomass production. At f i r s t i t was thought that a l l the input data required to run the coppice model should be given as a function of stump diameter. However, an analysis of data on 142 sprout growth over an age sequence where no sprouts were thinned- 5 revealed that the r e l a t i o n s h i p between sprout growth and stump size i s quite complex. Because of t h i s , i t was decided to use a simple repre-sentation of t h i s r e l a t i o n s h i p i n the simulation of coppice growth. In the coppice model, the sprouts are related to the o r i g i n a l stump only i n the f i r s t simulation time step (e.g. one month). In t h i s f i r s t time step, the number of sprouts per hectare i s calculated based on the r e l a t i o n s h i p between stump siz e and number of sprouts per stump. In subsequent time steps, a l l cal c u l a t i o n s of sprout growth are made on a hectare basis. The study of i n d i v i d u a l sprouts would be d i f f i c u l t because the same root system i s shared by a d i f f e r e n t number of sprouts. Stemwood biomass due to i t s importance for management purposes i s the only component to be calculated on an i n d i v i d u a l basis (by using information on sprout size d i s t r i b u t i o n ) . Simulation of subsequent rotations are based on information given for the f i r s t coppice rotation but with the provision for specifying a percent reduction i n the vigour of sprout growth i n subsequent rotations. Figure 10 shows the major factors a f f e c t i n g coppice p r o d u c t i v i t y that are e x p l i c i t l y represented i n the coppice model: transf e r of carbohydrate and nutrient reserves for sprout growth from the estab-li s h e d root system; nutrient absorption by the established root system; number of sprout per stump and the diameter of the stumps that survive; and, nutrient depletion and f e r t i l i z a t i o n . A c e s i t a F l o r e s t a l , Itamarandiba, Minas Gerais, B r a z i l (Internal r e p o r t ) . Figure 10. Flowchart summarizing the major factors represented i n the coppice model, based on the concepts presented i n t h i s chapter. 4> 144 4.4.1 Data Requirements The present study has led to the following information being used i n the coppice model: 1. Information on the r e l a t i o n s h i p between tree s i z e before harvesting  and percentage of stump mortality following harvest The diameter at breast height (DBH) of the trees before harvest i s used i n the model instead of stump diameter. In t h i s case, the user needs to e s t a b l i s h the r e l a t i o n s h i p between stump diameter and DBH i n order to give a l l the information for the coppice input f i l e based on tree size before harvest. At the end of the f i r s t r o t a t i o n , the biomass of each tree stem grown from a seedling i s known i n the case of the FORCYTE model. By using the empirical relationships of stem biomass to tree s i z e , the diameter of a l l trees following the f i r s t harvest can be simulated. Using the above information on stump mortality as a function of tree s i z e , a percentage of trees i n each diameter class w i l l be eliminated at each successive coppice harvest. It i s assumed that the percentage stump mortality i n each tree size class w i l l be the same i n each succeeding coppice r o t a t i o n . 2. Information on the r e l a t i o n s h i p between the number of sprouts and  the si z e of trees at harvest, for the f i r s t time step This information w i l l be used to calculate the number of sprouts following a harvest as a function of stump s i z e at the time of harvest. The r e l a t i o n s h i p between stem biomass and tree diameter described above w i l l be used to obtain tree size before harvest. 145 3. Density of sprouts per hectare for d i f f e r e n t ages over the f i r s t  coppice r o t a t i o n This i s used to determine the sprout mortality curve. The i n i t i a l density ( f i r s t time step) must also be included. It i s assumed that this mortality curve i s v a l i d f o r succeeding coppice r o t a t i o n s . 4. Biomass (t/ha) of sprout stems and branches, and the biomass of  sprout f o l i a g e , over age sequences Because of the f a s t early growth of the sprouts, the pattern of biomass accumulation over the age sequence of sprouts w i l l be d i f f e r e n t from that of the f i r s t (seedling) r o t a t i o n . For t h i s reason, data on biomass accumulation i n coppice sprouts are required i n addition to the data on f i r s t r o t a t i o n biomass accumulation that are already required by any model. The biomass of each sprout component except f o l i a g e i s calculated using the o r i g i n a l seedling biomass component r a t i o s assuming that for a given tree size the r a t i o s w i l l be s i m i l a r f o r both seedlings and sprouts. Foliage biomass i s given separately because the stem:foliage r a t i o of sprouts i s expected to be d i f f e r e n t from that of seedlings (see r e s u l t s of the f i e l d experiment). The growth i n the f i r s t simulation time step i s considered to be due to the "growth power" of the root system as leaves do not e x i s t when sprouting s t a r t s . The growth that i s contributed by carbohydrates and nutrient reserves i n the root system (root "growth power") i s calculated by subtracting from the actual growth of the sprouts, the growth that would have been expected based on the nutrient increment r e l a t i o n s h i p established for the seedlings. 146 5. A variable that defines the decrease i n coppice vigour over  successive rotations There i s usually a decline i n productivity over a series of consecutive coppice r o t a t i o n s . This decline may be due to a loss of vigour of the root system. For example, root mortality a f t e r coppice can be greater than production of new roots. This can decrease the sprout growth i n the subsequent r o t a t i o n . This vigour-decline variable i s based on growth of coppice rotations after the f i r s t coppice. Thus, i f a vigour decrease value of 5% i s used, the t h i r d coppice r o t a t i o n w i l l produce about 10% less biomass as com-pared to the f i r s t coppice rotation; the fourth r o t a t i o n about 15% l e s s . The decline i n subsequent coppice productivity could also be due to the e f f e c t of s o i l nutrient depletion instead of a decline of rootstock vigour. There i s a higher requirement for nutrients by coppice than by seedlings due to the fast e a r l y growth. If a coppice crop i s always harvested with very short rotations, the rate of nutrient replenishment by the s o i l may be smaller than the plant demand and consequently growth w i l l be reduced. If the user of the model believes that nutrient depletion i s the major cause of an observed decrease i n productivity with successive rotations, the factor to decrease coppice vigour should be set close to 1.0. 6. Percentage root death at time of coppice After the seedling shoot i s cut, root growth stops and, i n addition, root mortality increases (see r e s u l t s of the greenhouse experiment). The percentage of root mortality should be s p e c i f i e d 147 for each root s i z e class separately, since mortality might be d i f f e r e n t i n the d i f f e r e n t root classes. Because of the d i f f i c u l t i e s i n obtaining suitable data about root mortality, the f i r s t version of the coppice model assumes that a l l mortality occurs i n the f i r s t simulation time step. After the second time step, root mortality w i l l continue to occur using the root death rate already given for seedlings. In the greenhouse experiment, i t was observed that roots started to grow before the o r i g i n a l root:shoot r a t i o was attained. However, because t h i s information i s not available for f i e l d conditions, the growth of new roots after coppicing can be assumed to s t a r t when the r a t i o of each root component to the sprout shoot i s s i m i l a r to that of a seedling of the same aboveground biomass. 7. Table of s i z e d i s t r i b u t i o n of sprouts as a function of diameter of  sprout and age The r e l a t i o n s h i p between stump diameter, number of sprouts and sprout s i z e i s very complex and cannot be e a s i l y represented i n a model. As a simplication, the v a r i a b i l i t y of sprout size i s considered on an area basis, i r r e s p e c t i v e of the d i s t r i b u t i o n of sprouts of d i f f e r e n t size on stumps of d i f f e r e n t s i z e . This table of s i z e d i s t r i b u t i o n of sprouts defines the v a r i a b i l i t y i n sprout s i z e on a hectare basis ( i . e . the sprouts w i l l be grown as i n d i v i d u a l s without taking into account the character of the parent stump). Because of the e f f e c t s of the parent stump size v a r i a t i o n , i t i s expected that there w i l l be a greater v a r i a t i o n i n sprout s i z e as compared to seedling s i z e . Consequently a separate table i s required for sprouts. 148 Information on height of dominant sprout for d i f f e r e n t ages would be desirable because t h i s information i s d i f f e r e n t from that for seedlings. Height/age i s used to evaluate mortality due to l i g h t competition i n the FORCYTE model. However, i n order to prevent the input f i l e from becoming unreasonably long, i t was decided to use the r e l a t i o n s h i p between height and stemwood biomass of the seedlings to calculate the height of sprouts. It i s assumed that a sprout with a given biomass would have the same height as a seedling with the same biomass, even though the age w i l l be d i f f e r e n t . The simulation of sprout growth as described above represents a f i r s t approximation. As more information becomes available on the dynamics of coppice growth, the model w i l l be modified to incorporate any appropriate changes. 4.5 Summary and Conclusions In order to obtain basic information f or the development of a coppice model that could be implemented i n any model that evaluate long-term effects of intensive forest management on s i t e p r o d u c t i v i t y , the factors determining coppice productivity were reviewed. In addition, two experiments were conducted to provide guidance for the development of the coppice model: one i n the greenhouse to analyse root dynamics and the r e l a t i v e contributions of root nutrient reserves to sprout production, and one i n the f i e l d to compare the growth of sprouts with that of seedling and to quantify the effects of stump diameter and number of sprouts per stump on biomass production of sprouts. 149 The r e s u l t s of the greenhouse experiment support the idea that root reserves are important for the i n i t i a l growth of sprouts. With time, the dependence on s o i l reserves increases, but rapid sprout growth s t i l l continues due to the e f f e c t of the high root:shoot r a t i o on nutrient absorption. Roots ceased to grow immediately a f t e r the tops were harvested. S i g n i f i c a n t growth of new roots did not star t 1) u n t i l the dependence of new sprout growth on s o i l nutrients increased and 2) u n t i l the root:shoot r a t i o began to approach the r a t i o that existed before coppicing. The importance of root reserves and root:shoot r a t i o on sprout growth i s emphasized by the r e s u l t s of the f i e l d study on the effects of stump diameter on sprout growth. Larger stump diameter led to greater sprout growth as the reserves and i n i t i a l root:shoot r a t i o were greater than for small stump diameter. Also, the s l i g h t increase i n stem: foliage r a t i o with increasing stump diameter provides a d d i t i o n a l evidence of the importance of root reserves on sprout growth. If the s o i l were the most important source of nutrients for sprout growth t h i s r a t i o should be the same, independent of diameter. The contribution of the root reserves and the increased surface area for nutrient absorption by the established root system are probably responsible for the observed advantage of sprout growth as compared to seedlings, at e a r l y growth stages. Additional studies should be developed i n order to confirm the hypothesis that root reserves are important for the i n i t i a l growth of sprouts. Because of the great v a r i a b i l i t y between plants i n terms of sprout growth and root systems, such research should involve a larger number of plants than was possible to use i n this study. 150 In the case of the greenhouse experiment, the s o i l used should have been the same i n order to avoid e f f e c t s of other properties, such as physical conditions, on plant growth and s o i l moisture. D i f f e r e n t i a -t i o n i n s i t e q u a l i t y should have been obtained by applying d i f f e r e n t f e r t i l i z e r l e v e l s . A study with radioisotopes (e.g. 3^P) would be useful to define the exact time and proportion of nutrient reserves that are used for sprout growth. In the case of the f i e l d experiment, i t was d i f f i c u l t to obtain rooted cuttings for E. grandis and there was a great v a r i a b i l i t y i n sprout growth within the same diameter c l a s s . It would therefore have been preferable to use more stumps per diameter class and to produce seedlings from seeds instead of cuttings. In the future, more s p e c i f i c research could be conducted e n t i r e l y on clones to have closer c o n t r o l over genotype. A more detailed study of the early stage of growth (1 or 2 years), without thinning of the sprouts, should be made i n order to provide better information on the importance of root reserves. Root studies should be done at the same time to evaluate root dynamics i n the f i e l d . The conceptual model of coppice as described i n t h i s chapter represents a f i r s t approximation. The actual model emphasizes the importance of the established root system through transfer of carbohydrate and nutrient reserves and nutrient absorption for sprout growth. Because of the complexities of the relationships between i n d i v i d u a l sprout growth and the parent rootstock, the simulation of sprout growth i s done on a hectare basis. As more information becomes available on the dynamics of coppice growth, the model w i l l be modified to incorporate any appropriate changes. 151 CHAPTER 5 FACTORS DETERMINING THE DYNAMICS OF P IN THE SOIL SYSTEM AND A SIMULATION OF P CYCLING IN THE ECOSYSTEM 5.1 Introduction As already noted e a r l i e r , simulation models developed to predict the effects of intensive forest management on the s i t e nutrient c a p i t a l and productivity of the ecosystem being simulated are developed using concepts and data for nitrogen, which i s considered to be the l i m i t i n g nutrient for most temperate f o r e s t s . However, the "cerrado" s o i l s on which eucalypts are being widely planted i n B r a z i l are s e r i o u s l y d e f i c i e n t i n P, and pl a n t a t i o n growth cannot be predicted based on a consideration of N alone. Consequently, any model need to be modified to include a d e s c r i p t i o n of those aspects of nutrient c y c l i n g that are unique for phosphorus before the model can be applied with confidence to eucalypt forests i n t h i s region. Descriptions of the processes determining the organic phase of nutrient c y c l i n g are usually well developed i n these models. The area i n which they are weakest with respect to the cy c l i n g of P (and other nutrients) i s the processes occurring i n the mineral s o i l system. This chapter reviews the dynamics of P i n mineral s o i l s and proposes a way i n which this can be represented i n a model. The chapter s t a r t s with reviews of the P status of "cerrado" s o i l s and of P dynamics i n s o i l s , and then describes how t h i s knowledge can be applied i n a P model. 152 Because the P model to be presented here w i l l be implemented i n the FORCYTE model the approach used to develop the conceptual model of P r e f l e c t s two of the philosophies that have guided the development of FORCYTE: (1) the use of as simple a data set as possible, so that the model i s r e a d i l y usable for management purposes, and (2) a l l assumptions concerning compartment size s , process rates and the form of r e l a t i o n -ships between variables are controlled by the user via an input data f i l e . 5.2 Phosphorus Status of "Cerrado" S o i l s , and Implications f or  Eucalypt Plantations i n the "Cerrado" Region As mentioned e a r l i e r , the "cerrado" s o i l s are predominantly yellow red and dark red l a t o s o l s (Freitas and S i l v e i r a , 1977). Latosols are highly weathered s o i l s with high le v e l s of Fe and A l oxides, and k a o l i n i t e . According to Weaver (1974), ( c i t e d by Goedert, 1980) the average percentages of hematite, gibbsite and k a o l i n i t e i n yellow red lat o s o l s located i n the "cerrado" are 7.9; 42.6 and 26% res p e c t i v e l y ; the dark red l a t o s o l s contain 13.4; 31.3 and 27%, re s p e c t i v e l y . Aluminium ion le v e l s i n these s o i l s are also very high. These c h a r a c t e r i s t i c s of l a t o s o l s have a great influence on P a v a i l a b i l i t y to plants. According to Kamprath (1977), P can be adsorbed on the surfaces of hydrated Fe and A l oxides or can form a p r e c i p i t a t e with exchangeable A l . Such adsorbed and pre c i p i t a t e d P forms are not r e a d i l y a v a i l a b l e to plants. Bahia F i l h o and Braga (1975) found an average of 55.3; 76.4; 106.6 and 135.2 ppm of calcium-phosphate (Ca-P), aluminium-phosphate (Al-P), iron-phosphate (Fe-P) and occluded-phosphate, 153 r e s p e c t i v e l y , for 20 samples of "cerrado" s o i l s ; the extractable P (Mehlich method) was so low that i t could not be detected. Lopes (1975) observed that of a t o t a l of 518 samples of "cerrado" s o i l s analysed, 92% had a value for extractable P lower than 2 ppm, the average being only 0.4 ppm. These r e s u l t s show that only a small proportion of the P i n these s o i l s i s r e a d i l y available to plants. N u t r i t i o n a l studies developed i n the "cerrado" region showed that P i s one of the most important nutrients determining the productivity of eucalypt plantations. Mello et a l . (1970) observed a s i g n i f i c a n t increase i n volume of a 5-year-old Ej_ saligna with P a p p l i c a t i o n but there was no response to N. Barros et a l . (1981) pointed out that addi-t i o n of N, P and K had a highly s i g n i f i c a n t e f f e ct on the growth rate of E. grandis and E. saligna during the f i r s t two years and P was consider-ed as the most e f f e c t i v e nutrient. Rezende et a l . (1982b) reported a volume of 45.7 and 49.7 m3/ha for a 26-month-old Ej_ grandis when 150 g/plant of NPK (10-28-6) was applied at planting on two d i f f e r e n t "cer-rado" s i t e s . When 2 tonnes of rock phosphate were applied i n addition to the above f e r t i l i z e r , the volume attained was 86.8 and 92.3 m3/ha, resp e c t i v e l y . 5.3 Evaluation of S o i l P A v a i l a b i l i t y to Plants The d e f i n i t i o n of P a v a i l a b i l i t y to plants i s quite c o n t r o v e r s i a l . Barros (1974) observed that available P, determined by Mehlich method, was one of the s o i l c h a r a c t e r i s t i c s most strongly correlated with E.  alba productivity i n B r a z i l . However, Crane (1978) stressed that over the long term a l l forms of P should be included i n nutrient cycling 154 studies. He pointed out that the use of available P data i n the assess-ment of s o i l f e r t i l i t y can lead to a misinterpretation of the e f f e c t s of tree harvesting on s i t e p r oductivity. Recently, there has been a trend towards using data on t o t a l s o i l P i n nutrient c y c l i n g studies ( F e l l e r , 1984; Hingston et^ a l . , 1981). Total s o i l P i s subdivided into two f r a c t i o n s : organic and inorganic. Organic P i s considered to be e s p e c i a l l y important i n highly weathered s o i l s where i t can represent up to 80% of t o t a l P (Sanchez, 1976). Organic P has been considered i n nutrient c y c l i n g studies ( A t t i w i l l , 1980) and i s adequately represented i n ecosystem simulation models i n terms of plant uptake, i n t e r n a l c y c l i n g i n the plant, losses, l i t t e r f a l l and the dynamics of l i t t e r decomposition (Aber and M e l i l l o , 1982; Kimmins and Scoullar, 1983). The inorganic f r a c t i o n i s composed of Ca-P, Al-P, Fe-P and occluded forms (Chang and Jackson, 1957). In less weathered s o i l s , Ca-P i s the most commom form, while i n the case of t r o p i c a l s o i l s (highly weathered) P i s represented mainly by Al-P, Fe-P and occluded forms (Sanchez, 1976). Khanna (1981) considers that the p a r t i t i o n i n g of t o t a l inorganic P into i t s various forms i s useful because one can apply d i f f e r e n t rates of m o b i l i z a t i o n to the diferent forms and thus gain a more dynamic view of P a v a i l a b i l i t y . 5.3.1 Adsorption Isotherms Phosphorus can be adsorbed on the surfaces of hydrated Fe and A l oxides or can form p r e c i p i t a t e s with exchangeable A l (Kamprath, 1977), which a f f e c t s P a v a i l a b i l i t y f o r plant growth. The sorption processes can be represented through an adsorption isotherm. A P adsorption 155 isotherm relates the amount of P sorbed by the s o i l to the remaining concentration i n the s o i l s o l u t i o n . The slope of the adsorption isotherm i s the phosphate buffer capacity of the s o i l . This indicates the a b i l i t y of the sorbed P to maintain the P concentration i n the s o i l s o l u t i o n . Recently, P adsorption isotherms have been extensively studied as a possible t o o l by which to assess the a v a i l a b i l i t y of P f o r plant growth, and by which to predict the amount of f e r t i l i z e r required on a given s o i l (Fox and Kamprath, 1970; Robertson, 1980). Beckwith (1965) and Ozzane and Shaw (1969) showed that measurement of P sorption helps to explain the v a r i a t i o n i n the rel a t i o n s h i p between P f e r t i l i z e r requirements of plants and available P. Woodruff and Kamprath (1965) and Rajan (1973) found that the degree of saturation of the maximum P adsorption was related to the y i e l d of m i l l e t (Pennisetum  glaucum). They observed an inverse r e l a t i o n s h i p between percentage of P saturation of the adsorption maximum and the equilibrium concentration fo r maximum growth with maximum P adsorption. This occurs as a result of the high buffering capacity of s o i l s with a high P adsorption maximum (clayey s o i l s ) as compared with s o i l s with a low P adsorption maximum (sandy s o i l s ) . Vasconcelos et a l . (1975) obtained maximum y i e l d and P uptake for sorghum (Sorghum vulgare) when the amount of P added was equal to 0.75 of the maximum P adsorption for two B r a z i l i a n l a t o s o l s . Bahia F i l h o and Braga (1975) obtained a maximum y i e l d of oats (Avena  sativa) with values ranging from 0.79 to 0.98 of the maximum P adsorp-t i o n f or several l a t o s o l types i n B r a z i l , where maximum adsorption ranged from 255 to 627 ppm (average of 534 ppm). In a review of some studies of P sorption and y i e l d of barley i n Western Canada, Robertson 156 (1980) observed that i n general there i s a good r e l a t i o n s h i p between P requirement based on adsorption isotherms and r e l a t i v e y i e l d s obtained i n a greenhouse experiment. However, Ayodele et a l . (1982) found no co r r e l a t i o n s between y i e l d of maize and sorption values for Nigerian savannah s o i l s . They mention that this lack of c o r r e l a t i o n i s due to the d i f f i c u l t y of obtaining an accurate measurement of the low buffer capacity of the s o i l s studied. The studies mentioned above were f o r a g r i c u l t u r a l crops. A s i m i l a r study was carried out by Tiarks (1982) using seedlings of l o b l o l l y pine (Pinus taeda). He found that P extracted using the Bray 2, Mehlich 1 and Mehlich 2 methods indicated that P f e r t i l i z e r was required but no c o r r e l a t i o n was observed between these measures of s o i l P and the growth of l o b l o l l y pine. However, P i n s o i l solution estimated from the adsorption isotherms was related to P concentration i n the tops and roots of the seedlings, and consequently to y i e l d . The disadvantages of the adsorption isotherm as a method of assessing P a v a i l a b i l i t y and f e r t i l i z e r requirement are the time and labour costs involved i n determining the isotherm. I t also requires information on the minimum concentration of P i n s o i l solution needed fo r maximum plant growth which i s not well known for most crops and i s d i f f i c u l t to measure, because the P concentration i n equilibrium s o l u -t i o n i s very low. However, once established for a type of s o i l i t only requires measurement of the actual available pool (Ozanne and Shaw, 1968; Robertson, 1980). Despite those disadvantages, adsorption isotherms are useful way of assessing P a v a i l a b i l i t y for plant growth and the amount of f e r t i -157 l i z e r needed to increase the P i n s o i l solution to the minimum required f o r maximum plant growth. I t also involves a more dynamic approach as compared with the use of available P only. For these reasons, adsorp-t i o n isotherms w i l l be discussed i n d e t a i l i n t h i s chapter as the basis f o r representing the dynamics of P i n the s o i l ecosystem i n the P model. a. Desorption Phosphate buffering capacity represents the a b i l i t y of the s o i l solid-phase to replenish the s o i l s o l u t i o n (Fox and Kamprath, 1970). In using P adsorption isotherms, i t i s assumed that P desorption occurs when the concentration of P i n the s o i l solution decreases as a r e s u l t of plant uptake. It has also been assumed by many that desorption w i l l follow the adsorption isotherms. However, several studies have shown that P desorption curves are d i f f e r e n t from P adsorption curves; that adsorption isotherms overestimate the desorption process (Kafkafi et a l . 1967; Fox and Kamprath, 1970; Oka jima et al_., 1983). Okajima et a l . (1983) studied desorption and readsorption of P for two s o i l s . They observed that the desorption curves are d i f f e r e n t from the adsorption curves and that the readsorption curves are d i f f e r e n t i n the two s o i l s s tudi ed. There i s s t i l l considerable uncertainty about t h i s method of assessing P a v a i l a b i l i t y . One problem relates to the s e n s i t i v i t y of isotherms to the methodology used to develop them. According to Barrow (1983b), d i f f e r e n t desorption curves can be obtained according to the a n a l y t i c a l procedure employed. If the sample i s shaken too strongly 158 while obtaining data on desorption, new f i x a t i o n s i t e s may be exposed which causes sorption rather than desorption. 5.4 E f f e c t s of Residual P on Plant Growth The residual P would a f f e c t future productivity of the s i t e by decreasing the number of s i t e s for P f i x a t i o n of future P f e r t i l i z a t i o n additions, and/or by being slowly released i n subsequent years. Ozzane and Shaw (1967) found that s o i l s that have received f e r t i l i z e r over 16 years adsorbed less P than u n f e r t i l i z e d s o i l s . Fox and Kamprath (1970) observed a decrease i n adsorption with increase i n f e r t i l i z e r l e v e l . Smyth and Sanchez (1980) obtained a s i m i l a r result for "cerrado" s o i l s . B a l l a r d (1978) found that P added 19 years e a r l i e r i n the f i r s t r o t a t i o n of Pinus radiata had an e f f e c t on the productivity of the second rota-t i o n . Spratt et a l . (1980) observed that ap p l i c a t i o n of 100; 200 and 400 kg P/ha l a s t about 6; 9 and 13 years, respectively, before further f e r t i l i z e r i s required to maintain wheat production. These r e s u l t s emphasize the importance of the e f f e c t s of residual P i n subsequent crop rota t i o n s . 5.5 E f f e c t s of Root Growth on P Uptake Fine root growth i s a very important determinant of P uptake, as P d i f f u s i o n through the s o i l i s very slow. P absorption by plants occurs from a very short distance around the root system, only a f r a c t i o n of mm wide (Russell, 1973). If we accept that P i s absorbed only over a very short distance, new root growth i s very important i n "searching" f o r P i n presently unoccupied s o i l . And this i s e s p e c i a l l y important i n s o i l s 159 very low i n P. Plants with more extensive root systems can be maintain-ed at lower s o i l P l e v e l s (Kamprath, 1977; van Noordwijk, 1983). Hence, at low s o i l P solution concentrations, the plant i s able to achieve the same growth as at high s o i l P s o l u t i o n concentrations by having a more extensive root system, i f the roots are able to continue to grow. The e f f i c i e n c y of the root system i n absorbing P i s dependent not only on the biomass of the fi n e roots expressed as a proportion of the maximum biomass of f i n e roots observed f o r the s i t e (assumed to equal 100% s i t e occupancy), but also on the percentage of s o i l volume that i s exploited by roots when f i n e root biomass i s at i t s maximum. The f i n e root biomass might reach i t s maximum for a given s i t e and not ex p l o i t 100% of the s o i l volume. 5.6 Simulation of P i n the S o i l System Based on the above review, one can see that the P adsorption isotherms integrate the e f f e c t s of several s o i l parameters that deter-mine P a v a i l a b i l i t y and which, i f not accounted f o r , are responsible for great v a r i a b i l i t y i n growth response. For t h i s reason, the P adsorption isotherm was selected as the basis on which to simulate the a v a i l a b i l i t y of P. i n the mineral s o i l for plant growth. Modelling of P f i x a t i o n i n the mineral s o i l i s very complex. For example, Barrow (1983a) b u i l t a model to simulate P sorption processes taking into consideration the effects of pH, temperature, time of con-tact, and concentration of P. The model also simulates the desorption process. 160 Because the present coppice model i s being developed to be imple-mented i n models f or management purposes, a complex data set such as i s required i n Barrow's model i s not s u i t a b l e . On the other hand, one of the philosophies being used to b u i l d t h i s model, as i n FORCYTE, i s f l e x i b i l i t y i n terms of the user's a b i l i t y to control the q u a l i t a t i v e and quantitative aspects of processes. A s t a t i c approach, such as the use of a fixed amount of available s o i l P, by i t s e l f i s not appropriate for models being used with management purpose. Hence, the approach used here to simulate P w i l l be dynamic but with as simple a data set as possible. A flowchart of the conceptual model i s presented i n Figure 11. In simulation i t i s important to i d e n t i f y the order i n which processes occur. Usually there i s a very small transfer of P from decomposing f o r e s t f l o o r materials down into the mineral s o i l , as was observed f o r t r o p i c a l forest by using 32p (Luse, 1970; Stark and Jordan, 1978). Therefore, uptake by plants of P released from decom-posing organic matter i s simulated before s o i l sorption. In contrast, P released to the mineral s o i l by inputs such as f e r t i l i z a t i o n , weathering and p r e c i p i t a t i o n undergoes s o i l sorption before the model simulates plant uptake; the P that remains available after mineral s o i l sorption i s then absorbed by plants i f required. 5.6.1 Data Requirements The following are the data being used i n the development of the P model based on the review presented i n th i s chapter. 161 Figure 11 - Flowchart summarizing P dynamics in the ecosystem as represented in the P model, based on the concepts presented in this chapter 162 1. Information on the capacity of the soil to adsorb phosphorus, by  defining the soil adsorption isotherm The basic information that will be required in the input file is a set of data that define the P adsorption isotherm: the quantity of P added and the quantity of P in solution (kg/ha). The amount of P adsorbed will then be calculated for each time step, whenever there is any P input using the relationship between P added and P remaining in soil solution as defined by these information given in the input f i l e . 2. Proportion of sorbed P that will not be desorbed The amount of P sorbed after a fertilizer addition that will "never" (i.e not within the next several rotations) be desorbed after a fertilizer addition will be included in the input f i l e . This datum will be used to construct the desorption curve by connecting the present position on the isotherm to the given y axis value, using a straight line (Figure 12). A linear relationship for desorption is used because there is a lack of information concerning the shape of this line. In the case where a l l sorbed P is considered capable of desorption, the desorption process will follow the original adsorption curve. The approach adopted in the present model allows the user to select any degree of reversibility of the sorption process: total irreversibility, total reversibility following the original adsorp-tion curve, and reversibility following a different curve because not a l l adsorbed P might be desorbed. 163 LEGEND 00 •a M O c f l c 3 O r i g i n a l Adsorption Isotherm Desorption Line Re-adsorption Curve '1 x 2 x 3 Amount of P i n s o i l s o l u t i o n (kg/ha) x^ - I n i t i a l s o i l s o l u t i o n P x^ - S o i l s o l u t i o n P a f t e r f e r t i l i z a t i o n X.J-X2 - Decrease i n s o i l s o l u t i o n P due to uptake and subsequent desorption i n one time step - Amount of sorbed P that w i l l not be desorbed a f t e r f e r t i l i z e r a p p l i c a t i o n y^ - Amount of P sorbed a f t e r f e r t i l i z e r addition y_-y_ - Amount of P desorbed i n one time step Figure 12 - Graphic representation of P adsorption-desorption-re-adsorption processes 164 The adsorption of a d d i t i o n a l P f e r t i l i z e r i n the next time step w i l l follow another curve with a s i m i l a r shape to that of the o r i g i n a l adsorption isotherm. The new adsorption curve w i l l s t a r t at the lowest point reached i n the desorption l i n e and w i l l end at the maximum adsorption point of the o r i g i n a l adsorption isotherm (Figure 12). This method w i l l take the residual e f f e c t into consi-deration as there w i l l be change i n the slope. 3. Rate of desorption Because not a l l the sorbed P w i l l be desorbed i n one time step, e s p e c i a l l y i f the time step i s as short as a month, a rate of desorption must be given i n the input f i l e . By using a rate of desorption smaller than 1.0, a minimum amount of P w i l l never be desorbed, even though t o t a l r e v e r s i b i l i t y of the sorption process i s being considered. 4. Minimum P concentration i n the s o i l s o l u t i o n at which plants are not  able to absorb P This w i l l prevent the curve from reaching the point where P i n s o i l s o l u t i o n i s too low for plant uptake. 5. I n i t i a l a v a i l a b l e s o i l P At the i n i t i a l stage, before any input i s considered, there i s P i n the s o i l that i s considered r e a d i l y available to the plants and has to be given i n the input f i l e on a per hectare basis. Because available P i s not taken up immediately, i t w i l l then be released for plant uptake according to the rate of desorption given i n the input f i l e . I f the user wants to include P very slowly a v a i l a b l e , i t can be included as part of the amount released through weathering. 165 6. Root e f f i c i e n c y i n r e l a t i o n to P uptake The e f f i c i e n c y of the root system should be given i n the Input f i l e i n order to indicate the proportion of s o i l P considered a v a i l -able that could be absorbed by plants when the biomass of f i n e roots has reached i t s maximum. As most of the f i n e roots are located at the upper s o i l layer and i n the forest f l o o r , e s p e c i a l l y i n nutrient poor s o i l s (Stark, 1971) i t w i l l be considered that a l l P released from decomposition w i l l be absorbed by the plants to supply requirement, according to the root e f f i c i e n c y factor given i n the input f i l e . Thus, by having a root e f f i c i e n c y factor lower than 1.0 some P from decomposition w i l l undergo f i x a t i o n despite the plant needs. The effect of the presence of mycorrhizae on P absorption can be accounted for by increasing the root e f f i c i e n c y f a c t o r . 7. Rate of d i s s o l u t i o n of f e r t i l i z e r over time The rate of d i s s o l u t i o n can be given as a constant or as d i f f e r e n t rate over time ( i f the pattern of d i s s o l u t i o n over time i s known). The rate of d i s s o l u t i o n w i l l then c o n t r o l the amount of P from f e r t i l i z e r that enters the available pool i n each time step. 8. Volume of s o i l i n contact with f e r t i l i z e r The volume of s o i l i n contact with f e r t i l i z e r must be given i n the input f i l e . I f f e r t i l i z e r i s placed i n a band or spot there w i l l be a higher saturation of P i n that given s o i l volume. P a v a i l a b i l i t y w i l l be greater than i f the f e r t i l i z e r i s broadcast due to smaller number of f i x a t i o n s i t e s i n contact with the f e r t i l i z e r (Tiarks, 1982; Kamprath, 1977). 166 The c a l c u l a t i o n s i n FORCYTE are made on a per hectare basis, which i s appropriate for N. However, i f P i s to be applied to only a proportion of the area, the s o i l has to be subdivided into two compartments: f e r t i l i z e d and u n f e r t i l i z e d . The s o i l f i x a t i o n process w i l l not be overestimated after f e r t i l i z e r placement i f the calcul a t i o n s are done i n r e l a t i o n to the proportion of the area i n contact with f e r t i l i z e r . At e a r l i e r stages, the roots w i l l be growing i n the area with high P saturation. The roots w i l l get as much nutrient as they are allowed i f i t i s available from the f e r t i l i z e d area, as determined by the percentage of root occupancy. A f t e r the roots s t a r t to spread over the u n f e r t i l i z e d area, they w i l l s t i l l u t i l i z e as much P as they need from the f e r t i l i z e d area. A proportion of available P i n the u n f e r t i l i z e d area can also be used according to percentage of root occupancy. By considering that roots w i l l be e f f i c i e n t i n the usage of the P from the f e r t i l i z e d area we take into account the fact that most roots w i l l be located i n the f e r t i l i z e d area (Sheppard and Racz, 1980) and/or that there i s no l i m i t i n the uptake rate. This second assumption w i l l be considered despite the fact that Borkert and Barber (1983) mentioned that the exposure of only part of the root system to higher P concentration cannot supply the plant with s u f f i c i e n t P f o r maximum y i e l d due to li m i t a t i o n s i n the uptake rate. The d i f f e r e n t i a t i o n into two s o i l compartments w i l l disappear when a new f e r t i l i z a t i o n i s applied because the new a p p l i c a t i o n might be placed i n a spot d i f f e r e n t from that of the o r i g i n a l one. 167 Residual P (not desorbed), P to be desorbed and actual available P by that time w i l l be then transformed into a per hectare basis. Calculations related to the new addition of f e r t i l i z e r w i l l then depend on the P status of the l a s t time step on a per hectare basis. 9. Natural inputs and outputs Atmospheric inputs and weathering of P can also be included i n the input f i l e i f data are a v a i l a b l e . However, the contribution of these processes i s very small i n terms of P. If a short r o t a t i o n i s considered i n highly weathered s o i l s , these values might be n e g l i g i b l e . Leaching of P w i l l be simulated as a function of root e f f i c i e n c y (e.g. i f root e f f i c i e n c y i s 0.95 there w i l l be a leaching of 5% of P i n s o l u t i o n i n each time step). If the user considers that P leaching does not occur for a given s i t e , leaching can be switched o f f . 5.7 Summary and Conclusions A review of P i n the s o i l system was presented as a basis on which to define the parameters required to simulate P i n the ecosystem. The adsorption isotherm was selected as the tool by which to assess the a v a i l a b i l i t y of s o i l P f o r plant growth because i t integrates the effects of several s o i l parameters that determine P a v a i l a b i l i t y for plant growth. Uptake by plants of P released from decomposition i s simulated before s o i l sorption. In contrast, P added to the mineral s o i l by inputs such as f e r t i l i z a t i o n , weathering and p r e c i p i t a t i o n undergoes 168 s o i l sorption before the model simulates plant uptake from the P that remains a v a i l a b l e . The effect of f e r t i l i z e r type w i l l be taken into account by using a s p e c i f i c rate of d i s s o l u t i o n . F e r t i l i z e r placement w i l l be c o n t r o l l e d by having information on the volume of s o i l i n contact with P f e r t i l i z e r and by making a l l calculations based on t h i s volume. The e f f i c i e n c y with which the plant absorbs P w i l l be related to the percentage of s o i l that i s exploited by the roots when fine root biomass has reached i t s maximum, and the proportion of f i n e root biomass i n each time step as compared to i t s maximum. Atmospheric inputs and weathering are very small for P. However, they can be included i n the input f i l e i f data are a v a i l a b l e . Leaching i s simulated as a function of the root e f f i c i e n c y . The simulation of P i n the ecosystem as presented above can be improved i n the future by including more d e t a i l s on P processes i n the s o i l and/or based on r e s u l t s of c a l i b r a t i o n and v a l i d a t i o n tests of the model for d i f f e r e n t s i t e conditions i n order to obtain more accurate information on the effects of intensive management on s i t e nutrient status. 169 CHAPTER 6 EVALUATION OF THE TRADITIONAL METHODS OF PREDICTING THE LONG-TERM CONSEQUENCES OF INTENSIVE MANAGEMENT OF EUCALYPTS 6.1 Introduction Intensive forest management has been widely adopted i n B r a z i l i n recent years because of increasing wood demand. C u l t u r a l techniques such as s i t e preparation, f e r t i l i z a t i o n , weeding and close spacing (e.g 2x1 m) are now common practice i n the management of fast-growing species such as eucalypts. Short rotations have also been adopted and there i s a trend towards whole tree harvesting. Most of the recent plantations i n B r a z i l have been established i n "cerrado" s o i l s that are n a t u r a l l y very i n f e r t i l e , and which have been degraded by 100 years of repeated tree cutting (for charcoal produc-t i o n ) , burning and grazing. Eucalypt plantation management techniques such as reduced spacing, short rotations and whole tree harvesting i n combination w i l l probably have a further impact on the nutrient status of these s o i l s . There i s growing concern that the productivity of future rotations w i l l be reduced unless there i s a s i g n i f i c a n t increase i n f e r t i l i z e r a p p l i c a t i o n s , which may render these forest ventures uneconomic. The most commonly used method for the determination of the e f f e c t s of intensive management on s i t e nutrient status has been the s t a t i c biogeochemical inventory approach. This involves the estimation of nutrient 170 content of biomass components, l i t t e r and s o i l at a given time and an evaluation of the degree to which s i t e nutrient c a p i t a l w i l l be changed by various levels of management i n t e n s i t y . However, understanding the s u s t a i n a b i l i t y of productivity on a s i t e over several subsequent rotations i s rather complex, req u i r i n g the consideration of many factors other than just the magnitude of harvest removals r e l a t i v e to s i t e nutrient inventories (Kimmins, 1977). Ecologically-based computer simulation models have been suggested as the best way to handle the extensive knowledge required to predict the long-term consequences of forest management a c t i v i t i e s f o r s i t e nutrient status and p r o d u c t i v i t y (e.g. Aber and M e l i l l o , 1982; Kimmins and Scoullar, 1983, 1984a). In t h i s concluding chapter, data on the accumulation and dynamics of biomass and nutrients over age sequences of E. grandis plantations growing on good and poor "cerrado" s i t e s (Chapters 2 and 3) w i l l be used to evaluate the effects of shortening r o t a t i o n length and i n t e n s i f y i n g u t i l i z a t i o n on the nutrient status of the two s i t e s studied. 6.2 E f f e c t s of Intensive Management of E. grandis Plantations on P and  N removal: A T r a d i t i o n a l Evaluation The e f f e c t s of harvesting on P and N removal w i l l be evaluated i n two ways: 1. on the basis of data on biomass and nutrient accumulation over age sequences alone; and 171 2. on the basis of the f u l l data set on biomass, nutrient accumulation and on dynamics of nutrients ( i n t e r n a l c y c l i n g , l i t t e r f a l l and l i t t e r decomposition). By comparing the r e s u l t s of these two approaches, the value of having data on nutrient dynamics can be assessed. 6.2.1 S t a t i c Inventory Assessment a. Proportion of P and N i n tree biomass removed at harvest Figure 13 and Table 27 summarize the r e s u l t s obtained f o r biomass and for P and N content over age sequences for both s i t e s studied. A more detailed presentation of this Information was given i n Tables 5 (page 42), 9 (page 52) and 10 (page 53). Figure 13 i l l u s t r a t e s the cumulative e f f e c t of increasing i n t e n s i t y of u t l i z a t i o n . The bottom section of each bar represents the amount of biomass, P and N that would be removed by harvesting stemwood only. The subsequent sections of each bar represent the a d d i t i o n a l removal that would accompany successive increases i n u t i l i z a t i o n : whole stems, a l l aboveground biomass, and t o t a l tree. This figure also i l l u s t r a t e s the effects of shortening the r o t a t i o n on nutrient export from the s i t e s . On the good s i t e , i f only stemwood i s harvested for pulpwood (leaving stembark on the s i t e ) i n stands older than 51 months, about 25% of the P contained i n the t o t a l tree biomass would be removed. For charcoal production, t h i s proportion would be increased to 48% at age 51 months, and to 56% at age 73 months, because stembark i s also harvested. Hingston e_t a l . (1979) also emphasized the importance of on-site co JZ co co CO E o 00 80 60 40 20 Legend O Roots CZ3 Foliage d Branches CD Bark E 3 Wood 15 26 38 51 62 73 Age (Months) CD CD CO 3 o JZ a CO o JZ 0. 25~ 20 15-10 15 26 CD JZ '— CD C a) cn o 360 300 240 180-120 60 51 62 73 Age (Months) Age (Months) 60 -i CD JZ CO CO CO e O CO 40 20 (b) Figure 13. 222 21 32 43 56 67 Age (Months) CO sz CD CO Z> L_ O JZ a CO O JZ Q_ 10-| (b) 21 32 43 56 67 Age (Months) CD JZ CD c QJ cn o 180 120 60-21 32 43 56 67 Age (Months) ho Biomass (t/ha) and P and N content (kg/ha) over an age sequence of grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l : (a) Bom Despacho - good s i t e and, (b) Carbonita - poor s i t e . TABLE 27. A summary of P and N that would be removed (g/ha) at harvest for different ages and levels of u t i l i z a t i o n of E^ . grandis plantations growing on two different "cerrado" soils in Minas Gerais, Bra z i l . Level of U t i l i z a t i o n Age Stemwood Stem (wood + bark) Whole tree 1 Total tree (months) P N P N P N P N (a) Bom Despacho (good site) 38 4.8 (19A) 3 45 (1802) 8.0 (267) 61 51 6.A (110) 97 (1668) 12.2 (177) 134 62 6.6 (112) 121 (2064) 13.7 (198) 157 73 6.6 (107) 148 (2386) 15.0 (206) 185 i Carbonita (poor site) 32 0.8 (75) 14 (1341) 1.5 (108) 22 A3 0.7 (72) 14 (1364) 1.4 (107) 21 56 1.3 (67) 27 (1392) 2.4 (102) 40 67 1.4 (63) 33 (1429) 2.7 (97) 49 (2037) 20.9 (437) 285 (5946) 23.6 (420) 332 (5913) (1943) 22.3 (242) 291 (3152) 25.4 (267) 342 (3576) (2281) 23.9 (290) 310 (2768) 26.4 (277) 362 (3786) (2541) 24.4 (286) 323 (3789) 27.6 (281) 274 (2806) (1647) 5.4 (241) 98 (4393) 7.1 (222) 129 (4019) (1694) 4.9 (246) 87 (4413) 6.1 (213) 116 (4042) (1704) 4.4 (155) 108 (3798) 6.3 (151) 154 (3687) (1726) 6.1 (179) 110 (3232) 8.1 (164) 167 (3333) Whole tree = entire aboveground biomass: stembark, stemwood, branches and foliage. Total tree = Whole tree + stump and roots. Numbers within brackets represent removal of P and N (g) per tonne of biomass removed. 174 debarking as the bark of E. d i v e r s i c o l o r i n t h e i r study contained 30% of t o t a l stem P. The e f f e c t of on-site debarking w i l l be smaller f o r N as the stembark contains less than 11% of t o t a l tree N, while stemwood contains 28, 33 and 39% at ages 51, 62 and 73 months, r e s p e c t i v e l y . If branches and leaves are also harvested ( i . e . i f whole tree harvesting i s conducted), the t o t a l P removal would represent about 90% of t o t a l tree P and about 85% of t o t a l tree N, as the root system contains only about 10 and 15% of the t o t a l tree P and N, r e s p e c t i v e l y . S i m i l a r l y , on-site debarking on the poor s i t e at ages older than 56 months would have more ef f e c t on P than on N. However, harvesting stems (wood + bark) on the poor s i t e would remove only 38 and 30% of t o t a l tree P and N, r e s p e c t i v e l y , compared with up to 58% for both P and N on the good s i t e , because the root system on the poor s i t e contains much more P and N than the root system on the good s i t e . The root system contained about 30 and 26% of t o t a l tree P and 30 and 34% of t o t a l tree N, at ages 56 and 67 months, res p e c t i v e l y , on the poor s i t e , compared to 10% for P and 15% for N at ages older than 51 months on the good s i t e . Branches and leaves represented up to 41% of t o t a l tree P and 44% of t o t a l tree N on the poor s i t e , compared with good s i t e values of 40 and 46% at ages older than 51 months. b. Amount of P and N removed per tonne of biomass The amount of P and N removed per tonne of biomass harvested i s given i n Table 27 for d i f f e r e n t levels of u t i l i z a t i o n , for stands older than 30 months. 175 On the good s i t e , there i s a s i g n i f i c a n t decline i n P removed per tonne of stemwood biomass at 51 months for most levels of u t i l i z a t i o n , suggesting that stands should not be harvested much before t h i s age on this s i t e . A decrease i n P removal per unit biomass also occurs between 38 and 51 months with stem (wood + bark) harvesting. However, there i s a s l i g h t increase a f t e r 51 months because there i s a continuous increase i n P content of bark up to 73 months. There i s a general increase i n the amount of N removed per unit stemwood biomass with both stemwood and stem u t i l i z a t i o n l e v e l s a f t e r 51 months because of continuous accumula-tio n of N, even though stemwood growth had s t a b i l i z e d at t h i s age. One possible explanation for t h i s increase i n N i s that N a v a i l a b i l i t y i s probably not l i m i t i n g tree growth (because of the low levels of s o i l P), and therefore N continued to accumulate (luxury consumption) a f t e r growth had s t a b i l i z e d due to P shortage. For whole tree u t i l i z a t i o n , a s i g n i f i c a n t decrease i n N removal i s predicted a f t e r 51 months because of a decrease i n f o l i a g e N concentration. On the poor s i t e , a continuous decline i n P removed per unit bio-mass occurs for a l l u t i l i z a t i o n l e v e l s . However, there was no sharp d i f f e r e n t i a t i o n between 32- and 67-month-old stands, as occurred on the good s i t e . There i s a s l i g h t increase of N removal with age for stem-wood or stem (wood + bark) u t i l i z a t i o n . Older stands should be studied on the poor s i t e In order to e s t a b l i s h whether the observed pattern of change i n the rate of removal of nutrients continues with increasing age, because biomass was s t i l l increasing at 67 months. 176 Differences between the two s i t e s i n P and N removal, per unit of biomass harvested, can be seen i n Figure 13 and Table 27. The stemwood biomass produced per unit of P and N was greater on the poor s i t e as compared to the good s i t e . The r a t i o of stemwood biomass between the good and poor s i t e s was 2.5 at the oldest age studied, while the r a t i o of stemwood P and N was 4.5. If t o t a l tree harvesting were to be adopted, these r a t i o s would be 2.0, 3.3 and 2.2 for biomass, P and N, re s p e c t i v e l y . Hence, the removal of P and N per unit biomass w i l l be smaller on the poor s i t e as compared to the good s i t e . This i s e s p e c i a l l y important for the management of s o i l s with low P such as the "cerrado" s o i l s . c. E f f e c t s of harvesting on s o i l nutrient status The e f f e c t s of intensive management on s o i l nutrient status i s usually evaluated by comparing the amount of nutrients removed i n harvested materials with t o t a l (Hingston et a l . , 1979; Wise and Pitman, 1981) or available s o i l nutrients (Hingston et a l . , 1979; R u s s e l l , 1983). As discussed i n Chapter 5, there i s a l o t of controversy con-cerning whether t o t a l or available P should be used for such s t a t i c inventory evaluations, or i f a more dynamic approach should be adopted. Phosphorus i s considered to be the major l i m i t i n g factor for plant growth on "cerrado" s o i l s ( l a t o s o l s ) . Because of the high P f i x a t i o n capacity of these s o i l s , the rate of replenishment of P from the s o i l f o r plant growth i s very low. Consequently, the use of t o t a l P w i l l probably overestimate the size of the P s o i l pool, espe c i a l l y i f short rotations are considered. On the other hand, the methods used to measure s o i l a v a i l a b l e P might not extract a l l the P that plants are 177 able to extract from s o i l , so i t i s possible that s o i l P could i n some cases be underestimated. In the present study, the amount of nutrients contained i n unfer-t i l i z e d 24-month-old E. grandis stands w i l l be used as a bioassay of the a v a i l a b i l i t y of P and N f o r plant growth. It was f e l t that the quantity of P a c t u a l l y sequestered by the plants i s the best measure of P a v a i l a -b i l i t y that could be obtained i n t h i s study. Only data f o r a 24-month-old stand were a v a i l a b l e . It i s recognized that t h i s w i l l probably underestimate s i t e a v a ilable P because under conditions of extreme P deficiency, the stand may not have produced enough f i n e root biomass to f u l l y e x p l o i t the s o i l by this age. In the absence of data on the f i n e root biomass of the s i t e , the degree of s i t e occupancy i s not known. For purposes of c a l c u l a t i o n , t h i s problem must therefore be ignored. The values obtained were: 1.4 and 0.4 kg/ha of P and 27 and 7 kg/ha of N, on the good and poor s i t e s , r e s p e c t i v e l y , over a period of 24 months. Assuming that these plants continue to grow slowly and at a constant rate, the amount that could be available for plant uptake (from s o i l or other sources, such as atmospheric inputs) per month was used to obtain P and N a v a i l a b i l i t y over a period of 18 years. The amount of f e r t i l i z e r added at planting time was 30.5 kg/ha of - P and 25 kg/ha of N. At present, no additional f e r t i l i z e r i s being applied to subsequent rotations when plantations are managed by coppice. Assuming that a l l the f e r t i l i z e r applied w i l l be available for plant uptake over the 18 year period, the t o t a l quantities of nutrients a v a i l -able from the s o i l plus the f e r t i l i z e r would be 43 and 34 kg/ha of P and 268 and 88 kg/ha of N on the good and poor s i t e s , r e s p e c t i v e l y . 178 In order to assess long-term effects of shortening rotations and increasing l e v e l of u t i l i z a t i o n of E_^_ grandis on the "cerrado" s o i l nutrient status, Table 28 was prepared for both s i t e s for a period of 18 years for three r o t a t i o n ages and three levels of u t i l i z a t i o n based on inventory data presented i n Chapter 2. Feedback between nutrient a v a i l a b i l i t y and productivity i n subsequent rotations was not consider-ed. It was assumed that the productivity would be the same i n subse-quent rotations, and that nutrient concentrations of tree components would remain the same. While these are not reasonable assumptions, they have frequently been made i n assessing the impacts of i n t e n s i f y i n g management on s o i l nutrient status, and are made here i n order to permit a comparison to be made between the r e s u l t s of this and a more dynamic approach. On the good s i t e , when stems (wood + bark) are harvested the best harvesting age i s predicted to be about 54 months because there i s a greater biomass production and the amount of P removed i n the bark i s smaller than for other r o t a t i o n ages. By harvesting only stemwood, longer rotations (72 months) are shown to be more appropriate i n order to minimize P removal. However, the predicted biomass harvest i s much lower compared to a 54-month ro t a t i o n . By harvesting stems (wood + bark) over the 18 year period, without a d d i t i o n a l f e r t i l i z e r , there i s l i k e l y to be a reduction i n productivity because the amount of P removed would be greater than 40 kg/ha while a v a i l a b l e P for the 18 year period i s assumed to be only 43 kg/ha (assuming that a l l P f e r t i l i z e r applied i s a v ailable for plant uptake). Whole tree harvesting would not be possible unless more f e r t i l i z e r i s supplied i n subsequent rota t i o n s . TABLE 28. A summary of biomass (t/ha) and P and N (kg/ha) removal over 18 years at three levels of utilization and rotation length in E. grandis plantations (managed by coppice) growing in the "cerrado" region, Minas Gerais, Brazil. Levels of u t i l i z a t i o n Age at Stemwood Stem (wood + bark) Whole tree harvest No. of j j- ^ (months) rotations Biomass P N Biomass P N Biomass P N (a) Bom Despacho (good site) 36 6 140 27.2 253 144 45.2 344 272 86.0 1619 54 4 247 27.3 411 258 41.5 566 348 94.6 1231 72 3 183 19.5 437 215 44.3 547 252 72.1 956 Carbonita (poor site) 36 6 72 5.4 97 91 9.8 150 150 36.3 661 54 4 74 4.9 102 91 9.2 155 110 17.0 417 72 3 66 4.7 106 83 8.9 157 110 19.6 355 Fert i l i z e r added at planting time was equivalent to: 30.5 kg/ha of P and 25 kg/ha of N. Nutrient available from soil over 18 years period were 12.6 and 3.6 kg/ha of P and, 243 and 63 kg/ha of N on the good and poor sites, respectively (see text for explanation on method to calculate nutrient availability). 1 For purposes of these calculations, i t is assumed that biomass yield remains constant in successive rotations (this assumption is tested in Table 30). 180 On the poor s i t e , the predicted proportion of available s o i l P removed (P removed/total available P) through harvesting i s very high for a 36-month r o t a t i o n . The proportion of P removed decreases s i g n i f i -cantly with age when whole tree harvesting i s adopted; t h i s decrease i s less s i g n i f i c a n t for stemwood and stem harvesting. Because stemwood biomass harvested over the 18 year period i s less when a 72 month ro t a t i o n i s adopted, the P removed per unit biomass i s much lower than by harvesting at 54 months. P removed through whole tree harvesting constitutes about 60% of a v a i l a b l e P when 54 and 72-month rotations are adopted, while harvesting stems only removes an amount equivalent to 30% of the a v a i l a b l e P, over the 18 year period. According to the s t a t i c inventory assessment there was an excess of P i n the s o i l ( a l l f e r t i l i z e r P i s assumed to be a v a i l a b l e ) on the porr s i t e as compared with the t o t a l demand f o r tree growth i n the f i r s t r o t a t i o n . However, there was a slowing down of biomass accumulation at an early age (51 months) i n the f i r s t r o t a t i o n on the good s i t e and the growth rate on the poor s i t e was very low. This suggests that not a l l P added to the system was u t i l i z e d by plants. As discussed i n Chapter 5, the "cerrado" s o i l s have a high P f i x a t i o n capacity that varies with s o i l type, and much of the P added to the s o i l i s probably sorbed by the s o i l p a r t i c l e s . The response of plants to f e r t i l i z a t i o n on these s o i l s w i l l thus depend i n part on the rate of P desorption. Other factors such as the rate of f e r t i l i z e r d i s s o l u t i o n , plant demand at the time f e r t i l i z e r i s released, and e f f i c i e n c y of the root system i n absorbing P are also very Important i n determining the proportion of P that w i l l be recovered by plants. 181 The amount of N harvested i n stemwood on both s i t e s i s much greater than N a v a i l a b l e f or plant growth according to the s t a t i c inven-tory assessment of s i t e nutrient status. In the case of P, atmospheric inputs and leaching are very low. P already present in the s o i l i s very low and w i l l be released slowly; probably being immediately absorbed by plants or, l e s s l i k e l y , remaining i n the s o i l s o l u t i o n . However, N that i s not taken up may be leached. By using the u n f e r t i l i z e d 24-month-old stand to assess N a v a i l a b i l i t y there i s i n fact an underestimation of N a v a i l a b i l i t y : i f plant growth i s limited due to lack of P, the absorp-t i o n of N w i l l be low and the excess w i l l be leached. One of the major problems with t h i s s t a t i c inventory assessment of nutrient removal through harvesting over the 18 year period (Table 28) i s the assumption of unchanging biomass y i e l d . Productivity w i l l be overestimated i f nutrient withdrawals lead to s o i l impoverishment, and t h i s i n turn w i l l overestimate nutrient withdrawals. Because planta-tions in the "cerrado" region have started only recently, information on coppice prod u c t i v i t y of subsequent rotations on such poor s o i l s i s i n -complete. However, the r e s u l t s of the f i e l d experiment (which examined the early stages of sprout growth; Chapter 4), showed that sprout growth declined s i g n i f i c a n t l y a f t e r about 20 months. This may have occurred because of either rootstock vigour decline or nutrient depletion, or both. Irrespective of the reason for the decline, i t occurred, and i t therefore seems l i k e l y that the productivity of successive rotations w i l l also decline, thus reducing the export of P and N i n harvested material. 182 6.2.2 Assessment Based on Both Inventory and Nutrient Dynamics Data The P budget for growth of E^ _ grandis over two 54 month rotations under two d i f f e r e n t i n t e n s i t i e s of management i s given i n Table 29. This table i s comparable to Table 28, except that i t was developed using a dynamic rather than a s t a t i c assessment. Phosphorus was selected for t h i s purpose because i t i s believed to be the most l i m i t i n g f a c t o r for eucalypt growth i n the "cerrado" region. The r o t a t i o n age of 54 months was selected because i t i s l i k e l y that harvesting would be done at about t h i s age, based on r e s u l t s already presented. The P demand by tree growth i n the f i r s t r o t a t i o n was supplied from: (a) P present i n the u n f e r t i l i z e d s o i l , the a v a i l a b i l i t y of which was calculated based on a bioassay (as already described; page 177) and, (b) P released from the P f e r t i l i z e r added at planting time (30.5 kg/ha). The P demand for growth i n the second r o t a t i o n would be supplied from the following sources. (a) M i n e r a l i z a t i o n of the forest f l o o r and logging slash. The rates of decomposition presented i n section 3.7.3 were used to determine the proportion of P i n these compartments that would be available within a r o t a t i o n of given length. The decomposition rate of slash was assumed to be the same as that for senesced leaves and branches because there was no information for decomposition of fresh leaves and branches of E. grandis. I t was assumed that a l l P released from l i t t e r 183 T A B L E 2 9 : T r e e g r o w t h P b u d g e t ( k g / h a ) o v e r t w o 5 4 m o n t h r o t a t i o n s , a t t w o l e v e l s o f u t i l i z a t i o n I n E _ . g r a n d i s p l a n t a t i o n s ( m a n a g e d b y c o p p i c e ) g r o w i n g I n t h e " c e r r a d o " r e g i o n , M i n a s G e r a i s , B r a z i l . B o m D e s p a c h o ( g o o d s i t e ) C a r b o n i t a ( p o o r s i t e ) F i r s t R o t a t i o n D e m a n d L i v e t r e e s D e a d t r e e s L i t t e r f a l l T o t a l S o u r c e o f P a b s o r b e d U n f e r t i l i z e d s o i l 1 F e r t i l i z e r 2 R e m o v a l t h r o u g h h a r v e s t i n g S t e m o n l y ( w o o d + b a r k ) W h o l e t r e e P A v a i l a b l e f o r S e c o n d R o t a t i o n F o r e s t f l o o r a n d s l a s h : S t e m o n l y h a r v e s t W h o l e t r e e h a r v e s t 1 3 U n f e r t i l i z e d s o i l 1 R e s i d u a l f e r t i l i z e r ' R o o t r e s e r v e s T o t a l 4 S t e m o n l y h a r v e s t W h o l e t r e e h a r v e s t 2 6 . 8 2 . 7 4 . 1 3 3 . 6 3 . 2 3 0 . 4 ( 1 0 0 ) 1 2 . 8 2 3 . 6 1 1 . 8 3 . 3 3 . 2 0 . 0 3 . 2 1 8 . 2 ( 4 6 ) 9 . 7 ( 7 1 ) 6 . 1 0 . 6 3 . 9 1 0 . 6 0 . 9 9 . 7 ( 3 2 ) 2 . 3 4 . 2 5 . 8 1 . 7 0 . 9 6 . 7 1 . 8 1 5 . 2 1 1 . 1 ( 1 4 3 ) ( 1 0 5 ) D e t e r m i n e d b a s e d o n b i o a s s a y ( s e e t e x t f o r e x p l a n a t i o n o f m e t h o d ) . N u m b e r s w i t h i n b r a c k e t s a r e p e r c e n t a g e o f f e r t i l i z e r a d d e d a t p l a n t i n g t i m e t h a t w a s r e c o v e r e d i n t h e f i r s t r o t a t i o n ( P a d d e d - 3 0 . 5 k g / h a ) . P r o p o r t i o n o f r e m a i n i n g f e r t i l i z e r t h a t w o u l d b e a v a i l a b l e f o r p l a n t g r o w t h ( t h e e f f i c i e n c y o f f e r t i l i z e r r e c o v e r y w a s a s s u m e d t o b e t h e s a m e a s f o r t h e f i r s t r o t a t i o n ) . N u m 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 t h e r e d u c t i o n ( 5 ) o f P a v a i l a b i l i t y f o r t h e s e c o n d r o t a t i o n s a s c o m p a r e d t o t h e t o t a l d e m a n d f o r t h e f i r s t r o t a t i o n . 184 decomposition i s taken up by plants. This may be an over-estimate since some of the mineralized P may be sorbed by the mineral s o i l . The extent of sorption of t h i s P w i l l depend p a r t l y on the extent of f i n e root mortality a f t e r logging. The analysis of the contribution of nutrients returned through l i t t e r f a l l i s very complex because each l i t t e r com-ponent has a d i f f e r e n t s i t e - s p e c i f i c rate of decomposition (see Chapter 3) and the forest f l o o r consists of l i t t e r of d i f f e r e n t age c l a s s . The amount of nutrients released from decomposing l i t t e r w i l l not n e c e s s a r i l y be i n synchrony with plant demand. Both the t o t a l P mineralization rate and the tree uptake demand w i l l vary over the r o t a t i o n . Such complex interactions can only be f u l l y analysed through a computer simulation model. (b) Desorption and weathering i n the mineral s o i l . The bioassay method was used to determine the a v a i l a b i l i t y of P from u n f e r t i l i z e d s o i l f o r plant growth as discussed e a r l i e r . (c) Residual f e r t i l i z e r . A t o t a l of 30.5 kg/ha of P was added at planting time. A proportion of this was recovered in the f i r s t r o t a t i o n . Some proportion of the remaining f e r t i l i z e r would be available for plant growth i n the second rotation; t h i s proportion was assumed to be equal to the proportion of f e r t i l i z e r taken up by trees i n the f i r s t r o t a t i o n . For example, 9.7 kg/ha of P were taken up over 54 months in the f i r s t r o t a t i o n on the poor s i t e (32% of 30.5 kg/ha). The 185 remaining P f e r t i l i z e r i n the s o i l amounted to 20.8 kg/ha. Assuming that 32% of t h i s r e s idual P w i l l be recovered i n the second r o t a t i o n , the P available from the r e s i d u a l f e r t i l i z e r w i l l be 6.7 kg/ha. The desorption of P i s very complex as discussed i n Chapter 5. The a v a i l a b i l i t y of P for plant growth depends on a series of s o i l c h a r a c t e r i s t i c s that can be accounted for by using the P adsorption isotherms. However, the use of the adsorption isotherms i s more appropriate to a computer simu-l a t i o n assessment than to a manual assessment because of the complexity involved. For example, following each addition of P to the s o i l and uptake by plants there w i l l be a new equi-l i b r i u m and d i f f e r e n t rates of P release. The time-trend of P release would thus be d i f f i c u l t to calculate manually. As a consequence, the above assessment of f e r t i l i z e r P a v a i l a -b i l i t y i n the second r o t a t i o n may well be i n e r r o r . Root Reserves. A l l P i n the root system at the end of the f i r s t r o t a t i o n was considered to be available for plant growth i n the second r o t a t i o n . As discussed i n Chapter 4, P reserves i n the established root system may be used for i n i t i a l sprout growth or may remain i n the root system, thereby reducing the demand for P by root production i n subsequent rotations. Loss of P through root mortality a f t e r coppicing was not taken into account i n producing Table 29 because P released through decomposition of dead roots was assumed to be immediately taken up by plants. 186 From Table 29 one can see that without further addition of f e r t i -l i z e r i n the second r o t a t i o n there i s l i k e l y to be a decline In t o t a l p r o d u c t i v i t y on the good s i t e . Using r o t a t i o n lengths of 54 months the t o t a l p roductivity would be reduced by 46% as compared to the f i r s t r o t a t i o n productivity for stem harvesting, assuming that plant t i s s u e P concentrations remain the same i n the second r o t a t i o n ( a c t u a l l y , they would probably drop somewhat). The adoption of whole tree harvesting would reduce the productivity by 71% i n the second r o t a t i o n . This reduction might be even greater i f we consider aboveground biomass only, because a reduction i n P a v a i l a b i l i t y might cause the root:shoot r a t i o to increase to adapt to the new l e v e l of P a v a i l a b i l i t y . On the other hand, the nutrient a v a i l a b i l i t y from u n f e r t i l i z e d s o i l might be greater than that reported for the s i t e based on bioassay. U n f e r t i l i z e d trees might not f u l l y occupy the s i t e , and therefore not absorb a l l s o i l a v a i l a b l e P. In addition, the l i t t e r f a l l would p a r t i a l l y decompose i n the f i r s t r o t a t i o n increasing P a v a i l a b i l i t y . According to t h i s method of assessment, there should be s u f f i c i e n t P a v a i l a b l e on the poor s i t e a f t e r the f i r s t r o t a t i o n to supply the tree growth demand for the second r o t a t i o n . This i s very d i f f e r e n t from the s i t u a t i o n f o r the good s i t e . Part of the difference might be explained by the fact that when the s i t e i s very poor there are much lower harvest outputs, which reduces the rate of s i t e depletion. In addition, the amount of P required to produce one unit of biomass on the the poor s i t e i s less than that required on the good s i t e (Figure 13 and Table 27), so that less P i s removed per unit of harvested biomass. Also, the root systems already established on the poor s i t e contain a greater propor-187 t i o n of t o t a l tree P as compared to those on the good s i t e which gives an advantage to sprout growth on the poor s i t e r e l a t i v e to that on the good s i t e . However, a major reason for t h i s apparent anomaly may be the method used to evaluate P a v a i l a b i l i t y from the residual f e r t i l i z e r for the second r o t a t i o n . The proportional uptake of P from f e r t i l i z e r i n the second r o t a t i o n on the poor s i t e w i l l probably be lower than that of the f i r s t r o t a t i o n whereas i n Table 29 i t i s assumed to remain the same. The s o i l on the poor s i t e has a higher P f i x a t i o n capacity than that on the good s i t e (based on the e f f i c i e n c y of recovery of f e r t i l i z e r P i n the f i r s t r o t a t i o n ) . This high f i x a t i o n capacity, which i s confirmed by the P sorption curves for the two s i t e s (Appendix 8) would hold the r e s i d u a l f e r t i l i z e r more fi r m l y than on the good s i t e . Other possible factors that are not considered i n Table 29 but which could prevent the predicted increased production i n subsequent rotations include other nutrients which might become l i m i t i n g i n the second r o t a t i o n on the poor s i t e , and the monthly d i s t r i b u t i o n of p r e c i p i t a t i o n which may be some-what more l i m i t i n g on the poor s i t e than on the good s i t e . I f p roductivity i n the second r o t a t i o n on the good s i t e , i s to be maintained at the same l e v e l as that of the f i r s t r o t a t i o n , t h i s assess-ment suggests that the following amount of P f e r t i l i z e r would be required: about 15 and 24 kg/ha when using stem and whole tree harvest-ing, r e s p e c t i v e l y , assuming the same recovery of f e r t i l i z e r observed i n the f i r s t r o t a t i o n . Table 30 compares the 18-year biomass y i e l d predictions produced by the s t a t i c and dynamic approaches. In order to determine the produc-t i v i t y of coppice i n the t h i r d and fourth r o t a t i o n the percentage change 188 TABLE 30. Summary of the predicted biomass y i e l d (t/ha) of E_. grandis plantations growing i n the "cerrado" region over an 18 year period, for a 54 month r o t a t i o n length, under two levels of u t i l i z a t i o n , produced by the s t a t i c and the dynamic approaches Levels of U t i l i z a t i o n Stem (wood + bark) Whole tree Approach of y i e l d assessment Rotation S t a t i c Dynamic S t a t i c Dynamic (a) Bom Despacho (good s i t e ) 1 65 65 87 87 2 65 35 87 25 3 65 21 87 13 4 65 12 87 9 T o t a l 258 133 (-48) 1 348 134 (-61) (b) Carbonita (poor s i t e ) 1 23 23 28 28 2 23 33 28 29 3 23 35 28 24 4 23 33 28 18 Total 92 124 (+35) 112 99 (-12) Numbers within brackets represent change i n productivity i n percentage over an 18 year period. 189 i n t o t a l P for each u t i l i z a t i o n l e v e l was used to determine the amount of P observed i n the previous rotations i n the forest f l o o r and slash, and i n the root system. It was assumed that P i n those components would change i n proportion to the change i n t o t a l P available to supply the demand for plant growth. For example, there was a 46% reduction i n t o t a l a v a ilable P for the second r o t a t i o n by using stem harvest, the pro d u c t i v i t y of the next r o t a t i o n was assumed to be reduced by 46%. Consequently, nutrient returned through forest f l o o r and slash at the end of the second r o t a t i o n would also be reduced i n the same proportion. The amount of P absorbed from s o i l and f e r t i l i z e r was determined using the approach already discussed for Table 29. Biomass harvested under the two u t i l i z a t i o n l e v e l s i n each r o t a t i o n was calculated based on the predicted amount of P available for that r o t a t i o n assuming that P concentrations i n plant tissue would remain the same. On the good s i t e , there i s a predicted reduction i n p r o d u c t i v i t y of 48% with stems-only harvest, while whole tree harvest leads to a 61% decrease over an 18 year period. In contrast, there was a predicted Increase i n productivity of 35% on the poor s i t e , when only the stems were harvested. This probably occurred because the harvest output of P i s low (Table 27) and the amount of P remaining i n the root system i s large compared to that on the good s i t e . In addition, the proportion of t o t a l P (required to supply demand for the plant growth) remaining i n the s o i l i s greater than that for the good s i t e . Consequently, the s i t e i s able to sustain the harvest losses of P and even increase i n produc-t i v i t y . However, i f the slash i s also harvested (as i n whole tree u t i l i z a t i o n ) , there i s a predicted decrease i n productivity of 12% over 190 the 18 years. However, i t should be noted that the f e r t i l i z e r P available i n each successive r o t a t i o n might be much lower than that being considered i n the present study, because P f i x a t i o n capacity i s very high. 6.3 Conclusions By using information on both accumulation and dynamics of biomass and nutrients over age sequences of E. grandis growing on two d i f f e r e n t "cerrado" s i t e s , the following conclusions can be reached. 1. Based on data c o l l e c t e d i n the f i e l d i t was observed that on the good s i t e , P removal per unit biomass had decreased a f t e r age 51 months and remained constant thereafter, i r r e s p e c t i v e of u t i l i z a t i o n l e v e l . Therefore i t appears that there would probably be l i t t l e change i n the impact of harvesting on P removed i n stands varying i n age from 51 to 73 months. This i s not true for N because stemwood N i s s t i l l increasing by age 73 months. However, the decision should preferably be made based on the re s u l t s of P removal as P i s the most l i m i t i n g factor for plant growth i n "cerrado" s o i l s . On the poor s i t e , a decline i n P removal per unit biomass was observed, with increasing age i r r e s p e c t i v e of u t i l i z a t i o n l e v e l . However, as occurred on the good s i t e , N removal per unit stemwood or stem (wood + bark) biomass increased up to 67 months. There was a decline i n N removed per unit of biomass with increasing age for t o t a l aboveground biomass harvest because there was a decreasing contribution of fol i a g e with increasing age. 191 2. By using a s t a t i c assessment of the effects of intensive management on s i t e nutrient status (assuming that every r o t a t i o n would have the same p r o d u c t i v i t y ) , i t was predicted that harvesting at age 54 months would give a greater biomass production over a period of 18 years on both s i t e s than shorter or longer r o t a t i o n s . However, feedback between nutrient a v a i l a b i l i t y and p r o d u c t i v i t y of subse-quent rotations i s not considered i n t h i s method. Therefore, the s t a t i c approach w i l l often lead to misinterpretation. For example, the decline i n productivity i n successive rotations might be lower using longer rotations and consequently biomass production with longer rotations would be greater than when shorter rotations are used. By using t h i s s t a t i c approach i t was predicted that for the good s i t e stemwood harvest would remove a l l the P being released from the s o i l and from the f e r t i l i z e r applied i n the f i r s t r o tation, over a period of 18 years. For the poor s i t e , i t was predicted that even with whole tree harvest, P removal would not deplete available P ( s o i l + f e r t i l i z e r ) . 3. The use of a more dynamic approach that considered feedback between P a v a i l a b i l i t y and y i e l d predicted that, on the good s i t e , there would be a decrease i n productivity of 48% by using stem harvest while whole tree harvest would lead to a decrease of 61% i n produc-t i v i t y over an 18 year period, using a r o t a t i o n length of 54 months. On the poor s i t e , a decrease i n p r o d u c t i v i t y of 12% was predicted when whole tree harvesting was done. However, by harvesting stems only an increase of 35% i n s i t e productivity was predicted because of a greater amount of P i n the root system and i n the forest f l o o r 192 and slash, as compared to the good s i t e . However, P a v a i l a b i l i t y from f e r t i l i z e r might be smaller than estimated i n t h i s method and the predicted increase i n p r o d u c t i v i t y might not occur. 4. From the data presented f o r both s i t e s studied i t i s evident that branches and leaves represent a large proportion of nutrients required for tree growth (up to 44% of whole tree P and N at ages older than 50 months, on both s i t e s ) . If whole tree harvesting i s adopted, the f e r t i l i z e r requirement i n the second r o t a t i o n would represent approximately 80% of that required i n the f i r s t r o t a t i o n , i n order to maintain the same l e v e l of productivity. This i s about 30% greater than the requirement using stem harvesting. An economic analysis would be needed to i d e n t i f y whether or not whole tree harvesting plus heavy f e r t i l i z a t i o n i s less expensive than stemwood harvesting with less f e r t i l i z a t i o n . 5. The amount of P required to produce one unit of biomass on the poor s i t e i s l e s s than that required on the good s i t e . In addition, the root systems already established on the poor s i t e contain a greater proportion of t o t a l tree P as compared to those on the good s i t e . This gives an advantage to sprout growth on the poor s i t e . Also, there seems to be a reasonable balance between inputs and harvest outputs on the poor s i t e . If t h i s i s the case, intensive biomass harvesting on the poor s i t e would have less impact on future s i t e p r o d u c t i v i t y compared to that on the good s i t e . However, the good s i t e can be expected to y i e l d a greater biomass than the poor s i t e . 193 6. The data on accumulation and dynamics of biomass and P and N was shown to be helpful in understanding the effects of intensive forest management on site nutrient status. However, the evaluation of long-term effects is rather complex because of the interaction of several factors when different management conditions are considered. Therefore, computer simulation models are required to evaluate the long-term effects of tree harvesting on site nutrient status. 7. The growth strategies of a tree growing from seed are different from that of a tree developing from a stump and this was shown to influence the nutrient balance in subsequent rotations. In addition, the dynamics of P in mineral soils (the most limiting nutrient for eucalypt growth in the "cerrado" region) differ from that of N. Consequently, any model to be used to evaluate the long-term effects of eucalypt management on nutrient site status should include these two aspects of eucalypt management. 194 REFERENCES Aber, J.D. and J.M. M e l i l l o . 1982. FORTNITE: a computer model of organic matter and nitrogen dynamics i n forest ecosystems. College of A g r i c u l t u r a l and L i f e Sciences Research B u l l e t i n R3130, 49 pp. Agren, G.I. 1983a. The concept of nitrogen productivity i n forest growth modelling. M i t t , der F o r s t l . Bundesversuchanstalt, Wien. 147:199-210. Agren, G.I. 1983b. The nitrogen productivity of some co n i f e r s . Can. J . For. Res. 13:494-500. Alvim, P.T. and J.E. S i l v a . 1980. Comparacao entre os cerrados e a regiao amazonica em termos agroecoldgicos. pp. 141-160. In: Simp6sio sobre o cerrado, V, 1979. E d i t e r r a , B r a s i l i a . Anderson, H.W. 1979. Poplar farming. In: Fayle, D.C.F. and L. Zsuffa (eds.), Poplar Research, Management and U t i l i z a t i o n i n Canada. Ontario M i n i s t r y of Natural Resources, For. Res. Info. Pap. No. 102. 3:1-9 Anderson, H.W., C S . Papadopol and L. Zsuffa. 1983. Wood energy plantations i n temperate climates. For. E c o l . Manage. 6(3):281-306. Ashton, D.H. 1975. Studies of l i t t e r i n Eucalyptus regnans f o r e s t s . Aust. J . Bot. 23(3):413-433. Ashton, D.H. 1976. Phosphorus i n forest ecosystems at Beenak, V i c t o r i a . J . E c o l . 64:171-186. A t t i w i l l , P.M. 1968. The loss of elements from decomposing l i t t e r . Ecology 49:142-145. A t t i w i l l , P.M. 1979. Nutrient c y c l i n g i n a Eucalyptus obliqua (L'Herit) f o r e s t . I I I . Growth, biomass and net primary produc-t i o n . Aust. J . Bot. 27(4):439-458. A t t i w i l l , P.M. 1980. Nutrient c y c l i n g i n a Eucalyptus obliqua (L'Herit) f o r e s t . IV. Nutrient uptake and nutrient return. Aust. J . Bot. 28(2):199-222. A t t i w i l l , P.M., H.B. Guthrie, and R. Leuning. 1978. Nutrient c y c l i n g i n a Eucalyptus obliqua (L'Herit) f o r e s t . I. L i t t e r production and nutrient return. Aust. J . Bot. 26(1):79-91. Ayodele, 0. J . , A. Agboola and R.A. Sobulo. 1982. Evaluation of P sorption capacity and P extractants for assessing P requirements of maize i n Western Nigeria savannah s o i l s . Commun. i n S o i l S c i . Plant Anal. 13(4):295-307. 195 Azevedo, L.G. and R.L. Caser. 1980. Regionalizacao do cerrado. pp. 213-229. In: Simpdsio sobre o Cerrado, 5o., 1979. E d i t e r r a , Bras I l i a . Bahia F i l h o , A.F.C. and J.M. Braga. 1975. Fosfo ro em l a tos solos do Estado de Minas Gerais. I I I . indices de disponibilidade de fdsforo e crescimento vegetal. Experientiae, 20(8):217-234. Baker, T.G. 1983. Dry matter, nitrogen, and phosphorus content of l i t t e r f a l l and b r a n c h f a l l i n Pinus radiata and Eucalyptus f o r e s t s . N. Z. J . For. S c i . 13(2):205-221. Ba l l a r d , R. 1978. E f f e c t of f i r s t r o t a t i o n phosphorus applications on f e r t i l i z e r requirements of second r o t a t i o n radiata pine. N. Z. J . For. S c i . 8(1):135-145. Barrichelo, L.E.G. and J.O. B r i t o . 1982. As madeiras de essencias nativas e a producao de celulose. S i l v i c u l t u r a , 7(27):32-36. Barros, N.F. 1974. Contribuicao ao relacionamento de c a r a c t e r i s t i c a s pedologicas e topograficas com a l t u r a de Eucalyptus alba na regiao de Santa Barbara, MG. M.S. Thesis, Universidade Federal de Vicosa, Vicosa. 88 pp. Barros, N.F., J.M. Braga, R.M. Brandi and B.V. Defelipo. 1981. Producao de eucalipto em solos de cerrados em resposta a aplicacao de NPK e de B e Zn. Revista Arvore, 5(1):90-103. Barrow, N.J. 1983a. A mechanistic model for describing the sorption and desorption of phosphate by s o i l . J . S o i l S c i . 34(4):733-750. Barrow, N.J. 1983b. On the r e v e r s i b i l i t y of phosphate sorption by s o i l s . J . S o i l S c i . 34(4):751-758. B a s k e r v i l l e , G.L. 1972. Use of logarithmic regression i n the estima-t i o n of plant biomass. Can. J . For. Res. 2:49-53. Beckwith, R.S. 1965. Sorbed phosphate at standard supernatant concen-t r a t i o n as an estimate of the phosphate needs of s o i l s . Aust. J . Exp. Agr. and An. Hus. 5(16):52-58. Bednall, B.H. 1968. The problem of lower volumes associated with second rotations i n Pinus radiata plantations i n South A u s t r a l i a . Woods and Forests Dept. B u l l No. 17. 12 pp. Be l l o t e , A.F.J., J.R. Sarruge, H.P. Haag and G.D. O l i v e i r a . 1980. Extracao e exportacao de nutrientes pelo Eucalyptus grandis. H i l l , ex-Maiden em funcao da idade. 1. Macronutrientes. IPEF (20):l-23. Bevege, D.I. 1978. BiOmass and nutrient d i s t r i b u t i o n i n indigenous forest ecosystems. Qld. Dep. For. Techn. Pap. no. 6, 20 pp. 196 Bhatnagar, H.P. and D.N. J o s h i . 1973. Vegetative propagation of Eucalyptus t e r e t i c o r n i s Sm. (mysore gum) lignotubers. Indian Forester, 99(7):508-9. Birk, E.M. and R.W. Simpson. 1980. Steady state and the continuous input model of l i t t e r accumulation and decomposition i n A u s t r a l i a n eucalypt f o r e s t s . Ecology, 61(3):481-5. Black, CA. (ed.). 1965. Methods of s o i l a n a l y s i s . Part 2. Chemical and microbiological properties. American Society of Agronomy, Madison, pp. 771-1572. Blake, T.J. 1972. Studies on the lignotubers of Eucalyptus obliqua L'Herit. I I I . The effects of seasonal and n u t r i t i o n a l factors of dormant bud development. New Phytol. 71:327-334. Blake, T.J. 1980. E f f e c t s of coppicing on growth rates, stomatal c h a r a c t e r i s t i c s and water relations in Eucalyptus camaldulensis Dehn. Aust. J . Plant P h y s i o l . 7:81-84 Blake, T.J. 1981. Growth related problems of aging and senescence i n fast growing trees grown on short r o t a t i o n s . IEA Report NE 1891:21, National Swedish Bo for Energy Source Development, Stockholm. 43 pp. Borkert, CM. and S.A. Barber. 1983. E f f e c t of supplying P to a portion of the soybean root system on root growth and P uptake k i n e t i c s . J . PI. N u t r i t i o n , 6(10):895-910. Bradstock, R. 1981. Biomass i n an age series of Eucalyptus grandis plantations. Aust. For. Res. 11:111-127. Braga, J.M. and B.V. Defelipo. 1974. Determinacao espectrofotometrica de fosforo em extratos de solos e plantas. Revista Ceres, 21:73-85. Brandao, L.G. and S.C L u p a t t e l l i (eds.). 1982. B r a s i l f l o r e s t a l : ano 2000; d i r e t r i z e s estrate'gicas para o setor f l o r e s t a l . B r a s i l i a , 77 pp. B r i x , H. 1983. E f f e c t s of thinning and nitrogen f e r t i l i z a t i o n on growth rate of Douglas-fir: r e l a t i v e contribution of f o l i a g e quantity and e f f i c i e n c y . Can. J . For. Res. 13:167-175. Camargo, A.P. 1963. Clima do cerrado. pp. 75-95. In: Simposio sobre o cerrado, I. Universidade de Sao Paulo, Sao Paulo. Camargo, A.P.; R.R. A l f o n s i ; H.S. Pinto and J.V. C h i a r i n i . 1976. Zoneamento da aptidao climatica para culturas comerciais em areas de cerrado. pp. 89-120. In: Simposio sobre o Cerrado, 4o., 1976. Universidade de Sao Paulo, Sao Paulo. 197 Canada S o i l Survey Committee, Subcommittee on s o i l c l a s s i f i c a t i o n . 1978. The Canadian system of s o i l c l a s s i f i c a t i o n . Can. Dep. Ag r i c . Publ. 1646. 164 pp. Carpanezzi, F. 1980. Deposicao de material organico e nutrientes em uma f l o r e s t a natural e em uma plantacao de eucaliptos no i n t e r i o r do Estado de Sao Paulo. M.S. Thesis, Piracicaba, ESALQ, 107 pp. Carter, W.G. 1974. Growing and harvesting eucalypts on short rotations f o r pulping. Aust. For. 36(1):214-225. Chang, S.C. and M.L. Jackson. 1957. Fractionation of s o i l phosphorus. S o i l S c i . 84:133-144. Chattaway, M.M. 1958. Bud development and lignotuber formation i n eucalypts. Aust. J . Bot. 6:103-115. Clark, F.B. and F.G. Liming. 1953. Sprouting of black jack oak i n the Missouri Ozarks. US For. Serv., Central States For. Exp. Sta. Tech. Paper 137. 22 pp. Cozzo, D. and Riveros, M.H.C.K. 1971. Registro de l a caida de materia organica y elementos minerales en una plantacion de ocho anos de Eucalyptus camaldulensis. pp. 590-598. In: Congreso F o r e s t a l Argentino, l o . Buenos A i r e s . Crane, W.J.B. 1978. Phosphorus s t a b i l i t y i n eucalypt f o r e s t s . Aust. For. 41(2):118-126. Crane, W.J.B. and R.J. Raison. 1980. Removal of phosphorus i n logs when harvesting Eucalyptus delegatensis and Pinus radiata forests on short and long r o t a t i o n s . Aust. For. 43(4):253-260. Cremer, K.W. 1973. A b i l i t y of Eucalyptus regnans and associated ever-green hardwoods to recover from cutting or complete d e f o l i a t i o n i n di f f e r e n t seasons. Aust. For. Res. 6(2):9-22. Cromer, R.N.; M. Raupach; A.R.P. Clarke and J.N. Cameron. 1976. Eucalypt plantations i n A u s t r a l i a - the po t e n t i a l for intensive production and u t i l i z a t i o n , pp. 29-40. In: Oslo Biomass Studies. University of Maine at Orono, College of L i f e Sciences and Ag r i c u l t u r e . Cromer, R.N. and E.R. Williams. 1982. Biomass and nutrient accumula-t i o n i n a planted E. globulus ( L a b i l l . ) f e r t i l i z e r t r i a l . Aust. J . Bot. 30:265-278. Crow, T.R. 1983. Comparing biomass regressions by s i t e and stand age for red maple. Can. J . For. Res. 13(2):283-288. Daniel, T.W., H.A. Helms and F.S. Baker. 1979. P r i n c i p l e s of s i l v i c u l t u r e . McGraw-Hill, New York. pp. 436-459. 198 Draper, N.R. and H. Smith. 1966. Applied regression a n a l y s i s . John Wiley, New York. 407 pp. Duvigneaud, P. and S. Denayer-De-Smet. 1967. Biomasses, productivity and mineral c y c l i n g i n deciduous mixed forests i n Belgium, pp. 167-186. In: Symposium on Primary Productivity and Mineral Cycling i n Natural Ecosystems. University of Maine at Orono, College of L i f e Sciences and A g r i c u l t u r e . Fagerstrom, T. and U. Lohm. 1977. Growth i n Scots pine (Pinus s y l v e s t r i s L . ) . Mechanism of response to nitrogen. Oecologia, 26:305-315. FAO. 1981. Eucalypts for planting. FAO Forestry Series, No. 11. 677 pp. F e l l e r , M.G. 1978. Nutrient movement into s o i l s beneath eucalypt and exotic conifer forests i n southern ce n t r a l V i c t o r i a . Aust. J . E c o l . 3:357-372. F e l l e r , M.C. 1979. Nutrient c y c l i n g i n eucalypt forest ecosystems and the effects of pine planting on nutrient c y c l i n g and s o i l f e r t i l i t y . F i n a l report. Faculty of A g r i c u l t u r e and Forestry, University of Melbourne. 237 pp. F e l l e r , M.C. 1980. Biomass and nutrient d i s t r i b u t i o n i n two eucalypt forest ecosystems. Aust. J . E c o l . 5(3):309-333. F e l l e r , M.C. 1984. E f f e c t s of an exotic conifer (Pinus radiata) plantation on forest nutrient c y c l i n g i n southeastern A u s t r a l i a . Forest E c o l . Manage. 7:77-102. F e r r i , M.G. 1977. Ecologia dos cerrados. pp. 15-36. In: Simposio sobre o cerrado, IV: bases para u t i l i z a c a o agropecuaria, 1976. Universidade de Sao Paulo, Sao Paulo. Fox, R.L. and E.J. Kamprath. 1970. Phosphate sorption isotherms for evaluating the phosphate requirements of s o i l s . S o i l S c i . Soc. Amer. Proc. 34(6):902-907. Freedman, B. 1981. Intensive forest harvest: a review of nutrient budget considerations. Maritimes Forest Research Centre, Information Report M-X-121. 78 pp. F r e i t a s , F.G. and CO. S i l v e i r a . 1977. P r i n c i p a l s solos sob vegetacao de cerrado e sua aptidao a g r i c o l a . pp. 155-194. In: Simposio sobre o Cerrado, 5o., 1976. Universidade de Sao Paulo, Sao Paulo. Goedert, W.J. 1980. Uso e manejo dos recursos naturals do cerrado: solo e clima. pp. 473-498. In: Simposio sobre o cerrado, 5o., 1979. E d i t e r r a , B r a s i l i a . 199 G o l f a r i , L. 1967. Con^feras aptas para replobaciones forestales en e l estado de Sao Paulo. S i l v i c u l t u r a em Sao Paulo, vol 6. G o l f a r i , L. 1971. Coniferas aptas para reflorestamento nos Estados do Parana, Santa Catarina e Rio Grande do Sul. I n s t i t u t o B r a s i l e i r o de Desenvolvimento F l o r e s t a l , Boletim Te'cnico, no. 1. G o l f a r i , L. 1975. Zoneamento ecoldgico do Estado de Minas Gerais para reflorestamento. PRODEPEF, Serie Tecnica, 11. 66 pp. G o l f a r i , L. and R.L. Caser. 1977. Zoneamento ecologico da regiao nordeste para experimentacao f l o r e s t a l . PRODEPEF, Serie Tecnica no. 10. 116 pp. G o l f a r i , L., R.L. Caser and V.P.G. Moura. 1978. Zoneamento ecologico esquematico para reflorestamento no B r a s i l . (2a. aproximacao). PRODEPEF, Serie Tecnica, 11. 66 pp. Goodland, R. 1971. Oligotrofismo e alumlnio no cerrado. pp. 44-60. In: Simpo'sio sobre o cerrado, I I I , 1971. Universidade de Sao Paulo, Sao Paulo. Goodland, R. 1979. Analise ecologica da vegetacao do cerrado. pp. 44-60. In: Goodland, R. and M.G. F e r r i (eds.), Ecologia do cerrado, Sao Paulo, I t a t i a i a . Greig, M. and J . B j e r r i n g . 1980. A general least square analysis of variance program. University of B r i t i s h Columbia, Vancouver. 55 pp. Grunwald, C. and R. Karschon. 1974. E f f e c t of seed o r i g i n on coppice regeneration i n Eucalyptus camaldulensis Dehn. Silvae Genet. 23:141-144. Haag, H.P.; J.V. de C. Rocha. G.D. de O l i v e i r a . 1978. Ciclagem de nutrientes em f l o r e s t a s implantadas de Eucalyptus e Pinus. I I . Contribuicao das especies de nutrientes na manta. 0 Solo, 70(2):28-31. Hansen, E.A. and J.B. Baker. 1979. Biomass and nutrient removal i n short r o t a t i o n i n t e n s i v e l y cultured plantations, pp. 130-151. In: Proceedings; Impact of intensive harvesting on forest nutrient c y c l i n g . State University of New York, Syracuse. Heringer, E.P. 1971. Propagacao e sucessao de especies arboreas do cerrado em funpao do fogo, do cupim, da capina e do aldrim ( i n s e t i c i d a ) . pp. 167-179. In: Simposio sobre o cerrado, I I I , 1971. Universidade de Sao Paulo, Sao Paulo. Hingston, J . , A.G. Turton and G.M. Dimmock. 1979. Nutrient d i s t r i b u -t i o n i n K a r r i (Eucalyptus d i v e r s i c o l o r F. Muell) ecosystems i n southwest eastern A u s t r a l i a . Forest E c o l . Manage. 2(2):133-158. 200 Hingston, F.J., G.M. Dimmock and A.G. Turton. 1981. Nutrient d i s t r i b u -t i o n i n a jarrah (Eucalyptus marginata Donn ex Sm.) ecosystem i n southwest western A u s t r a l i a . Forest E c o l . Manage. 3:183-207. Howland, P. 1969. E f f e c t s of s i n g l i n g coppice i n Eucalyptus saligna wood f u e l crops at Muguga, Kenya. E. Af. Agric. For. J . 35(l):66-67. Ingestad, T.; A. Aronsson and G.I. Agren. 1981. Nutrient f l u x density model of mineral n u t r i t i o n i n conifer ecosystems, pp. 61-71. In: Linder, S. (ed.), Understanding and predicting tree growth. Studia F o r e s t a l Suecica Nr 160. Jacobs, M.R. 1955. Growth habits i n eucalypts. Forestry and Timber Bureau, Canberra. 262 pp. Johnson, P.S. 1975. Growth and s t r u c t u r a l development of red oak sprout clumps. For. S c i . 21(4):413-418. Johnson, P.S. 1977. Predicting oak stump sprouting and sprout develop-ment i n the Missouri Ozarks. USDA For. Serv. North Central For. Exp. Sta. Research Paper NC-149. 11 pp. Johnson, P.S. 1979. Shoot elongation of black oak and white oak sprouts. Can. J . For. Res. 9(4):489-494. K a f k a f i , U., A.M. Posner and J.P. Quirk. 1967. Desorption of phosphate from k a o l i n i t e . S o i l S c i . Soc. Am. Proc. 31(3):348-353. Kamprath, E.J. 1977. Phosphorus f i x a t i o n and a v a i l a b i l i t y i n highly weathered s o i l s , pp. 333-347. In: Simposio sobre o Cerrado, 4o., B r a s i l i a , 1976. Belo Horizonte, I t a t i a i a . Kaul, I.N., D.C. Sharma and P.B.L. Srivastava. 1979. D i s t r i b u t i o n of organic matter and plant nutrients i n a s a l (Shorea robusta) coppice plantation. Indian Forester, 105(2):171-179. Kaumi, S.Y.S. 1983. Four rotations of a Eucalyptus f u e l y i e l d t r i a l . Commonw. For Rev. 62(1):19-24 Khanna, P.K. 1981. S o i l analysis for evaluation of forest nutrient supply, pp. 231-38. In: Proceedings, A u s t r a l i a n Forest N u t r i t i o n Workshop, Canberra, 10-14 Aug. 1981. Canberra, CSIRO. Keyes, M.R. and G r i e r , C.C. 1981. Above- and below-ground net produc-t i o n i n 40-year-old Douglas-fir stands on low and high producti-v i t y s i t e s . Can J . For. S c i . 11(3):599-605. Kimmins, J.P. 1977. Evaluation of the consequences for future tree p r o d u c t i v i t y of the loss of nutrients i n whole-tree harvesting. Forest E c o l . Manage. 1:169-183. 201 Kimmins, J.P. and K.A. Scoullar. 1983. F0RCYTE-10. A user's manual (second approximation). Kimmins, J.P. and K.A. Scoullar. 1984. F0RCYTE-11: a f l e x i b l e modelling framework with which to analyse the long-term conse-quences for y i e l d , economic returns and energy e f f i c i e n c y of a l t e r n a t i v e forest and agro-forest crop production s t r a t e g i e s . In: Bioenergy R and D Seminar, 5th, 1984. Proceedings. Nat. Res. Council of Canada, Ottawa (In Press). Kimmins, J.P. and K.A. Scoullar. 1984b. The role of modelling i n tree n u t r i t i o n research and s i t e nutrient management. In: Bowen, G.D. and E.K.S. Nambiar (eds.). N u t r i t i o n of forest trees i n planta-t i o n s . Academic Press, New York (In press). Koerper, G.J. and C.J. Richardson. 1980. Biomass and net annual primary production regressions for Populus grandidentata on three s i t e s i n northern lower Michigan. Can. J . For. Res. 10(1):92-101. Kozak, A. 1970. A simple method to test p a r a l l e l i s m and coincidence for c u r v i l i n e a r , multiple l i n e a r and multiple c u r v i l i n e a r regres-sions. In: 3rd Conference of the Advisory Group of Forest S t a t i s -t i c i a n s , Section 25, IUFRO, Jouy-en-Joses, France, Sept. 7-11. Kramer, P.J. and T.T. Kozlowski. 1960. Physiology of Trees. McGraw-H i l l , New York. 642 pp. Kreutzer, K. 1972. Uber den E i n f l u s s e der Streunutzung auf den Sticks toffhaushalt von Kiefernbestanden (Pinus S i l v e s t r i s L . ) . Forstwissenschftliches C e u t r a l b l a t t , 91:263-270. Lambert, M.J. Inorganic constituents i n wood and bark of New South Wales forest tree species. Beecroft, N.S.W., Forest Commission (Unpublished). L a y z e l l , D.B. and T.A. LaRue. 1982. Modelling C and N transport to developing soybean f r u i t s . Plant P h y s i o l . 70:1290-1298. Lee, D.K. 1979. The influence of the geometry and d i s t r i b u t i o n of root system on coppice regeneration and growth of hybrid poplars. D i s s e r t . Abr. International, B, 39(7):3072B. Lee, K.E. and R.L. C o r r e l l . 1978. L i t t e r f a l l and i t s r e l a t i o n s h i p to nutrient c y c l i n g i n a South A u s t r a l i a n dry s c l e r o p h y l l f o r e s t . Aust. J . E c o l . 3(3):243-252. Lees, J.C. 1981. Three generations of red maple stump sprouts. Canadian Forest Service, Maritimes Forest Research Centre, Information Report No. M-X-119. 9 pp. 202 Lopes, A.S. 1975. A survey of the f e r t i l i t y status of s o i l s under "cerrado" vegetation i n B r a z i l . M.Sc. Thesis, Raleigh, North Carolina State U n i v e r s i t y . 138 pp. Luckhoff, H.A. 1955. The establishment and regeneration of Eucalyptus  saligna plantations i n the coastal belt of Zululand. South A f r . For. J . (22):15-20. Luse, R.A. 1970. The phosphorus cycle i n a t r o p i c a l r a i n f o r e s t . In: Odum, H.T. and R.F. Pigeau (ed.). A t r o p i c a l r a i n f o r e s t . USAEC, 3 v. pp. H-161-McColl, J.G. 1966. Accession and decomposition of l i t t e r i n spotted gum f o r e s t s . Aust. For. 30(2):191-198. MacDonald, J.E. and G.R. Powell. 1983. Relationships between stump sprouting and parent-tree diameter i n sugar maple i n the f i r s t year following c l e a r - c u t t i n g . Can. J . For. Res. 13(3):390-394. Madgwick, H.A.I. 1983. Above-ground weight of forest plots -comparison of seven methods of estimation. N. Z. J . For. S c i . 13(1):100-107. Madgwick, H.A.I., P. Beets and S. Gallagher. 1981. Dry matter accumu-l a t i o n , nutrient and energy content of the aboveground portion of 4-year-old stands of Eucalyptus nitens and Ej_ f a s t i g a t a . N. Z. J . For. S c i . l l ( l ) : 5 3 - 9 , Mann, L.K. Stump diameter and sprout r e l a t i o n s h i p s one year a f t e r c l e a r - c u t t i n g with and without residue removal i n East Tennessee upland hardwoods. (Unpublished). Mello, H.A., J . Mascarenhas Sobrinho, J.W. Simoes and H.T.Z. Couto. 1970. Resultados da aplicacao de f e r t i l i z a n t e s minerals na producao de madeira de Eucalyptus saligna Sm. em solos de cerrado do Estado de Sfo Paulo. IPEF (l):7-26. Moraes, T.S.A. 1982. Controle de cortadeiras na Companhia Agrlcola e F l o r e s t a l Santa Barbara. S i l v i c u l t u r a , 7(25):35-39. Morrison, I . and N.W. Foster. 1979. Biomass and element removal by complete tree harvesting of medium ro t a t i o n forest stands, pp. 111-129. In: Proceedings; Impact of intensive harvesting on forest nutrient c y c l i n g . State University of New York, Syracuse. O'Connell, A.M. and P.M.A. Menage'. 1982. L i t t e r f a l l and nutrient c y c l i n g i n k a r r i (Eucalyptus d i v e r s i c o l o r F. Muell.) f o r e s t i n r e l a t i o n to stand age. Aust. J . E c o l . 7(l):49-62. O'Connell, A.M. and P.M.A. Menage. 1983. Decomposition of l i t t e r from three major plant species of jarrah (Eucalyptus marglnata Donn ex. Sm.) for e s t i n r e l a t i o n to s i t e f i r e h i s t o r y and s o i l type. Aust. J . E c o l . 8(3):277-286. 203 O'Connell, A.M., T.S. Grove and G.M. Dimmock. 1978. Nutrients i n the l i t t e r on jarrah forest s o i l s . Aust. J . E c o l . 3(3):253-260. Okajima, H., H. Kubota and T. Sakuma. 1983. Hysteresis i n the phosphorus sorption and desorption processes of s o i l s . S o i l S c i . Plant Nutr. 29(3):271-283. Ozanne, P.G. and T.C. Shaw. 1968. Advantages of the recently developed phosphate sorption test over the older extractant methods for s o i l phosphate. In: Int. Cong. S o i l S c i . Trans. 9th, Adelaide, Aust. 2:273-280. Paiva, H.N., F. Paula Neto, R.M. Brandi and A.B. Vale. 1983. Influencia das idades de corte e de desbrota e do numero de brotos sobre o desenvolvimento da brotacao de cepas de Eucalyptus spp. Revista Arvore 7(1):1-10. Perala, D.A. 1979. Regeneration and productivity of aspen grown on repeated short r o t a t i o n s . USDA, North Central For. Exp. Sta., For. Serv. Res. Paper NC-176. 7 pp. Pereira, A.R. and R.M. Brandi. 1981. Conducao da brotacao em povoamentos de eucalipto. Sociedade de I n v e s t i g a t e s F l o r e s t a i s , Vicosa, Boletim Tecnico, no. 6. pp. 1-14. Pereira, A.R., A.J. Regazzi, J.C. Ribeiro and L.R. Ramalho. 1980a. E f e i t o do diametro das cepas no desenvolvimento de brotacoes de Eucalyptus spp. Revista A'rvore, 4(2):215-220. Pe r e i r a , A.R., F. Paula Neto and L.R. Ramalho. 1980b. Determinacao do numero dtimo de brotos em brotacoes de Eucalyptus spp. Vicosa, Sociedade de I n v e s t i g a t e s F l o r e s t a i s , Vicosa, Boletim Tecnico, no. 10. pp. 1-5. Pereir a , A.R., E.J. Morais and M.B. Nascimento F i l h o . 1982. Implantacao de f l o r e s t a s de c i c l o s - c u r t o s sob novos modelos de espacamentos. S i l v i c u l t u r a , 8(28):429-432. Poggiani, F., H.T.Z. Couto and W. Suiter F i l h o . 1983. Biomass and nutrient estimates removal i n short r o t a t i o n i n t e n s i v e l y cultured plantation of Eucalyptus grandis. S i l v i c u l t u r a , 8(32):648-651. Poynton, R.J. 1983. The s i l v i c u l t u r a l treatment of eucalypt planta-tions i n southern A f r i c a . S i l v i c u l t u r a , 8(31):603-605. Pressland, A.J. 1982. L i t t e r production and decomposition from an overstory of Eucalyptus spp. on two catchments i n the New England region of New South Wales. Aust. J . E c o l . 7(2):171-180. Raison, R.J., P.K. Khanna and W.J.B. Crane. 1982. E f f e c t s of i n t e n s i -f i e d harvesting on rates of nitrogen and phosphorus removal from Pinus ra d i a t a and Eucalyptus forests i n A u s t r a l i a and New Zealand. N. Z. J . For. S c i . 12(2):394-403. 204 Raitanen, W.E. 1978. A g r i f o r e s t r y i n Eastern Ontario. In: World Forestry Congress, 8th, Jakarta, 16-28 Oct., 1978. Rajan, S.S.S. 1973. Phosphorus adsorption c h a r a c t e r i s t i c s of Hawaiian s o i l s and t h e i r r e l a t i o n s h i p s to equilibrium phosphorus concentra-tions required. Plant and S o i l , 39:519-532. Ranzani, G. 1963. Solos de cerrado. pp. 37-74. In: Simposio sobre o cerrado, I j 1963. Universidade de Sao Paulo, Sao Paulo. Reich, P.B., R.D. Teskey, P.S. Johnson and T.M. Hinckley. 1980. Periodic root and shoot growth i n oak. For. S c i . 26(4):590-598. Rennie, P.J. 1955. The uptake of nutrients by mature forest growth. Plant and S o i l , 7(l):49-95. Reis, A.C.S. 1971. Climatologia dos cerrados. In: Simposio sobre o cerrado, I I I , 1963. Universidade de Sao Paulo, Sao Paulo. Reis, G.G. 1984. Responses of intact and resprouted Eucalyptus camaldulensis Dehn. plants to drought. Ph.D. Thesis, Univ. of C a l i f o r n i a , Riverside. 134 pp. Rezende, G.C., W.S. Suiter F i l h o and C.J. Mendes. 1980. Regeneracao dos macicos f l o r e s t a i s da Cia. A g r l c o l a , e F l o r e s t a l Santa Barbara. Sociedade de I n v e s t i g a t e s F l o r e s t a i s , Vicosa, Boletim Tecnico, No. 1. 24 pp. Rezende, G.C., F.F. Castro, T.S.A. Moraes, C.J. Mendes, W. Suiter F i l h o and J.V. Drummond. 1982a. Explorapao e produtividade de f l o r e s -tas de eucalipto plantadas em espacamentos nao convencionais. S i l v i c u l t u r a , 8(28):594-600. Rezende, G.C., N.F. Barros, T.S.A. Moraes, C.J. Mendes and W. Suiter F i l h o . 1982b. Aplicacao de fosfatos naturals em plantios de Eucalyptus grandis W. H i l l (ex-Maiden). Revista Ceres, 6:84-89. Richards, B.N. and J.L. Charley. 1983. Mineral c y c l i n g processes and system s t a b i l i t y i n the eucalypt f o r e s t . Forest E c o l . Manage. 7(l):31-47. Robertson, J.A. 1980. Use of P sorption parameters to predict f e r t i l i z e r - P requirements. In: Proceedings, Western Canada Phosphate Sumposium, Calgary, March 11-12, 1980. Rocha, D., M.B. Nascimento F i l h o , J.G.R. Magalhaes and J.C. Fernandes. 1982a. E f e i t o do fosfato natural e c a l c a r i o , aplicados apo's o p l a n t i o , no crescimento de Eucalyptus grandis na regiao de Itamarandiba - MG. S i l v i c u l t u r a , 8(28):467-469. 205 Rocha, D., M.B. Nascimento F i l h o , J.C. Fernandes and J.G.R. Magalhaes. 1982b. Estudo de adubacao em Eucalyptus grandis W. H i l l , ex Maiden nos solos de cerrado na regiao de Itamarandiba - MG. S i l v i c u l t u r a , 8(28):470-472. Rocha, D., M.B. Nascimento F i l h o , J.G.R. Magalhaes and J.C. Fernandes. 1982c. Estudo de fontes naturais de fdsforo e c a l c i o na 2a. rotacao de Eucalyptus, na regiao de Itamarandiba - MG. S i l v i c u l t u r a , 8(28):473-475. Rocha, D., J.G.R. Magalhaes, J.C. Fernandes and S.M. Fonseca. 1982d. Contribuicao do t u f i t o , calcareo c a l c i t i c o e dolomitico para o crescimento do Eucalyptus grandis, na regiao de Itamarandiba - MG. S i l v i c u l t u r a , 8(28):476-478. Rogers, R.W. and W.E. Westman. 1977. Seasonal nutrient dynamics of l i t t e r i n a subtropical eucalypt f o r e s t , North Stradbroke Island. Aust. J . Bot. 25(l):47-48. R u s s e l l , E.W. (ed.). 1973. S o i l conditions and plant growth. Longman, London, pp. 555-603. Russ e l l , C.E. 1983. Nutrient c y c l i n g and productivity of native and plantation forests at J a r i F l o r e s t a l , Para, B r a z i l . Ph.D. Thesis, University of Georgia, Athens. Sanchez, P.A. 1976. Properties and Management of S o i l s i n the Tropics. New York, John Wiley. 618 pp. Sander, I.L., P.S. Johnson and R.F. Watt. 1976. A guide for evaluating the adequacy of oak advance reproduction. USDA For. Serv. North Central For. Exp. Sta. St. Paul., General Technical Report NC-23. 6 pp. Santatonio, D., R.K. Hermann and W.S. Overton. 1977. Root biomass studies i n forest ecosystems. Pedobiologia, 17(5):1-31. Sanyal, P. 1975. R e h a b i l i t a t i o n of degraded coppice s a l f o r e s t s v i s - a - v i s conversion to Eucalyptus plantations i n South Bengal - a f i n a n c i a l study. Indian Forester, 101(1):103:108. Satoo, T. and H.A.I. Madgwick. 1982. Forest biomass. The Hague, Martinus N i j h o f f . 151 pp. Schler, G.A. 1981. P h y s i o l o g i c a l research on adventitious shoot development i n aspen roots. US For. Serv. Interm. For. & Rge. Exp. Sta. Gen. Tech. Report INT-107. 12 pp. Schier, G.A. and J.C. Zasada. 1973. Role of carbohydrate reserves i n the development of root suckers i n Populus tremuloides. Can. J . For. Res. 3:243-250. 206 Sharma, R.P. 1979. Production po t e n t i a l and other crop characters of the f i r s t generation coppice of Eucalyptus hybrid. Indian Forester 105(2):89-100. Sheppard, S.C. and G.J. Racz. 1980. Phosphorus n u t r i t i o n of crops as affected by temperature and water supply, pp. 159-199. In: Proceedings of the Western Canada Phosphate Symposium, Calgary, March 11-12, 1980. Simoes, J.W., R.M. Brandi, N.B. L e i t e and E.A. B a l l o n i . 1981. Formacao, manejo e exploracao de f l o r e s t a s com especies de rapido crescimento. B r a s i l i a , I n s t i t u t o B r a s i l e i r o de Desenvolvimento F l o r e s t a l , 131 pp. Singh, R.P. and V.K. Sharma. 1976. Biomass estimation i n f i v e d i f f e r e n t aged plantations of Eucalyptus t e r e t i c o r n i s Smith i n Western Uttar Pradesh, pp. 145-51. In: Oslo Biomass Studies. University of Maine at Orono. Smith, D.M. 1962. The Practice of S i l v i c u l t u r e . 7th ed. John Wiley, New York. pp. 519-532. Smyth, T.J. and P.A. Sanchez. 1980. E f f e c t s of lime, s i l i c a t e , and phosphorus applications to an o x i s o l on phosphorus sorption and ion retention. S o i l S c i . Soc. Amer. J . 44(3):500-505. Spratt, E.D., F.G. Warder, L.D. Bailey and D.W.L. Read. 1980. Measure-ment of f e r t i l i z e r phosphorus residues and i t s u t i l i z a t i o n . S o i l S c i . Soc. Amer. J . 44(6):1200-1204 Stark, N. 1971. Nutrient c y c l i n g . I. Nutrient d i s t r i b u t i o n i n some Amazonian s o i l s . T r o p i c a l Ecology, 12:24-50. Stark, N.M. and Jordan, C.F. 1978. Nutrient retention by the root mat of an Amazonian r a i n f r o e s t . Ecology, 59(3):434-437. Steinbeck, K. and L.C. Nwoboshi. 1980. Rootstock mass of coppiced Platanus o c c i d e n t a l i s as affected by spacing and r o t a t i o n length. For. S c i . 26(4):545-547. Taylor, J.S., T.J. Blake and R.P. Pharis. 1982. The role of plant hormones and carbohydrates i n the growth and s u r v i v a l of coppiced Eucalyptus seedlings. P h y s i o l . Plant. 55:421-430. Tew, R.K. 1970. Root carbohydrates reserves i n vegetative reproduction of Aspen. For. S c i . 16(3):318-320. Tiarks, A.E. 1982. Phosphorus sorption curves for evaluating phosphorus requirements of l o b l o l l y pine (Pinus taeda). Commun. i n S o i l S c i . Plant Anal. 13(8):619-631. 207 T r i t t o n , 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. Turner, J . 1980. Nitrogen and phosphorus d i s t r i b u t i o n i n n a t u r a l l y regenerated Eucalyptus spp. and planted Douglas-fir. Aust. For. Res. 10:289-294. Turner, J . and M.J. Lambert. 1983. Nitrogen cy c l i n g within a 27-year-old Eucalyptus grandis stand. For. E c o l . Manage. 6:155-168. Upadhyay, R.K. 1982. Release of nutrient elements from decomposing leaf l i t t e r of Eucalyptus globulus. Indian Forester, 108(8):550-553. van Hook, R.I., D.W. Johnson, D.C. West and L.K. Mann. 1982. Environ-mental effects of harvesting forests for energy. For. E c o l . Manage. 4(1):79-94. van Noordwijk, M. 1983. Functional i n t e r p r e t a t i o n of root densities i n the f i e l d for nutrient and water uptake, pp. 207-226. In: "Root ecology and i t s p r a c t i c a l a p p l i c a t i o n " , Int. Symp. Gumpenstein, 1982, Bundesanstalt Gumpenstein, A-8952. Irdning. Vasconcelos, C.A., J.M. Braga, R.F. Novais and O.C.B. Pinto. 1975. Fosforo em dois latossolos do Estado de Mato Grosso. I I I . Relacoes entre planta, solo e fosforo. Revista Ceres, 22(119):22-49. Veloso, H.P. 1966. Atlas f l o r e s t a l do B r a s i l . Rio de Janeiro, M i n i s t e r i o da A g r i c u l t u r a . 82 pp. Venkataramanan, C , B. Haldorai, P. Samraj, S.K. Nalatwadmath and C. Henry. 1983. Return of nutrient by the l e a f - l i t t e r of bluegum (Eucalyptus globulus) and black wattle (Acacia mearnsii) plantations of N i l g i r i s i n Tamil Nadu. Indian Forester, 109(6):370-378. Wenger, K.F. 1953. The sprouting of sweetgum i n r e l a t i o n to season of cutting and carbohydrate content. PI. Phys. 28(l):35-49. Westman, W.E. and R.W. Rogers. 1977a. Biomass and structure of a subtropical eucalypt f o r e s t , North Stradbroke Island. Aust. J . Bot. 25(2):171-191. Westman, W.E. and R.W. Rogers. 1977b. Nutrient stocks i n a subtropical eucalypt f o r e s t , North Stradbroke Island. Aust. J . E c o l . 25(4):447-460. Wise, P.K. and M.G. Pitman. 1981. Nutrient removal and replacement associated with short-rotation eucalypt plantations. Aust. For. 44(3):142-152. 208 Woodruff, J.R. and E.J. Kamprath. 1965. Phosphorus adsorption maximum as measured by the Langmuir isotherm and i t s r e l a t i o n s h i p to phosphorus a v a i l a b i l i t y . S o i l S c i . Soc. Amer. Proc. 29:148-50. Woods, R.V. 1980. An i n v e s t i g a t i o n into the re l a t i o n s h i p between f i r e and nutrient depletion leading to decline i n produc t i v i t y between rotations of Pinus radiata plantations i n South A u s t r a l i a . Misc. Paper Wood and For. Dept. S. A u s t r a l i a . Woods, P.V. and R.J. Raison. 1983. Decomposition of l i t t e r i n sub-alpine forests of Eucalyptus delegatensis, E. p a u c i f l o r a and E.  dives. Aust. J . E c o l . 8(3):289-299. Zavitkovski, J . 1982. Juvenile coppice growth and production of several hybrid poplars i n Northern Wisconsin. In: Proc. North Am. Poplar Council Meeting, Rhinelander, Wisconsin, July 20-22, 1982. Kansas State U n i v e r s i t y , Manhattan. APPENDIX 1 S o i l data at two d i f f e r e n t depths for the stands studi TABLE Al.li Soil data at two different depths for the ataoda studied Available p Total P Kl Kg1 (PP») (ppa) (ppa) (ppa) Age Soli depth (ca) (aonthe) 0-10 10-30 0-10 10-30 0-10 10-30 O-10 10-30 0-10 10-30 0-10 10-30 a l * ( eq) P» (a) BOB Deapacho (good site) 15 2.7 (0.8) 8.1 (6.9) 533 (58) 167 (58) 62 (S) 40 26 1.2 (0.4) 3.8 (4.3) 567(121) 183 (75) 38(16) 22 38 2.6 (0.9) 2.8 (1.5) 600 (0) 233 (58) 31 (3) 22 31 1.4 (0.2) 1.1 (0.2) 367 (58) 167 (58) 24 (4) 18 62 1.8 (0.6) 1.3 (0.3) 633 ( 58) 267 (58) 29 (3) 22 73 1.3 (0.3) 1.0 (0.5) 600 (0) 633 (58) 20 (4) 16 (b) Carbonita (poor site) 21 8.2(10.9) 9.1(12.9) 633(321) 467(208) 22 (6) 17 32 2.1 (1.9) 2.5 (3.1) 417 (41) 367 (52) 15 (6) 14 43 0.9 (0.3) 2.9 (4.0) 500 (0) 433 (58) 17 (2) 13 56 1.4 (0.05) 1.8 (0.9) 500 (0) 300(173) 14 (0) 13 67 0.7 (0.05) 0.5 (0.05) 500 (0) 100 (0) 9(0.6) 9 23.1 10.2 5.8 (2.3) (5.2) (17) 7.1 (1.0) 5.6 (2.5) 3.6 (1.5) 3.7 (0.6) 3.2 (0.2) 12.1 (1.8 6.6 (2 5.8 (1 8.1 (3 1.5 (1.3) 3.2 (1.3) 4.2 (2.7) 3.0 (1.2) 2.0 (0.17) 1.9 (0.55) 1.4 (0.23) 1.4 (0.06) 1.4 (0.17) 1.2 (0.06) 1.0 (0.11) 1.6 (0.65) 0.9 (0.15) 1.5 (0.21) 1.4 (0.12) 2.1 (0.17) 1.6 (0.54) 1.1 (0.17) 1.0 (0.06) 1.1 (0.15) 1.2 (0.12) 1.0 (0.06) 1.3 (0.45) 0.6 (0.10) 1.2 (0.12) 1.1 (0.06) 4.7 4.7 4.7 4.7 4.7 4.7 (0.12) (0.16) (0.05) (0.04) (0.10) (0.17) 4.9 (0.01) 4.6 (0.07) 4.0 (0.03) 4.8 (0.08) 4.3 (0.02) 4.7 (0.07) 4.9 (0.03) 4.8 (0.08) 4.8 (0.01) 4.9 (0.13) 4.7 (0.09) 4.9 (0.01) 4.7 (0.10) 5.1 (0.05) 4.8 (0.03) 4.5 (0.03) Umbers within brackets represent standard deviations • Very high standard deviations for available P of soae young stands Bight be explained by aaaplea having been taken In pockets of soil that received fertiliser. The area was spot-fertilised, not broadcast fertilised. Values obtained for Ca were very low and could not be detected. 1 Exchangeable. 211 APPENDIX 2 Diameter d i s t r i b u t i o n per plot for the stands studied TABLE A2.1: Diameter distribution per plot (600 m2) over an age sequence of E. grandis plantations growing in the "cerrado" region - Bom Despacho (good site), Minas Gerais, Brazil. Stand Age (Months) 731 62 51 38 27 261 15 Average DAP H Frequency H Frequency H Frequency H Frequency H Frequency H Frequency H Frequency (co) O) I 11 III (m) I II III (m) I II III (m) I II HI (m) I II III (m) I II HI <m) I 11 III 2.71 4.7 0 1 5 4.7 0 2 4 4.7 2 1 I 3.7 5 2 3 3.7 9 11 6 5.1 3 4 2 5.1 34 32 43 4.48 7.5 0 1 3 7.5 0 1 3 7.7 1 1 3 6.5 3 5 10 6.5 15 22 14 6.7 11 9 9 6.7 65 63 66 5.41 10.0 2 2 2 10.0 3 4 3 10.0 3 2 4 7.4 9 4 7 7.4 25 21 25 8.3 19 9 13 8.3 32 25 16 6.68 11.7 5 2 6 11.7 5 2 6 12.3 1 4 3 8.4 16 15 18 8.4 34 44 29 9.0 13 22 26 9.0 2 2 1 7.96 12.9 2 5 4 12.9 3 5 5 13.3 6 4 6 9.1 24 17 25 9.1 29 27 26 10.0 42 23 32 9.23 14.3 5 6 4 14.3 4 9 8 14.3 5 8 5 9.8 22 17 26 9.8 11 22 15 10.7 19 28 23 10.50 15.3 7 6 8 15.3 9 6 15 15.7 12 6 11 10.4 22 32 24 10.4 3 2 10 11.5 16 16 10 11.78 16.5 11 9 17 16.5 10 10 15 16.7 16 9 23 10.8 11 12 18 10.8 1 1 1 11.8 4 5 6 13.05 16.2 12 9 13 16.2 18 15 12 17.0 12 28 14 11.3 4 11 5 12.4 0 1 0 14.32 17.0 8 16 5 17.0 10 14 11 17.5 17 15 14 12.0 1 3 1 15.60 17.7 9 4 16 17.7 12 12 9 18.5 9 9 11 12.5 0 1 1 16.87 18.0 10 16 2 18.0 4 9 3 19.4 5 4 2 18.14 18.4 3 5 2 18.4 6 1 1 19.3 0 3 1 19.42 20.3 5 1 1 20.3 1 1 2 20.69 21.0 0 1 2 21.0 1 0 1 21.96 21.8 0 0 1 TOTAL - 80 84 91 - 86 91 98 - 89 94 98 - 117 119 138 - 127 150 126 - 127 117 121 - 133 122 126 1 Data collected 1 In the same stand as the previous one. In the second 1 year of data col lec t Ion (1982) (I.e. data I for 15 and 26 months are from the same stand). TABLE A2.2: Diameter distribution per plot (600 m?) over an age sequence of E. grandis plantations growing In the "cerrado" region - Carbonita (poor site), Ninas Cerals, Brazil. Stand Age (Honths) 56 43 32 67l 321 21 Average Average DBH H Frequency H Frequency H Frequency DBH H Frequency 11 Frequency H Frequency (cm) (m) I II III (»> I II III (m) I II III (cm) (m) I II III (n>) I II III (a) I II III 2.71 4.8 2 6 0 2.8 6 0 3 5.0 9 10 13 2.55 4.8 1 3 0 4.8 8 2 3 4.8 33 4 14 4.48 5.2 6 3 2 5.9 11 0 6 6.7 22 13 18 3.82 5.2 2 2 2 6.5 21 4 8 6.5 45 47 40 5.41 7.5 6 8 10 7.8 25 0 17 7.5 36 27 34 5.09 7.5 7 5 6 7.5 24 23 25 7.5 47 56 48 6.68 9.2 10 17 13 8.1 19 0 15 8.3 40 42 33 5.37 9.2 8 16 8 8.4 34 36 32 8.4 7 19 20 7.96 10.0 21 15 24 8.9 23 0 19 8.9 14 26 21 7.64 10.0 13 10 22 9.3 33 38 32 9.23 10.9 20 28 25 9.7 7 0 20 9.4 5 12 6 8.91 10.9 22 20 17 9.7 8 16 18 10.50 12.0 20 15 13 10.4 4 0 5 10.0 0 1 0 10.19 12.0 21 25 18 10.3 3 7 3 11.78 12.6 7 3 5 10.9 3 0 1 11.46 12.6 21 5 10 13.05 13.2 5 5 3 12.73 13.2 4 4 4 14.32 14.01 13.7 1 6 5 15.60 15.28 14.0 I 0 1 TOTAL - 97 100 95 - 98 0 86 - 126 131 125 _ - 101 96 93 - 131 126 121 - 132 126 122 Data collected In the same stand as the previous one, in the second year of data collection (1982) (I.e. data for 21 and 32 months are from the same stand). APPENDIX 3 Phosphorus and nitrogen concentrations each component of the sampled trees TABLE A3.1: Phosphorus and nitrogen concentrations (Z) in each component of all sampled trees over an age sequence of E. grandis plantations growing In the "cerrado" region, Bom Despacho (good site), Minas Gerais, Brazil. Age Diameter Height Bark Wood Branches Foliage Fine roots Medium roots Large roots (months) (cm) ( m ) P N P N P N P N P N P N P N 15 5.1 7.1 - - 0.011 0.11 0.022 0.30 0.138 3.43 0.048 0.89 0.028 0.62 0.011 0.26 2.6 3.6 - - 0.011 0.14 0.033 0.43 0.115 2.35 - - - - - -5.1 7.0 - - 0.012 0.22 0.032 0.46 0.155 2.67 - - - - - -265 5.6 8.5 0.036 0.61 0.017 0.38 0.054 0.66 0.188 2.19 0.033 0.80 0.033 0.73 0.031 0.82 10.4 13.3 0.044 0.40 0.016 0.15 0.035 0.30 0.134 1.89 0.074 0.88 0.062 0.72 0.032 0.63 5.6 8.0 0.055 0.36 0.014 0.18 0.023 0.38 0.115 2.40 0.067 0.84 0.036 0.82 0.022 0.35 27 7.3 11.8 0.055 0.41 - _ 0.025 0.31 0.126 2.07 - - - - - -7.3 11.0 0.037 0.46 - - 0.053 0.43 0.098 2.04 0.038 0.58 0.031 0.51 0.009 0.21 5.1 7.2 0.070 0.48 - - 0.037 0.37 0.134 2.40 - - - - - -38 7.3 12.8 0.024 0.30 0.015 0.11 0.023 0.31 0.111 1.91 0.037 0.62 0.030 0.52 - -10.2 16.6 0.030 0.44 0.011 0.14 0.038 0.32 0.066 2.27 0.042 0.70 0.023 0.47 0.011 0.23 12.7 15.2 0.052 0.35 0.009 0.13 0.047 0.41 0.115 2.57 - - - - - -51 7.6 15.8 0.053 0.32 0.009 0.15 0.047 0.36 0.188 2.64 - - - - - -12.4 19.0 - - 0.012 0.17 0.032 0.28 0.115 2.12 0.051 0.49 0.031 0.40 0.015 0.29 16.9 22.0 - - 0.006 0.13 0.022 0.44 0.098 1.83 - - - - - -625 17.4 20.4 0.038 0.32 0.011 0.21 0.042 0.53 0.101 1.82 - - - - - -12.6 19.3 0.076 0.29 0.011 0.16 0.049 0.43 0.134 1.71 0.035 0.42 0.025 0.59 0.021 0.31 8.4 16.2 0.052 0.33 0.006 0.15 0.048 0.45 0.115 2.37 - - - - - -73 7.5 13.8 0.037 0.27 0.007 0.22 0.029 0.32 0.101 2.35 0.028 0.76 0.026 0.53 0.016 0.48 17.4 21.8 0.064 0.33 0.009 0.21 0.033 0.44 0.159 2.25 0.028 0.61 0.033 0.60 0.010 0.27 13.1 18.3 0.052 0.27 0.011 0.17 0.031 0.29 0.098 1.85 0.028 0.58 0.022 0.44 0.010 0.24 Including stems smaller than 3 cm diameter. Diameter less than 0.3 cm. Diameter between 0.3 and 1.0 cm. Diameter greater than 1.0 cm (including taproot). Data collected In the same stand as the previous one, In the second year of data collection (1982) (i.e. data for 15 and 26 months are from the 9ame stand). TABLE A3.2: Phosphorus and nitrogen concentrations (X) in each component of a l l sampled trees over an age sequence of E. grandis plantations growing In the "cerrado" region, Carbonita (poor site), Hlnaa Gerais, Brazil. 1 2 3 4 Age Diameter Height Bark Wood Branches Foliage Fine roots Medium roots Large roots (months) (cm) (a) P N P N F N F M F N F N F N 21 2.9 3.7 0.013 0.17 0.007 0.14 0.016 0.29 0.058 1.23 0.027 0.50 0.015 0.35 0.009 0.25 5.1 6.5 - - 0.010 0.23 0.010 0.30 0.081 1.71 0.030 0.76 0.033 0.77 0.008 0.18 5.1 6.1 - - 0.012 0.18 0.010 0.23 0.066 1.55 - - - - - -325 5.4 7.9 0.019 0.24 0.006 0.14 0.021 0.31 0.088 1.52 0.033 0.60 0.020 0.35 0.006 0.14 7.3 9.9 0.018 0.26 0.007 0.13 0.017, 0.34 0.066 1.61 0.025 0.38 0.022 0.37 0.009 0.19 7.6 9.5 0.022 0.21 0.006 0.13 0.015' 0.22 0.058 1.32 0.018 0.46 0.013 0.33 0.009 0.18 32 5.4 7.3 0.030 0.34 0.009 0.13 0.018 0.22 0.094 1.87 0.025 0.49 0.013 0.32 0.013 0.22 7.3 8.4 0.025 0.29 0.009 0.12 0.016 0.38 0.078 1.61 0.036 0.70 0.015 0.41 0.010 0.27 7.3 7.7 0.027 0.28 0.008 0.15 0.022 0.38 0.088 2.00 - - - - - -43 7.6 8.0 _ - 0.009 0.13 0.015 0.31 0.081 1.87 0.032 0.53 0.018 0.27 0.018 0.21 5.1 6.9 - - 0.006 0.15 0.010 0.31 0.072 1.41 - - - - - -9.9 10.0 - - 0.007 0.12 0.025 0.55 0.081 1.61 0.037 0.63 0.016 0.28 0.007 0.13 56 12.7 13.0 0.022 0.28 0.008 0.12 0.032 0.45 0.075 2.57 0.027 0.61 0.016 0.36 - -9.9 11.9 0.032 0.33 0.007 0.14 0.039 0.46 0.122 2.42 0.029 0.54 0.013 0.38 0.011 0.27 5.4 8.1 0.017 0.22 0.006 0.16 0.013 0.27 0.081 1.74 - - - - - -675 12.4 11.7 0.025 0.32 _ _ 0.030 0.48 0.101 2.31 0.022 0.54 0.016 0.40 0.008 0.27 9.9 12.3 0.023 0.24 0.004 0.15 0.032 0.44 0.119 2.43 0.021 0.58 0.014 0.35 0.009 0.19 7.6 11.1 0.015 0.30 0.006 0;17 0.030 0.36 0.104 2.14 0.017 0.52 0.012 0.48 0.005 0.23 1 Including stems smaller than 3 cm diameter. 2 Diameter less than 0.3 cm. 3 Diameter between 0.3 and 1.0 cm. 4 Diameter greater than 1.0 cm (Including taproot). 5 Data collected In the same stand aa the previous one, in the second year of data collection (1982) (i.e. data for 21 and 32 months are from the same stand). APPENDIX 4 Phosphorus and nitrogen concentrations of each component l i t t e r f a l l , f o r e s t f l o o r and l i t t e r i n the decomposition TABLE A4.1: Phosphorus and nitrogen concentrations (%) i n l i t t e r f a l l over an age sequence of E_. grandis plantations growing on two different "cerrado soils in Minas Gerais, Bra z i l . Bom Despacho (good site) Carbonita (poor site) 2 2 . Age Branches Foliage Age Branches Foliage Season (months) P N P N (months) P N P N 1 2 3 4 25 0.009 0.006 0.005 0.014 0.27 0.22 0.21 0.29 0.034 0.046 0.027 0.031 0.61 0.87 0.68 0.69 21 0.009 0.26 0.006 0.26 0.005 0.22 0.016 0.026 0.035 0.028 0.44 0.72 0.81 0.62 36 0.005 0.004 0.007 0.007 0.17 0.14 0.17 0.18 0.027 0.039 0.034 0.031 0.67 0.84 0.80 0.50 32 0.009 0.006 0.010 0.006 0.20 0.18 0.18 0.19 0.026 0.030 0.026 0.031 0.59 0.73 0.60 0.79 1 2 3 4 49 0.012 0.006 0.004 0.014 0.19 0.15 0.15 0.23 0.025 0.035 0.031 0.026 0.98 0.75 0.52 0.51 42 0.010 0.010 0.008 0.010 0.22 0.23 0.13 0.13 0.025 0.036 0.022 0.030 0.49 0.72 0.54 0.59 1 2 3 4 60 0.007 0.014 0.007 0.011 0.22 0.31 0.25 0.20 0.036 0.046 0.032 0.032 0.71 0.91 0.67 0.76 55 0.015 0.007 0.013 0.005 0.27 0.18 0.28 0.23 0.042 0.035 0.040 0.042 0.92 0.86 0.93 0.93 On the good sit e , l i t t e r traps were set in the f i e l d on May 5, 1981 and collected every 3 months: (a) May-July; (2) August-October; (3) November-January; (4) February-April. On the poor s i t e , l i t t e r traps were set in the f i e l d on June 5, 1981 and collected every 3 months: (1) June-August; (2) September-November; (3) December-February, (4) March-May. Age of the plantation at the commencement of l i t t e r sampling. 219 TABLE A4.2: Phosphorus and nitrogen concentrations (%) i n forest f l o o r branch and fol i a g e material over an age sequence of E_. grandis plantations growing on two d i f f e r e n t "cerrado" s o i l s i n Minas Gerais, B r a z i l Age Branches Foliage (months) P N P N (a) Bom Despacho (good s i t e ) 15 - - 0.032 (0.010) 0.90 (0.09) 26 0.012 (0.009) 0.32 (0.19) 0.030 (0.008) 0.74 (0.08) 38 0.007 (0.001) 0.20 (0.01) 0.018 (0.003) 0.68 (0.08) 51 0.006 (0.001) 0.20 (0.03) 0.030 (0.003) 0.76 (0.12) 62 0.011 (0.004) 0.25 (0.03) 0.038 (0.007) 0.74 (0.04) 73 0.008 (0.001) 0.15 (0.07) 0.037 (0.001) 0.87 (0.04) (b) Carbonita (poor s i t e ) 21 0.010 (0.016) 0.28 (0.49) 0.024 (0.009) 0.76 (0.04) 32 0.017 (0.008) 0.81 (0.41) 0.029 (0.005) 0.89 (0.13) 43 0.009 (0.003) 0.39 (0.08) 0.026 (0.003) 0.91 (0.12) 56 0.009 (0.001) 0.65 (0.17) 0.034 (0.002) 1.04 (0.05) 67 0.008 (0.002) 0.31 (0.04) 0.033 (0.005) 1.08 (0.07) Numbers within brackets represent standard deviations. 220 TABLE A4.3: Phosphorus and nitrogen concentrations (%) i n branch and foli a g e l i t t e r of E_. grandis decomposing i n l i t t e r bags fo r a 24 month period Length of the Branches''" Foliage^" decomposition period (months) N P N a. Bom Despacho (good s i t e ) 0 0.012 (0.008) 0.25 (0.09) 0.032 (0.004) 0.64 (0.04) 6 0.012 (0.005) 0.30 (0.07) 0.039 (0.006) 0.76 (0.14) 12 0.011 (0.004) 0.27 (0.11) 0.039 (0.006) 0.92 (0.11) 18 0.010 (0.002) 0.24 (0.08) 0.039 (0.007) 0.98 (0.09) 24 0.011 (0.002) 0.33 (0.09) 0.038 (0.005) 0.97 (0.13) b. Carbonita (poor s i t e ) 0 0.013 (0.003) 0.23 (0.10) 0.029 (0.004) 0.76 (0.13) 6 0.011 (0.003) 0.39 (0.10) 0.037 (0.007) 0.83 (0.14) 12 0.012 (0.003) 0.25 (0.04) 0.037 (0.004) 1.00 (0.19) 18 0.014 (0.007) 0.37 (0.15) 0.033 (0.004) 0.99 (0.15) 1 Numbers within brackets represent standard deviations. APPENDIX 5 Results of the test of s i g n i f i c a n c e of the d i f f e r i n means of l i t t e r f a l l and decomposing l i t t e r T A B L E A 3 . 1 : B i o m a s s ( k g / h a ) o f l i t t e r f a l l o v e r a n a g e s e q u e n c e o f E _ . g r a n d i s p l a n t a t i o n s g r o w i n g o n t w o d i f f e r e n t " c e r r a d o " s o i l s o n M i n a s G e r a i s , B r a z i l . B o m D e s p a c h o ( g o o d s i t e ) C a r b o n i t a ( p o o r s i t e ) 2 2 ^ A g e A g e S e a s o n ( m o n t h s ) B r a n c h e s F o l i a g e ( m o n t h s ) B r a n c h e s F o l i a g e 1 2 5 5 1 b 8 7 9 e 2 1 0 a 4 5 8 b e 2 1 2 9 a 4 5 9 b 6 a b 5 1 2 b e d 3 1 3 0 b 1 1 6 5 f 6 3 a b c 2 7 6 5 1 4 4 4 4 c 1 6 1 4 8 3 5 b e 6 8 0 d e f 1 3 6 3 8 5 c 1 0 5 5 f 3 2 3 1 a b c 4 0 8 b e 2 4 6 4 c 1 8 8 a 1 5 9 d e 8 1 4 f 3 3 9 2 c 1 2 7 3 f 3 4 4 e f 2 9 6 0 4 9 6 8 c 1 5 9 9 8 2 3 4 e f 3 5 3 a b 1 4 9 3 9 7 c 7 9 9 e 4 2 1 0 5 d e 5 0 8 c d e 2 5 4 0 c 4 7 0 b e 1 3 3 d e f 5 7 3 c d e f 3 7 4 9 c 1 5 5 7 8 2 0 2 d e f 1 8 1 9 g l 4 8 3 9 c 1 0 9 2 f 5 0 c d 2 5 5 a 1 6 0 5 3 0 c 6 1 1 d 5 5 1 7 6 e f 3 2 3 a b 2 4 2 4 c 5 5 6 c d 4 4 4 f 7 1 8 e f 3 8 2 6 c 1 5 5 2 8 2 9 7 e f 1 2 9 1 8 4 3 8 6 c 7 5 6 e 4 9 5 e f 4 0 9 b e O n e - w a y a n a l y s i s o f v a r i a n c e w a s p e r f o r m e d s e p a r a t e l y f o r e a c h l i t t e r c o m p o n e n t , f o r e a c h s i t e t y p e , a s a f u n c t i o n o f s e a s o n a n d s t a n d a g e . S e a s o n a n d a g e w e r e a r r a n g e d i n a f a c t o r i a l e x p e r i m e n t . M e a n s f o l l o w e d b y t h e s a m e l e t t e r f o r e a c h l i t t e r f a l l c o m p o n e n t , d o n o t d i f f e r s i g n i f i c a n t l y a t P < 0 . 0 5 . T e s t o f m e a n s i s b a s e d o n l o g a r i t h m i c t r a n s f o r m e d d a t a . 1 O n t h e g o o d s i t e , l i t t e r t r a p s w e r e s e t i n t h e f i e l d o n M a y 5 , 1 9 8 1 a n d c o l l e c t e d e v e r y 3 m o n t h s : ( a ) M a y - J u l y ; ( 2 ) A u g u s t - O c t o b e r ; ( 3 ) N o v e m b e r - J a n u a r y ; ( 4 ) F e b r u a r y - A p r i l . O n t h e p o o r s i t e , l i t t e r t r a p s w e r e s e t i n t h e f i e l d o n J u n e 5 , 1 9 8 1 a n d c o l l e c t e d e v e r y 3 m o n t h s : ( 1 ) J u n e - A u g u s t ; ( 2 ) S e p t e m b e r - N o v e m b e r ; ( 3 ) D e c e m b e r - F e b r u a r y , ( 4 ) M a r c h - M a y . 2 A g e o f t h e p l a n t a t i o n a t t h e c o m m e n c e m e n t o f l i t t e r s a m p l i n g . 223 TABLE A5.2: Percentage change in the nitrogen and phosphorus content of branch and foliage l i t t e r of E_. grandis, after 18 months of decomposition. Stand age (months) 25 36 49 60 Branches P Wet2 -85.8 (44.8)a 17.7 (5.0) be 28.4 be 40.0 c Dry -24.1 (45.6)a 7.9 (32.0)ab -29.5a 64.3 (1.6) c Mean -55.0 4.9 -0.6 52.2 N Wet 8.7 (11.8) 8.7 (4.8) -151.7 62.0 Dry 2.5 (22.1) 7.3 (26.6) -29.6 43.7 (1.3) Mean 5.6 a 8.0 a -90.7a 52.9 b Foliage p Wet 37.3 (2.7) b 40.1 (6.5) b 14.6a 27.6 b Dry 36.4 (8.8) b 39.2 (1.7) b 32.8 b 41.0 (6.2) b Mean 36.9 39.7 23.7 34.3 N Wet 25.6 (7.6) c 5.3 (3.8)a 1.4a 3.4 ab Dry 18.9 (4.4) c 24.9 (6.0) c 11.8 be 31.3 (12.2) c Mean 22.3 15.1 6.6 17.4 One-way analysis of variance was performed separately for each nutrient in each l i t t e r component, as a function of season (wet, dry) and stand age. Season and stand age were arranged in a f a c t o r i a l experiment. Numbers within brackets represent standard deviations. Numbers without corresponding standard deviations originated from one sample only. Means followed by the same letter for each nutrient in each component do not d i f f e r s i g n i f i c a n t l y at P <0.05. Test of means i s based on logarithmic transformed data. The interaction between season and stand age for branch N was not significant at P <0.05. 1 Age of the stands on which decomposition l i t t e r bags were located. Age was count from the date of commencement of l i t t e r sampling. 2 Wet - bags placed in the f i e l d in November 1981 (rainy season) using l i t t e r collected during the period May-October, 1981. Dry - bags placed in the f i e l d i n May 1982 (dry season) using l i t t e r collected during the period November, 1981 - A p r i l , 1982. 224 APPENDIX 6 Percentage change i n the nutrient content of decomposing l i t t e r TABLE A6.1: Percentage change In the nutrient content of decomposing branch and foliage l i t t e r of £. grandla growing In Bom Despacho (good alte). Numbers without corresponding standard deviations originated from one sample only. Loss as a I of i n i t i a l value Branchea Foliage Stand' Time of _ Z age exposure (months) (months) Wet1 Dry Wet Dry Wet Dry Wet Dry 25 6 -67.6 (39.5) - -99.9 (35.2) - 17.9 (15.9) - 33.4 (2.6) 12 - -48.9 (3.9) - -16.8 (52.3) - 9.9 (1.4) - 4.9 (11.4) 18 -85.8 (44.8) -24.1 (45.6) 8.7 (11.8) 2.5 (22.1) 37.3 (1.7) 36.4 (8.8) 25.6 (7.6) 18.9 (4.4) 24 -90.1 - -161.7 - 69.7 - 55.0 36 6 -4.5 (43.7) - -52.6 (35.8) - 12.6 (35.4) - 26.9 (11.5) 12 - -21.5 (26.3) - -9.4 (32.2) - 15.5 (8.5) - 2.8 (5.1) 18 17.7 (5.0) 7.9 (32.0) 8.7 (4.8) 7.3 (26.6) 40.1 (6.5) 39.2 (1.7) 5.3 (3.8) 24.9 (6.0) 24 3.9 (16.9) - 4.0 (22.7) - 51.8 (1.0) - 36.7 (0.1) 49 6 30.4 (12.7) - -65.1 (50.3) - 18.7 (5.7) - -5.0 (12.8) 12 - -27.3 (18.6) - 39.8 (28.8) - 14.6 (7.0) - -0.9 (3.2) 18 28.4 (20.8) -29.5 -151.7 (106.6) -29.6 14.6 (23.9) 32.8 1.4 (38.0) 11.8 24 36.3 (18.7) - -20.9 (8.4) - 49.7 (9.7) - 27.1 (10.8) 60 6 35.0 (28.9) - 22.6 (16.0) - 23.6 (13.3) - 17.7 (22.0) 12 . - 52.9 (32.0) - 19.0 (42.6) - 5.0 (9.9) - 6.3 (12.6) 18 40.0 64.3 (1.6) 62.0 43.7 (1.3) 27.6 41.0 (6.2) 3.4 31.3 (12.2) 24 49.5 (8.5) - 27.4 (11.3) - 56.4 (5.4) - 33.0 (10.6) 1 Numbers within brackets represent standard deviations. 2 Age of the stand at which the first litter collection started. 3 Wet - bags plsced In the field In November, 1981 (rainy season) using litter collected during the period Hay-October, 1981. Dry - bags placed in the field in May, 1982 (dry season) using Litter collected during the period November, 1981-Aprll, 1982. APPENDIX 7 Results of the test of s i g n i f i c a n c e of the differences means for biomass, and P and N content and concentrations of coppicing Ej_ grandis growing i n the greenhouse TABLE A7.1 Biomass (g) and phosphorus and nitrogen content (mg) of coppicing IS. grandis growing on two different soils in the greenhouse. Age of Fine roots Total roots Sprouts  sprouts (months) Biomass P N Biomass P N Biomass P N (a) Good s o i l (b) 0.0 6.67a 7.36 b 35.69a 13.18a 14.21ab 66.91a 0 0 0 1.5 6.56a 6.57 b 28.55a 12.72a 12.92ab 55.66a 0.23a 1.35a 6.52a 2.5 4.85a . 2.87a 24.25a 10.91a 6.33a 53.58a 4.69ab 9.15a 58.03a 3.52 5.22a 6.95 b 31.17a 13.15a 17.59 b 78.61a 9.36 b 23.61 b 125.07 b 4.5 6.05a 8.49 b 34.44a 13.80a 19.49 b 77.95a 17.40 c 32.17 b 165.46 b i Poor s o i l 0.0 6.17 c 5.90 be 22.53a 11.09a 10.79abc 42.81a 0 0 0 1.5 5.00abc 4.04ab 17.32a 10.48a 8.37ab 35.86a 0.19a 0.93a 5 .60a 2.5 3.53a 2.41a 14.46a 9.31a 6.12a 35.74a 1.42a 2.69a 13.20a 3.52 5.48 be 6.14 c 22.32a 11.47a 12.41 be 46.06a 3.97 b 7.72 b 40.66 b 4.5 4.35ab 6.87 be 18.23a 9.15a 14.66 c 38.00a 7.37 c 12.89 b 48.94 t One-way analysis of variance was performed separately for biomass, P and N i n each tree component, for each s o i l type, as a function of age. Means followed by the same letter in each column, for each s o i l type, do not d i f f e r significantly at P <0.05. * Roots smaller than 0.5 mm diameter. 2 Addition of 1.2 g (good soil) and 0.2 g (poor soil) of diammonium phosphate. 228 TABLE A7.2 Phosphorus and nitrogen concentration (%) i n roots and sprouts of E_. grandis growing on two d i f f e r e n t s o i l s i n the greenhouse. Age of Roots Sprouts sprouts (months) P N P N (a) Good s o i l (b) 0.0 0.11 (0.01) 1bc 0.53 (0.08)a 1.5 0.10 (0.03) b 0.44 (0.07)a 0.56 (0.10) b 2.65 2.5 0.06 (0.02)a 0.50 (0.16)a 0.20 (0.02)a 1.26 3.5 2 0.13 (0.01) d 0.61 (0.05)a 0.28 (0.08)a 1.56 4.5 0.14 (0.01) cd 0.57 (0.08)a 0.19 (0.05)a 0.97 i Poor s o i l 0.0 0.10 (0.03)a 0.39 (0.05)a 1.5 0.08 (0.04)a 0.33 (0.17)a 0.42 (0.26)a 2.56 2.5 0.07 (0.02)a 0.39 (0.15)a 0.19 (0.03)a 0.93 3.5 2 0.11 (0.04)ab 0.41 (0.06)a 0.21 (0.04)a 1.09 4.5 0.16 (0.03) b 0.42 (0.03)a 0.17 (0.01)a 0.67 One-way analysis of variance was performed separately f o r biomass, P and N i n each tree component, for each s o i l type, as a function of age. Means followed by the same l e t t e r i n each column, for each s o i l type, do not d i f f e r s i g n i f i c a n t l y at P <0.05. 1 Numbers within brackets represent standard deviations. 2 Addition of 1.2 g (good s o i l ) and 0.2 g (poor s o i l ) of diammonium phosphate. APPENDIX 8 Adsorption isotherms for the two s i t e s studied 230 2 4 6 3 P i n s o l u t i o n (ppm) Phosphorus adsorption isotherms for 30 cm s o i l depth f o r the s i t e s studied., These curves were hand-smoothed through the data points. Each point i s the mean of 9 r e p l i c a t i o n s . PUBLICATIONS Ferrelra, M.G.M..; Brandi, J*.M. and Schneider, G. 1977. Observacoes sobre floracab e frutlflcacao de Eucalyptus grandis de origem hibrlda, em Vicosa, Minas Gerais. Revista Ceres, 24(133):341-344. Ferrelra, M.G.M.; Candldo, J.F.; OUva Cano, M.A. and^ Conde, A.R. 1977. Efelto do sombreamento na producao de mudas de quatro especles florestais natlvas. Revista Arvore, 1(2):121-134. Ferrelra, M.G.M.; Candido, J.F.; Conde, A.R. and Brandi, R.M. 1978. Efelto do sombreamento na producao de mudas de quatro especles florestais nativas. I. Germ1naca"b. Revista Arvore, 2(1):61-67. Gomes, J.M.; Ferrelra, M.G.M.; Brandi, R.M. and Paula Neto, F. 1978. Influencla do sombreamento no desenvolvimento de Eucalyptus grandis W. Hill ex Maiden. Revista Arvore, 2(1):68-75. Pereira, A.R.; Gomes, J.M.; Ferreira, M.G.M. and Olivelra, A.C. de. 1980. Efelto do sombreamento parcial na produc/ao de mudas de Eucalyptus grandis W. Hill ex Maiden. Boletlm Tecnico da SIF, (9):1-6. Ferreira, M.G.M.; Candido, J.F.; Silva, D.A. da and Colodette, J.L. 1981. Efeito do sombreamento e densidade de sementes sobre o desenvolvimento de mudas de Pinus Insularis EndHcher e seu desenvolvimento Inicial no campo. Revista Floresta 2(1):53-61. Ferrelra, M.G.M.; Kimmins, J.P. and Barros, N.F. 1984. Impact of Intensive management on phosphorus cycling In Eucalyptus grandis plantations In the savannah region, Minas Gerais, Brazil. In: Symposium on Site and Productivity of Fast Growing Plantations, South Africa, April 30 - May 22, 1984. Proceedings. Vol. 2, pp. 847-856. 

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.831.1-0096642/manifest

Comment

Related Items