UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Moving towards sustainable livestock grazing in the Jujuy Model Forest, northwestern Argentina Ripley, Shannon Wynne 2007

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2007-0570.pdf [ 14.2MB ]
Metadata
JSON: 831-1.0101124.json
JSON-LD: 831-1.0101124-ld.json
RDF/XML (Pretty): 831-1.0101124-rdf.xml
RDF/JSON: 831-1.0101124-rdf.json
Turtle: 831-1.0101124-turtle.txt
N-Triples: 831-1.0101124-rdf-ntriples.txt
Original Record: 831-1.0101124-source.json
Full Text
831-1.0101124-fulltext.txt
Citation
831-1.0101124.ris

Full Text

M O V I N G TOWARDS SUSTAINABLE LIVESTOCK GRAZING IN THE JUJUY M O D E L FOREST, NORTHWESTERN ARGENTINA by S H A N N O N W Y N N E RIPLEY B.Sc, The University of Alberta, 2001 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Soil Science) THE UNIVERSITY OF BRITISH C O L U M B I A June 2007 © Shannon Wynne Ripley, 2007 Abstract Natural resource managers are concerned that domestic livestock grazing is negatively impacting subtropical Yungas forests, and subsistence-scale livestock owners report annual cattle deaths due to inadequate native forage in the Yungas of Jujuy province, Argentina. This study was carried out to assist livestock owners and landowners in the Jujuy Model Forest to move towards more sustainable livestock grazing. Annual herbaceous forage production was measured within grazing exclosures, and annual forage from woody plants was estimated at six sites within each of the deciduous forest, anthropogenic pasture, and highland pasture ecological zones. The anthropogenic pasture had significantly greater annual forage production and recommended stocking rates than the deciduous forest. A high proportion of unpalatable plants within the anthropogenic pasture limited potential forage production. Annual forage production in the highland pasture was not directly measured due to uncertainty regarding what portion of tussock grasses is annual growth. Forest structure, as measured by tree basal area, density, frequency, and importance value, suggests that heavy livestock grazing in the anthropogenic pasture, along with firewood and timber harvesting, has led to changes in arboreal composition and decreased sapling density in relation to the deciduous forest. Indicators of soil quality measured along four transects at each study site, including soil cover, organic C, total N , and penetration resistance, suggested that land use in the anthropogenic pasture is leading to declines in soil quality in relation to the deciduous forest. Potential solutions to current challenges related to livestock grazing include reducing livestock numbers to recommended stocking rates, and implementation of deferred, rotational grazing. There is interest among livestock owners and landowners in broadening their understanding of range management principles, and learning how to improve animal health and productivity. Agroforestry systems could provide benefits such as rehabilitation of degraded anthropogenic pastures, income diversification, increased food security, and improved demarcation of pastures. Open, unregulated access to public land, and precarious land tenancy for livestock owners, are the two most significant challenges to sustainable livestock grazing. The success of proposed solutions depends upon further community-based learning, and increased cooperation among land owners and livestock owners. Table of Contents Abstract u Table of Contents • iv List of Tables. viii List of Figures x Acknowledgements xi Agradecimientos xii Dedication xiv Co-Authorship Statement xv 1. INTRODUCTION AND LITERATURE REVIEW 1 1.1. Introduction ; 1 1.2. Jujuy Model Forest 1 1.2.1. Model Forest Concept and Establishment of the Jujuy Model Forest 1 1.2.2. Landscape Diversity 3 1.2.3. Yungas Subtropical Forest 3 1.2.4. Soils of the Yungas Subtropical Forest 6 1.2.5. Land Use History and Ownership 7 1.2.6. Socioeconomic Realities and Population 9 1.2.7. Current Livestock Grazing and Resourc e Use .' 10 1.2.8. Grazing History in Yungas Forest Ecosystems 12 1.3. Sustainable Livestock Grazing 13 1.3.1. Definition of Sustainable Livestock Grazing 13 1.3.2. Forage Production and Grazing Capacity 14 1.3.3. Rangeland Health Concept 14 1.4. Soil Quality 16 1.5. Community-Based Action Research Approach 17 1.6. Research Objectives 19 1.7. References 24 2. IMPACTS OF LIVESTOCK GRAZING ON SOIL QUALITY AND FOREST STRUCTURE IN YUNGAS FOREST ECOSYSTEMS OF SOUTHERN JUJUY, ARGENTINA 30 2.1. Introduction 30 2.2. Materials and Methods 32 iv 2.2.1. Study Sites ••••• 32 2.2.2. Sampling and Analyses ••• 34 2.2.2.1. Forage production and forest structure 34 2.2.2.2. Soil sampling • 35 2.2.2.3. Statistical analysis 37 2.3. Results and Discussion . . 38 2.3.1. Annual Forage Production and Grazing Capacity 38 2.3.2. Forest Structure . 40 2.3.3. Soil Quality 41 2.3.3.1. Soil cover 41 2.3.3.2. Soil chemical properties 42 2.3.3.3. Soil compaction 43 2.4. Conclusions 45 2.5. Acknowledgements 46 2.6. References * 55 3. OPPORTUNITIES AND C H A L L E N G E S IN M O V I N G T O W A R D S M O R E SUSTAINABLE L I V E S T O C K G R A Z I N G IN T H E J U J U Y M O D E L F O R E S T , A R G E N T I N A 62 3.1. Introduction 62 3.2. Study Area and Methodology 63 3.2.1. Location and Ecological Characteristics of Study Area 63 3.2.2. Local Livelihoods and Livestock Grazing Practices 64 3.2.3. Formulation of Research Questions and Data Collection 66 3.3. Sustainable Livestock Grazing Practices and Economic Alternatives 67 3.3.1. Reduction of Stocking Rates to Within Local Grazing Capacity 67 3.3.1.1. Subsistence-scale livestock owners 68 3.3.1.2. Land owners 70 3.3.2. Implementation of Deferred, Rotational Grazing Systems 72 3.3.2.1. Subsistence-scale livestock owners 72 3.3.2.2. Land owners 73 3.3.3. Community Pastures and Co-operative Grazing 73 3.3.4. Economic Diversification through Rural Tourism 76 3.3.4.1. Subsistence-scale livestock owners 77 3.3.4.2. Landowners 78 3.3.5. Implementation of Agroforestry Practices 78 3.3.5.1. Subsistence-scale livestock owners 81 3.3.5.2 Landowners 81 3.3.5.3. Common opportunities and challenges 82 v 3.3.6. Formation of a Grazing Management Group 83 3.7. Conclusions and Recommendations 84 3.7.1. Livestock Owners 84 3.7.2. Land Owners 85 3.7.3. Common Ground 86 3.8. References 88 4. CONCLUSIONS, R E C O M M E N D A T I O N S AND E V A L U A T I O N O F R E S E A R C H M E T H O D S 93 4.1. Chapter Overview 93 4.2. Research Conclusions and Recommendations 93 4.3. Evaluation of Study Methods and Recommendations for Future Research 97 4.3.1. Evaluation of Forage Production and Grazing Capacity Study 97 4.3.1.1. Forage quantity • 97 4.3.1.2. Forage quality • 99 4.3.2. Evaluation of Soil Quality Indicators 99 4.3.2.1. Soil cover and soil litter layer depth 100 4.3.2.2. Organic C and total N 103 4.3.2.3. Bulk density and penetration resistance 104 4.3.3. Evaluation of Forest Structure Study 106 4.4. Community Participation and Educational Activities 108 4.4.1. Participation of Livestock Owners and Landowners in Field Research 108 4.4.2. Participation of University Students 108 4.4.3. Evaluation of Community Workshops 109 4.4.4. Evaluation of the High School Field-Trip Program I l l 4.4.5. Reflections on Community-Based Research 112 4.5. A Journey of Learning Together 114 4.6. References 116 APPENDIX I Forest regions of Argentina, and detailed map of the Yungas-Chaco transition forest zone in northwestern Argentina 121 APPENDIX II Field site names and locations 123 APPENDIX III Soil sampling layout at study sites 124 APPENDIX IV Comparison of 2005-2006 precipitation levels at San Salvador de Jujuy with 98-year average precipitation levels 125 APPENDIX V Schedule of activities for community-based rural tourism pilot activity.. 126 APPENDIX VI Field workbook used for 'Living with the Forest' school field trips 127 APPENDIX VII Typical schedule of activities for a "Living with the Forest" school vi field trip 135 APPENDIX VIII Analysis of variance (ANOVA) tables for the forage production, soil quality and forest structure study , 136 List of Tables Table 1.1 Elevation, landscape position, annual precipitation and vegetation for Yungas forest ecological zones in northwestern Argentina 20 Table 1.2 Elevation, slope and dominant vegetation for three ecological zones in the Yungas forest of southern Jujuy, Argentina 21 Table 1.3 Characteristics of soil subgroups found at field sites in the Jujuy Model Forest, Argentina 22 Table 2.1 Annual forage production, recommended stocking rate, and estimated current stocking rate for anthropogenic pasture, deciduous forest and highland pasture ecological zones in the Yungas forest, Jujuy province, Argentina 47 Table 2.2 Basal area (dominance), frequency, density, and importance value of tree species found in deciduous forest and anthropogenic pasture ecological zones of the Yungas forest, Jujuy province, Argentina 48 Table 2.3 Soil physical and chemical properties measured in anthropogenic pasture, deciduous forest, and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina \ 49 Table A . l Field site names and locations 123 Table A.2 A N O V A table for the effect of ecological zone on annual forage production 136 Table A.3 A N O V A table for the effect of ecological zone on unpalatable biomass production 136 Table A.4 A N O V A table for the effect of ecological zone on the proportion of forage consisting of grasses ...136 Table A.5 A N O V A table for the effect of ecological zone on the proportion of forage consisting of forbs 136 Table A. 6 A N O V A table for the effect of ecological zone on the proportion of forage consisting of woody biomass 136 Table A.7 A N O V A table for the effect of ecological zone on the number of saplings 137 Table A.8 A N O V A table for the effect of ecological zone on % bare soil 137 Table A.9 A N O V A table for the effect of ecological zone on % litter cover 137 Table A. 10 A N O V A table for the effect of ecological zone on % rock cover 137 Table A . l 1 A N O V A table for the effect of ecological zone on soil litter depth 137 Table A. 12 A N O V A table for the effect of ecological zone on % organic C 137 Table A. 13 A N O V A table for the effect of ecological zone on % total N 137 Table A. 14 A N O V A table for the effect of ecological zone on the ratio of organic C to total N(C:N) . . . . 138 Table A. 15 A N O V A table for the effect of ecological zone on soil pH 138 Table A. 16 A N O V A table for the effect of ecological zone on bulk density 138 Table A. 17 A N O V A table for the effect of ecological zone on soil penetration resistance from 0-5 cm depth in November, 2005 138 Table A. 18 A N O V A table for the effect of ecological zone on soil penetration resistance from 5-10 cm depth in November, 2005 138 Table A. 19 A N O V A table for the effect of ecological zone on soil penetration resistance from 0-5 cm depth in February-March, 2006 138 Table A.20 A N O V A table for the effect of ecological zone on soil penetration resistance from 5-10 cm depth in February-March, 2006 138 Table A.21 A N O V A table for the effect of ecological zone on soil penetration resistance from 0-5 cm depth in April, 2006 139 Table A.22 A N O V A table for the effect of ecological zone on soil penetration resistance from 5-10 cm depth in April, 2006......... 139 List of Figures Figure 1.1 Location of the Jujuy Model Forest, within Jujuy province, northwestern Argentina, South America 23 Figure 2.1 Location of the Jujuy Model Forest, within Jujuy province, northwestern Argentina, South America 50 Figure 2.2 Forage composition for anthropogenic pasture, deciduous forest, and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina 51 Figure 2.3 Unpalatable biomass production for anthropogenic pasture, deciduous forest, and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina 52 Figure 2.4 Soil organic C (a) and soil total N (b) at 0-10 cm depth in three ecological zones of the Yungas forest, Jujuy province, Argentina 53 Figure 2.5 Soil penetration resistance at depths of a) 0-5 cm and b) 5-10 cm, measured in November 2005, February-March 2006, and April 2006 within anthropogenic pasture, deciduous forest and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina .' 54 Figure 3.1 Location of the Jujuy Model Forest, within Jujuy province, northwestern Argentina, South America 87 Figure A . l Forest regions of Argentina 121 Figure A.2 Detail of the Yungas - Chaco transition zone in northwestern Jujuy 122 Figure A.3 Soil sampling layout around each grazing exclosure at study sites 124 Figure A.4 Monthly precipitation from June 2005 to May 2006 in comparison with the 98-year average for San Salvador de Jujuy, Argentina 125 x Acknowledgements I am grateful to my supervisors Dr. Maja Krzic and Dr. Art Bomke, and committee member Dr. Gary Bradfield, who provided such tremendous support, helpful advice, encouragement and learning opportunities throughout the journey of this project - always with a great sense of humour. Thank you to Dr. Roy Turkington for helpful recommendations during the project design phase, and for kindly participating in the defense. Thank you to Dr. Les Lavkulich for so generously sharing his time, experience, insight and enthusiasm in so many 'Learning with Les' sessions, when I would pop by his office with one question, and leave with new perspectives and insight on many more topics. I am grateful to Dr. Chuck Bulmer for kindly sharing his soil sampling expertise, and providing me with an opportunity to gain sampling practice prior to leaving for Jujuy, and to Dr. Tom Sullivan for a fascinating trip to his field sites and valuable project design advice. I deeply appreciate the statistical assistance generously provided by Dr. Tony Kozak. Financial support for this project was provided by the John G. Bene Fellowship in Community Forestry of the International Development Research Centre (IDRC), the Natural Sciences and Engineering Research Council of Canada (NSERC), and the Organization of American States. I would like to particularly acknowledge the John G. Bene Fellowship for providing resources that enabled this to be a community-oriented project that combined research and education activities. Thank you to fellow students in Land and Food Systems and Forestry for such marvelous friendship, support, kindness, and shared learning which has tremendously enriched my experience throughout this journey. Thank you to Rhianna Nagel for bringing such enthusiasm and dedication to her volunteer work with us in Jujuy, and for her friendship, and many insightful conversations about Latin America over shared mate back here in Vancouver. I am grateful to Art for enabling my life path to cross with Rhianna's. Thank you to Mom, Dad and Steph, and other family and friends for their love and tremendous support, and for coming to share in experiences in Argentina. xi Agradecimientos Estoy muy agradecida a todas las familias e individuos quienes participaron en las actividades de investigacion, talleres y eventos educativos durante el transcurso de este proyecto. Estoy profundamente agradecida a Rolando Braun Wilke y Sara Villafane por su tremendo apoyo y ayuda. Agradezco mucho a ellos por haber compartido sus diversos conocimientos sobre la ecologia local, sus sugerencias acerca de metodos de investigacion, por haber facilitado el uso del laboratorio a la Universidad Nacional de Jujuy, por haberme dado oportunidades enriquecedoras para participar en cursos, talleres y salidas con la Catedra de Ecologia, y sobre todo por su amistad y apoyo personal. Muchisimas gracias a Virginia Outon y Ralf Schillinger por su valiosa ayuda, sus sugerencias y perspectivas, y por su gran dedicacion al Bosque Modelo Jujuy. Agradezco el apoyo de la Asociacion Bosque Modelo Jujuy, y la oportunidad de colaborar con tantas personas y organizaciones quienes estan dedicados a trabajar juntos en el fomento del manejo sustentable de los ecosistemas locales. Gracias a Claudia Chauque por haber sido un miembro integral del equipo principal del proyecto, por haber compartido sus diversos conocimientos sobre la cultura y las realidades locales, y por haber generosamente compartido el carifio de su familia conmigo. Estoy muy agradecida a Maria Ines Bonansea por su tremendo aporte a la planificacion, al trabajo de campo y a las actividades educativas de este proyecto, por su dedicacion, gran sentido de humor y su profesionalismo. Agradezco mucho a Ivan Escalier por siempre estar tan dispuesto a compartir sus conocimientos y valiosas perspectivas sobre la agricultura local, y por sus diversos apoyos durante el transcurso del proyecto. Gracias a Helena Arraya (Lati) por haber contribuido un aporte enorme al trabajo de campo, y en particular por haber compartido su gran entusiasmo y sus conocimientos sobre los arboles nativos. Agradezco mucho la ayuda, determination, interes y sentido de humor de los "clausureros": Martin Chauque, Pepe Diaz, Daniel Colque y Beto Diaz. Estare siempre muy agradecida a todos los estudiantes de la Universidad Nacional de Jujuy quienes contribuyeron una ayuda enorme con el trabajo de campo y actividades educativas, con entusiasmo y dedicacion inspiradora: Haydee Albornoz, Monica Burgos, Erica Cuyckens, Juan de Pascuale, Veronica Encinas, Emanuel Gonzalez, Pedro Gonzalez, Carina Puca Saavedra, Maggie Reader, Alejandra Tintilay, y Alcira Torres. Gracias a Chantelle Leidl por su gran ayuda con el muestreo de suelos, por su fabulosa amistad, y por no haber advertido a otros quienes iban a visitarme que les esperaba! Gracias a "Mama Linda", una experta en excavar pozos sobre laderas pedregosas. Gracias a Aaron Bates por su acompanamiento a traves este camino de aprendizaje. Estoy tremendamente agradecida a Kimberly Olson por la oportunidad de participar en su proyecto de maestria en education ambiental en 2002, lo que fue una de las inspiraciones principales que me llevo al camino hacia Jujuy. Estoy agradecida a Fanny Altamirano por haber facilitado la participation de estudiantes universitarios en el proyecto. Gracias a Liliana Lupo, Gustavo Guzman, y todos en la Facultad de Ciencias Agrarias por su ayuda y apoyo. Agradezco el uso del laboratorio de la Catedra de Ecologia en la Universidad Nacional de Jujuy. Agradezco consejos sobre el diseno de muestreo y perspectivas sobre la ecologia regional generosamente provistos por el Dr. Lucio Malizia. Muchas gracias a Ana Carranza por su ayuda en la identification de plantas. Estoy muy agradecida por el gran apoyo de Ernesto Reyes, y por su entusiasmo y dedication a la idea de implementar un programa educativo para estudiantes secundarios. Gracias a los docentes de E l Carmen y Perico por su interes y colaboracion en hacer el programa educativo una realidad. Para siempre agradecere la amistad, apoyo y carino de miembros de la comunidad de E l Carmen quienes compartieron generosamente las actividades de sus vidas conmigo, y me hicieron sentir que tengo familia y raices en la comunidad. xiii Dedication To each of the individuals who shared in the learning of this project. To Grandad, who I will always think of when admiring a mountain view or beautiful rock, and Grandma, who cherished time spent in her community. A todos los individuos quienes compartieron en el aprendizaje de este proyecto. A mi abuelo, en quien siempre pienso cuando admiro una vista hermosa de una montana o una piedra linda, y a mi abuela, quien disfrutaba mucho del tiempo pasado en su comunidad. x i v Co-Authorship Statement Chapter 2: I was responsible for the identification and design of the research program, data collection, all statistical analyses, and writing the manuscript. Maja Krzic and Art Bomke provided guidance in experimental design, and in the interpretation and presentation of the manuscript. Gary Bradfield provided assistance with statistical analyses, data interpretation and presentation of the manuscript. xv 1. INTRODUCTION AND LITERATURE REVIEW 1.1. Introduction Rangelands are commonly defined as uncultivated land, which has the potential to be grazed or browsed by wild or domestic animals, and is managed as a natural ecosystem (Holechek et al. 1998). Approximately 26% of the global terrestrial land surface is currently grazed by domestic animals; this area includes grasslands, shrublands, pasturelands, riparian . areas, and forest stands (Steinfeld et al. 2007). With expansion of cultivated agricultural land and rapid human population growth over the past decades, many subsistence livestock owners have been forced to move their grazing animals to more marginal lands, often found in montane and/or forested regions. Degradation of forest ecosystems occurs when forested lands, not adapted to grazing pressure, are subjected to livestock grazing or when grazeable forest lands are subjected to excessive stocking rates in relation to the forage resources available. Forests that are used to provide forage and browse for domestic livestock often provide many other . goods to local populations such as firewood for cooking and heating, timber, fertilizer in the form of the forest leaf litter layer, food and medicinal plants. The subtropical Yungas forest ecosystems in Jujuy province, northwestern Argentina, are an example of a region where livestock grazing is widespread, and occurs in conjunction with a wide variety of other forest uses. Local community members and natural resource managers have identified current domestic livestock grazing practices as being one of the greatest challenges to sustainable management of forest resources in the region (Braun Wilke et al. 2001, Outon 2002). -1.2. Jujuy Model Forest 1.2.1. Model Forest Concept and Establishment of the Jujuy Model Forest In response to concerns about the sustainability of forest management in Canada, in 1992 the Canadian Forest Service created a network often 'model forests' in six of the main forest ^ regions of Canada (CMFN 2007). This has now expanded to 11 model forests and three special project areas (CMFN 2007). The primary goal of the Canadian Model Forest Network is to 1 facilitate discussion and dialogue among the diverse forest users of a region so individuals and organizations can work together towards a long-term plan for sustainable forest management. A unique feature of model forests is that the designation of an area as a 'model forest' does not change land ownership or the rights of participating landowners and land managers in the area. Rather, the model forest consists of a working-scale landbase where forest users and managers are encouraged to come together to share diverse viewpoints, and combine expertise and resources to work towards sustainable management of the model forest area. Following the success of the Canadian Model Forest Network in facilitating dialogue among forest users, a number of other countries expressed interest in establishing their own model forest networks. Currently there are 41 model forests in existence or being developed in 18 countries around the world (IMFN 2007a). Although each model forest faces unique local challenges and realities, the International Model Forest Network provides a valuable forum for the sharing of local experiences and approaches in sustainable forest management among participating model forests. In 2002, the Jujuy Model Forest Association (Asociacion Bosque Modelo Jujuy) was established to work towards the long-term sustainable management of 130 000 ha of the Los Pericos-Manantiales watershed located at approximately 24°S and 66°W in the departments of El Carmen and San Antonio, Jujuy province, northwestern Argentina (IMFN 2007b) (Figure 1.1). Resource management within the model forest is based on a watershed management concept. In the spirit of the International Model Forest Network, the Jujuy Model Forest Association recognizes as part of it's mandate that meeting the social, cultural, and economic needs of local communities is an integral part of sustainable forest management. The Jujuy Model Forest Association has more than 40 institutional and individual members who collaborate and participate in forest management initiatives (Outon 2005). These members include local individual landowners, resource users, educators, provincial ministries of the government of Jujuy, local municipal governments, non-governmental organizations, three local hospitals, the National University of Jujuy, local schools and the school district, agricultural producer cooperative organizations, and neighbourhood associations. 2 1.2.2. Landscape Diversity The Los Pericos and Manantiales watersheds that comprise the Jujuy Model Forest range in elevation from 750 to 5000 m above sea level, and therefore include a diverse array of ecosystems and associated livelihoods (Outon 2002). Lowland valleys in the eastern part of the watershed have been progressively cleared of Chaco and Yungas forest vegetation since the 1600s, and currently 30 000 ha are dedicated to irrigated farming (IMFN 2007b). Native Yungas subtropical forest ecosystems extend along steep mountainous slopes within the middle elevations (-1000 - 2000 m) of the Jujuy Model Forest, and have an area of 30 000 ha (Outon 2002). At elevations immediately above the Yungas forest, highland pasture communities, dominated by large tussock grasses, extend up the mountain slopes. Within the 2000 - 3000 m elevation range, these moderately humid, highland pastures form a transition zone that extends into the dry, sparsely vegetated pre-puna and puna highland zones found at higher elevations and characterized by a cold, dry desert climate (Braun Wilke et al. 2001). This study will focus on livestock grazing dynamics within the Yungas forest and adjacent highland pasture ecological zones at elevations of 1200 - 1800 m. 1.2.3. Yungas Subtropical Forest Along with subtropical forests in northeastern Argentina, the Yungas occupy less than 2% of the surface area of Argentina, but contain approximately 50% of the biological diversity (Brown et al. 2001). 'Yungas' is a word used to describe tropical and subtropical montane forests found along the eastern slopes of the Andes mountains in South America, extending from Colombia in the north, to the province of Catamarca, Argentina in the south (Brown et al. 2001). In general, the Yungas form an ecologically diverse transition zone between the high Andean peaks to the west, and lowland Amazon rainforest and dry Chaco forest zones to the east (Appendix I). As warm, moist air from the Amazon rainforest moves westward, it rises up the eastern side of the Andes, cools and condenses, forming a zone of high precipitation, and seasonally or annually persistent cloud cover. Precipitation levels and type vary along steep altitudinal gradients, and result in the formation of different forest types and associated fauna (Brown and Grau 1993). The Yungas of northwestern Argentina are characterized by a subtropical climate with 3 distinct wet and dry seasons, with the majority of precipitation occurring from the end of November until the first part of April (Brown et al. 2001). Precipitation varies between 700 -2000 mm/year across strong altitudinal gradients (Braun Wilke et al. 2001). Average summer temperatures vary between approximately 20-26°C and winter temperatures vary from 10-15°C (Braun Wilke et al. 2001). Ecological changes in the Yungas forests have far-reaching hydrological implications, as these forests form the headwaters and water source for a large extent of northwestern Argentina. Steep mountain slopes, intense summer precipitation, and thin soils make the Yungas forests susceptible to erosion when the forest canopy cover is reduced or land degradation occurs (Braun Wilke et al. 2001). Therefore, although the Yungas forest system covers a relatively limited geographical area, land use decisions made in the Yungas have widespread regional and continental ecological, health, social and economic implications. In northwestern Argentina, the Yungas are customarily divided into four altitudinal zones based on elevation, precipitation, and forest type as outlined in Table 1.1. It is important to emphasize that clear boundaries do not exist among these ecological types, and forest types may extend beyond the approximate elevation range outlined due to interactions of anthropogenic modifications to the ecosystem (such as livestock grazing and logging), slope and aspect (Braun Wilke et al. 2001). The study area included deciduous Yungas-Chaco transition forest, as well as highland pasture, exemplifying the mosaic of forest and pasture types that can form across a narrow elevation range due to the interplay of topography and land use. The three ecological zones of focus in our study were deciduous forest, anthropogenic pasture and highland pasture (Table 1.2). In contrast to traditional range science terminology which defines 'pasture' as grazing land that produces introduced (agronomic) forage species and receives periodic cultural improvements (USDA-NRCS 2003), the word 'pasture' (pastizal) is used in this study to refer to ecosystems dominated by native herbaceous vegetation. Deciduous forest sites were composed of secondary growth forest found at elevations of 1200 - 1500 m along the base of the Andes, and included vegetation characteristics of both the Yungas and Chaco forests. In contrast to the Yungas forest which extends in a narrow band 4 along the Andes, and is characterized by high levels of precipitation, the Chaco is a dry forest characterized by xerophytic trees and shrubs (Braun Wilke et al. 2001), many of which bear barbs and spines to discourage herbivory. Extensive logging was carried out in the Yungas-Chaco transition forest within our study area in the mid-1900s through to the 1970s, with particularly intensive selective logging around 1975 (Braun Wilke et al. 2001). Vervoorst (1982) estimated that the Yungas-Chaco transition forest requires approximately 60 to 100 years to reach it's climax community. Our anthropogenic pasture sites were a mosaic of areas at the base of the Andes foothills found at elevations of 1200 - 1450 m that have predominantly been converted from deciduous Yungas-Chaco transition forest to open savanna landscape through intensification of timber extraction, firewood harvesting, and livestock grazing since the 1970s. Grasses, forbs, and shrub species that are resistant to grazing pressure dominate the open slopes of anthropogenic pasture sites, and are interspersed with scattered stands of trees or single trees. Highland pasture sites were open grass / shrublands found at elevations above the deciduous forest on exposed north and northwesterly facing slopes. These sites were dominated by large tussock grasses with rigid leaves up to 1 m long, interspersed with shorter grasses and forbs, and patches of bare ground. In moister draws near our highland pasture sites there were some stands of Alnus acuminata (aliso) present, which is characteristic of this highland pasture transition zone (Cabrera 1976). Highland pasture zones are often found at elevations above 1800 m (Braun Wilke et al. 2001); however, some of our highland pasture sites were at elevations of 1650 m. Domestic livestock grazing, along with the local practice of burning highland pastures at the end of the dry season to stimulate forage regrowth, and remove unpalatable species, are two activities that tend to lower the elevation at which highland pastures are found (Grau 2002). Although our study was restricted to the deciduous Yungas-Chaco transition forest and highland pasture Yungas forest types, it is important to note that there were areas of mountain cloud forest within our general study area. These were found on deeply incised draws and steep slopes with a south aspect at elevations of approximately 1300 - 1800 m. In some instances where highland pasture sites were found on steep, north- or north-west facing slopes near the top 5 of foothills, there were mountain cloud forests found on the steep south-facing slope within 50 m opposite the treeless highland pasture! These mountain cloud forests were dominated by perennial trees in the Myrtaceae family, many of which had tnmks covered in thick layers of moss and epiphytes. 1.2.4. Soils of the Yungas Subtropical Forest Soils in the Yungas forest zone of the Jujuy Model Forest have been classified as members of the La Quesera - La Quesera Chica and La Cruz soil associations (Nadir and Chafatinos 1990). These soils have formed on sedimentary (sandstone and siltstone) and metamorphic (shale, slate and quartzite) parent materials from the Precambrian, Ordovician and Tertiary geological time periods (Nadir and Chafatinos 1990). Volcanic ash has been periodically deposited and incorporated into regional soil profiles (Vargas Gi l 1990) as a result of volcanic eruptions in the Puna highlands to the west of the Yungas forest. The most recent local volcanic eruptions likely occurred between 400 and 700 years ago (Grau 1985). There were no distinguishable layers of volcanic ash present in the six soil pits excavated during our study. Soils at study sites within the three ecological zones that were part of this investigation were classified as members of the Typic Haplustalf, Ultic Haplustalf and Lithic Ustorthent subgroups (Soil Survey Staff 1999) (Table 1.2). Soils in the Alfisol Order are characterized by the translocation of silicate clays without excessive depletion of bases, and are not dominated by processes that lead to a mollic (chernozemic) epipedon. Soils in the Entisol Order lack distinct pedogenic horizons, and are dominated by mineral soil materials. In the case of our study area, limited pedogenic development has occurred due to the presence of lithic contact within 50 cm of the soil surface. Soils in our study area have an ustic moisture regime in which moisture is limited, but is present at the time when conditions are suitable for plant growth (Soil Survey Staff 1999). Ustic moisture regimes are typical in subtropical regions with a distinct rainy season of three months or more (Soil Survey Staff 1999). The soil temperature regime for our study area is classified as thermic, with a mean annual soil temperature of 15-22°C and the difference between mean summer and mean winter soil temperatures greater than 6°C (Soil Survey Staff 1999). / 6 Soils throughout the Yungas forest tend to have an abrupt limit with the underlying bedrock and contain large rock fragments throughout the profile (Vargas Gi l 1990). Primary limiting factors to land use in forested regions of the Jujuy Model Forest are steep altitudinal gradients, shallow soil depth, and rockiness. Land capability for agriculture on all subgroups (apart from rocky outcrops) is rated as VI according to the United States Department of Agriculture classification system (Vargas Gi l 1990). Class VI lands have severe limitations to the long-term production of common crops. They are generally unsuited to cultivation and are restricted in their use to mainly pasture, range, forestland or wildlife food and cover (USDA NRCS 2003). 1.2.5. Land Use History and Ownership A variety of aboriginal cultures have inhabited the Yungas forests of northwestern Argentina for the past 11,000 years (Yacobaccio et al. 2004). The Jujuy Model Forest region was home to the Churumata First Nation, and archaeological artefacts indicate they carried out hunting and gathering activities in the area 2,000 years before the present time (Outon 2002). In addition to it's role as an ecological transition zone, the Yungas forests served as an important transition and trading zone for aboriginal cultures in the dry highland valleys and mountains to the west, and lowlands to the east (Ventura 1999, Yacobaccio et al. 2004). During the time of the Incas around 1400 A.C. , the lowland valleys were used by agriculture-based indigenous cultures whose crops included corn, potatoes and squash (Brown et al. 2001, Braun Wilke et al. 2001). Initiation of the concept of private land ownership began during the mid-1500s, when the Spanish viceroy for the region gave large tracts of forested land to Spanish colonists. During the same time period, Spanish administrators 'distributed' indigenous people among the Spanish colonists to be used as slave labourers (Rutledge 1996). During the colonial era from the 1500s to the 1800s, forest areas predominantly in the lowland mountain forest area were cleared and the first sawmills in the region were established (Outon 2002). Cleared lands were used for expanding agricultural activities ranging from the cultivation of cereal crops, sugar cane, fruits, and vegetables to domestic livestock production including cattle, sheep, horses, and goats (Braun Wilke and Picchetti 2006). Towards the end of 7 the 1700s, and first part of the 1800s, domestic livestock production increased markedly in the lowland valleys of Jujuy, to meet the demand for mules and other pack animals in the mines of Chile and Bolivia (Braun Wilke et al. 2001). In the early 1900s, there were waves of immigrant arrivals to northwestern Argentina from Spain, Italy, eastern Europe, Syria, Lebanon, Armenia, and Turkey (Outon 2002). When Spanish landholders in the eastern areas of the Jujuy Model Forest area were unable to pay immigrant workers employed on their farms, many ceded small tracts of their land in exchange for work performed (Outon 2002). This led to the division of large landholdings into smaller parcels within the agricultural zones of the region. In contrast, in the montane forest zones, relatively large tracts of land have remained primarily in the hands of a few private landowners. This trend continues today with the largest proportion of landowners in lowland agricultural areas owning 10-25 ha, with landholdings varying from less than 5 to 500 ha (Borgogno 2003). In contrast, many forested land holdings vary from 1000 to 5000 ha (Outon 2005). This situation is particularly relevant to livestock grazing management as in many cases landowners do not own livestock, and therefore rent their land on an annual or seasonal basis to livestock owners. In 1925, an irrigation dam, Dique La Cienaga, was established, which led to expanded cultivation of a variety of fruits and vegetables within the Jujuy Model Forest (Borgogno 2003). In the 1930s, Argentinean federal government policy led to the initiation of tobacco growing, and by the 1960s tobacco production dominated the agricultural landscape of the Jujuy Model Forest region (Outon 2002). This industry continues to dominate the regional landscape and economy in the present day, augmented by other agricultural crops. Impacts of tobacco production in adjacent forest ecosystems include extraction of firewood to fuel tobacco drying ovens and removal of the forest leaf litter layer as an organic amendment for tobacco seedlings. Although the majority of land within the Jujuy Model Forest region is privately owned, the periphery of irrigation dams and canals in the region is owned by the province of Jujuy (Outon 2002). In the 1970s, the provincial state expropriated land from a single farm owner to build an irrigation dam, "Las.Maderas". Many families who had occupied small tracts of land in this area for decades, with the permission of the previous landowner, subsequently found 8 themselves in a precarious land ownership situation. Although some individuals have lived in the zone of the Las Maderas dam for 40 to 90 years, and consider themselves rightful landowners, the province considers these families and individuals illegal land occupants. During the past two decades, there has been additional movement of families onto this provincial land. In some cases, people have been forced to leave land they occupied in other locations, and in other cases people occupy land beside the Las Maderas dam because they enjoy living in a rural landscape. Complex land use conflicts in the zone surrounding the Las Maderas dam currently pose many social, economic, and resource management challenges. Most of the families who occupy land adjacent to Las Maderas dam own livestock which they graze on the provincial lands surrounding the dam, and on adjacent privately owned forested land. Almost all of the approximately 40 families in this zone heat their houses and cook with firewood cut from provincial or privately owned forested land. 1.2.6. Socioeconomic Realities and Population Resource management decisions, including those pertaining to livestock grazing, are highly dependent upon the social, political and economic context in a region. Jujuy is one of the most impoverished provinces in Argentina, with the proportion of residents with unmet basic needs reaching 39 % in some zones of the Jujuy Model Forest region (INDEC 2001). Unemployment rates in the lowland valley region of Jujuy, which encompasses communities within the Jujuy Model Forest, was approximately 20% at the time of the last national census in 2002 (DiPPEC 2002). Widespread poverty affects the resiliency of households and communities to cope with change, underlining the importance that resource management plans encompass current social, economic, political, and biophysical realities. The population within the Jujuy Model Forest region is approximately 100,000 (IMFN 2007b) with 80% of this population found in the city of Perico and other urbanized towns, primarily in the lowland valleys to the east of the montane native forest (Outon 2002). Potable water, electricity, natural gas, and communications are generally available in urban areas; however, these services are not always within economic reach for families, particularly on the periphery of urban areas. Families who are not able to afford natural gas or propane for cooking and heating rely on firewood cut from forested areas to fulfill their heating needs; thus 9 economic challenges are tightly linked to forest resource use. In rural areas, many families do not have regular access to clean drinking water, and lack of adequate sanitation has been identified as a priority health concern by the local hospital in E l Carmen (C. Chauque, personal communication, 2004). The population of the Jujuy Model Forest region consists of a diverse mix of descendents of Spanish, Italian and other European and Middle Eastern immigrants, descendents of aboriginal inhabitants of the region, and Bolivian immigrants primarily of the Quechua culture. Spanish is the primary language in Jujuy, with some Bolivian immigrant associations working to maintain Quechua language and culture among Bolivian immigrants. The local economy is based primarily on tobacco production, with increasing emphasis on diversification into other agricultural crops such as beans, fruits, vegetables, sugar cane, and flowers. There is some regional tourism from nearby towns and cities to visit and fish in the Las Maderas and La Cienaga dams. In the lowland agricultural region there are a few family-run cattle and horse-breeding operations, where landowners are also livestock owners. Within the montane Yungas forest areas, livestock production on a subsistence basis is the predominant. economic activity. c 1.2.7. Current Livestock Grazing and Resource Use In the Jujuy Model Forest, extensive, season-long livestock grazing is one of the primary uses of the Yungas forest ecosystems (Outon 2002). Cattle grazing is the most common activity, with some goat, horse, sheep and pig grazing also occurring. Other local uses of the Yungas forest ecosystem include fuelwood harvesting, timber extraction, removal of the forest litter layer for use as a soil amendment, hunting and collection of edible and medicinal plants. Local resource managers are concerned that increasing livestock densities within forested areas, and burning of highland pastures to stimulate regrowth, have led to a variety of forest degradation processes (Outon 2002, Braun Wilke et al. 2001). Primary concerns include a lack of woody species regrowth, changes in forest structure and composition, soil erosion, and decreasing water quality due to sedimentation and nutrient loading (Outon 2002). However, there have been few formal studies to investigate these concerns in the Yungas forest of southern Jujuy. Potential social and economic implications of these ecological changes include 10 health risks due to reduced water quality, increased hydrological extremes (flooding during the summer and drier soil conditions during the winter), decreased fuelwood and forage production, reduced quality of livestock, and land use conflicts. During personal visits with livestock owners in the Las Maderas dam area of the Jujuy Model Forest from May - July, 2004, livestock owners told me about their concerns regarding 'the flowering of rocks' in the adjacent forested lands they used for grazing, caused by loss of vegetative cover and land degradation. They also discussed the problem of lack of forage during the dry winter period, which led to declines in animal weight and health, and in severe circumstances, livestock death. . Livestock management within the Jujuy Model Forest is complicated by a precarious land ownership situation in which many livestock owners do not own their own grazing land, and have annual agreements with owners of forested lands to graze their animals. Estimates of stocking rates for forest ecosystems vary between 4-8 ha/animal unit (Lamas et al. 2003). At a first glance, this fits relatively well within the estimation of Braun Wilke et al. (1995) that the grazing capacity of Yungas forest ecosystems varies from 1-5 ha / animal unit. However, many landowners report that although some livestock owners have formal land-use agreements, many others graze their animals clandestinely. This has led to highly variable livestock densities, severe overgrazing in some regions, and a lack of information regarding true livestock grazing pressure (Braun Wilke et al. 2001). Many livestock owners live on the periphery of forested areas, including approximately 40 families in the Las Maderas dam area who do not have tenancy of the land they live on. The 2002 provincial agricultural census and surveys indicate that livestock owners in the Jujuy Model Forest region own, on average, 50-80 cattle (Lamas et al. 2003). Grazing is primarily transhumant, with animals occupying lower elevation forest areas (down to approximately 1100 m) in the summer months (December until March) when seasonal rains stimulate forage production. During the dry winter season, animals move to higher elevations (up to -2000 m) in search of forage. Cattle grazed in the Jujuy Model Forest are predominantly of the Creole (criollo) race of European cattle (Bos taurus), which is a descendent of Spanish livestock introduced to Jujuy 11 during the 1500s (Martinez 2000). Over 400 years this breed of cattle has become adapted to "mountainous local conditions and is smaller, hardier, and often considered to be of lower quality for meat production than breeds found in other regions of Argentina (Lamas et al. 2003). 1.2.8. Grazing History in Yungas Forest Ecosystems The resiliency of ecosystems to disturbances caused by livestock grazing is often related to the type and density of native grazers and browsers that have caused a natural disturbance regime in the forest (Bengtsson et al. 2000, Milchunas & Lauenroth 1993). Dynamic interactions between herbivores and vegetation communities affect soil cover, physical properties, nutrient cycling, and soil organic matter inputs. As a result, an examination of the impact of livestock grazing on soil quality and forest structure in the Yungas must consider the historical role of native herbivores and the natural disturbance regime. The dense multi-story canopy of the lower elevation Yungas forest types results in low light levels in the forest understory, and corresponding low herbage production. Native terrestrial herbivores that consume vegetation in the, understory include a variety of small to medium sized mammals such as the brown brocket (Mazama gouazoubira), and Peruvian huemul (Hippocamelus antisensis) (Aprile 2003). These small members of the Cervidae family weigh between 15-50 kg, and are less than 1 m tall (Csomos 2001, Haralson 2004). Other similar sized herbivores include the pig-like white-lipped peccary (Tayassu pecari), and collared peccary (Pecari tajacu), and flightless bird, the Greater rhea (Rhea americana) (Aprile 2003, Braun Wilke et al. 1995). Herbivores in the Yungas are primarily browsers, with diets consisting mostly of fruits, nuts and leaves, supplemented by roots, twigs, fungi, flowers, buds, bark, cacti, and bromeliads (Csomos 2001, Haralson 2004).. Highland pastures of the Yungas, typically found at elevations of 1800 to 3500 m, consist of large tussock grasses interspersed with mountain cloud forest stands, and have a different historic grazing regime than lowland forests. The vicuna (Vicugna vicugna) (40 - 50 kg), and guanaco (Lama guanicoe) (50 - 100 kg) are members of the Camelid family that are native grazers in highland pastures of the Andes above approximately 3 000 m (Canevari and Fernandez Balboa 2003). Thousands of years ago, indigenous people domesticated these native grazers, forming the lama (Lama glama) and alpaca (Lama pacos). These two domesticated 12 I Camelids have been raised in large herds in the highland pastures and puna of the Andes for thousands of years, and therefore have played a role in the disturbance regime and evolution of native plant communities (Pucheta et al. 1998). Impacts of herbivores on soil properties in the forests and highland pastures of the Yungas include localized soil compaction from hooves, and possible horizon mixing and exposure of bare soil due to digging, foraging and grazing activities. In particular, the gregarious white-lipped peccary, native to the lowland Yungas forests, is known for leaving localized areas filled with holes and churned soil following as little as a few hours of group foraging (Csomos 2001). As with many mammals native to the Yungas, the white-lipped peccary has been extirpated from northern Argentina in the past century due to loss of habitat (Csomos 2001). Cattle were introduced to the Yungas region during the colonial period in the 1500s (Martinez 2000). Cattle and horses are much larger than native herbivores found in the Yungas, and therefore can potentially cause greater soil compaction and changes in hydrologic functioning compared to native herbivores. Cattle, horses, sheep, and pigs are primarily grazers, and most often prefer herbaceous vegetation over woody plant parts (Braun Wilke et al. 1995). This contrasts with the majority of native herbivores found in forested ecosystems of the Yungas, who are predominantly browsers. Although some native herbivores such as the white-lipped peccary can cause soil disturbance, their impacts are generally spatially localized and temporally sporadic (Csomos 2001). Current seasonal or year-long livestock grazing practices in the Yungas often do not give vegetation and soil resources time to recover between disturbance events. 1.3. Sustainable Livestock Grazing 1.3.1. Definition of Sustainable Livestock Grazing For the purpose of this thesis, livestock grazing is 'sustainable' i f it does not impair the structure or function of the natural ecosystem where livestock graze in the long term, contributes positively to social relationships and'the development of social capital in the community, and is economically viable. 13 1.3.2. Forage Production and Grazing Capacity In an extensive livestock grazing system, humans introduce their animals into an ecosystem so the animals can graze (consume herbaceous plants) or browse (consume the palatable portion of woody plants) to meet their energy and nutritional requirements. In the field of range management, stocking rate is defined as the amount of land allocated to each animal unit during the grazeable period of the year (Society for Range Management 1989). 'Animal unit month' (AUM) refers to the amount of forage consumed over the period of a month by "one mature 1000 lb. (450 kg) cow with or without an unweaned calf, or equivalent", based on the assumption that this animal consumes 26 lb. (11.8 kg) of dry matter per day (PFRA 2003). One A U M is therefore equivalent to the consumption of 355 kg of dry forage matter. The amount of forage consumed by an animal depends on the animal species, as well as the size and maturity of the animal. Animal unit equivalent charts are used to calculate animal units for livestock other than adult cattle. For example, a mature horse is considered to be 1.5 animal units, and 5 ewes are equivalent to 1 animal unit (PFRA 2003). . Grazing capacity refers to the maximum number of animals a given land area can support over the long term without causing ecosystem deterioration (Holechek et al. 1998). In order to predict the number of animals a given ecological site can support, the total annual forage production at the site must be known. Total annual forage production is the annual production of plant species that are edible for an identified kind of livestock (USDA NRCS 2003). Once the total annual forage production for a given site has been determined, it is necessary to estimate the 'safe use factor', which is the proportion of total forage production that can be used by livestock, without causing detrimental effects to the ecosystem. A portion of total forage production must be left for wild browsers and grazers, for maintenance and reproduction of forage plants, and for litter production and other ecosystem services (USDA NRCS 2003). 1.3.3. Rangeland Health Concept In the 1990s, the concept of rangeland health evolved from previous systems to evaluate range condition in which the composition of a vegetation community was compared to the climax community for a given range site (Pellant et al. 2005). The purpose of range health 14 assessment is to assist land and livestock managers to holistically evaluate the impact of livestock grazing on ecosystem processes, in efforts to improve livestock management practices (Pellant et al. 2005). Rangeland health is defined as "the degree to which the integrity of the soil, vegetation, water, and air, as well as the ecological processes of the rangeland ecosystem, are balanced and sustained" (Task Group on Unity in Concepts and Terminology 1995). In the discussion of rangeland health, integrity is the "maintenance of the functional attributes characteristic of a locale, including normal variability" (USDA, NRCS 2003). In assessing rangeland health for a given site where livestock grazing occurs, three interrelated attributes are examined and defined as follows (Pellant et al. 2005): . • soil / site stability - capacity of an area to limit redistribution and loss of soil resources by wind and water • hydrologic function - capacity of a location to capture, store, and safely release water, to resist a reduction in this capacity, and to recover this capacity when a reduction occurs • biotic integrity - capacity of the biotic community to support ecological processes within the normal range of variability, to resist a loss in the capacity to support these processes, and to recover this capacity when losses do occur. The broad, interrelated nature of these attributes, and the complexity of ecological processes, makes it necessary to use a set of indicators that can be measured in the field to assess rangeland health. Indicators are components of the rangeland system whose characteristics can be used as an index of one of the rangeland health attributes (Pellant et al. 2005). A set of multiple quantitative indicators should be selected to evaluate the three range health attributes at a given site. The concept of rangeland health served as an important guiding set of principles as we examined the sustainability of livestock grazing management practices in ecosystems of the Jujuy Model Forest. Soil quality assessment is an integral component of a rangeland health evaluation. 15 1.4. Soil Quality Although recently there has been increasing discussion and measurement of soil quality in the fields of agriculture and resource management (Doran and Parkin 1994; Karlen et al. 1997; Karlen et al. 2003), the notion of 'soil quality' is not new: people have discussed and evaluated the characteristics of the soils they use for millennia (Carter et al. 1997). Increasing awareness of global patterns of soil degradation in the latter half of the 20 t h century prompted the soil science community to more formally define the concept of soil quality in the 1990s (Doran and Parkin 1994). Many scientists and resource managers felt that a clear definition of soil quality could improve soil management efforts by encouraging a more holistic, integrative concept of soil evaluation (Herrick et al. 2002). A commonly accepted definition of soil quality is "the capacity of a specific kind of soil to function, within natural or managed ecosystem boundaries, to sustain plant and animal productivity, maintain or enhance water and air quality, and support human health and habitation" (Karlen et al. 1997; SSSA 1997). Within the Yungas forest ecosystem, critical soil functions include provision of a rooting medium for diverse subtropical forest associations and their fauna, regulation of the hydrological cycle through partitioning of water into surface and subsurface flow, and provision of food, fuel and timber for local human populations. Implicit in any definition of soil quality is an underlying idea that value is placed upon a soil in relation to the desired functions of the soil, which depend upon human perception, and the context in which the soil is used (Carter et al. 1997; Schoenholtz et al. 2000). Soil properties that may confer high value for one given land use (e.g. high clay content to prevent leaching of contaminants at a waste storage site), may reduce the quality of the soil for a differing use (e.g. agriculture). This has caused considerable debate within the field of soil science, with some feeling that the concept of soil quality is too subjective, and value-laden (Sojka and Upchurch 1999). Others feel that despite its limitations, soil quality is a useful tool for enhancing communication about soil resources and sustainable management practices (Karlen et al. 1997). Because of the diverse array of uses a given soil can be used for, it is important that the concept of soil quality be considered relational rather than absolute (Karlen et al. 1997). Sojka and Upchurch (1999) caution that soil quality at a particular site should be 16 defined in terms of distinct management and environmental circumstances specific to the soil being evaluated, with social, economic, biological and other value judgments clearly stated. Assessment of the ability of a soil to carry out a desired set of functions is dependent upon a complex and dynamic array of physical, chemical and biological processes, and their variation in time and space (Larson and Pierce 1991). Examples of such soil processes are the decomposition of organic matter, leaching of nutrients, and infiltration and percolation of water. It is often difficult to measure such processes directly, and therefore indicators that relate information about soil processes must be used. A soil quality indicator is a measurable soil property, which acts as a surrogate of a soil attribute, and determines how well a soil functions (Doran and Parkin 1994). Effective soil quality indicators integrate soil physical, chemical and biological properties and processes, are easily measured and verifiable, and are sensitive to variations in management and climate (Doran and Parkin 1994; Carter et al. 1997). Measurement of soil indicators in a given ecosystem or agro-ecosystem over time can provide a useful tool for evaluating the sustainability of land management and agricultural systems. 1.5. Community-Based Action Research Approach Investigating the sustainability of a livestock grazing system involves examining human livelihoods, social and power dynamics, and economic realities, in addition to ecological processes. It is most often the people who carry out their daily lives at the heart of a resource management challenge or conflict who can shed the most light on reasons for conflict, tension over resource sharing, and resource scarcity. It is also these people who have the most to gain through equitable resolution of the challenge, or lose, should tensions intensify or resources become more degraded. In contrast to traditional research where there is a strong separation between 'researcher' and 'research subject', action research seeks to engage those people who have direct involvement in the topic of research to be active participants in the search for solutions to the research question of concern to themselves and their community. As outlined by Stringer (1999), community-based action research (CBAR) favours participatory procedures that enable people to: 17 • investigate systematically their problems and issues • • form powerful and detailed accounts of their situation • create plans to deal with the problems at hand The basic format for C B A R is an iterative cycle described as Took, think, act', whereby community members: (1) gather information, and describe their situation; (2) explore and interpret what is occurring; (3) plan, implement and evaluate (Stringer 1999). A central premise of C B A R within the context of natural resource management is that research processes must value and respect the traditional and local knowledge of resource users (Tyler 2006). In a review of a 7-year program focusing on the use of C B A R approaches in Asia, Tyler (2006) showed the ability of this research style to strengthen livelihoods, build capacity in local decision-making and spark policy change. Community-based action research has also been promoted as a valuable tool in building community capacity to resolve challenges in watershed management (FAO 2006). As a research facilitator, it was important for me to examine and acknowledge my personal paradigms, as these influenced my approach to research, and my interactions with community members. I believe that there are many ways of learning about and understanding the world. I value the process of formal scientific inquiry, and at the same time I believe that traditional knowledge and interpretation of the dynamics of local ecosystems are extremely vafuable. Growing up in a small, agriculture-based town in southern Alberta, and spending time on family members' farms has led me to believe strongly in the importance of supporting rural livelihoods. As a research facilitator within a culture and community that was not where I was originally from, it was of utmost importance to me that the decision-making processes regarding land and livestock management clearly remain in the hands of local community members with whom I was collaborating. 18 1.6. Research Objectives Although natural resource managers.and ecologists have expressed concerns regarding the impacts of livestock grazing on the subtropical Yungas forest ecosystem in general (Braun Wilke et al. 2000, Brown et al. 2001), and the Jujuy Model Forest in particular (Outon 2002), there have been no studies carried out to determine how many domestic animals this forest ecosystem can support on a long-term basis without becoming degraded. As a contribution towards assisting livestock owners and landowners to move towards more sustainable livestock grazing practices in the Jujuy Model Forest, this research project seeks to: 1. Determine the annual forage production and recommend stocking rates for deciduous forest, anthropogenic pasture, and highland pasture ecological zones 2. Evaluate the impact of livestock grazing on selected physical and chemical soil quality indicators, and forest structure parameters 3. Describe opportunities and strengths within local communities that will support movement towards more sustainable livestock grazing practices, and outline challenges that may pose barriers to such change 4. Carry out these research goals in a participatory manner that involves local community members and students directly in research and educational activities 19 Table 1.1 Elevation, landscape position, annual precipitation and vegetation for Yungas forest ecological zones in northwestern Argentina Yungas Forest Zone Elevation Range (m) Annual Landscape Position Precipitation (mm) Vegetation Yungas-Chaco transition forest (Selva de transition / Selva pedemontana) 350/500 700/1000 Base of the Andes; forms transition zone with dry Chaco forest to east 7 0 0 - 1000 Multi-canopy rainforest with many vines and epiphytes; 70% deciduous trees Mountain rainforest (Selva montana) 600/800 1200/1500 M i d - and low-mountain slopes 1000-2000 ; increases • with altitude Dense, humid, multi-canopy rainforest of deciduous and evergreen species Mountain cloud forest (Bosque montano) 1500 2000/2500 Upper mountain slopes to altitudinal tree line 1 0 0 0 - 1500; in some areas fog present most o f the year Open mixed forest and single-species stands of deciduous and evergreen trees; dense epiphyte, orchid and fern growth on tree trunks Highland pastures (Pastizales andinos) 1800/2000 3000/3500 Mountain slopes above mountain cloud forest zone -400 - 700;' fog is an important additional source of moisture Pasturelands of large tussock grasses and sedges; can be interspersed with mountain cloud forest stands (Data f rom B r a u n W i l k e et al. 2001 , B r o w n et al. 2001). 20 Table 1.2 Elevation, slope and dominant vegetation for three ecological zones in the Yungas forest of southern Jujuy, Argentina. Ecological Zone Elevation (m) Slope at Field Sites (°) Dominant vegetation1 Deciduous forest 1200-1500 18-37 Trees: cebil Colorado (Anadenanthera colubrina ("Veil.') Brenan), horco cebil (Parapiptadenia excelsa (Griseb.) Burkart), urundel (Astronium urundeuva Engl.), duraznillo bianco (Ruprechtia apetala Weddell), zapallo caspi (Pisonia . zapallo Griseb.), chai chai (Allophylus edulis (A. St.-Hil.) Niederl., piquillin (Condalia buxifolia Reiss.), and quebracho (Schinopsis spp. Engl.) Shrubs: garabato (Acacia praecox Griseb.). ortiguilla (Urtica urens L.), clavillo (Barnadesia odorata Griseb.), Eupatorium spp. Herbs: pasto bianco (Panicum trichoides Sw.) Anthropogenic pasture 1200-1450 13-30 Trees: churqui (Acacia caven (Molina) Molina), tusca (Acacia aroma Gillies ex Hook. & Arn.), lecheron (Sapium haematospermum Mull. Arg.) Shrubs: rosa de monte (Eupatorium spp.), clavillo (Barnadesia odorata Griseb.), romerillo (Baccharis coridifolia DC), carqueja (Baccharis pingraea DC ), Hyptis spp. Jacq. Herbs: Paspalum spp., Setaria spp., Cvnodon spp., pasto bianco (Panicum trichoides Sw.) Highland pasture 1650- 1775 18-33 Shrubs: Baccharis spp. L. , Eupatorium spp., Stevia spp. Large bunchgrasses: sivinguilla (Lamprothvrsus hieronvmi (Kuntze) Pile.), guailla (Deyeuxia spp.) Other common grasses: Festuca spp., Stipa spp., Chloris spp. 1. Botanical nomenclature is according to the International Plant Names Index (2004). Table 1.3 Characteristics of soil subgroups found at field sites in the Jujuy Model Forest, Argentina Soil Subgroup1 Typic Haplustalf Ultic Haplustalf Lithic Ustorthent U.S. Taxonomic Order Alfisol Alfisol Entisol Canadian Taxonomic Equivalent2 Luvisol Luvisol Regosol Typical Horizon Sequence LFH, Ah, Btl ,Bt2,C LFH, Ah, Btl,Bt2, C LFH, Ahe, BC, R Key Characteristics • Freely drained soils • Argillic horizon • Moderate to high base saturation • Moderately deep or deep soil to hard rock • Common on relatively recent erosional surfaces • Similar to the Typic Haplustalf subgroup, but with moderately low base saturation throughout the argillic horizon • Horizon development limited by lithic contact within 50 cm of the soil surface • Most often found under deciduous forest and savanna vegetation on relatively recent erosional surfaces 1 According to USDA Soil Taxonomy (Soil Survey Staff 1999), which is widely used to classify soils in Argentina. 2 According to The Canadian System of Soil Classification (Soil Classification Working Group 1998) 22 This material has been removed due to copyright restrictions. This page contained a series of three maps depicting the location of the study area: A map of Argentina within South America, a map of Jujuy within Argentina, and a map of the study area within Jujuy. Figure 1.1 Location of the Jujuy Model Forest, within Jujuy province, northwestern Argentina, South America Maps adapted from Direction de Bosques, Secretaria de Ambiente y Desarrollo Sustentable (2003). 23 1.7. References Aprile, G. 2003. Fauna silvestre de la cuenca Los Pericos-Manantiales. In P. Laclau (ED.). Manejo sustentable de ecosistemas forestales de la cuenca Los Pericos-Manantiales: Informe diagnostico. Food and Agricultural Organization of the United Nations Project TCP/ARG/2802. Bengtsson, J., S.G. Nilsson, A . Franc and P. Menozzi. 2000. Biodiversity, disturbances, ecosystem function and management of European forests. Forest Ecology and Management 132: 39-50. Borgogno, C. 2003. Informe social. In P. Laclau (ED.). Manejo sustentable de ecosistemas forestales de la cuenca Los Pericos-Manantiales Informe Diagnostico. Buenos Aires, Argentina: Direction de Bosques - Secretaria de Desarrollo Sustentable y Politica Ambiental de la Nation. 123 pp. Braun Wilke, R., and L. Picchetti. 2006. Ecologia humana. In: Vision interdisciplinaria del ambiente. San Salvador de Jujuy, Argentina: Fundacion A V E S . 51 pp. Braun Wilke, R.H., E.E. Santos, L.P.E. Picchetti, M.T. Larran, G.F. Guzman, and C R . Colarich. 2001. Carta de aptitud ambiental de la provincia de Jujuy. San Salvador de Jujuy, Argentina: Departamento de Suelos y Ecologia, Facultad de Ciencias Agrarias, Universidad Nacional de Jujuy. 245 pp. Braun Wilke, R.H., B.S. Villafane, and L.P.E. Picchetti. 1995. Plantas de interes ganadero de Jujuy y Salta Noroeste Argentine San Salvador de Jujuy, Argentina: Secretaria de Ciencia, Tecnica y Estudios Regionales and Universidad Nacional de Jujuy. 309 pp. Brown, A .D . and H.R. Grau. 1993. La naturaleza y el hombre en las selvas de montafia. Salta, Argentina: Proyecto GTZ - Desarrollo Agroforestal en Comunidades Rurales del Noroeste Argentina. 143 pp. Brown, A.D. , H.R. Grau, L.R. Malizia and A. Grau. 2001. Argentina. In: M . Kappelle and A.D. Brown [EDS]. Bosques nublados del neotropico. Santo Domingo de Heredia, Costa Rica: Instituto Nacional de Biodiversidad. p. 623-659. Cabrera, A . L . 1976. Enciclopedia argentina de agricultura y jardineria: Fasciculo 1 - Regiones fitogeograficas argentinas. Buenos Aires, Argentina: Editorial Acme S.A.C.I. 82 p. 24 Canadian Model Forest Network. 2007. "About Model Forests. Available at: http://www.modelforest.net/cm Accessed 4 May 2007. Canevari, M . , and C. Fernandez Balboa. 2003. 100 Mamiferos argentinos. Buenos Aires, Argentina: Editorial Albatros. 159 pp. Carter, M.R., E.G. Gregorich, D.W. Anderson, J.W. Doran, and H.H. Janzen. 1997. Concepts of soil quality and their significance. In: E.G. Gregorich and M.R. Carter [EDS.]. Soil quality for crop production and ecosystem health. Developments in Soil Science Series #25. Amsterdam, Holland: Elsevier, p. 1-18. Chauque, C. 2004. Nurse, Hospital Nuestra Sefiora del Carmen, Volunteer with the Jujuy Model Forest Association. Personal Communication. July 2004. Csomos, R. 2001. Tayassu pecari. Animal Diversity Web. Available at: http://animaldiversitv.ummz.umich.edu/site/accounts/information/Tayassu_pecari.html. Accessed 15 April 2005. Direction Provincial de Planeamiento, Estadistica y Censos (Jujuy) (DiPPEC). 2002. Tasa bruta de participacion en la actividad economica, tasa de desocupacion abierta, tasa de demandantes de empleo ocupados y tasa de subocupacion horaria, por sexo. Available at: http ://www. dippec. i u j uy. go v. ar/. Accessed 29 May 2007. Direction de Bosques Nativos. 2003. Atlas de los bosques nativos argentinos. Available at: http://www2.medioambiente.gov.ar/bosques/bosques nativos/2004 atlas/default.htm Accessed 20 May 2007. Doran, J.W., and T.B. Parkin. 1994. Defining and assessing soil quality. In J.W. Doran, D.C. Coleman, D.F. Bezdicek, and B.A. Stewart [EDS.]. Defining soil quality for a sustainable environment. SSSA Special Publication Number 35. Madison, WI: SSSA. Food & Agriculture Organization of the United Nations (FAO). 2006. The new generation of watershed management programmes and projects. FAO Forestry Paper 150. Rome, Italy: FAO. 128 pp. Grau, A. 1985. La expansion del Aliso del Cerro (Alnus acuminata H.B.K. subsp> Acuminata) en el noroeste de Argentina. Universidad Nacional de Tucuman. Lilloa 36 (2). Grau, H.R. 2002. Ecologia del fuego en los ecosistemas montanos superiores del subtropico. In: C. R. Kunst, S. Bravo and J.L. Panigatti [EDS.]. Fuego en los ecosistemas 25 argentinos. Santiago del Estero, Argentina: Instituto Nacional de Tecnologia Agropecuaria. p. 181-188. Haralson, C. 2004. Mazama gouazoupira. Animal Diversity Web. Available at: http://animaldiversity.ummz.umich.edu/site/accounts/informatioi-i/Tavassu pecari.html. Accessed 15 April 2005. Hensen, I. 2002. Impacts of anthropogenic activity on the vegetation of Polylepis woodlands in the region of Cochabamba, Bolivia. Ecotropica 8:183-203. Herrick, J.E., J.R. Brown, A.J . Tugel, P.L. Shaver, and K . M . Havstad. 2002. Application of soil quality to monitoring and management: Paradigms from rangeland ecology. Agronomy Journal 94:3-11. Holechek, J.L., R.D. Pieper and C H . Herbel. 1998. Range management: Principles and practices. Prentice Hall, Upper Saddle River, NJ. 542 pp. Instituto Nacional de Estadistica y Censos (INDEC). 2001. Provincia de Jujuy segun departamento: Hogares y poblacion - Totaly con necesidades bdsicas insatisfechas. Ano 2001. Buenos Aires, Argentina: INDEC. Available at: www.indec.gov.ar/webcenso/provincias/JUJUY/JUJTJY.asp. Accessed 30 May 2007. International Model Forest Network. 2007a. About the International Model Forest Network. Available at: http://www.idrc.ca/imfn/ev-22893-201-l-DO TOPIC.html. Accessed 4 May 2007. International Model Forest Network. 2007b. Jujuy Model Forest. Available at: http://www.idrc.ca/en/ev-23367-201-l-DO TOPIC.html. Accessed 16May 2007. International Plant Names Index. 2004. The International Plant Names Index. Available at: http://www.ipni.or g/index .html. Accessed 15 January 2007. Karlen, D.L., M.J . Mausbach, J.W. Doran, R.G. Cline, R.F. Harris and G.E. Schuman. 1997. Soil quality: A concept, definition, and framework for evaluation (A guest editorial). Soil Science Society of America Journal 61: 4-10. Karlen, D.L., C A . Ditzler, and S.S. Andrews. 2003. Soil quality: why and how? Geoderma 114: 145-156. Lamas, H.E., C. Borgogno, and M . del Carpio. 2003. Informe sobre la situation actual y perspectivas de los agrosistemas de la cuenca Los Pericos-Manantiales. In P. Laclau (ED.). Manejo sustentable de ecosistemas forestales de la cuenca Los Pericos-Manantiales Informe 26 Diagnostico. Buenos Aires, Argentina: Direccion de Bosques - Secretaria de Desarrollo Sustentable y Politica Ambiental de la Nacion. 73 pp. Larson, W.E. and F J . Pierce. 1994. The dynamics of soil quality as a measure of sustainable management. In J.W. Doran, D.C. Coleman, D.F. Bezdicek, and B.A. Stewart [EDS.]. Defining soil quality for a sustainable environment. SSSA Special Publication Number 35. Madison, WI: SSSA. Martinez, R.D., E.N. Fernandez, E.R. Genero, and F.J.L. Rumiano. 2000. E l ganado bovino criollo en Argentina. Archivos de Zootecnia 49: 353-361. Milchunas, D.G. and W.K. Lauenroth. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecological Monographs 63: 327-366. Nadir, A. , and T. Chafatinos. 1990. Los suelos del Noroeste Argentino (Salta y Jujuy). Vol . 3. Salta, Argentina: Universidad Nacional de Salta / Direccion General Agropecuaria Salta. p. 283-284. Outon, V . 2002. Bosque Modelo Jujuy: Gestion integral de la cuenca hidrografica -propuesta. E l Carmen, Argentina. 41 p. Outon, V . 2005. President of the Jujuy Model Forest Association. Personal communication. February, 2005. Pellant, M . , P. Shaver, D.A. Pyke, and J.E. Herrick. 2005. Interpreting indicators of rangeland health. Technical Reference 1734-6, Version 4. Denver, CO: United States Department of the Interior, Bureau of Land Management, National Science and Technology Centre. 122p. Prairie Farm Rehabilitation Administration. 2003. Animal unit months, stocking rate and carrying capacity. Available at: http://www.agr.gc.ca/pfra/land/fftl.htm. Accessed 27 May 2007. Pucheta, E., M . Cabido, S. Diaz, and G. Funes. 1998. Floristic composition, biomass, and aboveground net plant production in grazed and protected sites in a mountain grassland of central Argentina. Acta Oecologia 19(2): 97-105. Rutledge, I. 1996. De la sociedad indigena a la independencia de Espana: un modelo historico de formation de clases sociales en el noroeste. In: M . Manzanal [ED.]. E l desarrollo rural en el noroeste argentino. Salta, Argentina: Proyecto Desarrollo Agroforestal en 27 Comunidades Rurales del Noroeste Argentino. p. 45-51. Schoenholtz, S.H., H . Van Miegroet and L A . Burger. 2000. A review of chemical and physical properties as indicators of forest soil quality: challenges and opportunities. Forest Ecology and Management 138:335-356. Society for Range Management. 1989. A glossary of terms used in range management. 3 r d Ed. Denver, CO - Society for Range Management. Sojka, R.E., and D.R. Upchurch. 1999. Reservations regarding the soil quality concept. Soil Science Society of America Journal 63:1039-1054. Soil Science Society of America. 1997. Glossary of soil science terms. Madison, WI: SSSA. Soil Survey Staff. 1999. Soil Taxonomy: A basic system of soil classification for making and interpreting soil surveys. 2nd ed. Washington, D.C.rUnited States Department of Agriculture, Natural Resources Conservation Service. 869 p. Steinfeld, H. , C. de Haan, and H . Blackburn. 2007. Livestock - environment interactions: Issues and options. Food and Agricultural Organization of the United Nations. Available at: http://www.fao.org/ag/aga/lspayLXEHTML/policv/index.htm. Accessed 15 April 2007. Stringer, E.T. 1999. Action research. Thousand Oaks, C A : Sage Publications. 229 pp. Task Group on Unity in Concepts and Terminology. 1995. New concepts for assessment of rangeland condition. Journal of Range Management 48: 271-282. Tyler, S.R. 2006. Community based natural resource management: A research approach to rural poverty and environmental degradation. In S.R. Tyler [ED.]. Communities, livelihoods and natural resources: Action research and policy change in Asia. Ottawa, ON: International Development Research Centre. 456 pp. USDA-NRCS. 2003. National range and pasture handbook. Grazing Lands Technology Institute. Available at: http://www.glti.nrcs.usda.gov/technical/publications/nrph.html Accessed J5 April 2007. Vargas Gi l , J.R. 1990. Jujuy. In: G. Moscatelli [ED.]. Atlas de suelos de la Repiiblica Argentina. Buenos Aires, Argentina: Instituto Nacional de Tecnologia Agropecuaria. p. 686-731. Ventura, B .N. 1999. Los ultimos mil afios en la arqueologia de las Yungas. In E.E. Berberian and A.E . Nielsen [EDS.]. Historia Argentina prehispanica: Tomo I. Argentina: 28 Editorial Brujas. p.447-492. Vervoorst, F. 1982. Noroeste. In: Conservation de la vegetation natural en la Republica Argentina. Serie Conservation de la Naturaleza. Tucuman, Argentina: Fundacion Miguel Lillo. p. 9-24. [Cited in: Braun Wilke et al. 2001]. Yacobaccio, H.D., P.S. Escola, F .X. Pereyra, M . Lazzari and M.D. Glascock. 2004. Quest for ancient routes: Obsidian sourcing research in Northwestern Argentina. Journal of Archaeological Science 31: 193-204. \ 29 2. IMPACTS OF LIVESTOCK GRAZING ON SOIL QUALITY AND FOREST STRUCTURE IN YUNGAS FOREST ECOSYSTEMS OF SOUTHERN JUJUY, ARGENTINA1 2.1. Introduction Expansion of cultivated agricultural land and rapid human population growth in subtropical and tropical regions around the world have forced many livestock owners to move their grazing animals to more marginal lands, often found in forested regions. Extensive, season-long domestic livestock grazing has been identified as one of the principal uses of Yungas forest ecosystems located at the base of the Andes mountains in Jujuy province, northwestern Argentina (Braun Wilke et al. 2000). Livestock grazing occurs in a variety of ecological zones within the Yungas, including deciduous Yungas-Chaco transition forest, anthropogenic pastures, where deciduous forest has been cleared, and native highland pastures dominated by tussock grasses at elevations above the deciduous forest. Natural resource managers and ecologists have expressed concerns that overgrazing by domestic livestock is negatively impacting soil and water quality, and forest composition and structure in the Yungas forest (Braun Wilke et al. 2000; Grau and Brown 2000). Subsistence-scale livestock owners in southern Jujuy report that some cattle perish annually due to inadequate native forage availability during the dry season. There have been no formal studies to determine the carrying capacity and appropriate stocking rates within Yungas forest ecosystems of Jujuy. Excessive stocking rates on a range site over the long-term can lead to declines in forage production and ecosystem functioning (Martinez and Zinck 2004). Annual forage production is often included in assessments of rangeland health, along with indicators of hydrologic functioning, biotic integrity, and soil quality / site stability (Herrick et al. 2002). The amount and distribution of bare soil are among the most important factors influencing site stability, since an increase in bare ground increases soil susceptibility to water and wind erosion (Wischmeier and Smith 1978; Blackburn and Pierson 1994; Gutierrez and Hernandez 1996). Soil litter plays a critical role in site stability by slowing overland flow, 1 A version of this chapter will be submitted for publication to Rangeland Ecology and Management. Ripley, S.W., M. Krzic, A.A. Bomke, and G.E. Bradfield. Impacts of livestock grazing on soil quality and forest structure in Yungas forest ecosystems of southern Jujuy, Argentina. 30 promoting water infiltration, serving as a source of soil nutrients and organic matter, and protecting the soil surface from the erosive force of raindrops (Wood et al. 1989; Belsky and Blumenthal 1997). Livestock grazing can therefore reduce site stability i f it leads to litter decline and increased bare soil (Chartier and Rostagno 2006). Tree removal from forested ecosystems has the potential to reduce soil cover by reducing the annual addition of leaves and tree biomass to the litter layer, and through disturbance of the litter layer during tree harvesting (Page-Dumroese et al. 2000). Organic C and total, organic or mineralizable N are often cited as baseline chemical indicators of soil quality (Schoenholtz et al. 2000; Doran and Parkin 1994). Increased levels of organic C tend to improve soil aggregate stability, porosity, gas exchange, infiltration, water-holding capacity, cation exchange capacity and microbial activity, and reduce erosion (Weil and Magdoff 2004). Consumption of plant biomass by livestock can reduce soil litter production, and therefore lower soil organic matter (Heady and Child 1994), which is the principal source of soil organic C and N . However, when livestock grazing does not lead to declines in the thickness of the soil litter layer, organic C and total N do not decline (Krzic et al. 2001, 2003). Other activities, such as timber and firewood harvesting, which remove vegetation biomass and reduce litter production, can also lead to declines in organic C and N (Thompson et al. 2000). Soil compaction is of particular interest in the Yungas forest characterized by a monsoon climate, since compaction generally leads to reduction of soil macropores, which in turn decreases water infiltration and hydraulic conductivity (Greacen and Sands 1980; Ballard 2000). The degree of soil compaction that occurs at a site depends on soil texture, water content, organic matter (Van Haveren 1983), as well as livestock type, grazing duration and grazing intensity. This helps explain why some studies found that livestock grazing increased soil compaction (as measured by bulk density) (Bezkorowajnyj et al. 1993; Greenwood et al. 1997), while others reported no significant impacts of livestock grazing on bulk density (Abdel-Magid etal. 1987, Krzic et al. 1999). Soil penetration resistance is often a more sensitive indicator of soil compaction than bulk density (Krzic et al. 1999; Rodd et al. 1999; Chanasyk and Naeth 1995). Vertical vegetation structure and relative dominance of vegetation functional groups are important indicators of biotic integrity in grazed ecosystems (Pellant et al. 2005). In forest 31 ecosystems, heavy livestock grazing can cause trampling damage to seedlings, and browsing of woody plants and seedlings (Mayer et al. 2006, Hensen 2002), which can lead to changes in forest composition and structure (Hensen 2002; Pellant et al. 2005). At moderate stocking rates, forest grazing may lead to increased overall plant diversity (Mayer et al. 2006, Krzic et al. 2003). Selective timber and firewood harvesting are also important factors in forest structural and functional change within the Yungas forest (Brown et al. 2001). In 2002, the Jujuy Model Forest Association was established to. work towards the long-term sustainable management of 1300 km of the Los Pencos-Manantiales watershed located in the southern part of Jujuy province, northwestern Argentina (Figure 2.1). This study was carried out in cooperation with the Jujuy Model Forest Association. The study objectives were: (1) to determine annual forage production and recommend stocking rates for deciduous forest, anthropogenic pasture, and highland pasture ecological zones of the Yungas forest, (2) to evaluate the impact of livestock grazing, on selected physical and chemical soil quality indicators, and forest structure parameters. The hypotheses were that there is currently greater annual forage production, and potential stocking rates, in the two pasture ecosystems that are not shaded by tree canopy. We also hypothesized that the anthropogenic pasture has less soil cover, greater compaction, and lower levels of nutrient cycling than the deciduous forest and highland pasture. 2.2. Materials and Methods 2.2.1. Study Sites The study was carried out in cooperation with local landowners and livestock owners at 18 sites in the biogeoclimatic zones of Yungas / Chaco transition forest and montane forest (Cabrera 1976) in southern Jujuy province, northwestern Argentina. Field sites were located in the foothills of the Andes mountains, within the Los Pericos-Manantiales watershed of the Jujuy Model Forest, and extended from 24° 24' - 24° 30' S latitude and 65° 17' - 65° 21' W longitude (Appendix III). Yungas forest ecosystems of southern Jujuy are characterized by a subtropical montane climate. The average December-February (summer) temperature in our study area is 21°C, while the average June-August (winter) temperature is 11°C (Braun Wilke et al. 2001). 32 Average annual precipitation is 780 mm; however, the study area is located along a strong precipitation gradient with the Las Maderas meteorological station just to the east of the study area reporting an average annual precipitation of 621 mm, and the San Antonio meteorological station just to the west of the study area reporting 943 mm (Braun Wilke et al. 2001). This region has a monsoon precipitation regime with approximately 90% of precipitation falling in the rainy summer season between November and March (Braun Wilke et al. 2001). Typical soils in the study area have formed from sedimentary (sandstone and siltstone) and metamorphic (shale, slate and quartzite) parent materials on moderate to steep slopes (13-37°) on mountain foothills. Soils are characterized by the presence of argillic horizons, loam to sandy loam surface textures, moderate to high base saturation (90%) (Nadir and Chafatinos 1990) and are slight to very acidic (pH 4.9 - 6.0). Soils are classified as members of the Haplustalf Great Group (Luvisolic Order, Canadian system), with some more shallow and weakly developed soils on steep slopes classified as part of the Ustorthent Great Group (Regosolic Order, Canadian system) (Soil Survey Staff 1999; Soil Classification Working Group 1998). Six sites were located, using a randomized draw of pre-selected site locations, within representative areas of three ecological zones currently grazed by livestock: deciduous forest, anthropogenic pasture, and highland pasture. Domestic livestock grazing has occurred in the region since the 1600s; however, grazing pressure has increased since the 1970s. Deciduous forest sites (1200 - 1500 m elevation) are composed of secondary growth forest as extensive logging was carried out in this region in the mid 1900s through to the 1970s, with particularly intense selective logging around 1975 (Braun Wilke et al. 2001). Tree diversity is high in this ecotone region, with shrub and herbaceous strata found in the understory. Current land uses include livestock grazing, hunting, firewood cutting, and medicinal / edible plant gathering. The anthropogenic pasture ecological zone (1200 - 1450 m elevation) is a mosaic of areas at the base of the foothills that have been converted from deciduous forest to open savanna landscape through intensification of timber extraction, firewood harvesting, and livestock grazing since the 1970s. Grasses, forbs, and shrub species that are resistant to grazing pressure dominate the open slopes of anthropogenic pasture sites, and are interspersed with scattered stands of trees or single trees. Current land uses include heavy livestock grazing pressure, timber / firewood cutting, and 33 recreation. Highland pasture sites (1650 - 1775 m elevation) are open grass / shrublands found at elevations above the deciduous forest on exposed north and northwesterly facing slopes. These sites are characterized by large tussock grasses up to 1 m tall, that have rigid, sclerenchymous leaves, and are interspersed with shorter grasses and forbs, and patches of bare ground. Livestock grazing is the predominant land use in this ecological zone, and many highland pastures are periodically burned by local residents at the end of the dry season to stimulate regrowth. 2.2.2. Sampling and Analyses 2.2.2.1. Forage production and forest structure In October 2005, prior to the start of the rainy growing season, one 3 x 3 m fenced grazing exclosure (1.5 m in height) was installed at each study site. The locations of study sites and grazing exclosures were chosen using a randomized draw of pre-selected site locations. A combination of chicken wire, smooth and barbed wire was used to prevent the entry of all domestic animals. In April 2006, following the rainy growing season, annual forage production measurements were carried out. Herbaceous and short shrub (< 50 cm tall) annual biomass was collected within five 50 x 50 cm quadrats at locations inside the exclosure that were chosen using a random number draw. Within each quadrat sampled, biomass was clipped to ground level, and separated into the following categories: (1) grass / grass-like species, (2) forbs, (3) current annual growth of shrubs / saplings < 50 cm tall, and (4) species known to be unpalatable and/or poisonous for at least one domestic livestock species (cows, horses, sheep, and goats). A l l biomass samples were oven-dried at 75°C for 48 hours, and weighed. Tall shrubs and saplings were defined as woody plants that had a height > 50 cm, and a diameter at breast-height (dbh) < 10 cm. Annual forage production of tall shrubs and saplings was estimated at each study site within four 3 x 5 m quadrats that were placed within an 18 m radius outside the exclosure, at locations chosen using a random number draw. Annual biomass production of woody plants was estimated up to a height of 1.5 m, as few domestic animals access forage above this height. Within each sampling quadrat, individual woody species were differentiated. For each species, a 'sample unit' (e.g. one new twig with 10 leaves) of the current 34 year's growth was clipped, and placed in a paper bag. Using the clipped sample unit as a guide, the number of sample units / species within the quadrat were visually estimated and recorded. The weight of each sample unit was multiplied by the estimated number of sample units in each quadrat to calculate annual forage production for each species. Forage production of trees at each study site was measured within a 1000 m 2 circle around each grazing exclosure. Tree forage production was defined as leaves and current annual growth of twigs found below a height of 1.5 m on trees that had a dbh of > 10 cm. The same 'sample unit' method described above was used to estimate annual forage production of trees. Forest structure and regeneration were measured within the 1000 m 2 circle by identifying and recording the dbh of all trees with a dbh > 10 cm, and counting the number of saplings (dbh < 10 cm and height > 50 cm), without identifying saplings by species. These data were used to calculate the basal area (cross-sectional area of tree stems at 1.3 m height), density (number of stems/ha), frequency (number of field sites where a species was found) and importance value (IV) for each tree species by ecological zone. Importance value is widely used to assess the biological contribution of arborescent species to a forest community, where IV = relative basal area coverage + relative density + relative frequency (Curtis and Mcintosh 1951). Canopy cover was measured between 9 - 3 0 November, 2005, with a sighting tube and internal crosshair densiometer at 10 points along each of 4 transects that extended 18 m from the corners of each grazing exclosure (Appendix IV). 2.2.2.2. Soil sampling The majority of soil sampling was carried out at the end of the dry season from November 9 - 30, 2005. Samples were collected along 4 transects, each 18 m in length, that extended out from the corners of the grazing exclosure at a 45° angle (Appendix IV). Percent bare soil (exposed soil that was not covered with litter, rocks, manure, or live vegetation), litter, manure, and rock cover were visually estimated within a Daubenmire frame (20 x 50 cm) at 5 points along each of the 4 transects. In addition, at each sampling point, the depth of the L F H horizon was measured at the 4 corners of the Daubenmire frame. Soil samples for chemical properties and particle size distribution analysis were collected 35 at a 0-10 cm depth at 5 locations along each of the 4 sampling transects (Appendix IV), with one composite sample analyzed per transect. Samples were sieved to remove coarse fragments (> 2 mm) and air-dried. Organic C was determined by wet oxidation using 0.066 M potassium dichromate solution (Tiessen and Moir 1993), while total N was measured using the Kjeldahl procedure (McGill and Figueiredo 1993). Soil pH was determined on a 1:2 (v/v) soil to distilled water saturated paste (Hendershot and Lalande 1993). Soil particle size distribution was determined by the hydrometer method (Sheldrick and Wang 1993). Soil bulk density was measured by collecting intact soil cores (Culley 1993) at a depth of 0-4.9 cm with a single-cylinder 5.2-cm-diameter by 4.9-cm-deep core. Three cores were collected along each of the 4 transects at each study site (Appendix TV). Bulk density samples were dried to constant weight in a forced-air oven at 105°C for 48 hours. Coarse fragments (diameter > 2 mm) were screened out, weighed, and their volume was determined by displacement. Fine-fraction soil bulk density was calculated as the mass of dry, coarse fragment-free mineral soil per volume of field-moist soil, where volume was also calculated on a coarse fragment-free basis (Culley 1993). Soil penetration resistance (Lowery and Morrison 2002) was measured from 9 - 3 0 November, 2005 (before the rainy season), 25 February to 11 March, 2006 (during the rainy season), and 4-24 April, 2006 (at the end of the rainy season). Penetration resistance was measured with a HFG-45 hand-held force gauge with a 4-mm-basal diameter (30°) cone (Transducer Techniques, Temecula, CA, USA). Measurements were made at 3 locations along each of the 4 soil sampling transects / study site (Appendix TV). Five replicate measurements were taken within 0-5 and 5-10 cm depths at each sampling location by horizontally inserting the force gauge into the soil profile at a constant velocity. For both sampling depths, soil samples were also collected for gravimetric water content determination (Topp 1993). Since soil penetration resistance is strongly affected by the soil water content at the time of measurement, correction to a reference soil water content was done using the method proposed by Busscher and Sojka (1987). This method applies an empirical power function relationship among bulk density, gravimetric water content, and soil penetration resistance allowing comparisons of absolute soil penetration resistance independent of the original soil 36 water content. The Busscher and Sojka (1987) method assumes that soil penetration resistance (PR) depends on bulk density (pb) and gravimetric water content (w) as follows: m = cpbawb [1] where a, and b are coefficients, and c is constant. When the empirical equation for log (PR), given by log (PR) = a log (pb) + Mog(w) + c [2] at one water content (1, corrected) is subtracted from another at a different water content (2, uncorrected), the bulk density term, which was assumed to remain constant for a given site, cancels yielding PRi /PR 2 =(wi /w 2 ) b [3] The coefficients a and b were determined by measuring PR, pb, and w in the laboratory. Soils of the 2 textural classes found at our study sites (loam and sandy loam) were passed through a 2 mm sieve and wet to varying moisture contents representative of the seasonal field variation by adding water to samples and mixing. These samples were then re-packed to varying bulk densities in 7.3 cm diameter by 7.5 cm high soil cores, and PR was measured using the same penetrometer as used in field measurements. A l l data were substituted into the Busscher-Sojka model, giving b coefficients of-0.45 for loam and -0.28 for sandy loam (i.e., log (PR) = 6.78 log (pb) - 0.45 log (w) + 0.49; r 2 = 0.49; P < 0.01 for loam and log (PR) = 4.94 log (pb) 7 0.28 log (w) + 0.42; r 2 = 0.51; P < 0.01 for sandy loam). Corrected soil penetration resistance measurements were adjusted to water content of 0.2 kg kg"1, the overall average water content for all sites across the three measurement periods. Only corrected penetration resistance data are reported in the manuscript. 2.2.2.3. Statistical analysis Annual forage production, the proportion of grasses, forbs, shrubs, and unpalatable biomass among the three ecological zones, and sapling density were analyzed as a completely randomized design with six replications (study sites). It was not possible to statistically compare annual forage production or sapling density in the highland pasture with the other two ecological zones because there is no local information about tiller longevity for the dominant 37 tussock grasses in the highland pasture, and no trees or saplings present. Soil cover, litter layer depth, organic C, total N , C:N, pH and bulk density were analyzed as completely randomized designs with six replications (study sites) and 4 multiple measurements (transects) per site. Soil penetration resistance was analyzed as a completely randomized design with 6 replications (study sites) and 12 multiple measurements per site. The SAS general linear model procedure was used (SAS Institute 2004) for all analyses, and an a value of 0.05 was considered significant. Following a significant F-test, least square means were evaluated using the Bonferroni correction. 2.3. Results and Discussion 2.3.1. Annual Forage Production and Grazing Capacity Annual forage production in the anthropogenic pasture ecological zone was significantly greater (2970 kg/ha), and more variable, than in the deciduous forest (476 kg/ha) where the ground surface was shaded by the tree canopy (Table 2.1). It was not possible to directly measure annual forage production for the highland pasture, due to uncertainty regarding what portion of the tussock grass biomass measured consists of annual growth, and how much is carryover from previous years. The above-ground live biomass measured in the highland pasture was 2200 kg/ha (data not shown). There have been very few studies of tiller longevity in highland perennial tussock grasses (Kdrner et al. 2006). Meurk (1978) found that tillers ranged in age from 2.6 to 3.2 years at highlands in New Zealand, and Hnatiuk (1978) found that tussock tillers lived for 7 to 16 months in a wet tropical zone near the treeline in New Guinea. Until studies have been conducted on tiller longevity in the highland pasture of our study area, we chose to select a conservative estimate of 3-year tiller longevity based on the tiller study by Meurk (1978). Therefore, we have assumed that one-third of the above-ground live biomass for the highland pasture is annual forage production (733 kg/ha), given the dominance of tussock grasses in this ecological zone. Forage within the anthropogenic and highland pastures was composed predominantly of grasses, whereas the deciduous forest had a greater proportion of woody species contributing to forage production (Figure 2.2). Although the anthropogenic pasture currently has the greatest forage production, it is 38 important to note that this zone also has more than 8 times greater unpalatable biomass (310 kg/ha) than the deciduous forest (38 kg/ha) and 6 times more than the highland pasture (48 kg/ha) (Figure 2.3.3). There was no significant difference in unpalatable biomass production between deciduous forest and highland pasture ecological zones. Species that are toxic or unpalatable to livestock have a selective advantage over edible species under grazing pressure, and therefore long-term heavy grazing pressure can lead to increased proportions of unpalatable species (Bastin et al. 1993; Todd and Hoffman 1999; Riginos and Hoffman 2003) resulting in forage declines. Forage production is highly dependent upon precipitation during the growing season. In the period from June 2005 to May 2006, there were 1099 mm of precipitation at the San Salvador de Jujuy meteorological station, within the Yungas-Chaco forest transition zone approximately 30 kilometres north of our study area, while the 98-year average for this station was 863.2 mm (Universidad Nacional de Jujuy 2006) (Appendix V). As a result, forage production values reported in this study may be slightly higher than for growing seasons with lower precipitation levels. Based on annual forage production and above-ground live biomass measurements, preliminary stocking rate recommendations have been calculated for the three ecological zones in this study (Table 2.1). Stocking rate recommendations are based on the assumption that 354 kg of forage are required to support one animal unit for one month (1 A U M ) (USDA, NRCS 2003), and that 50% of the annual forage produced is available for domestic livestock. The remaining 50% of biomass production will remain to support ecosystem services, including forage for wildlife, and ground cover to prevent erosion and contribute to nutrient cycling (Holechek et al. 1998). Studies of tussock grass tiller longevity in the highland pasture are needed to make more precise stocking rate recommendations in this ecological zone. Based on field observations, consultations with livestock owners and landowners, and a previous study in the Jujuy Model Forest carried out by Lamas et al. (2003), we estimated current stocking rates, with the anthropogenic pasture having the highest stocking rate, and the deciduous forest the lowest (Table 2.1). Comparing these estimations to the recommended stocking rates, overstocking is occurring in both the deciduous forest and anthropogenic pasture. It is difficult to estimate exact stocking rates within the highland pasture, as there is limited 39 fencing and cattle are often grazed clandestinely without the permission of landowners. Overstocking is particularly acute in the anthropogenic pasture, with estimated current stocking rates at 2 to 3.5 times the recommended stocking rate. Declines in forest cover within the anthropogenic pasture have led to increased forage production and higher recommended stocking rates (Table 2.1), which could be an incentive for further conversion of Yungas forest to anthropogenic pasture. However, given the monsoon climate and steep slopes, and high levels of forest biodiversity within the montane forests of southern Jujuy, it is important to examine how current livestock grazing practices impact soil function and biotic integrity in the long term. 2.3.2. Forest Structure On a scale of 0 to 1 (where 0 = no canopy cover and 1 = complete canopy cover) the anthropogenic pasture ecological zone had a mean canopy cover of 0.04 ± 0.084 SD, the deciduous forest had 0.88 + 0.099 SD, and the highland pasture had no canopy cover (0) (n = 24). There were no trees found in the highland pasture ecological zone, and therefore forest structure analysis was not carried out in this zone. A total of 22 tree (dbh > 10 cm) species were identified in total, with 15 tree species found in the deciduous forest ecological zone, and 10 tree species in the anthropogenic pasture ecological zone (Table 2.2). Tree species diversity within the Yungas forest is highly dependent upon latitude and altitude (Brown et al. 2001). The estimated regional tree diversity in the Yungas forest, across an altitudinal gradient at the latitude of our study area (24°S), is 94 tree species (Morales et al. 1995). Local tree diversity within a given forest type at this latitude averages 34 species (Brown et al. 2001). These data, combined with our identification of distinct tree species within the deciduous forest that were not present at our field sites, indicate that more study sites would be needed to capture the entire local tree diversity. The trees within the deciduous forest with highest importance values were Anadenanthera colubrina (Cebil Colorado), Parapiptadenia excelsa (Horco cebil), Pisonia zapallo (Zapallo caspi), Schinopsis lorentzii (Quebracho santiaguefio) and Allophylus edulis (Chal chal). The arboreal structure of the anthropogenic pasture zone was markedly different, 40 with Sapium haematospermum (Lecheron), Acacia aroma (Tusca), Phyllostylon rhamnoides (Palo lanza), Acacia caven (Churqui), and Schinopsis marginata (Horco quebracho) having the highest importance values. Only 4 tree species were found within both ecological zones, suggesting that intensive livestock grazing and timber / firewood harvesting have changed not only the tree density, but also the arboreal composition of the anthropogenic pasture. Three tree species found exclusively in the anthropogenic pasture ecological zone are indicative of heavy livestock grazing: Acacia aroma, Acacia caven, and Acacia praecox. The seed pods of these species are preferred by livestock, which facilitate their dispersal (Braun Wilke et al. 1995). Although livestock encourage the dispersal of a select few tree species, overall there are fewer saplings present in the anthropogenic pasture, where heavy livestock grazing occurs. The deciduous forest ecological zone had an average of 1524 saplings/ha, while sapling numbers in the anthropogenic pasture were significantly lower (130 saplings/ha). Sapling density is an important indicator of future forest composition and structure, and the capacity of a forest to regenerate itself. Under heavy grazing pressure, livestock often consume saplings, which can lead to severe declines in forest regeneration (Hensen 2002; Sagar and Singh 2004). Under light or moderate stocking rates, livestock grazing may not cause declines in canopy species, and can increase understory species richness (Krzic et al. 2003). 2.3.3. Soil Quality 2.3.3.1. Soil cover The proportion of bare soil was significantly different in each of the 3 ecological zones: the anthropogenic pasture zone had the greatest proportion of bare soil, and the deciduous forest the least (Table 2.3). Percent bare soil was inversely related to litter cover and litter layer depth. Deciduous forest sites had the greatest litter cover and thickest litter layer, while anthropogenic pasture sites had the lowest litter cover and thinnest litter layer (Table 2.3). The annual addition of deciduous leaves to the litter layer, slow decomposition rates of woody litter materials, and lower intensity livestock use contributed to lower percent bare soil in the deciduous forest, in comparison with the other 2 ecological zones. Although the anthropogenic pasture zone has scattered deciduous tree cover, it had significantly more bare soil 41 than the highland pasture that had no trees. This difference was likely due to heavy livestock grazing in the anthropogenic pasture, while the highland pasture was lightly grazed. Livestock grazing tends to reduce litter cover and increase the proportion of bare ground through the consumption of plant biomass that could otherwise be available for conversion into litter, trampling and compaction of the litter layer, and trailing (Wood et al. 1989; Belsky and Blumenthal 1997; Talbot et al. 2003). Livestock owners in the study area commented that in the heavily grazed anthropogenic pasture zone it appeared that over approximately the past 10 years the 'the rocks were flowering' (land degradation was causing more rocks to appear on the soil surface). In the montane region of Cordoba province, Argentina, bare rock exposed by erosion has previously been used as an indicator of long-term grazing pressure, with greater bare rock exposure indicating higher long-term grazing pressure (Renison et al. 2006). In this study, rock cover in the anthropogenic pasture (10%) was higher than for the deciduous forest (5%), while rock cover was greatest in the highland pasture (15%); however, these differences were significant at a = 0.10 due to high variability in rock cover within ecological zones (Table 2.3.3). 2.3.3.2. Soil chemical properties Both soil organic C and total N were significantly higher in the deciduous forest ecological zone than in either the anthropogenic or highland pastures (Figure 2.4). The ability of tree roots to access nutrients deep within the soil profile, and annual deposition of deciduous leaves and woody debris onto the soil surface, are important nutrient-cycling processes (Atttiwill and Adams 1993; Buschbacher et al. 1988) that were greatly reduced in both pasture ecosystems of our study. Three leguminous trees, Acacia aroma, Acacia coven, and Acacia praecox, occupied 30% of the basal area at deciduous forest study sites, which likely increased soil N content (Pons et al. 2007), and enabled microorganisms to decompose greater amounts of C-rich leaf litter. Heavy livestock grazing pressure (Roder et al. 2002; Abril et al. 2006) and tree removal (Ballard 2000) can lead to declines in soil nutrient levels due to biomass removal from the ecosystem. In this study, however, the heavily grazed anthropogenic pasture did not show significantly lower organic C and total N levels than the lightly grazed highland pasture. This 42 may be partially due to livestock facilitating the expansion of three leguminous, pioneer tree species of the genus Acacia, and the expansion of unpalatable woody shrubs in the anthropogenic pasture, which helped to offset C and N losses in the forage removed. The highland pasture ecological zone had a significantly higher C:N compared with the anthropogenic pasture and deciduous forest, although the C:N ratios of the three ecological zones only varied from 9.6 to 11.6 (Table 2.3.3). The tussock grasses that dominate the highland pasture have rigid, long-lived, sclerenchymatous leaves that are low in N . Nitrogen is often a key limiting nutrient in highland pastures found at low latitudes (Korner et al. 2006), explaining the higher C:N in the highland pasture. Although the pH levels for all three ecological zones were within a close range, from 5.1 to 5.5, the deciduous forest had a significantly higher pH than the highland pasture zone (Table 2.3.3). Soils at highland pasture study sites tended to be more coarsely textured (sandy loam) than deciduous forest soils, which were predominantly loams. The coarser texture implies a lower cation exchange capacity, and greater potential for leaching of base cations. Within the elevation range of this study, precipitation tends to increase with elevation, contributing to increased leaching, and lower soil pH levels in the highland pasture. 2.3.3.3. Soil compaction The anthropogenic pasture ecological zone consistently exhibited the highest soil strength as measured by penetration resistance of the three ecological zones. This difference was significantly greater than at least one of either the deciduous forest or highland pasture zone at both depths and all times of measurement, except for the 5 - 10 cm depth in February-March 2006 (Figure 2.5.). There were no significant differences in penetration resistance between deciduous forest and highland pasture ecological zones for any depth or time of measurement. Penetration resistance measurements were higher in the anthropogenic pasture ecological zone primarily due to more intensive livestock grazing in this zone. A number of other studies have found that grazed areas exhibited higher soil penetration resistance than adjacent ungrazed areas (Krzic et al. 1999; Martinez and Zinck 2004), and that heavily grazed areas had higher soil strength than lightly grazed areas (da Silva et al. 2003; Xie and Wittig 2004). 43 Although timber and firewood harvesting occurs in the anthropogenic pasture zone, there was no evidence at any anthropogenic pasture sites to indicate that trees had recently been felled in close proximity to where soil sampling was carried out. Alternate periods of soil wetting and drying, root penetration, earthworm activity and microbial activity has likely reduced soil compaction (Barzegar et al. 1995) that occurred due to previous cutting events, particularly as trees are usually felled with an axe, and no machinery or large equipment passes over the soil surface. The impact of increased soil penetration resistance on plant growth varies according to many factors including plant species, organic matter content, soil texture and the soil water regime (Greacen and Sands 1980). Compaction-induced changes in soil-water relationships are often more important to plant growth than absolute changes in soil physical properties (Unger and Kaspar 1994; Lipiec and Hakansson 2000). Most often soil compaction has a negative effect on plant growth and can cause declines in forest productivity (Kozlowski 1999), and restricted root growth and elongation (Laboski et al. 1998), however this is not always the case. In a study of soil compaction on young ponderosa pine (Pinus ponderosa P. Lawson & C. Lawson) in California, Gomez et al. (2002) found that compacted soils with clay and loam textures reached critical water potentials sooner than uncompacted soils and exhibited reduced tree growth, while compaction of sandy loam soils extended the period of plant-available water and increased tree growth. In our study area, 5 of 6 anthropogenic pasture field sites had a loam texture, and therefore plants at these sites may be experiencing increased moisture stress due to compaction, particularly during the 7 - 8 month dry season. Soil compaction in the anthropogenic pasture is also a concern for soil-water relations during the rainy season when intense rainstorms are frequent. A reduction of soil macropores due to compaction tends to lead to decreased water infiltration and hydraulic conductivity (Greacen and Sands 1980, Savadogo et al. 2007), which increases the likelihood of soil loss through water erosion. Soil bulk density values were not significantly different among the three ecological zones (Table 2.3); however, the trend in soil bulk density data was similar to that of soil penetration resistance. The anthropogenic pasture exhibited higher bulk density values than the deciduous forest and highland pasture. Bulk density has repeatedly been shown to change less in response to soil compaction than penetration resistance, especially in forest and rangeland soils 44 that commonly have high spatial variability (Chanasyk and Naeth 1995; Krzic et al. 1999; Rodd et al. 1999). To determine treatment impacts on soil bulk density, a large number of samples are needed. Courtin et al. (1983) found that sample sizes of 14-28 were needed on 0.8 to 3.3 ha plots i f a 95% confidence and 10% error were used. 2.4. Conclusions In the Yungas forest of southern Jujuy province, forage production and recommended stocking rates within the anthropogenic pasture ecological zone were higher than in adjacent deciduous forest. Based on annual forage production measurements, overstocking of domestic livestock was occurring in both deciduous forest and anthropogenic pasture zones, with the problem being particularly acute in the anthropogenic pasture. A high proportion of unpalatable plant species within the anthropogenic pasture limits potential forage production in this zone, and poses a hazard to livestock health. Heavy livestock grazing in the anthropogenic pasture, along with firewood and timber harvesting, have led to changes in forest structure and composition, and forest regeneration is currently limited by the intensity of livestock grazing. Indicators of soil quality, including soil cover, organic C, total N , and penetration resistance suggest that land use, including heavy livestock grazing pressure, in the anthropogenic pasture is leading to declines in soil quality in relation to the deciduous forest. These changes have the potential to increase soil loss due to water erosion in the monsoon climate of this region. Reducing livestock numbers to recommended stocking rates will enable improved ecosystem functioning including increased litter production, nutrient cycling, and reduced erosion susceptibility. Implementation of a deferred, rotational grazing system that limits livestock movement on steep montane slopes during the rainy season would reduce the potential for soil compaction, and livestock injury due to falls. Continued measurement of forage production over a number of years would be helpful for determining long-term average forage production values, and measurement of forage quality would improve stocking rate recommendations. Studies of tiller longevity for tussock grasses in the highland pasture are needed to better understand rates of annual forage production in this ecological zone. 45 2.5. Acknowledgements We are very grateful to Dr. Tony Kozak for statistical advice, and Dr. Les Lavkulich for assistance with soil classification. Thank you to Maria Ines Bonansea, Claudia Chauque, Ivan Escalier, Helena Arraya and other local community members for assistance with field work. Funding was generously provided by The John G. Bene Fellowship in Community Forestry (International Development Research Centre, Canada), the Natural Sciences and Engineering Research Council of Canada, and the Organization of American States. 46 Table 2.1 Annual forage production, recommended stocking rate, and estimated current stocking rate for anthropogenic pasture, deciduous forest and highland pasture ecological zones in the Yungas forest, Jujuy province, Argentina. Ecological Zone Annual Forage Production (kg/ha) Recommended Stocking Rate ( A U M 3 / ha) Current Stocking Rate ( A U M / ha) Anthropogenic pasture 2970 a1 4.2 10-15 . Deciduous forest 476 b , 0.7 0 .9- 1.84 Highland pasture 7332 1.0 N A 5 1 Annual forage production means followed by different letters are significantly different (P < 0.05). 2 Annual forage production for the highland pasture was estimated due to lack o f information about the longevity of tussock grasses, and therefore was not statistically compared to the other two ecological zones. 3 A U M = Animal unit month 4 Estimated by Lamas et al. (2003). 5 Not estimated due to prevalent clandestine grazing, limited fencing and scarce information. 47 Table 2.2 Basal area (dominance), frequency, density, and importance value of tree species found in deciduous forest and anthropogenic pasture ecological zones of the Yungas forest, Jujuy province, Argentina. Deciduous Forest Anthropogenic Pasture Tree species Basal Area (m2/ha), (Frequency)1 Density (Stems/ha) Importance Value2 Basal Area (nrVha), (Frequency)1 Density (Stems/ha) Importance Value2 Acacia aroma Gillies ex Hook. & Am. - - - 1.11(2) 25.0 0.75 Acacia caven (Molina) Molina - - -- 0.0779 (3) 6.7 0.27 Acacia praecox Griseb. - - 0.0144 (1) 1.7 0.079 Allophylus edulis (A. St.-Hil.) Niederl. 1.03 (4) 30.0 0.22 - - -Anadenanthera colubrina (Veil.) Brenan 2.96(5) 96.7 0.47 - - _ Aspidosperma quebracho Schltdl. 0.160(1) 1.7 0.04 - - -Astronium urundeuva Engl. 0.549 (3) 28.3 0.16 - - -Ceiba insignis (Kunth) P.E. Gibbs & Semir 3.06(1) 5.0 0.18 - - _ Condalia buxifolia Reiss. 0.796 (1) 30.0 0.13 0.0212 (1) 1.7 0.080 Gleditsia amorphoides Taub. - - - 0.109(1) 3.3 0.13 Jodina rhombifolia (Hook. & Am.) Reissek 0.281 (1) 3.3 0.05 - - _ Myrcianthes cisplatensis (0. Berg) 0.025(1) 1.7 0.03 - -Parapiptadenia excelsa (Griseb.) Burkart 3.746 (5) 55.0 0.42 0.0766(1) 1.7 0.10 Phyllostylon rhamnoides (Poiss.) Taub. - - 0.617(2) 8.3 0.39 Pisonia zapallo Griseb. 3.873 (2) 68.3 0.37 - -Ruprechtia apetala Weddell 0.225 (2) 18.3 0.10 - -Ruprechtia triflora Griseb. 0.810(4) 33.3 0.21 - -Sapium haematospermum Mull. Arg. - - 0.685 (5) 23.3 0.76 Schinopsis lorentzii Engl. 2.234 (3) 30.0 0.25 - - -Schinopsis marginata Engl. 0.393 (3) 18.3 0.14 0.520 (1) 1-7 0.22 Scutia buxifolia Reissek 0.015 (1) 1.7 0.03 - -Tipuana tipu (Benth.) Kuntze 0.017(1) 1.7 0.03 0.117(1) 1.7 0.11 Unidentified spp.3 1.984 (5) 46.7 NA 0.144(1) 1.7 0.12 1. Frequency (in brackets) denotes the number of field sites, out of a total of 6, where the species was located. 2. Importance value (IV) = relative basal area (dominance) + relative frequency + relative density, therefore maximum IV = 3; TV cannot be calculated for multiple unidentified species. 3. There were between 2 and 5 unidentified species within the deciduous forest ecological zone, and 1 unidentified species in the anthropogenic pasture ecological zone. . Table 2.3 Soil physical and chemical properties measured in anthropogenic pasture, deciduous forest, and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina. Property Anthropogenic Deciduous Highland n P (F - test) Pasture Forest Pasture Bare soil (%) 33 (2.9)1 a 2 1 (2.8) b 13 (2.9) c 6 < 0.0001 Litter cover (%) "55 (3.6) a 94 (3.6) b 72 (3.6) c 6 < 0.000.1. Rock cover (%) 10 (3.0) a 5 (2.9) a 15 (3.0) a 6 0.0959. Litter depth (cm) 0.4 (0.34) a 4.6 (0.33) b 1.4 (0.34) a 6 < 0.0001 C:N 9.6 (0.37) a 10.2 (0.36) a 11.6 (0.36) b 6 0.0048 Soil pH 5.4 (0.10)ab 5.5 (0.10) a 5.1 (0.10) b 6 0.0185 Bulk density (Mg 0.92 (0.059) a 0.78 (0.059) a 0.87 (0.059) a 6 ' 0.2912 m-3) 1 Standard error of the mean in parentheses 2 Means followed by the same small letter are not significantly different (P > 0.05). 49 This material has been removed due to copyright restrictions. This page contained a series of three maps depicting the location of the study area: A map of Argentina within South America, a map of Jujuy within Argentina, and a map of the study area within Jujuy. i Figure 2.1 Location of the Jujuy Model Forest, within Jujuy province, northwestern Argentina, South America. Maps adapted from Direccion de Bosques, Secretaria de Ambiente y Desarrollo Sustentable (2003). 50 Anthropogenic Deciduous Highland Pasture Forest Pasture Ecological Zone Figure 2.2 Forage composition for anthropogenic pasture, deciduous forest, and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina. The proportion of grasses, forbs and woody plants within forage is indicated by the stacked bars; the same small letters beside a stacked bar type indicate the mean forage production for this forage type is not significantly different (P > 0.05, n - 6). Forage composition values for the highland pasture were based on measurement of standing above-ground live biomass. 51 <0 c o 600 500 o f 400 i _ Q_ S 300 to E •§ 200 a> •g 100 •+-> ro re a c 0 Anthropogenic Deciduous Highland Pasture Forest Pasture Ecological Zone Figure 2.3 Unpalatable biomass production for anthropogenic pasture, deciduous forest, and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina. Error bars represent standard deviation (n = 6). Means with the same letter are not significantly different (P > 0.05). 52 60 -I , „ • U) 50 -40 -O o 30 -"E (0 O) 1_ 20 -O o 10 -w 0 -Anthropogenic Deciduous Highland Pasture Forest Pasture Ecological Zone z 15 •+•> o I— o if) 6 5 4 3 2 1 0 (b) Anthropogenic Deciduous Highland Pasture Forest Pasture Ecological Zone Figure 2.4 Soil organic C (a) and soil total N (b) at 0-10 cm depth in three ecological zones of the Yungas forest, Jujuy province, Argentina. Error bars represent standard deviation (n = 6). Means with the same letter for the same soil parameter are not significantly different (P > 0.05). Ch 2005 2006 2006 Time of Measurement Time of Measurement Figure 2.5 Soil penetration resistance at depths of a) 0-5 cm and b) 5-10 cm, measured in November 2005, February-March 2006, and April 2006 within anthropogenic pasture, deciduous forest and highland pasture ecological zones of the Yungas forest, Jujuy province, Argentina. Error bars represent standard error of the mean (n = 6). Means with the same letter at the same time of measurement are not significantly different (P > 0.05). 2.6. References Abdel-Magid, A .H . , G.E. Schuman, and R.H. Hart. 1987. Soil bulk density and water infiltration as affected by grazing systems. Journal of Range Management 40(4):307-309. Abril, A. , P. Barttfeld, and E.H. Bucher. 2005. The effect of fire and overgrazing disturbances on soil carbon balance in the Dry Chaco forest. Forest Ecology and Management 206: 399-405. Attiwill, P .M. , and M . A . Adams. 1993. Tansley Review No. 50. Nutrient cycling in forests. New Phytologist 124(4): 561-582. Ballard, T .M. 2000. Impacts of forest management on northern forest soils. Forest Ecology and Management 133:37-42. Barzegar, A.R., P. Rengasamy, and J.M. Oades. 1995. Effects of clay type and rate of wetting on the mellowing of compacted soils. Geoderma 68: 39-49. Bastin, G.N., G. Pickup, V . H . Chewings, and G. Pearce. 1993. Land degradation assessment in central Australia using a grazing gradient method. Rangeland Journal 15:190-216. Belsky, A.J . and D . M . Blumenthal. 1997. Effects of livestock grazing on stand dynamics and soils in upland forests of the interior west. Conservation Biology 11:315-327. Bezkorowajnyj, P.G., A . M . Gordon, and R.A. McBride. 1993. The effect of cattle foot traffic on soil compaction in a silyo-pastoral system. Agroforestry Systems 21:1-10. Braun Wilke, R.H., B.S. Villafane, and L.P.E. Picchetti. 1995. Plantas de interes ganadero de Jujuy y Salta Noroeste Argentino. San Salvador de Jujuy, Argentina: Secretaria de Ciencia, Tecnica y Estudios Regionales and Universidad Nacional de Jujuy. 309 pp. Braun Wilke, R.H., E.E. Santos, L.P.E. Picchetti, M.T. Larran,G.F. Guzman, and C.R. Colarich. 2001. Carta de aptitud ambiental de la provincia de Jujuy. San Salvador de Jujuy, Argentina: Departamento de Suelos y Ecologia, Facultad de Ciencias Agrarias, Universidad Nacional de Jujuy. 245 pp. Brown, A.D. , H.R. Grau, L.R. Malizia and A. Grau. 2001. Argentina. In: M . Kappelle and A.D. Brown [EDS]. Bosques nublados del neotropico. Santo Domingo de Heredia, Costa Rica: Instituto Nacional de Biodiversidad. p. 623-659. Buschbacher, R., C. Uhl, and E.A.S. Serrao. 1988. Abandoned pastures in eastern Amazonia II. 55 Nutrient stocks in the soil and vegetation. Journal of Ecology, U.K. 76(3): 682-699. Busscher, W.J. and R.E. Sojka. 1987. Enhancement of subsoiling effect on soil strength by conservation tillage. Transactions of the AS'AE 30(4): 888-892. Cabrera, A . L . 1976. Enciclopedia argentina de agricultura y jardineria: Fasciculo 1 - Regiones fitogeograficas argentinas. Buenos Aires, Argentina: Editorial Acme S.A.C.I. 82 p. Chanasyk, D.S. and M . A . Naeth. 1995. Grazing impacts on bulk density and soil . strength in the foothills fescue grasslands of Alberta, Canada. Canadian Journal of Soil Science 75:551-557. Chartier, M.P. and C M . Rostagno. 2006. Soil erosion thresholds and alternative states in northeastern Patagonian rangelands. Rangeland Ecology and Management 59:616-624. Courtin, P., M.C. Feller, and K . Klinka. 1983. Lateral variability in some properties of disturbed forest soils in southwestern British Columbia. Canadian Journal of Soil Science 63:529-539. Culley, J.L.B. 1993. Density and compressibility. In M.R. Carter [ED.]. Soil sampling and methods of analysis. Boca Raton, FL: Canadian Society of Soil Science, Lewis Publishers, p. 529-539. Curtis, J.T., and R.P. Mcintosh. 1951. An upland forest continuum in the prairie-forest border region of Wisconsin. Ecology 32:476-496. da Silva, A.P., S. Imhoff, and M . Corsi. 2003. Evaluation of soil compaction in an irrigated short-duration grazing system. Soil & Tillage Research 70: 83-90. Direccion de Bosques Nativos. 2003. Atlas de los bosques nativos argentinos. Available at: http://www2.medioambiente.gov.ar/bosques/bosques nativos/2004 atlas/default.htm Accessed 20 May 2007. Doran, J.W., and T.B. Parkin. 1994. Defining and assessing.soil quality. In: J.W. Doran, D.C. Coleman, D.F. Bezdicek, and B.A. Stewart. [EDS.]. Defining soil quality for a sustainable environment. Madison, WI: Soil Science Society of America, American Society of Agronomy, p. 3-21. Gomez, A. , R.F. Powers, M.J. Singer, and W.R. Horwath. 2002. Soil compaction effects on growth of young ponderosa pine following litter removal in California's Sierra Nevada. Soil Science Society of America Journal 66: 1334-1343. 56 Grau, A. , and A.D. Brown. 2000. Development threats to biodiversity and opportunities for conservation in the mountain ranges of the Upper Bermejo River Basin, N W Argentina And SW Bolivia. Ambio 29:445-450. Greacen, E.L. and R. Sands. 1980. Compaction of forest soils: A review. Australian Journal of Soil Resources 18:163-189. Greenwood, K . L . , D.A. MacLeod, and K.J . Hutchinson. 1997. Long-term stocking rate effects on soil physical properties. Australian Journal of Experimental Agriculture 37:413-419. Gutierrez, J., and I.I. Hernandez. 1996. Runoff and interrill erosion as affected by grass cover in a semi-arid rangeland of northern Mexico. Journal of Arid Environments 34: 287-295. Heady, H.F. and R.D. Child. 1994. Rangeland ecology and management. Boulder, CO: Westview Press. 519 p. Hendershot, W.H., and H . Lalande. 1993. Soil reaction and exchangeable acidity. In M.R. Carter [ED.]. Soil sampling and methods of analysis. Boca Raton, FL: Canadian Society of Soil Science, Lewis Publishers, p. 141-145. Hensen, I. 2002. Impacts of anthropogenic activity on the vegetation of Polylepis woodlands in the region of Cochabamba, Bolivia. Ecotropica 8:183-203. Herrick, J.E., J.R. Brown, A.J . Tugel, P.L. Shaver, and K . M . Havstad. 2002. Application of soil quality to monitoring and management: Paradigms from rangeland ecology. Agronomy Journal 94:3-11. Holechek, J.L., R.D. Pieper and C H . Herbel. 1998. Range management: Principles and practices. 3 r d Edition. Upper Saddle River, NJ: Prentice Hall. 542 pp. Hnatiuk, R.J. 1978. The growth of tussock grasses on an equatorial high mountain and on two sub-antarctic islands. In: C. Troll, and W. Lauer [EDS.]. Geoecological relations between the southern temperate zone and the tropical mountains. Erdwiss Forschung 11, Franz Steiner Verlag, Wiesbaden, p. 159-188. Kdrner, C. 2006. High-elevation land use, biodiversity, and ecosystem functioning. In: E . M . Spehn, M . Liberman, and C. Kdrner [EDS.]. Land use change and mountain biodiversity. Boca Raton, FL: Taylor & Francis, p. 3-21. Kozlowski, T.T. 1999. Soil compaction and growth of woody plants. Scandinavian Journal of Forest Research 14(6): 596-619. 57 Krzic, M . , K. Broersma, R.F. Newman, T .M. Ballard and A. A. Bomke. 2001. Soil quality of harvested and grazed forest cutblocks in southern British Columbia. Journal of Soil and Water Conservation 56(3): 192-197. Krzic, M . , R.F. Newman, K . Broersma and A. A. Bomke. 1999. Soil compaction of forest plantations in interior British Columbia. Journal of Range Management 52:671-677'. Krzic, M . , R.F. Newman, and K. Broersma. 2003. Plant species diversity and soil quality in harvested and grazed boreal aspen stands of northeastern British Columbia. Forest Ecology and Management 182:315-325. Laboski, C . A . M . , Dowdy, R.H., Allmaras, R.R. and J.A. Lamb. 1998. Soil strength and water content influences on corn root distribution in a sandy soil. Plant Soil 203: 239-247. Lamas, H.E., C. Borgogno, and M . del Carpio. 2003. Informe sobre la situation actual y perspectivas de los agrosistemas de la cuenca Los Pericos-Manantiales. In P. Laclau (ED.). Manejo sustentable de ecosistemas forestales de la cuenca Los Pericos-Manantiales Informe Diagnostico. Buenos Aires, Argentina: Direction de Bosques - Secretaria de Desarrollo Sustentable y Politica Ambiental de la Nation. 73 pp. Lipiec, J. And I. Hakansson. 2000. Influences of the degree of compactness and matric water tension on some important plant growth factors. Soil and Tillage Research 53(2): 87-94. Lowery, B. , and J.E. Morrison. 2002. Soil penetrometers and penetrability. In: J.H. Dane and G.C. Topp [EDS.]. Methods of soil analysis: Part 4 Physical methods. Madison, WI: Soil Science Society of America, p. 363-388. Martinez, L.J. , and J.A. Zinck. 2004. Temporal variation of soil compaction and deterioration of soil quality in pasture areas of Colombian Amazonia. Soil & Tillage Research 75:3-17. Mayer, A.C. , C. Huovinen, V . Stoeckli, and M . Kreuzer. 2006. Plant species diversity, forest structure, and tree regeneration in subalpine wood pastures. In: E . M . Spehn, M . Liberman, and C. Kdrner [EDS.]. Land use change and mountain biodiversity. New York, N Y : Taylor and Francis, p.249-259. McGil l , W.B., and C.T. Figueiredo. 1993. Total nitrogen. In M.R. Carter [EDS.]. Soil sampling and methods of analysis. Boca Raton, FL: Canadian Society of Soil Science, Lewis Publishers, p. 201-211. Meurk, C D . 1978. Alpine phytomass and primary productivity in Central Otago, New Zealand. 58 New Zealand Journal of Ecology 1:27-50. Milchunas, D.G. and W.K. Lauenroth. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecological Monographs 63: 327-366. Morales, J .M., M . Sirombra, and A.D. Brown. 1995. Riqueza de arboles en las Yungas argentinas. In: A .D. Brown, and H.R. Grau. [EDS.]. Investigacion, conservation y desarrollo en las selvas subtropicales de montana. Tucuman, Argentina: Laboratorio de Investigaciones Ecologicas de las Yungas, Universidad Nacional de Tucuman. p. 163-174. Nadir, A. , and T. Chafatinos. 1990. Los suelos del Noroeste Argentino (Salta y Jujuy). Vol . 3. Salta, Argentina: Universidad Nacional de Salta / Direction General Agropecuaria Salta. p.283-284. Page-Dumroese, D., M . Jurgensen, W. Elliot, T. Rice, J. Nesser, T. Collins and R. Meurisse. 2000. Soil quality standards and guidelines for forest sustainability in northwestern North America. Forest Ecology and Management 138:445-462. Pellant, M . , P. Shaver, D.A. Pyke, and J.E. Herrick. 2005. Interpreting indicators of rangeland health. Technical Reference 1734-6, Version 4. Denver, CO: United States Department of the Interior, Bureau of Land Management, National Science and Technology Centre. 122p. Pons, T.L., K . Perreijn, C. van Kessel, and M.J.A. Werger. 2007. Symbiotic nitrogen fixation in a tropical rainforest: 1 5 N natural abundance measurements supported by experimental isotopic enrichment. New Phytologist 173: 154-167. Renison, D., I. Hensen, R. Suarez, and A . M . Cingolani. 2006. Cover and growth habit of Polylepis woodlands and shrublands in the mountains of central Argentina: human or environmental influence? Journal of Biogeography 33: 876-887. Riginos, C. and M.T. Hoffman. 2003. Changes in population biology of two succulent shrubs along a grazing gradient. Journal of Applied Ecology 40: 615-625. Rodd A .V . , Y . A . Papadopoulos, L.F. Laflamme, K .B . McRae, S.A.E. Fillmore, and R.W. Wilson. 1999. Effect of rotational grazing on selected physical properties of a Gleyed Brunisolic Gray Luvisol loam in Nova Scotia. Canadian Journal of Soil Science 79:117-125. Roder, W., G. Gratzer, and K. Wangdi. 2002. Cattle grazing in the conifer forests of Bhutan. 59 Mountain Research and Development 22(4): 368-374. Sagar, R. and J.S. Singh. 2004. Local plant species depletion in a tropical dry deciduous forest of northern India. Environmental Conservation 31: 55-62. SAS Institute, Inc. 2004. SAS. 9.1.3. Cary, N C : SAS Institute Inc. Savadogo, P., L. Sawadogo, and D. Tiveau. 2007. Effects of grazing intensity and prescribed fire on soil physical and hydrological properties and pasture yield in the savanna woodlands of Burkina Faso. Agriculture, Ecosystems & Environment 118: 80-92. Schoenholtz, S.H., H . Van Miegroet and J.A. Burger. 2000. A review of chemical and physical properties as indicators of forest soil quality: challenges and opportunities. Forest Ecology and Management 138:335-356. Sheldrick, B.H. , and C. Wang. 1993. Particle size distribution. In M.R. Carter [ED.]. Soil sampling and methods of analysis. Boca Raton, FL: Canadian Society of Soil Science, Lewis Publishers, p. 499-511. Soil Classification Working Group. 1998. The Canadian system of soil classification. 3 r d ed. Agriculture and Agri-Food Canada Publication 1646. Ottawa, Canada: National Research Council of Canada. 187 p. Soil Survey Staff. 1999. Soil Taxonomy: A basic system of soil classification for making and interpreting soil surveys. 2nd ed. Washington, D.C.: United States Department of Agriculture, Natural Resources Conservation Service. 869 p. Talbot, L . M . , S.M. Turton, and A.W. Graham. 2003. Trampling resistance of tropical rainforest soils and vegetation in the wet tropics of north east Australia. Journal of Environmental Management 69: 63-69. Thompson, D.J., R.F. Newman, G. Hope, K. Broersma, and D.A. Quinton. 2000. Nitrogen cycling in silvopastoral systems in the Pacific Northwest: A review. Canadian Journal of Plant Science 80:21-28. Tiessen, H., and J O . Moir. 1993. Total and organic carbon. In M.R. Carter [ED.]. Soil sampling and methods of analysis. Boca Raton, FL: Canadian Society of Soil Science, Lewis Publishers, p. 187-199. Todd, S.W. and M.T. Hoffman. 1999. A fence-line contrast reveals effects of heavy grazing on plant diversity and community composition in Namaqualand, South Africa. Plant Ecology 60 142: 169-178. Topp, G.C. 1993. Soil water content. In M.R. Carter [ED.]. Soil sampling and methods of analysis. Boca Raton, FL : Canadian Society of Soil Science, Lewis Publishers, p. 541-557. Unger, P.W., and T.C. Kaspar. 1994. Soil compaction and root growth: A review. Agronomy Journal 86(5): 759-766. Universidad Nacional de Jujuy. 2006. Serie de precipitataciones para San Salvador de Jujuy, Periodo 1908-2006. Catedra de Agroclimatologia, Facultad de Ciencias Agrarias, Universidad Nacional de Jujuy. Unpublished data obtained June 2006. USDA, NRCS. 2003. National range and pasture handbook. Grazing Lands Technology Institute. Available at: http://www.glti.nrcs.usda.gov/technical/publications/nrph.html Accessed 15 April 2007. Van Haveren, B.P. 1983. Soil bulk density as influenced by grazing intensity and soil type on a shortgrass prairie site. Journal of Range Management 36:586-588. Wischmeier, W.H., and D.D. Smith. 1978. Predicting rainfall erosion losses - A guide to conservation planning. Agriculture Handbook No. 537. Hyattsville, M D : United States Department of Agriculture. 58 p. Wood, J.C., W.H. Blackburn, H.A. Pearson, and T.K. Hunter. 1989. Infiltration and runoff water quality response to silvicultural and grazing treatments on a longleaf pine forest. Journal of Range Management 42(5):378-381. Xie, Y . and R. Wittig. 2004. The impact of grazing intensity on soil characteristics of Stipa grandis and Stipa bungeana steppe in northern China (autonomous region of Ningxia). Acta Oecologica 25: 197-204. 61 3. OPPORTUNITIES AND CHALLENGES IN MOVING TOWARDS MORE SUSTAINABLE LIVESTOCK GRAZING IN THE JUJUY MODEL FOREST, ARGENTINA 3.1. Introduction Expansion of cultivated agricultural land and rapid human population growth in subtropical and tropical regions around the world have forced many impoverished livestock owners to move their grazing animals to more marginal lands (Sere et al. 1995). When livestock numbers exceed the carrying capacity of the local ecosystem, ecological processes such as site stability, hydrologic functioning, and biotic integrity can be negatively affected (Pellant et al. 2005), as has been reported in the Yungas forest ecosystems of northwestern Argentina (Braun Wilke et al. 2001, Grau and Brown 2000). Overgrazing tends to be particularly serious in regions where traditional patterns of grazing management have deteriorated, and where open, unregulated access to common lands occurs (Angassa 2005; Steinfeld et al. 1997). Since the early 1970s, there have been substantial increases in the demand for meat, eggs and dairy products, in nations of the global south (developing nations), with this demand primarily being met by increased livestock herd sizes in what is termed the "Livestock Revolution" (Delgado et al. 1999). This will intensify the pressure for livestock production on marginal agricultural lands, and emphasizes the importance of assisting livestock owners and landowners in moving towards sustainable livestock grazing practices. In 2002, the Jujuy Model Forest Association was established to work towards the long-term sustainable management of 1300 km of the Los Pericos-Manantiales watershed located in the southern part of Jujuy province, northwestern Argentina (IMFN 2007). Resource managers and Jujuy Model Forest volunteers have expressed concerns that heavy livestock grazing pressure in local Yungas forest ecosystems is causing land degradation, including soil erosion and declines in water quality (Outon 2002; Diaz Benetti et al. 2004). Livestock owners have noted that on heavily grazed hillsides soil erosion is causing rocks to emerge. Subsistence-scale livestock owners suffer economic consequences due to overgrazing, as many report that during the dry winter season livestock lose weight, and some animals perish because native forage \ 62 production is not adequate to support all grazing animals. The unauthorized presence of grazing animals on provincially owned land designated as a protected natural area, along roadways, and on private forested land, is a local source of social tension and conflict (Outon 2002; Diaz Benetti et al. 2004). In response to these concerns, in 2005 a research and education program entitled "Living with the Forest" was initiated in the Jujuy Model Forest to support livestock owners and land owners in moving towards more sustainable livestock grazing. Ripley et al. (in press) found that overstocking of domestic livestock is occurring in deciduous forest and anthropogenic pasture range sites, with the problem being particularly acute in the anthropogenic pasture, where growth of unpalatable plant species limits forage production and poses a hazard to livestock health. Indicators of soil quality, including soil cover, organic C, total N , and penetration resistance, suggested that land use practices in the anthropogenic pasture, including heavy livestock grazing, are leading to declines in soil quality in relation to the deciduous forest. Forest regeneration in the anthropogenic pasture is currently limited by heavy livestock grazing. Recommendations, which arose from this research, included reducing livestock numbers to within the grazing capacity, implementation of deferred, rotational grazing to enable forage regrowth, and limiting livestock movements on steep slopes during the wettest months of the rainy season. Although recommendations based on biophysical research are important tools to support livestock management decisions, current social and economic realities for livestock owners and landowners play a decisive role in determining whether changes to livestock grazing practices will be adopted. The objectives of this study are to identify the social and economic opportunities that exist in the Jujuy Model Forest that will support movement towards more sustainable livestock grazing practices2, and describe challenges that may pose barriers to such change. 3.2. Study Area and Methodology 3.2.1. Location and Ecological Characteristics of Study Area This study was carried out in Yungas forest ecosystems of the Jujuy Model Forest, in the 'Sustainable' livestock grazing practices do not degrade the natural environment, promote the development of social capital and are economically viable. 63 foothills of the Andes mountains, at approximately 24°S and 66°W, in southern Jujuy province, Argentina (Figure 3.1). The Yungas forest of southern Jujuy has a subtropical montane climate with an average December-February (summer) temperature of 21°C, and an average June-August (winter) temperature of 11 °C (Braun Wilke et al. 2001). This region has a monsoon precipitation regime with approximately 90% of the 780 mm annual precipitation falling in the rainy summer season between November and March (Braun Wilke et al. 2001). Data collection and educational activities were carried out in three ecological zones within the Yungas forest: deciduous forest, anthropogenic pasture and highland pasture. Deciduous forest sites (1200 - 1500 m elevation) are composed of secondary growth forest as extensive logging was carried out in this region in the mid 1900s through to the 1970s, with particularly intense selective logging around 1975 (Braun Wilke et al. 2001). Current land uses include livestock grazing, hunting, firewood cutting, and medicinal / edible plant gathering. Anthropogenic pastures (1200 - 1450 m elevation) are a mosaic of areas at the base of the foothills that have been converted from deciduous forest to open savanna landscape through intensification of timber extraction, firewood harvesting, and livestock grazing since the 1970s. Grasses, forbs, and shrub species that are resistant to grazing pressure dominate the open slopes of anthropogenic pasture sites, and are interspersed with scattered stands of trees or single trees. Current land uses include heavy livestock grazing pressure, timber / firewood cutting, and recreation. Highland pastures (1650 - 1775 m elevation) are open grass / shrublands found at elevations above the deciduous forest on exposed north and northwesterly facing slopes. These sites are dominated by large tussock grasses, interspersed with shorter grasses and forbs, and patches of bare ground. Livestock grazing is the predominant land use in this ecological zone. 3.2.2. Local Livelihoods and Livestock Grazing Practices Within the Yungas forest of the Jujuy Model Forest, extensive, season-long livestock grazing is the primary economic activity, with firewood extraction, timber harvesting, leaf litter removal, and medicinal and edible plant collection as important secondary activities. There are two basic types of land occupancy/ownership and livestock production systems occurring within our study area. Three families own large tracts of land (approximately 1000 - 5000 ha) (Outon 2005), primarily within the deciduous forest and highland pasture 64 zones, although some anthropogenic pasture is also present on one family's land. The primary economic activity for these families is tobacco farming on land they own that is located in nearby lowland valleys. Two landowners also raise cattle and livestock, and use their Yungas forest land for extensive grazing. The third family that does not own livestock rents their forested land to local subsistence-scale livestock owners. The second group of livestock producers within the study area consists of approximately 40 families who occupy tracts of provincially owned land adjacent to the Las Maderas irrigation dam, and produce livestock on a predominantly subsistence basis. Most families own a variety of livestock including cattle, horses, sheep, goats, pigs, and chickens. Herd sizes are variable, with examples of common household herd sizes being: a) 30 cattle, 7 horses, 1 pig; b) 20 cattle, 3 horses, 70 goats; c) 80 sheep, 20 goats; d) 17 cattle, 10 horses, 20 goats. Most families who occupy land adjacent to Las Maderas dam have fenced off up to 3 ha of land that they graze year-round (C. Chauque, personal communication, April 2007). Because this enclosed land base is too small to support their livestock herds, livestock are released to graze on unclaimed provincial land. This region adjacent to Las Maderas dam is designated as part of the anthropogenic pasture ecological zone. In addition, many livestock owners graze their animals in the adjacent deciduous forest and highland pastures, which are privately owned by the landowners who are part of this study. Some subsistence-scale livestock owners seek permission and pay annual rent to graze their animals on private land, while other livestock owners introduce their livestock onto private land clandestinely without paying rent. Although this study focused on a relatively limited geographic area, concentrated land ownership (DiPPEC 2002), subsistence-scale livestock production, and unregulated livestock grazing are common realities in the Yungas of northwestern Argentina (Cafferata 1996). There are strong cultural traditions surrounding livestock production and ownership in rural regions of northwestern Argentina. Many livestock owners are members of local social organizations that parade their horses in community festivals and regularly organize barbeques and events to celebrate rural lifestyles. There are bustling turnouts at the local annual doma (rodeo), and special national radio programs dedicated to celebrating the rural livelihood of the gaucho (cowboy) through music, coplas (sung poetry), and stories. Meat, and in particular beef, is a staple of the Argentine diet for all social classes, with Argentina having the highest per 65 capita beef consumption in the world at an estimated 64 kg/yr (AAFC 2006). 3.2.3. Formulation of Research Questions and Data Collection Ideas and reflections discussed within this chapter have been generated through a variety of experiences and diverse learning. M y understanding of socioeconomic realities and community dynamics in the Jujuy Model Forest has been primarily informed through my experience living in the town of El Carmen, and working as a volunteer and researcher within the Jujuy Model Forest. From January to July, 2004,1 began to meet and interact with local livestock owners and landowners, as I helped facilitate a community composting program, met with families who occupy land near the Las Maderas dam to explore their interest in using composting toilets, and assisted health promoters from Nuestra Senora del Carmen hospital conduct a survey of medicinal and edible plant use among rural families. During informal visits, subsistence-scale livestock owners expressed their concern over the death of livestock during the dry season, theft of livestock, and the fines or bribes they were required to pay to the local police when their animals were confiscated from the roadway on provincially-owned land. Landowners showed frustration at their inability to control clandestine grazing. Resource managers explained their concerns regarding land degradation due to livestock grazing. Informal visits about all of these topics led me to base my Masters research around the issue of livestock grazing within the Jujuy Model Forest. During August 2005 to May 2006,1 returned to live in E l Carmen to carry out field research in cooperation with local livestock owners and landowners. The location of the 18 field sites where biophysical data were collected were chosen not only based upon ecological criteria, but also upon the opportunity to directly involve subsistence-scale livestock owners and large landowners directly in research activities. Of the approximately 40 subsistence-scale livestock owning households who occupy land adjacent to Las Maderas dam, four families were selected to be 'guardians' of a field site located near their home and participated directly in field research activities. I had the opportunity to engage in informal conversations with members of approximately 19 additional households in this region through educational and community events that enabled further shared learning, including: 66 • A n interactive workshop in the Los Naranjos area near Las Maderas dam to stimulate sharing and discussion among subsistence-scale livestock owners regarding the goods and services the forest provides, along with any observations of change they have noticed in the local landscape (December, 2005; six participants). • Rural tourism pilot activity (January, 2006; ten households participated directly in planning and carrying out activities; members of five additional households joined in the evening social event) • Hands-on workshop on "Getting to Know the Soils Where We Live" in the Los Naranjos area, near Las Maderas dam (March, 2006; seven participants) The remaining 14 field sites were located on private land owned by three local families. An understanding of their agricultural goals, challenges, and livestock production systems was learned through informal visits in their homes, phone conversations and visits on the street when we met one another when going about our lives in E l Carmen. A l l field research and educational activities were planned and carried out by a core team of four local university students and myself, with the enthusiastic assistance of 15 additional students and community members. Interacting and working with this diverse group of individuals provided additional opportunities to learn about local and regional resource dynamics and human ecology. Experiential learning has been supplemented by secondary data sources including Jujuy Model Forest documents, ecological reports compiled at the National University of Jujuy, and statistics from the 2002 agricultural census conducted by the National Institute of Statistics and Censuses (INDEC). 3.3. Sustainable Livestock Grazing Practices and Economic Alternatives 3.3.1. Reduction of Stocking Rates to Within Local Grazing Capacity Grazing capacity refers to the maximum number of animals an area of land can support over the long term without causing ecosystem deterioration (Holechek et al. 1998). The number of animals using the forage resources within an area of land over a set period of time is known 67 as the stocking rate (PFRA 2003). If the stocking rate for an ecological site exceeds the grazing capacity over the long-term, declines in ecosystem functioning and forage production can result (Martinez and Zinck 2004; Holechek et al. 1998). Forage production and soil quality research carried out in 2005-2006 indicated that the stocking rates for domestic livestock need to be reduced in the anthropogenic pasture and deciduous forest, and that further research of tiller longevity and grazing intensity is needed in the highland pasture of the Jujuy Model Forest (Ripley et al., in press). Landowners and livestock owners face unique opportunities and challenges regarding their propensity and ability to make changes to their current livestock management practices. Therefore, the opportunities and challenges for each of these groups will be discussed separately. 3.3.1.1. Subsistence-scale livestock owners One of the driving forces for the initiation of this project was the concern of some subsistence-scale livestock owners about annual livestock deaths and weight loss during the dry season, due to excess livestock numbers relative to native forage availability. The fact that the current imbalance between stocking rates and forage production is severe enough to be causing such a direct, visible consequence for livestock owners in terms of income or meat production can be considered an opportunity. The gravity of the situation likely increases the interest of livestock owners in considering management alternatives that could alleviate this problem, and help to improve the long-term economic, social and environmental viability of the grazing system. Adjusting the number of grazing animals to meet grazing capacity recommendations will improve productivity on a per animal basis (Holechek 1998): This should result in higher meat production / animal when slaughter takes place at the household level for personal consumption, and greater income to livestock producers per animal unit sold. A reduction in animal deaths during the dry season improves the 'inflation-proof saving' feature of livestock ownership, which is often highly valued in subsistence-scale systems (Ayalew et al. 2003). However, given that current stocking rates in the anthropogenic pasture are 2 to 3.5 68 times higher than recommended stocking rates, in order for productivity gains on an individual animal basis to be realized, the number of animals currently grazing in this ecological zone will likely need to be reduced substantially. This increases the likelihood that in the short term livestock owners would suffer a decline in income and/or sustenance that they currently gain from livestock. Continuation of current grazing practices and stocking rates will likely lead to future more long-term declines in livestock production, given the current prevalence of toxic and unpalatable plants within the anthropogenic pasture, compared with the adjacent deciduous forest (Ripley et al. in press). One df the greatest opportunities that exists within the community of livestock owners is the interest of a small group of households in expanding their learning about livestock grazing systems, and experimenting with alternatives to improve livestock productivity. In order for change to occur at a broader community level, it is imperative that the Jujuy Model Forest Association, or another organization, assist livestock owners in satisfying their desire to learn more about livestock management systems and techniques. The Jujuy Model Forest Association has indicated it would be interested in supporting livestock owners in such learning efforts (V. Outon, personal communication, January 2006). It is important for individuals involved in cultivating these seeds of change to be aware of current social realities that may present barriers to more sustainable livestock management. The current precarious land-tenancy situation of subsistence-scale livestock owners who occupy provincial land has created a sense of uncertainty and fear that members of the provincial legislature could attempt to evict community members from the land at any moment. Argentina is considered one of the more corrupt countries in the Americas, with a corruption perception index of 2.8 (± 0.3, a = 0.10) on a scale of 0 (highly corrupt) to 10 (highly clean) (Transparency International 2005). One of the visible impacts of widespread corruption at all government levels, and within public and private institutions, is an underlying mistrust among individuals in society and of government processes. Widespread mistrust of government and 'community outsiders' may potentially cause subsistence-scale livestock owners to view initiatives to reduce livestock numbers as a threat to their local livelihoods, particularly i f there are low levels of awareness that reducing stocking rates can improve animal health and quality. Therefore, it is very important that educational activities be carefully planned in close cooperation with local 69 community members who have close ties with subsistence-scale livestock owners, and carried out in a patient manner that encourages trust-building, transparency and confidence. An additional barrier to reducing livestock grazing intensity and improving livestock quality for subsistence-scale livestock owners relates to the current open and unregulated access to provincial land. During field research and education activities in 2005-2006, many subsistence-scale livestock owners commented that they found it difficult to control grazing intensity within the land they occupy because they were not able to prevent neighbouring families' animals from grazing on their land. Although families use barbed wire to fence the land they occupy, goats, sheep and pigs are often able to make their way through fences, and therefore graze throughout large areas of the provincial land adjacent to Las Maderas dam. This reality serves as a disincentive to subsistence-scale livestock owners who are interested in reducing their livestock numbers and improving livestock quality. Even i f they reduce their household stocking rates on land they occupy, it will be difficult for their animals to improve in health and gain weight if grazing pressure from other families' animals remains high; a frequent issue experienced by many other communities around the world (Campbell et al. 2000). Improved fencing would be needed to mitigate this problem. 3.3.1.2. Land owners A key opportunity that exists for land owners who were part of this study is their existing knowledge base, and on-going interest in improving the management of livestock on their land. The two landowners in our study area who also own livestock demonstrated a strong understanding of range management principles in informal discussions held with them. They already practice fenced, rotational grazing in some seeded pastures they own in valleys at the base of the foothills. These land/livestock owners produce and use hay as a supplemental feed for livestock during the dry season, regularly treat animals for health issues, and raise high-quality animals. Both land/livestock owners openly discuss their perspectives regarding livestock market prices, national policies relating to meat prices, and how these impact their family livestock operations. The landowner who did not own livestock expressed interest in more closely controlling activities of grazing animals on family-owned land. Associated with this, landowners are financially much more stable than subsistence-scale 70 livestock owners, and do not appear to view discussions of livestock numbers as a potential threat to their livelihood. As with most landowners in our study area, tobacco growing is the primary economic activity for the three landowners who participated in this study. Livestock production, or rent of land to other parties for livestock production, is of secondary economic importance. As a result, any potential economic fluctuations that could be incurred as a result of reducing livestock numbers would not likely pose a serious economic burden to the overall operations of these landowners. It is important to keep in mind that the economic stability and production of relatively high-quality animals could also pose potential deterrents to changing current livestock management practices for landowners. Adjusting stocking rates according to forage production tends to improve rates of animal productivity (Holechek 1998), however, the potential incremental gain/animal and improvement in animal health for land owners is less pronounced than for subsistence-scale livestock owners. A l l three landowners have expressed interest in expanding their knowledge base of range management, experimenting with new techniques, and improving livestock distribution. Perhaps the most salient advantage that landowners possess in regards to improving livestock management practices is that they have title to the land they possess and use. This enables land owners to make long-term farm management plans with the security that they will still have a land-base to operate on, and that they and/or their family members will be able to benefit from infrastructural improvements and investments they make on their land. In conversations throughout the period of this study, landowners expressed that one of the greatest challenges they face in regards to livestock management is unauthorized livestock, grazing on their land in deciduous forest and highland pasture zones. Due to the extensive nature of their land holdings on montane, forested landscapes, it is difficult for landholders to monitor the movement of all livestock on their land. The prevalence of clandestine grazing on private land poses a serious challenge to efforts to reduce stocking rates. Even i f landowners reduce the grazing intensity of animals they have authorized to graze on their land, clandestine grazing poses a real potential for overgrazing to continue. Within the past two years, two of the landowners have expanded perimeter fencing around their land to attempt to reduce the entry of 71 unauthorized animals. 3.3.2. Implementation of Deferred, Rotational Grazing Systems In addition to reducing stocking rates to correspond with the grazing capacity of ecological zones, the timing and season of grazing are also important considerations in improving the sustainability of livestock grazing practices (Holechek 1998). Season-long, continuous grazing is the most common livestock grazing system currently utilized by livestock owners and land owners in the Jujuy Model Forest. Deferred, rotational grazing is an alternative in which a number of herds of livestock are rotated among multiple pastures during the grazing season in such a manner that each pasture also receives a rest from grazing to allow for re-growth (USDA-NRCS 2003). Deferred, rotational grazing systems tend to improve forage productivity in mesic to humid moisture regimes (Holochek 1998), and can also improve watershed protection and wildlife habitat (USDA-NRCS 2003). 3.3.2.1. Subsistence-scale livestock owners Deferred, rotational grazing systems can lead to improvements in livestock productivity and health (Buttolph and Coppock 2004), and therefore greater meat production or financial gain when animals are sold. However, implementation of deferred, rotational grazing requires some financial investment to build fences, which poses a barrier for subsistence-scale livestock owners. Rotational grazing normally requires pastures to be fenced, so the movement of animals across the landscape can be timed and controlled. In the anthropogenic pasture ecological zone, families who are subsistence-scale livestock owners occupy a small area of land they have fenced off, which in many cases does not exceed 3 ha. Most livestock owners rely on forage in anthropogenic pasture common lands, and adjacent privately-owned deciduous forest and highland pasture to sustain their animals throughout the year. Common lands are. not fenced, apart from the plots claimed by subsistence-scale livestock owners. Some private deciduous forest and highland pasture lands have perimeter fences, but internal fencing is scarce to non-existent. As a result, before a system of rotational grazing can be implemented, more perimeter and cross-fencing will be required. It is important to note that in the deciduous forest where the 72 primary livestock to be controlled are horses and cattle, the lower wire of fences should be strung at a height to allow unobstructed movement of a small native deer (corzuela), Mazama gouazoubira, and other native herbivores through local ecosytems. A potential, more 'immediate alternative to increased fencing would be to train and employ individuals to act as shepherds to control the timing and distribution of livestock grazing: 3.3.2.2. Land owners Two of the three landowners who participated in this study own livestock, and have already implemented a system of rotational grazing on planted pastures in lowland valleys. A l l three landowners have recently expanded the network of fences on their land. Therefore opportunities exist to build on this knowledge and infrastructure base to support and encourage rotational-grazing of native forage resources in the Yungas forest region. Such changes in management practices will require local research to support decisions regarding optimal rotation and deferment times, and what rotational grazing systems would most benefit grazing animals and ecosystem integrity. Possible sources of extension support include the National Institute of Agricultural Technology (INTA) and Jujuy Model Forest Association. As with adjustments to grazing intensity, successful management of grazing timing and distribution on privately owned land wil l depend upon the ability of landowners to limit clandestine grazing practices. 3.3.3. Community Pastures and Co-operative Grazing The issue of livestock management and land use on a societal scale has been resolved using a broad array of human social constructions around the world. For example, in rural communities in the Himalayas (Richards 1999), and the Andean altiplano (Buttolph and Coppock 2004), common property tenure was used in areas where people grazed livestock for hundreds or thousands of years. In many regions, common property tenure provided subsistence-scale livestock owners with access to land, and complex social norms required that the land be used in a manner that ensured the long-term sustainability of the grazing system (Richards 1999). In Canada, one mechanism by which federal and provincial public land is made available 73 to small-scale farmers and ranchers to graze their livestock is through community pasture programs. Grazing space on community pastures is allotted in inverse proportion to the amount of land a livestock producer owns, leases or rents, so that supplemental grazing on public land is preferentially provided to small farm or ranch operators (PFRA 2003b). If a livestock owner applies, and is accepted to graze his/her animals on a community pasture, a grazing fee must be paid based on the type and number of livestock to be grazed and the total time the animals spend on public land (PFRA 2003b). In the case of federally-managed community pastures, pastures are managed by a full-time pasture manager, and supporting technical, field and range conservation staff. On provincial lands, livestock owners are responsible for managing their own livestock, and must submit and follow a grazing management plan that outlines grazing intensity, animal rotations, and other management techniques that will be used to ensure the range resources are not degraded. Failure to follow this plan can result in the livestock owner having his/her grazing permit revoked. (B.C. Ministry of Forests Forest Practices Branch 2000; Alberta Sustainable Resource Development 2005) An additional community-based initiative that assists livestock owners to access grazing land is grazing cooperatives / grazing associations. The principle behind a grazing cooperative is that a group of livestock owners pool their resources to buy or lease land which can then be used by all members of the group, under a set of rules the group has established. There is normally a membership fee to join a grazing cooperative, in addition to annual fees which are paid based upon the number of animals a livestock owner grazes on cooperative-held land. These fees are used to cover land payments / rent, taxes, insurance, equipment, and for some cooperatives, the salary of a pasture manager who tends and rotates the animals grazing on cooperative land. (D. Hardy, personal communication, May 2007). Both community pastures and grazing cooperatives allow small-scale livestock owners to have access to grazing resources they may not otherwise be able to utilize if they needed to lease or purchase land as an individual operator. These cooperative structures allow many livestock owners to be more diversified, as during the time their livestock graze on community pasture or cooperative land, livestock owners can spend more time producing crops and undertaking other activities. The cooperative nature of grazing cooperatives also encourages 74 linkages and social networks among livestock owners. Community pastures and grazing cooperatives provide a potential model for livestock owners, land owners, and resource managers to work together to resolve some of the challenges associated with livestock grazing within the Jujuy Model Forest. One of the largest obstacles to establishment of a community pasture within the provincially owned land adjacent to the Las Maderas dam is that since 2003 the land has been designated as a "Protected Natural Area", with primary land use goals including protection of land and water resources, ecotourism and recreation (Legislatura de Jujuy 2003). According to proposed zonation for the protected area, the zone where livestock owners occupy land has been designated as a 'transition zone', where tourism is encouraged and cattle and horse production are discouraged (Legislatura de Jujuy 2003b). However, given the politically and socially-sensitive nature of land tenancy and occupancy around the Las Maderas dam, there is hesitation by provincial politicians to address the issue, and for the moment the status quo of land occupancy and livestock grazing continues. Before a provincially-run community pasture system could be established, it would be imperative to resolve the issue of land tenancy, occupancy and management, and principal land use objectives for the region would need to change to be more supportive of agricultural activities. Given this situation, it appears that the feasibility of establishing a community pasture on provincial land near Las Maderas dam is currently relatively limited. However, the concept of cooperative grazing may be a useful model that could be employed to improve relationships between subsistence-scale livestock owners and landowners and to decrease the incidence of unauthorized livestock grazing on private land. If a number of subsistence-scale livestock owners were to partner together, and plan the number of animals they would like to graze on adjacent privately owned deciduous forest / highland pasture land, they could reach a cooperative agreement with the landowner regarding the number of animals to be grazed and timing of grazing. There are already a number of subsistence-scale livestock owners who have individual agreements with landowners to graze their livestock on privately-owned land. An advantage livestock owners would gain by grazing their animals cooperatively is that they could share the work of checking on grazing animals, and herding animals among different zones of forest and pasture. This may make it more feasible for livestock owners to have greater control over where their animals are grazing, and open up the possibility of 75 working towards implementation of some rotational grazing and deferred grazing practices. There were previously strong social traditions of community members working cooperatively to carry out agricultural work in Jujuy (C. Chauque, personal communication, April 2007). Claudia Chauque, one of the research team members, and a resident of E l Carmen whose family own livestock and lives in the Las Maderas region, tells how her family remembers a time period when people used to work together as a community to carry out agricultural activities, in what was known as minga: " M y mother remembers that my grandfather knew to go to another family's house, and he stayed for two or three days, helping to clean enclosures to be planted, and that was the way almost all the heads of households who lived nearby would go to one house and help. When they finished they would barbeque and eat a small animal (sheep) and they drank corn liquor to celebrate the activities that they had completed together" Claudia's family remembers this occurring until into the 1960s, and her mother recalls that friends of their family used to come to their house to cultivate the soil and seed. Although there may be less minga occurring in the present day, community members in the Las Maderas region have formed a neighbourhood association and meet regularly to work together to solve issues of importance to their community. This process of social gathering presents an opportunity for development of more cooperative livestock management. 3.3.4. Economic Diversification through Rural Tourism Diversification in agricultural production systems is often promoted as a manner to increase income and/or food consumption, reduce risks by alleviating dependency on one or a few agricultural commodities, and increase income stability (Warren 2002). Rural tourism, or agrotourism, encourages visitors to experience agricultural life first-hand by observing and/or participating in agricultural activities with rural residents, and can serve as a form of income diversification (Dolors Garcia-Ramon 1995). Other benefits of rural tourism include an opportunity to market agricultural products, creation of social and economic relationships between urban and rural dwellers, education of urban people regarding the contribution of agriculture to their livelihoods (Akpinar et al. 2004), and the formation of bonds among residents of the community undertaking rural tourism (Saxena 2006). 76 Since the formation of the Jujuy Model. Forest in 2002, Model Forest volunteers and institutional members have frequently suggested rural or eco-tourism as a potential source of economic diversification (Diaz Benetti et al. 2004). During the past 10 years, Jujuy has become a more common destination for national and international tourists, with many travel agencies conducting tours along a highway that connects the cities of Salta and Jujuy and passes through the Jujuy Model Forest, near Las Maderas dam. Some subsistence-scale livestock owners and landowners reported experiences in rural tourism prior to the initiation of this research and ^ education program. 3.3.4.1. Subsistence-scale livestock owners For approximately 10 years, some subsistence-scale livestock owners have rented their horses for horseback rides, and sold homemade bread near the local irrigation dams where visitors from nearby communities gather on weekends. In January, 2006, ten families who are subsistence-scale livestock owners, living in one valley near Las Maderas dam, cooperated together to plan and conduct a day-long rural tourism activity for four international visitors with assistance from project team members (Appendix VI). The objectives of this activity were to provide a positive learning environment for subsistence-scale livestock owners to gain first-hand experience in community-based rural tourism, and the opportunity to share their rural culture and daily activities with visitors in a participatory manner. A few days following the rural tourism day, a woman who had participated told one of the project team members that she had been able to buy sugar and other household staples with the extra income she had earned through her participation in the rural tourism day. However, in February, 2006, at the next valley neighbourhood meeting, community members narrowly voted to discontinue rural tourism on a community-scale. Some community members were fearful that if rural tourism became successful the provincial land they currently occupy may be privatized, families may be evicted, and livestock grazing may be banned. In 2006, a new provincial land administration unit began operating on the provincially owned land surrounding the irrigation dams within our study area. The administrative unit has been working to establish positive and respectful relationships with subsistence-scale livestock owners who occupy provincial land, and rural tourism has been discussed again as a potential 77 economic opportunity. Local community members have again expressed interest in this idea, yet the same concerns previously expressed remain, and the concept of rural tourism is still a potential source of dissention within the local community. 3.3.4.2. Landowners In conversations about income diversification between October 2005 and May 2006, two of three large landowners who participated in the forage production and soil quality study described their interest in leading trail rides through the montane forest and grasslands on their land, hosting visitors overnight and demonstrating agricultural and livestock-rearing activities. Landowners possess the infrastructure and financial resources to establish such an operation. An on-going challenge for landowners is that they are often very busy with agricultural operations, and do not have a lot of extra time to plan new initiatives. Given that financial need is not a pressing issue, landowners may feel less of an economic drive to pursue rural tourism. Landowners who expressed interest in rural tourism, spoke particularly about their interest in meeting visitors from other regions. 3.3.5. Implementation of Agroforestry Practices Agroforestry is generally known as the practice of purposefully growing trees in combination with crops and /or livestock in a complementary manner to achieve environmental, social and economic benefits (World Agroforestry Centre 2007; Wojtkowski 1998). There has been a great expansion of formal agroforestry research since the 1970s (Young 1989); however, human populations have practiced forms of agroforestry for thousands of years (McNeely 2004; Kumar and Nair 2004). Some commonly cited ecological; social and economic benefits of agroforestry include: • Improved soil fertility and reduced erosion susceptibility (Young 1989) • Improved food security through the production of fruits, nuts and edible oils, as well as improvements in farm soil fertility (Kumar and Nair 2004) • Diversification of rural income and poverty reduction through production of agroforestry products for household consumption and sale (Garrity 2004) 78 • Reducing deforestation and pressure on woodlands by providing fuelwood grown on farms (Shavanas and Kumar 2003) The anthropogenic pasture ecological zone of our study area has reduced soil quality, and is the ecological zone where most intensive land use and inhabitation occurs. For these reasons, this zone is most suited to agroforestry development. Examples of some potentially useful agro forestry systems in the anthropogenic pasture of the Jujuy Model Forest include: • Silvo-pastoral systems Tree and livestock production can be purposefully combined in a variety of manners. Although the current practice of grazing domestic livestock within Yungas forests can be considered a form of agro forestry, there are opportunities for additional silvo-pastoral arrangements in the anthropogenic pasture. There are currently a number of leguminous trees growing in anthropogenic pastures that are known for their ability to provide forage for livestock {Acacia caven, Acacia aroma, Acacia praecox). Planting high quality forage-producing trees in a scattered arrangement throughout the anthropogenic pasture could increase arboreal forage production, particularly i f branches above the reach of grazing animals are pollarded during the dry season to provide additional forage for animals (Wojtkowski 1998). Forage trees also provide shade for livestock, which reduces temperature stress, and can increase animal productivity (Cajas-Giron and Sinclair 2001). Addition of biologically fixed nitrogen to the soil by leguminous trees should eventually increase herbaceous forage production and protein content, further improving livestock carrying capacity. It is imperative that subsistence-scale livestock owners and landowners be directly involved in the selection of leguminous tree species for the anthropogenic pasture. Although the three Acacia species currently present in the pasture are known as forage-producers, they are also indicators of heavy grazing pressure^ and Acacia praecox is regarded as quite a nuisance due to its sharp barbs and low, brushy growth form. Agroforestry projects have the greatest likelihood of success if the people who will be managing the systems have actively participated throughout the entire planning and design process, including tree selection (Warner 1995). A n additional silvo-pastoral option is to grow a 'forage bank' consisting of a fenced plot dedicated to the production of superior biomass trees that can be used as forage. Biomass of 79 these forage trees can be cut, and transferred to animal pastures during the dry season when forage is most scarce. (Wojtkowski 1998) In our study area, indicators of soil quality, including soil cover, organic C, total N , and penetration resistance suggest that land use, including heavy livestock grazing pressure and tree removal, in the anthropogenic pasture is leading to declines in soil quality in relation to the deciduous forest (Ripley et al. 2007). Planting scattered leguminous trees that have high biomass production and good erosion control characteristics in the anthropogenic pasture, would likely help to increase canopy and litter cover, soil N levels and nutrient cycling, i f carried out in conjunction with a reduction of stocking rates to local carrying capacity (Young 1989). • Living fences The agroforestry technique of planting trees to support wire fencing along an area to be demarcated provides an opportunity to jointly satisfy multiple needs within the anthropogenic pasture of the Jujuy Model Forest. The living fences would assist with the need to improve animal containment, and could also serve as a source of biomass production i f a tree species that can be regularly pollarded for firewood or forage is used. (Wojtkowski 1998). • Homegardens In this system, a diverse variety of trees is cultivated in conjunction with other edible and medicinal crops near a family's home. Trees are selected to provide a variety of goods and services to the household, such as fruit, nuts, forage for domestic animals, and/or timber and firewood (Kumar and Nair 2004). Some subsistence-scale livestock owners have already transplanted edible and /or medicinal plants from the forest understory, such as the pepper Capsicum microcarpum (Aji kitucho) to garden plots they tend near their houses. Further homegarden development would present an opportunity for increased and diversified food production, particularly i f the issue of household water scarcity can be addressed (C. Chauque, personal communication, April 2007). It is beyond the scope of this thesis to fully evaluate each of these options, or provide a definitive list of potential options. I will focus on identifying some community-level strengths and opportunities that could support development of agroforestry initiatives in a general sense, and outline challenges that could pose barriers within the Yungas of the Jujuy Model Forest. 80 Some opportunities and challenges differ between subsistence-scale livestock owners and landowners, whereas others are shared between these two community groups. 3.3.5.1. Subsistence-scale livestock owners One of the advantages of agroforestry initiatives is that they can be initially started on a small pilot-scale with producers who are enthusiastic and dedicated to experimenting with new ideas. Mercer (2004) suggests that this is the most common pathway of agroforestry adoption. Five livestock owners and their families consistently attended interactive educational workshops, and expressed interest in experimentation and the use of non-timber forest products. Among livestock owners there is a rich and diverse knowledge of the growth patterns and characteristics of local trees, and other medicinal and edible plants. This knowledge would be instrumental in planning the components of potential agroforestry systems. One of the potential barriers to adoption of agroforestry programs in areas such as Las Maderas dam is that trees are a form of long-term investment, and therefore particularly sensitive to property rights (Place and Otsuka 2002). The livestock owners who occupy land do not have the security that long-term benefits of planting and tending trees will be realized by themselves or their families. Some potential benefits from agroforestry systems could be realized on a short to medium time frame ( 3 - 5 years), such as fruit production and increased forage production from fast-growing species, which could assist with the willingness of residents to dedicate time and effort to such an initiative. 3.3.5.2 Landowners Given the large land bases held by landowners, dedication of a small area to an' agroforestry pilot project would likely not be a,problem. Land ownership implies security that the resources invested in planting trees and establishing agroforestry systems will benefit either the land owner or his/her family. One landowner has expressed particular interest in conducting experimental trials to improve forage production on anthropogenic pastures. This enthusiasm presents an opportunity for a pilot-scale trial, which could potentially serve as an example for other landowners. Landowners run very full schedules, particularly during tobacco planting and harvesting. 81 Time availability to tend to an agro forestry project is therefore an important constraint to take into account when examining agroforestry options. A potential option would be for landowners to employ one or a number of livestock owners to assist with maintenance of the agroforestry project. Payment for livestock owners' services could be in cash, or in exchange for grazing rights on the private land. Such a trueque, or traditional barter arrangement, could help to create more positive relationships between landowner and livestock owners, and encourage more knowledge sharing amongst these two communities. 3.3.5.3. Common opportunities and challenges Planning and designing agroforestry systems can be a knowledge and research-intensive undertaking (McNeely 2004; Mercer 2004), however this knowledge can be acquired and transferred through diverse pathways. Many indigenous cultures have a long history of maintaining diverse agroforestry systems: Javanese homegardens are thought to have originated in the seventh millennium B.C. (Hutterer 1984), and agroforestry was likely the method by which plants were first domesticated in Central America (McNeely 2004). These systems have evolved through biological and cultural transformations for thousands of years, and represent a diverse accumulation of knowledge that has been learned, transferred and expanded upon without the use of inputs from outside the regional area, capital or formal scientific skills (Kumar and Nair 2004). However, subsistence-scale livestock owners and landowners in the study area are not currently practicing agroforestry techniques, apart from grazing livestock in forested ecosystems, and do hot have a recent local history of practicing agroforestry. Therefore, planning, education, research, and organizational support would likely be required to support livestock owners and landowners in initiating an agroforestry pilot project. Jujuy Model Forest volunteers are likely candidates to assist with this, yet as in many volunteer organizations, human resources are often stretched quite thin. Current collaboration between the Jujuy Model Forest and National University of Jujuy presents an opportunity to establish an agroforestry internship, or encourage undergraduate thesis research on agroforestry and agrosilvopastoral systems. It is important to emphasize that although technical and research assistance for producers is important, greater adoption of agroforestry methods generally occurs when land users play a key role in final agroforestry design decisions and are able to undertake 82 their own on-farm modifications of agroforestry designs (Wiersum 1994; Warner 1995) Current livestock grazing pressure within the anthropogenic pasture poses a potential barrier to the success of an agroforestry system. Any plans to plant trees, or cultivate crops below trees will need to be accompanied by measures to protect the young trees and/or crop from being consumed or trampled by grazing livestock. Options for protecting seedlings include fencing, and piling thorny, leafless branches around young trees (Wojtkowoski 1998). An additional constraint to seedling and crop establishment is severe water shortage during the dry season. Livestock owners consistently cite water scarcity during the dry season as one of their largest concerns, and there are currently no irrigation systems in place in the Las Maderas dam area where livestock owners occupy land. Although land owners use irrigation for their tobacco operations in the lowland valleys, water scarcity would likely still be an issue at anthropogenic pasture sites on their land where an agroforestry pilot project would be the most feasible. 3.3.6. Formation of a Grazing Management Group A common theme among many of the potential solutions to support more sustainable livestock management within the Jujuy Model Forest is the need for community-based education, and increased cooperation among all livestock owners and landowners. During the past decade, 'watershed management groups' have formed in many rural areas across Canada, as people join together to resolve issues of common and inter-related interest such as water quality, land use and riparian habitat protection. These groups are sometimes initiated by a local resident who wishes to bring fellow community members who live in the vicinity of a water body together to exchange ideas, obtain educational resources to learn more about watershed management, and access funding for community-based watershed conservation initiatives (D. Hardy, personal communication, May 2007). This model, adapted to the local context of the Jujuy Model Forest as a 'Grazing Management Group' could serve very well to provide a social space where livestock owners and landowners could meet to discuss issues of mutual interest. One potential challenge that would need to be overcome to form such a group would be the social barrier that exists between livestock owners and landowners due to the strong class system in northern Argentina, and the great disparity in income, and political and social 83 influence, that exists between these two groups. It would be extremely important to choose meeting styles and locations that were considered neutral and comfortable for both communities. It may be helpful to seek a mutually-trusted facilitator to mediate, and assist with initial meetings. It is possible that livestock owners may feel hesitant about participating in such a group due to unequal social and power dynamics where land owners have secure land rights, social and political influence, and livestock owners are concerned about the uncertainty of their land tenancy. However, theoretically, an opportunity for collaboration should exist given that both communities could attain benefits from increased communication and learning about livestock grazing. The Jujuy Model Forest would be an appropriate organization to help facilitate the formation of a livestock management group, as one of the principal objectives of model forests is to encourage diverse stakeholders on a landbase to work together to solve problems. However, ultimately, the impetus and enthusiasm to form a livestock management group, and ensure its long-term functioning, will need to come from livestock owners and landowners themselves. 3.7. Conclusions and Recommendations Although there are many challenges which pose potential barriers to the adoption of more sustainable livestock grazing practices within the Yungas of the Jujuy Model Forest, there are also many valuable opportunities for change, learning and increased dialogue among livestock owners, landowners and resource managers. Many of the opportunities and challenges for subsistence-scale livestock owners differ from those of landowners. 3.7.1. Livestock Owners Key opportunities for livestock owners that would support movement towards more sustainable livestock management are the interest of some livestock owners in learning more about alternative range management practices, and improving the health and quality of their animals, as well as the potential which exists for improvement to be made in animal health and production. Practical hands-on workshops and educational activities that require meaningful input and active participation of livestock owners are key tools that could be used to increase 84 awareness about techniques in more sustainable livestock management. Viable economic alternatives that can assist livestock owners to diversify their income have the potential to play a role in supporting a reduction of livestock numbers. Many livestock owners have expressed interest in rural tourism, while others are very hesitant and feel this economic option poses a threat to their livelihoods due to precarious land tenancy. This option will need to be explored carefully, respectfully, and in a sensitive manner to ensure that it does create division and conflict among livestock owners. Agroforestry techniques could potentially diversify incomes and diet for livestock owners, in addition to providing other benefits such as rehabilitation of degraded anthropogenic pasture, and improved delimitation of pastures. However, insecure land tenancy for livestock owners reduces the incentive for long term investment in the landbase. Educational and research support would be needed to support the community in learning about agroforestry techniques. Open, unregulated access to public land, and the currently precarious nature of land tenancy are the two most significant challenges to more sustainable livestock grazing for livestock owners. Although deferred, rotational grazing systems present an opportunity for improving forage production and quality, which should lead to gains in livestock productivity, this option currently has limited viability within the anthropogenic pasture near Las Maderas dam where subsistence-scale livestock owners live due to complex land tenancy dynamics and open access to common lands. It appears that the current feasibility of establishing deferred, rotational grazing systems or a community pasture on provincial land near Las Maderas dam is relatively limited. However, the concept of cooperative grazing may be a useful model that could be employed to improve relationships between subsistence-scale livestock owners and landowners and to decrease the incidence of unauthorized livestock grazing on private land. 3.7.2. Land Owners A strong knowledge base of range management principles, interest in experimentation and improving livestock management practices, existing infrastructure, and financial stability 85 are all strengths that landowners have that will assist them in moving towards more sustainable livestock grazing. Perhaps the greatest advantage for landowners is that they have title to their land, and therefore the security that improvements in land management have the potential to provide benefits to themselves or their family members over the long-term. Although land owners have secure title to their land, given the large expanses of land they own, and montane, forested topography, it has been difficult for them to control unauthorized entry of livestock onto their landbase. Efforts to alter grazing pressure or timing -on their land may not result in improvements in range or livestock health if this unauthorized grazing cannot be controlled. Agroforestry presents a viable option for rehabilitating degraded anthropogenic pasture, provided that landowners are able to dedicate their time, or contract assistance, to carry out the agroforestry development. 3.7.3. Common Ground The success of proposed solutions to enable both landowners and livestock owners to move towards more sustainable livestock grazing depends upon the stimulation and support of further community-based learning, as well as increased cooperation among land owners and livestock owners. Formation of a 'Grazing Management Group', could provide a public forum for interaction, learning and problem-solving for livestock owners and landowners, i f there is a desire within these communities to apply this model in the local context. Although this analysis of opportunities and challenges has focused in detail on the Jujuy Model Forest, many other communities within the Yungas forest are in similar situations, and may find it useful to apply and interpret these lessons according to their own needs and realities. 86 This material has been removed due to copyright restrictions. This page contained a series of three maps depicting the location of the study area: A map of Argentina within South America, a map of Jujuy within Argentina, and a map of the study area within Jujuy. Figure 3.1 Location of the Jujuy Model Forest, within Jujuy province, northwestern Argentina, South America. Maps adapted from Direction de Bosques, Secretaria de Ambiente y Desarrollo Sustentable (2003). 87 3.8. References Agriculture and Agri-food Canada (AAFC). 2006. Argentina. In: J. Wang [Ed.]. Bi-weekly Bulletin. Vol . 19 No. 11. August 4, 2006. Available at: http://www.agr.gc.ca/mad-dani/. Accessed 30 May 2007. Akpinar, N . , I. Talay, C. Ceylan, and S. Gunduz. 2004. Rural women and agrotourism in the context of sustainable rural development: A case study from Turkey. Kluwer Journal 6: 473-486. Alberta Sustainable Resource Development. 2005. Range and pasture management. Available at: http://www.srd.gov.ab.ca/lands/managingpublicland/rangemanagement/ rangepasturemanagement.aspx. Accessed 25 May 2007. Angassa, A , 2005. The ecological impact of bush encroachment on the yield of grasses in Borana rangeland ecosystem. African Journal of Ecology 43: 14-20. Ayalew, W., J .M. King, E. Bruns, and B. Rischkowsky. 2003. Economic evaluation of smallholder subsistence production: Lessons from an Ethiopian goat development program. Ecological Economics 45:473-485. Braun Wilke, R.H., E.E. Santos, L.P.E. Picchetti, M.T. Larran, G.F. Guzman, and C R . Colarich. 2001. Carta de aptitud ambiental de la provincia de Jujuy. San Salvador de Jujuy, Argentina: Departamento de Suelos y Ecologia, Facultad de Ciencias Agrarias, Universidad Nacional de Jujuy. 245 pp. British Columbia Ministry of Forests Forest Practices Branch. 2000. Range management and the Forest Practices Code. Version 3.0. Victoria, BC: Forest Practices Branch, Ministry of Forests. 32 pp. Buttolph, L.P., and D.L. Coppock. 2004. Influence of deferred grazing on vegetation dynamics and livestock productivity in an Andean pastoral system. Journal of Applied Ecology 41:664-674. Cafferata, A. 1996. Una caracterizacion global y actual de la actividad forestal y ganadera en zonas marginales. In: M . Manzanal [ED.]. E l desarrollo rural en el noroeste argentino. Salta, Argentina: Proyecto Desarrollo Agroforestal en Comunidades Rurales del Noroeste Argentino. p.98-106. 88 Cajas-Giron, Y.S. , and F.L. Sinclair. 2001. Characterization of multistrata silvopastoral systems on seasonally dry pastures in the Caribbean Region of Colombia. Agroforestry Systems 53: 215-225. Campbell, B . M . , D. Dore, M . Luckert, B. Mukamuri, and J. Gambiza. 2000. Special section: Land use options in dry tropical woodland ecosystems in Zimbabwe. Ecological Economics 33: 413-438. Chauque, C. Nurse, Hospital Nuestra Senora del Carmen, Volunteer with the Jujuy Model Forest Association. Personal communication. April 2007. Costas Otero, S. Director, Livestock Development Department, Department of Agricultural Development, Province of Jujuy. Personal communication. May, 2006. Delgado, C , M . Rosegrant, H. Steinfeld, S. Ehui, and C. Courbois. 1999. Livestock to 2020: The next food revolution. Nairobi, Kenya: International Food Policy Research Institute. 83 pp. Diaz Benetti, W., M.R. Echenique, A.S. Munoz, and J.J. Kulemeyer. 2004. Plan maestro de manejo sustentable: Cuenca Los Pericos-Manantiales. Proyecto FAO - TCP / A R G 2802 - Manejo sustentable de ecosistemas forestales de la cuenca Los Pericos-Manantiales. San Salvador de Jujuy, Argentina: Direccion de Bosques, Secretaria de Ambiente y Desarrollo Sustentable de la Nation. Direction Provincial de Planeamiento, Estadistica y Censos (DiPPEC). 2002. Explotaciones agropecuarias con limites definidos, por escala de extension, segun departamento - Total provincia de Jujuy. Censo Nacional Agropecuario. San Salvador de Jujuy, Argentina: DiPPEC. Dolors Garcia-Ramon, M . , G. Canoves, and N . Valdovinos. 1995. Farm tourism, gender and the environment in Spain. Annals of Tourism Research 22(2): 267-282. Garrity, D.P. 2004. Agroforestry and the achievement of the Millenium Development Goals. Agroforestry Systems 61:5-17. Grau, A. , and A.D. Brown. 2000. Development threats to biodiversity and opportunities for conservation in the mountain ranges of the Upper Bermejo River Basin, N W Argentina And SW Bolivia. Ambio 29:445-450. Hardy, D. 2007. Grazing cooperative member, and watershed management group organizer, 89 south-western Alberta. Personal communication. May 2007. Holechek, J.L., R.D. Pieper and C H . Herbel. 1998. Range management: Principles and practices. 3 r d ed. Upper Saddle River, NJ: Prentice Hall. 542 pp. Hutterer, K . L . 1984. Ecology and evolution of agriculture in Southeast Asia: In: An Introduction to Human Ecology Research on Agricultural Systems in Southeast Asia. Los Banos, Phillippines: University of Phillippines. [Cited in: Kumar and Nair 2004]. International Model Forest Network (IMFN). 2007. Jujuy Model Forest. Available at: http://www.idrc.ca/en/ev-23367-201-l-DO TOPIC.html. Accessed 16May 2007. Kumar, B . M . , and P.K.R. Nair. 2004. The enigma of tropical homegardens. Agroforestry Systems 61:135-152. Legislatura de Jujuy. 2003. Ley N°5365 Area NaturalProtegida "Los Diques". Available at: http://www.estrucplan.com.ar/Legislacion/Result.asp?Origen=Juiuy Accessed 25 May 2007. Legislatura de Jujuy. 2003b. Ley N°5378. Legislatura de Jujuy, Argentina. Available at: http://www.estrucplan.com.ar/Legislacion/Result.asp?Origen=Jujuy Accessed 25 May 2007. Martinez, L.J. , and J.A. Zinck. 2004. Temporal variation of soil compaction and deterioration of soil quality in pasture areas of Colombian Amazonia. Soil & Tillage Research 75:3-17. McNeely, J.A. 2004. Nature vs. nurture: managing relationships between forests, agroforestry and wild biodiversity. Agroforestry Systems 61:155-165. Mercer, D.E. 2004. Adoption of agroforestry innovations in the tropics: A review. Agroforestry Systems 61:311-328. Outon, V . 2002. Bosque Modelo Jujuy: Gestion integral de la cuenca hidrografica -propuesta. E l Carmen, Argentina. 41 p. Outon, V . 2005. President of the Jujuy Model Forest Association. Personal communication. February, 2005. Outon, V . 2006. President of the Jujuy Model Forest Association. Personal communication. January, 2006. Pellant, M . , P. Shaver, D.A. Pyke, and J.E. Herrick. 2005. Interpreting indicators of rangeland health. Technical Reference 1734-6, Version 4. Denver, CO: United States Department 90 of the Interior, Bureau of Land Management, National Science and Technology Centre. 122p. Place, F., and K . Otsuka. 2002. The role of tenure in the management of trees at the community level: Theoretical and empirical analyses from Uganda and Malawi. In: R. Meinzen-Dick, A. Knox, F. Place, and B. Swallow [EDS.]. Innovation in natural resource management. Baltimore, M D : John Hopkins University Press. Prairie Farm Rehabilitation Administration. 2003. Animal unit months, stocking rate and carrying capacity. Available at: http://www.agr.gc.ca/pfra/land/fftl .htm. Accessed 27 May 2007. Prairie Farm Rehabilitation Administration. 2003b. PFRA Community pasture program. Available at: http://www.agr.gc.ca/pfra/cpprog e.htm. Accessed 25 May 2007. Richards, C. 1999. Management of mountain commons in the Hindu Kush Himalaya's. International Centre for Integrated Mountain Development. Newsletter No. 35. Available at: http://www.icimod.org/archive/icimod/publications/newsletter/news35/n35toc.htm. Accessed 16 May 2007. Ripley, S.W., M . Krzic, A . A . Bomke, and R.H. Braun Wilke. In press. Impacts of livestock grazing on soil quality and forest structure in Yungas forest ecosystems of southern Jujuy, Argentina. Salafsky, N . , H . Cauley, G. Balachander, B. Cordes, J. Parks, C. Margoluis, S. Bhatt, C. Encarnacion, D. Russell, and R. Margoluis. 2001. A systematic test of an enterprise strategy for community-based biodiversity conservation. Conservation Biology 15:1585-1595. Saxena, G. 2006. Beyond mistrust and competition - the role of social and personal bonding processes in sustainable livelihoods of rural tourism businesses: A case of the Peak District National Park. International Journal of Tourism Research 8:263-277. Sere, C , H. Steinfeld, and J. Groenewold. 1995. World livestock production systems. Current status, issues and trends. FAO Animal Production and Health Paper No. 127. Rome, Italy: Food and Agriculture Organization of the United Nations. 58 pp. Shavanas, A. , and B . M . Kumar. 2003. Fuelwood characteristics of tree species in the homegardens of Kerala, India. Agroforestry Systems 58:11-24. 91 Steinfeld, FL, C. de Haan, and H. Blackburn. 2007. Livestock - environment interactions: Issues and options. Food and Agricultural Organization of the United Nations. Available at: http://vvww.fao.org/ag/aga/lspa/LXEHTML/policv/index.htm. Accessed 15 April 2007. Transparency International. 2005. Informe annual 2005 - Seccion especial para Latinoamerica y el Caribe. Available at: http://www.transparencv.org/publications/annual report. Accessed 11 May 2007. USDA, NRCS. 2003. National range and pasture handbook. Grazing Lands Technology Institute. Available at: http://www.glti.nrcs.usda.gov/technical/publications/nrph.html Accessed 15 April 2007. Young, A. 1989. Agroforestry for soil conservation. Wallingford, United Kingdom: C A B International and the International Council for Research in Agroforestry. 276 pp. Warner, K . 1995. Selecting tree species on the basis of community needs. Community Forestry Field Manual 5. Food and Agriculture Organization of the United Nations. Rome, Italy: FAO. 158 pp. Warren, P. 2002. Livelihoods diversification and enterprise development: An initial exploration of concepts and issues. LSP Working Paper 4: Livelihoods Diversification and Enterprise Development Sub-Program. Food and Agricultural Organization of the United Nations. Available at: ftp://ftp.fao.org/docrep/fao/008/i2816e/i2816e00.pdf Accessed 28 May 2007. Wiersum, K.F. 1994. Farmer adoption of contour hedgerow intercropping: a case study from east Indonesia. Agroforestry Systems 27:163-182. Wojtkowski, P. A . 1998. The theory and practice of agroforestry design: A comprehensive study of the theories, concepts and conventions that underlie the successful use of agroforestry. Enfield, N H : Science Publishers, Inc. 282 pp. World Agroforestry Centre (ICRAF-Peru). 2007. Sistema dindmico de manejo de recursos naturales. Available at: http://www.icraf-peru.org/pages/agroforesteria.php. Accessed 22 May 2007. 92 4. CONCLUSIONS, RECOMMENDATIONS AND EVALUATION OF RESEARCH METHODS 4.1. Chapter Overview Current livestock grazing practices are often considered a threat to ecological functioning and sustainable forest management within the Yungas forests of northwestern Argentina. This research project was undertaken in collaboration with local livestock owners, landowners, students, Jujuy Model Forest volunteers, and other community members to improve our understanding of the relationship between livestock grazing and Yungas forest ecosystem processes. This chapter will begin by outlining conclusions from the research. This will be followed by an evaluation of strengths and weaknesses of our study methods. Particular emphasis is placed on evaluating soil quality and forest structure indicators, and reflections on community participation and educational activities associated with research activities. The chapter will conclude with personal reflections on my role as a research facilitator and Jujuy Model Forest community member throughout this study. 4.2. Research Conclusions and Recommendations This one-year study of annual forage production within the deciduous forest, anthropogenic pasture and highland pasture ecological zones of the Jujuy Model Forest showed that forage production and recommended stocking rates are higher in the anthropogenic pasture (2970 kg/ha) than in the deciduous forest (476 kg/ha). It was not possible to directly measure annual forage production for the highland pasture, due to lack of information on tussock grass tiller longevity, however, a conservative estimate of forage production was made (733 kg/ha) for the purpose of making a preliminary stocking rate recommendation. Based on annual forage production measurements, and estimates of current stocking rates, overstocking of domestic animals is occurring in both deciduous forest and anthropogenic pasture zones, with the problem being particularly acute in the anthropogenic pasture. Scarce information regarding current stocking rates in highland pastures precluded a comparison of 93 current and recommended stocking rates in this ecological zone. Although the anthropogenic pasture currently has the greatest forage production, this zone also has 8 times more unpalatable biomass than the deciduous forest and 6 times more unpalatable biomass than the highland pasture. If heavy grazing pressure continues, the proportion of unpalatable species will likely increase further (Milchunas and Lauenroth 1993), leading to declines in forage production and grazing capacity (Abule et al. 2005; Angassa 2005), and increased hazards to livestock health. Currently, reduced forest cover within the anthropogenic pasture has resulted in increased forage production, which could act as an incentive for further conversion of deciduous forest to anthropogenic pasture. However, measurements of soil cover, organic C, total N , and penetration resistance suggest that land use in the anthropogenic pasture, including heavy livestock grazing pressure, is leading to declines in soil quality in relation to the deciduous forest. These declines in soil quality have the potential to increase soil loss due to water erosion given the monsoon climate and steep slopes found in the Yungas forest region. Soil loss at a site implies a loss of soil nutrients, reduced plant rooting medium, and an altered hydrological regime (Lai et al. 1999). Over time, these changes can lead to declines in plant health and productivity (Chartier and Rostagno 2006). Therefore, although forage production may be currently higher on anthropogenic pasture sites, reductions in soil quality caused by continued intensive land use, including heavy livestock grazing pressure, pose the potential to eventually reduce forage production. Measurement of forest structure parameters indicated that heavy livestock grazing in the anthropogenic pasture, along with firewood and timber harvesting, have led to changes in forest structure and composition. Of 15 tree species found in the deciduous forest, and 10 species found within the anthropogenic pasture, only 4 species were found within both ecological zones, indicating differing arboreal compositions. Three tree species found exclusively in the anthropogenic pasture ecological zone as trees (Acacia aroma, Acacia caven and Acacia praecox) are indicative of heavy livestock grazing (Braun Wilke et al. 1995). Livestock encourage the dispersal of these three members of the Acacia genus, by consuming their seed pods (Demaio et al. 2002). Overall there are fewer saplings present within the anthropogenic 94 pasture than the deciduous forest, most likely due to consumption of saplings by livestock. This constraint to forest regeneration limits the future capacity of the forest ecosystem to provide the array of goods and services local residents currently depend upon including timber, firewood, medicinal and edible plants, water quality and quantity. Based upon the results of the forage production, soil quality and forest structure study, potential solutions include reducing livestock numbers to recommended stocking rates, deferred, rotational grazing to allow forage plants an opportunity for re-growth, and avoiding grazing on steep slopes during the wettest portion of the rainy season. Although recommendations based on biophysical research are important tools to support livestock management decisions, current social and economic realities of local livestock owners and landowners wil l play a decisive role in determining whether changes to livestock grazing practices will be adopted. Analysis of opportunities that will support movement towards more sustainable livestock grazing practices, and challenges that may pose barriers to such change, supports the following recommendations: • Participatory, community-led educational activities There is interest among livestock owners and landowners to increase their understanding of range management principles, and learn how to improve animal health and productivity. This presents a tremendous opportunity to collaborate with these two communities to design educational workshops that can address specific issues people would like to learn about, and present information about deferred, rotational grazing techniques. It would also be helpful to facilitate the transfer of knowledge among different communities in the Jujuy Model Forest. For example, some landowners having years of experience with rotational grazing, and some livestock owners possessing intricate knowledge of local trees and plants. Educational activities will likely have the greatest opportunity for success i f they are driven by interests of local community members, and require their direct and active participation. • Economic diversification and alternatives In conjunction with educational efforts to raise awareness about the importance of 95 reducing stocking rates according to the grazing capacity, it would be valuable to facilitate exploration of economic alternatives which could help diversify income, and reduce economic dependence on livestock. The option of rural tourism as a community-based economic alternative for livestock owners will need to be further discussed within that community in a careful, respectful and sensitive manner to ensure that it does not create conflict among livestock owners. Two of the three landowners whom project team members spoke with were interested in carrying out rural tourism activities. The proximity of their land adjacent to a major tourist route suggests that rural tourism is an alternative worth exploring further. Agroforestry systems could also provide economic diversification, particularly for livestock owners. • Agroforestry Agroforestry systems could potentially provide a variety of benefits to livestock owners and landowners including rehabilitation of degraded anthropogenic pastures, income diversification, increased food security, and improved demarcation of pastures. Within the community of livestock owners there is a broad knowledge of local trees and their potential uses. Encouraging the sharing of this knowledge could support the design of an appropriate agroforestry system, and emphasize the value and importance of local knowledge. Community learning about agroforestry systems will require a medium to long-term commitment of educational and research facilitation on the part of Jujuy Model Forest or another organization. • Formation of a'Grazing management group' Increased communication and cooperation among livestock owners, and between livestock owners and landowners will be needed to achieve more sustainable livestock grazing. A potentially useful model which could be used to create a social space where livestock owners and land owners could engage in dialogue about issues of mutual interest is the concept of a 'Grazing management group'. The Jujuy Model Forest would be an appropriate organization to help facilitate the formation of a livestock management group, as one of the principal objectives of model forests is to encourage diverse stakeholders who use a landbase to work together to solve problems. However, ultimately, the impetus and enthusiasm to form a livestock management group will need to come from livestock owners and landowners themselves. 96 Although this study has focused in detail on the dynamics of livestock grazing within the Yungas forest of the Jujuy Model Forest, the challenges faced in this zone surrounding the issue of livestock grazing are common around the world. The study methods, analysis, reflections, and recommendations could provide an impetus for other communities to carry out work within their own local context. In particular, the International Model Forest Network provides a forum through which the results and experience gained in the Jujuy Model Forest can be shared with other forest communities. 4.3. Evaluation of Study Methods and Recommendations for Future Research The investigation of forage production, soil quality indicators and forest structure was carried out over one field season between October 2005 and April 2006. The International Model Forest Network promotes the use of forest indicators to measure and assess forest management practices (IMFN 2006), and in March, 2007, Jujuy Model Forest volunteers learned about criteria and indicators for sustainable forest management through a local workshop. It is therefore particularly important to reflect upon the strengths and weaknesses of the field techniques used, evaluate the suitability of indicators measured for long-term monitoring, and make recommendations for future research. These evaluations may provide some insight for Jujuy Model Forest volunteers as they plan a local sustainable forest management monitoring program. 4.3.1. Evaluation of Forage Production and Grazing Capacity Study 4.3.1.1. Forage quantity For the purpose of this thesis, it was only possible to measure forage production over one growing season, from November, 2005 to April, 2006. Based on these measurements, preliminary estimations of grazing capacity were made. Forage production is highly dependent upon precipitation during the year of current growth, and during past growing seasons (USDA-NTRCS 2003), therefore it is recommended that forage production be measured over a number of years prior to calculating grazing capacity. It is advisable that the Jujuy Model Forest and/or National University of Jujuy continue with the forage production measurements to improve the 97 quality of grazing capacity estimations. Forage production measurements for this study were based upon the growth of plants within grazing exclosures that were built prior to the growing season at each field site. It was necessary to use grazing exclosures for this study because there were few to no areas where livestock grazing does not occur in the study area. Although the grazing exclosures prevent grazing by livestock and large wild herbivores, some forage biomass may have been lost due to consumption by insects, rodents, and other small animals, in addition to some weathering throughout the growing season (USDA-NRCS 2003). When exclosures are used to measure forage production, it is not possible to measure the compensatory growth of plants, which is the re-growth that occurs after a grazing event, often resulting in higher total forage production that i f no grazing had occurred (McNaughton 1983). For this reason, exclosures can slightly underestimate the total potential forage production. Due to financial constraints, and difficult terrain and access at field sites, it was not feasible to build grazing exclosures larger than 9 m 2 . As a result, woody plant biomass was measured outside the exclosures, and therefore may have been browsed by domestic or wild herbivores prior to measurement. This was especially a concern in anthropogenic pasture areas where even during the rainy growing season there is a shortage of grass and forbs for grazing animals. If some browsing of woody plants occurred prior to our measurements, browse estimation may have been lower than true values. Throughout the duration of the project, the project team came to recognize the role that grazing exclosures can play as an education tool; a sentiment also experienced by others who have carried out similar enclosure-based forage production studies (Sonneveld et al. 2005). . Livestock owners and landowners were curious to see how high the grass would grow inside the exclosures, what plants were present, and to imagine what quantity of forage was being produced across the landscape. The exclosures served as a research focal point and discussion starter, and were a physical demonstration that investigation was being carried out in the landscape. 98 4.3.1.2. Forage quality The recommended stocking rates in this study are based only on forage quantity, and do not take forage quality into consideration. It would be very useful to expand this study to include determination of crude protein and digestibility, as livestock feed intake rates are most affected by forage quality. If the nutritional level of vegetation is low, more forage is required to support a grazing or browsing animal. (USDA-NRCS 2003; Holechek 1998) The long, rigid, sclerenchymatous leaves of tussock grasses in the highland pasture have low nutrient concentrations (Chapin 1980; Reich et al. 1992), and are poor quality forage (Cabrera 1968). As a result, livestock would likely require more forage in the highland pasture relative to the anthropogenic pasture where herbaceous species contain less sclerenchyma. A . study of forage quality in the three ecological zones would enable more precise stocking rate recommendations. Plants that are known to be poisonous or unpalatable to livestock were not included in forage production measurements. The knowledge and identification of these species was based upon visits with livestock owners, discussions with mentors at the National University of Jujuy, and local publications. Identification of all unpalatable species was attempted; however, there is still much to learn. As part of further forage production studies, it would be useful for the research team to organize a series of field extension walks through the deciduous forest and anthropogenic pasture with livestock owners and landowners, to facilitate further knowledge exchange and learning about unpalatable plants. 4.3.2. Evaluation of Soil Quality Indicators Soil quality was assessed as part of this study to help examine how current land uses, and in particular livestock grazing, are affecting ecological functioning within the Yungas of the Jujuy Model Forest. Soil quality assessment has become integrated into many rangeland monitoring programs (Herrick et al. 2002), to'assist land managers in evaluating the impact of land management decisions and guide land management planning. The purpose of a soil quality assessment is to determine the ability of soil to perform a desired set of functions. Within the Yungas forest ecosystem, critical soil functions include 99 provision of a rooting medium for diverse subtropical forest associations and their fauna, regulation of the hydrological cycle through partitioning of water into surface and subsurface flow,, and provision of food, fuel and timber for local human populations. These soil functions are dependent upon a variety of interrelated soil processes, such as decomposition, infiltration, and leaching of nutrients. It is difficult to assess these ecological processes or the functions they contribute to directly, so we measured a set of soil quality indicators to act as surrogates for processes and functions that were less easily measured (Doran and Parkin 1994). For the purpose of this study, and in consideration of future monitoring activities within the Jujuy Model Forest, the following characteristics of a good soil quality indicator were felt to be most relevant and important: • integrate soil physical, chemical and biological properties and processes; • be sensitive to management-induced changes; • maintain relevance across various temporal and spatial scales; • be easily measured by different trained observers over time; • be inexpensive - financial resources are very limited for the Jujuy Model Forest Association (Doran and Parkin 1994; Karlen et al. 1997; Schoenholtz et al. 2000; Herrick et al. 2002). It is important to evaluate the performance and relevance of the soil quality indicators used, to assist Jujuy Model Forest volunteers in deciding whether these indicators should be re-measured in the future as part of long-term sustainable forest management monitoring. 4.3.2.1. Soil cover and soil litter layer depth Percent bare soil has been identified as a basic indicator of soil quality and rangeland health (Herrick et al. 2002; Pellant et al. 1995). Litter layer depth measurements give information about the quantity and variation in litter quantity across the landscape. Percent bare soil and litter cover / depth are generally inversely related. Bare soil is highly correlated with runoff and susceptibility to water erosion (Wischmeier and Smith 1978). The litter loss associated with increasing bare soil implies reduced nutrient cycling, water infiltration and habitat for soil microorganisms (Weil and Magdoff 2004). Due to all of these relationships, 100 measurement of bare soil within the Jujuy Model Forest will allow livestock owners, landowners and resource managers to make inferences about a variety of soil processes and soil functions. In this study, there were significant differences in percent bare soil and litter layer depth among deciduous forest, anthropogenic pasture and highland pasture ecological zones, with litter layer depth inversely related to percent bare soil. Litter quantity and percent bare soil are impacted by management practices such as cattle grazing (Talbot et al. 2003; Belsky and Blumenthal 1997), and timber harvesting (Ballard 2000). A useful aspect of percent bare soil measurements is that they can be used in a predictive manner to estimate site conservation thresholds, beyond which there is a marked increase in soil erosion with increasing bare soil. Forexample, following studies of semi-arid grassy woodland ecosystems of Australia, land managers recommended that bare ground levels not exceed 30-40% in order to reduce soil erosion (Mclvor and Mclntyre 2002). The monsoon precipitation regime, steep (13-37°) slopes, and susceptibility of Alfisols to erosion (Lai 1999) mean that a lower bare ground threshold would likely be necessary in our study area. There is currently 33% bare soil in the anthropogenic pasture zone of the Jujuy Model Forest, suggesting that erosion may be a serious threat to site stability. Perhaps some of the most valuable features of bare soil and litter layer depth as indicators are that both parameters are straightforward to measure, and can be measured very inexpensively. Many of the university students who assisted with the soil quality study learned and practiced how to measure bare soil and litter layer depth. For livestock owners and landowners who have demonstrated interest in learning more about rangeland management and ecosystem processes, it would be interesting to organize a field day to share with them how they could monitor and record percent bare soil and litter layer depth on the land they own or occupy. This would also provide a great opportunity for livestock owners and landowners to share with us some of the techniques they use to describe and characterize soils. Livestock owners in the study area commented that in the heavily grazed anthropogenic pasture zone it appeared that over the past approximately ten years the 'the rocks were flowering' (land degradation was causing more rocks to appear on the soil surface). Our 101 estimations of rock cover suggested that there is more exposed rock on anthropogenic pastures compared to the deciduous forest (a = 0.10). It may therefore be useful to add percent rock cover to the base set of soil quality indicators that are monitored in the Jujuy Model Forest. Visual estimation of rock cover in larger, more numerous plots should improve the ability to detect differences between ecological zones. The highland pasture ecological zone is characterized by the natural presence of large boulders on the land surface (Braun Wilke et al. 2001), and therefore cannot be readily used in comparisons of the impact of land management practices on rock cover. Technical expertise regarding the use of rock exposure as an indicator of long-term grazing pressure could be solicited from Daniel Renison, a researcher at the National University of Cordoba, Argentina who has experience with the use of this indicator (Renison et al. 2006). When looking closely at the emerging rocks in the anthropogenic pasture near Las Maderas dam, it was possible to observe that there is a common pattern of lichen colonization on rocks. The edges of large rocks that are close to the soil surface, and have been more recently exposed, did not contain lichens, while lichens were relatively abundant on the centres of the rocks that had been exposed for a longer period of time. The number of lichens on a rock surface, and total lichen cover tend to increase with surface age, and have been used to date recent erosional surfaces and glacially-deposited rocks (Matthews 2005; Bull and Brandon 1998). Rates of lichen growth, and patterns of colonization are dependent upon many environmental variables, and most particularly moisture regimes (Matthews 2005). If there was interest within the local community, further research to implement a rock lichen monitoring program could help to deepen understanding regarding how quickly rock surfaces are being exposed by erosion. Apart from monitoring lichens, a permanent marker could be used to draw a thin line along the soil-rock edge of a number of rocks, and these rocks could be monitored to measure the rate of rock exposure, and soil loss, overtime. An additional recommendation to the Jujuy Model Forest regarding soil cover measurements is that it is important that bare soil and soil litter depth measurements be consistently measured at the same time of the year. The amount of bare soil at a site can vary seasonally, according to changes in vegetation canopy and litter quantity (Gutierrez and Hernandez 1996). In this study, litter cover was measured at the end of the dry season, when 102 deciduous foliar cover and litter is at a minimum. This is also the time period during which the role of litter cover in reducing soil water erosion is most critical. In mountainous regions with a monsoon climate, such as our study area, the greatest soil erosion is often measured during the first few large rainstorms that mark the end of the dry season, prior to the re-growth of vegetation (Schreier 2004). 4.3.2.2. Organic C and total N Organic C and total N are often listed as basic indicators of soil quality (Doran and Parkin 1994; Schoenholtz et al. 2000). Organic C is an excellent example of a soil quality indicator that integrates physical, chemical and biological processes. Organic C tends to improve soil porosity, and therefore gas exchange and water relations (Schoenholtz et al. 2000), and it affects biological activity and many chemical processes due to its role in nutrient release and availability (Nambiar 1997). Most soil N occurs as part of organic molecules, and N is a common limiting nutrient in terrestrial ecosystems (Brady and Weil 2002). Soil microorganisms require N to decompose organic matter, and therefore C : N is a useful indicator of potential decomposition rates. In this study, both soil organic C and total N were significantly higher in the deciduous forest ecological zone, than for either the anthropogenic or highland pastures. Given that approximately 30 years ago, the anthropogenic pasture zone was predominantly deciduous forest, measurement of C and N levels helped to show that nutrient cycling processes have changed under management practices in the anthropogenic pasture. Collection of soil samples for organic C and total N determination is straightforward, and can be easily carried out by different people over time who have been trained in soil sampling techniques. Analysis of organic C cost approximately 6 pesos ($2.40)/sample and total N cost 5 pesos ($2.00)/sample. If an agroforestry system consisting of leguminous trees is implemented to rehabilitate anthropogenic pastures, it would be useful to monitor organic C and total N levels to assess the ability of the agroforestry system to increase levels of these nutrients. 103 4.3.2.3. Bulk density and penetration resistance Soil compaction is of particular interest in the Yungas forest characterized by a monsoon climate, since compaction generally leads to reduction of soil macropores, which in turn decreases water infiltration and hydraulic conductivity (Greacen and Sands 1980; Ballard 2000). The degree of soil compaction that occurs at a site depends on soil texture, water content, and organic matter (Van Haveren 1983), as well as livestock type, grazing duration and grazing intensity. Soil strength is a measure of the resistance a soil offers to compaction, as well as to penetration by roots (Greacen and Sands 1980). Bulk density is a commonly measured parameter to determine the level of compaction of a soil. Doran and Parkin (1994) suggested bulk density as a basic indicator of soil quality, despite the fact that on it's own it can be a relatively insensitive measure of soil processes that are affected by compaction (Hakansson and Lipiec 2000). For this study, bulk density was measured by collecting intact soil cores. The rocky nature of the soils in the study area made this method difficult, and often it was necessary to insert the core multiple times before it was possible to fully introduce the core without hitting any rocks. This made sampling quite time-consuming. An alternative method to bulk density sampling in rocky soils is by excavation (Grossman and Reinsch 2002). A pilot trial of this method was attempted at a deciduous forest site in November, 2005, and it was found that due to the dry, friable and loose nature of the soil at this time of the year, it was difficult to obtain a sample without the sides of the excavation hole collapsing. For this reason, intact soil core sampling was used. Bulk density has repeatedly been shown to change less in response to soil compaction than penetration resistance, especially in forest and rangeland soils that commonly have high spatial variability (Chanasyk and Naeth 1995; Krzic et al. 1999, Rodd et al. 1999). These observations were in agreement with this as no significant differences in bulk density were found among the three ecological zones studied. However, the trend of bulk density was similar to the trend for penetration resistance data, with the anthropogenic pasture exhibiting higher bulk density values than the deciduous forest or highland pasture. A n advantage to using bulk density as an indicator of soil compaction was that the 104 materials required to sample were very inexpensive, with the National University of Jujuy kindly lending us soil cores, and donating the use of a drying oven and scale.. This made it possible for project team members to conduct bulk density determination without any additional laboratory support, and at a low cost. Bulk density sampling can be carried out by different field workers over time without seriously compromising data quality, provided that the people conducting the sampling are given some training in bulk density sampling methods. If bulk density is adopted as a soil quality indicator in the future, it is recommended that sample size be increased to improve the likelihood of detecting a significant difference among land management practices and/or ecological zones. Soil strength is a valuable soil physical parameter to measure as it affects the ability of plant roots to grow and explore the soil for water and nutrients, and therefore is closely related to plant growth (Clark et al. 2003). In this study, soil penetration resistance measurements were a more sensitive indicator of the impact of land management practices on physical soil properties than bulk density. Soil penetration resistance in the anthropogenic pasture was significantly greater than at least one of either the deciduous forest or highland pasture zone at both depths and all times of measurement, except for the 5 - 10 cm depth in February-March 2006. Penetration resistance data for this study were collected with a hand-held force gauge (penetrometer) with a 4-mm basal diameter (30°) cone. Although penetration resistance data show significant trends upon statistical analysis, when we were collecting penetration resistance measurements in the field, we questioned the validity of this method. The greatest challenge to using the penetrometer was that it was difficult to insert the penetrometer cone without hitting any rocks. This made collection of penetrometer resistance data quite time-consuming as for each successful penetrometer reading, multiple measurements were often attempted. A particular concern regarding the collection of penetration resistance data for long-term soil quality monitoring is data consistency. The penetrometer must be inserted into the soil at a constant velocity. At the beginning of the study, project team members took turns inserting the penetrometer into the soil, and results obtained in the same location were compared for different observers. There were some marked differences in results according to observer, and therefore 105 one person ended up collecting penetration resistance data at each field site to reduce the potential for observer variation. If penetration resistance is adopted as a long-term soil quality indicator, it would be important to write a clear, detailed protocol on measurement techniques, and attempt to have some consistency over time in field workers who are carrying out measurements. Another recommendation would be to ensure that a similar-sized penetrometer, designed for the same range of soil strength, be used across times of measurement. Hand-held penetrometers, such as the one used in this study, have been found to give higher penetration resistance readings than depth-recording penetrometers, however the trends in penetration resistance measured over time were similar (Bulmer et al. in press). In order to accurately compare penetration resistance values for a site over the long term, it is essential that measurements be taken at the same time of the year, when the soil has a relatively constant water content. In this study, it was found that soil penetration resistance was higher with lower field moisture content, and that differences in penetration resistance among ecological zones were most pronounced at the end of the dry season when soil had the lowest moisture content. In an examination of soil penetration resistance data at five sites in British Columbia over a five-year period, Bulmer et al. (in press) found that seasonal variation tends to have more impact on soil water content and penetration resistance values than treatments imposed through compaction or soil rehabilitation. 4.3.3. Evaluation of Forest Structure Study Assessment of forest structure, including the regrowth of perennial species, is an important component of rangeland forest health assessments (Pellant et al. 2005). In forest ecosystems, heavy livestock grazing can cause trampling damage to seedlings, and browsing of woody plants and seedlings (Mayer et al. 2006, Hensen 2002). Selective timber and firewood harvesting are also important factors in forest structural and functional change within the Yungas forest (Brown et al. 2001). Local tree diversity within a given forest type at the latitude of our study averages 34 species (Brown et al. 2001). Only 22 species were identified within the six 0.1 ha study sites for each of the deciduous forest and anthropogenic pasture (12 sites in total) that were assessed for 106 forest structure. Additional tree species that had not been identified in the 0.1 ha plots were observed outside of the study sites, suggesting that more sites would need to be assessed to gain an understanding of full tree diversity within the study area. For future forest structure assessments, it would be useful to conduct a more detailed assessment of tree size classes. For this study we separated trees into: (1) adult trees: dbh > 10 cm (2) saplings: dbh < 10 cm and height > 50 cm. An example of a more detailed size class range would be: (1) main canopy ( > 5 m) (2) sub-canopy (1.5 - 5m) (1) small saplings (< 1.5 m). A more detailed size class assessment would allow for more detailed monitoring of recruitment trends within the deciduous forest and anthropogenic pasture. For example, i f an entire size class is missing for a species, it can alert observers to potential recruitment problems for a species. When possible, it would also be helpful to identify saplings by species, which was not carried out in this study due to time constraints, and difficulties with identification. Forest structure assessment for this study was facilitated greatly by the enthusiastic field assistance of students from the National University of Jujuy who shared a great wealth of knowledge, and identification skills for local tree species. The strong botanical and conservation interests of these students are a great resource that could be used by the Jujuy Model Forest i f undertaking future monitoring studies. A number of students expressed direct interest in conducting a long-term tree monitoring project in the area. Forest structure assessment would be an important component of forest monitoring . within the Jujuy Model Forest, as it wil l allow people to observe long-term trends in forest regeneration, learn more about impacts of livestock grazing on sapling growth, and track rates of firewood and timber harvesting. There are a number of livestock owners that live near Las Maderas dam who have excellent tree identification skills and extensive knowledge about uses of local trees. These individuals are very generous in sharing their knowledge, and may be interested in assisting with forest structure monitoring. Their direct involvement in monitoring activities could also serve as an educational and awareness-raising tool about forest dynamics within the community. 107 4.4. Community Participation and Educational Activities One of the primary objectives of this project was to directly involve community members in research activities and to provide opportunities for skill-building, learning and community dialogue about sustainable livestock grazing and forest management. Community participation and educational activities took a variety of forms to encourage the involvement of a broad array of citizens. 4.4.1. Participation of Livestock Owners and Landowners in Field Research In addition to satisfying ecological criteria, the exact locations of field sites were chosen to facilitate the direct participation of subsistence-scale livestock owners and landowners in field research activities. At two of the anthropogenic pasture field sites, located near the homes of livestock owners, local residents came to visit us when we were conducting field work. Some livestock owners felt most comfortable observing, while others were more directly involved in soil or vegetation sampling. At two sites, livestock owners did not regularly participate in field research due to other commitments, or lack of interest. At the outset of field research, landowners who had kindly allowed the project team to establish field sites on their property told the project team that they would likely be too busy to participate in field research, but were interested in the work being carried out. To stimulate dialogue with them, and a sense of inclusion in project work, the project team communicated with landowners by phone and made visits to their homes to share updates of project work and activities. One landowner who demonstrated particular interest made a visit to field sites on his land to observe soil sampling activities and visit with project team members. 4.4.2. Participation of University Students An objective of this project was to actively involve members of the broader community, including local students. The four members of the core project team are local students in the fields of biology, agronomy, forestry and nursing. From November, 2005 to May, 2006, the core project team was assisted by 12 students from the National University of Jujuy, in San . Salvador de Jujuy, who contributed enthusiastic field assistance and acted as education facilitators for the high-school field trip program. Students were primarily from the biology and 108 agronomy programs at the university, with one student from computer science also participating. In April, 2006, a student from the Global Resource Systems program, at the University of British Columbia, contributed valuable field assistance. In addition to generously sharing diverse knowledge of local ecosystems, local university students brought enthusiasm, and creative ideas to educational activities. University students expressed gratitude for the opportunity to gain practical field experience, which is often quite limited in undergraduate programs. 4.4.3. Evaluation of Community Workshops Two workshops were held in the Los Naranjos area near Las Maderas dam, where the majority of livestock owners live, to share information about research activities and stimulate dialogue about sustainable forest use. Workshops were held outside in the pasture, at a location that was considered neutral, and not occupied by a family. There was a strong emphasis during all workshops on the valuation and importance of local beliefs, knowledge and traditions. The first workshop, held in December, 2005, was entitled "What does the forest provide for us?" ^iQue nos da el bosque?"). The workshop was attended by six participants, and facilitated by two members of the core project team and myself. The purpose of the workshop was to stimulate reflection and dialogue about the different forest uses that occur in the area, and enquire i f community members felt that the forest was 'as it has always been' or if they had observed changes over time. The project team felt that it was important to create a community space to encourage dialogue about such topics, with an emphasis on valuation of local perspectives. It was important that community members have an opportunity to share their own perspectives about their local ecosystems prior to introduction to topics such as soil quality monitoring and discussion of the ways livestock grazing can affect the local ecosystem. The second workshop was a hands-on activity entitled "Getting to Know the Soils Where We Live" ("Conociendo a los suelos donde vivimos "), and was held in March, 2006. During this workshop, two core project team members and the research leader briefly described the objective and methods involved in the soil quality study, and then all workshop participants had the opportunity to participate in describing the soil pit that the project team had dug at the Los Naranjos field site near Las Maderas dam. Community members helped measure pH, 109 learned hand-texturing, described soil colour using a Munsell colour chart, and estimated coarse fragments. One local resident enriched the workshop by telling participants and facilitators about his experience digging holes in the area for pit toilet (latrine) construction. The soil pit examined was only one metre deep, but through personal experience this resident was able to tell us about some of the local soil characteristics and parent material up to four metres in depth. This also facilitated a valuable discussion about the importance of careful site selection for pit toilets to avoid water contamination, and the use of dehydrating or composting toilets as an alternative sanitation option. This experience underlined the interest of some local livestock owners in expanding their learning, and the richness of local knowledge. Given that there are at least 15 families, consisting of approximately 50 people, that live in the Los Naranjos area, attendance at community workshops was low. Some of the livestock owners attended primarily due to a personal visit and encouragement by one of the workshop facilitators immediately prior to the workshop. A few livestock owners who attended workshops showed great interest in learning more about the forest, livestock management, and soils. It would be valuable to create further learning opportunities and collaboration with these livestock owners, as they may be able to serve as a catalyst to increase the interest of other livestock owners. At a neighbourhood meeting in the Los Naranjos area, a number of livestock owners stated they would be interested in forest management workshops for their children, which could open another opportunity for increased learning on a household level. It is important to recognize that all of the activities described here occurred during the first year of this project, and building trust with a community is a long-term endeavour. Due to issues of mistrust, the university students who participated as field assistants and educators were not able to take part in community workshops near Las Maderas dam. There is likely valuable information to be shared between university students and subsistence-scale livestock owners. A number of the students are keen to learn more about traditional plant uses from community elders, and frequently demonstrated excellent capabilities in sharing their university-based knowledge in a manner that was accessible and interesting to others at a broader community level. Future decisions with regards to how to organize community workshops for livestock owners will be in the hands of local community members. 110 4.4.4. Evaluation of the High School Field-Trip Program In order to encourage dialogue and learning about the role of livestock grazing in sustainable forest management on a broader community scale, a high school field-trip program entitled "Living with the forest - Exploring livestock grazing and other uses of forest ecosystems together" was initiated. This educational activity was based upon principals of "Teaching Ecology in the School Playground" (EEPE) (Arango et al. 2002), and inspired by similar educational activities facilitated by Olson (2003), in southern Argentina. The objectives of the program were to: • Give local high school students an opportunity to form questions and conduct their own first-hand investigative research in the local forest ecosystem • Encourage critical and creative thinking about the links between local agricultural and forest ecosystems • Create an opportunity for local university students, high school students and high school teachers to interact, and learn from one another The field trip program started with a one hour visit by two or three project team members to the high school classroom to introduce the "Living with the Forest" program, facilitate brainstorming about local uses of forest ecosystems, and encourage students to make hypotheses about some of the topics they would investigate on the field trip. A few days later, the class took part in a five-hour field trip to the deciduous forest near one of our field sites. At the start of the field trip, students were separated into six teams. Teams were named after a local species, such as the "pica/lores cola larga" (long-tailed hummingbirds) and "bichos palo" (stick insects). In their teams, students rotated among the following six interactive activities related to forest ecology, that were facilitated by core project team members and university students: • Research Tools (Herramientas de Investigation) • Native Trees / Trees and Erosion (Arboles Nativos I Arboles y Erosion) • From Rocks... to Soil - The Search for Lichens, Fungi and Moss) 111 (\De la Piedra... al Suelo! En la Busqueda deLiquenes, Hongosy Musgos) • The Leaf Litter Layer (La Hojarasca) • Making an Herbarium (Herbarid) • Alternative Forest Products (Productos Alternativos del Bosque) Each activity was planned and researched by the team member that led it, with an emphasis on links between agricultural and forest ecosystems. During each activity, the high school student researchers noted their observations and reflections in a field workbook designed for the program (Appendix VII). Appendix VIII contains an outline of a typical schedule of activities for a school field trip. The field trip program was designed to encourage high school students to view themselves as researchers, and think critically about how their daily actions impact the environment around them. Between November, 2005, and May, 2006, three forest field trips were conducted for high schools in the towns of E l Carmen and Perico, in the Jujuy Model Forest. Project team members and local teachers remain in contact, and in April, 2007, they conducted two additional field trips. The educational field trip program has served to catalyze further curiosity and interest in El Carmen and Perico about research activities we had been conducting. High school teachers expressed their gratitude at an opportunity to conduct a field trip to the local forest, as funding is extremely limited at schools. University students and project team members showed enthusiasm and dedication in researching and preparing their individual activities, and high school students responded with great interest. At least one high school student on each field trip exclaimed that it was the first time he/she had visited the local forest, even though there are similar forests within less than 1 km of El Carmen and 10 km of Perico. This program holds promise to encourage greater understanding of livestock grazing and sustainable forest management in the Jujuy Model Forest. 4.4.5. Reflections on Community-Based Research Although the research style used in this investigation was based on the philosophy of community-based action research (CBAR) (Stringer 1999), it is important to acknowledge that 112 the research conducted deviated from true C B A R in a few areas. If this had been a true C B A R project, subsistence-scale livestock owners and landowners would have actively developed the. research questions with the research facilitator (in this case, myself). In reality, after spending time with various members of these communities, and visiting with them regarding their concerns, specific research questions were developed that I felt would help to address these concerns, in collaboration with mentors at the National University of Jujuy and Jujuy Model Forest Association. Under true C B A R conditions, field work would have been predominantly carried out by livestock owners and landowners, and the research facilitator would have acted more as a resource person, assisting community members to define their own problems, and work towards solutions. In this research project, livestock owners and landowners took part in some research and learning activities, but were not actively instigating the entire research process. Core project team members, who are local community members, were actively involved in on-going research and education decisions; however, for the most part they do not personally own livestock or land in the study area. A key reason that this research process departed from true C B A R was that I did not act only as a 'research facilitator', but also as a researcher. M y goal in this process was not solely to support the community in a problem-solving process they had already chosen to initiate, as it would be in a true C B A R approach. Because one of my goals within this process was to work towards a masters' degree, I ended up making some of the research decisions that should otherwise have been made by the community under true C B A R conditions. To conduct research according to a true C B A R approach in the Jujuy Model Forest would require the research facilitator to work with community members for a lengthy period of time prior to the initiation of research. This would enable trust building, and encouragement of community members that they have the ability to bring about change themselves. Although all conventions of C B A R were not followed in this research, it is hoped that through encouraging community participation in investigative activities, and efforts to emphasize the value of local knowledge, community members will start working toward formulating some of their own research questions. I will return to the Jujuy Model Forest in August, 2007 to share research results with livestock owners, landowners, and members of the broader community in a series of workshops. I look forward to reflecting with all project 113 participants and community members about where we have come so far in this research and education process, and to further learning together... 4.5. A Journey of Learning Together Although the more formal learning outcomes of this project are summarized within the pages of this thesis, there were many other captivating, enriching, and at times challenging moments of learning. These were wrapped into the activities of daily life as a resident of E l Carmen, and through many fascinating days spent with local community members and students in the Yungas forest. I felt very fortunate to be able to take some magnificent walks through the forest with livestock owners who grew up in the local area, and have learned throughout their lives about diverse uses for the forest. It was a privilege to learn from them about trees and plants that could be used for medicinal purposes, how to make the best use of trees for firewood and timber, and how plants were used for ceremonial purposes or to drive bad spirits away. It also struck me in comparison how few of the high school students who have grown up in towns or cities very close to the forest had learned these skills, as demonstrated in the school field trips. I contemplated the similarity of this pattern in Canadian society. An on-going challenge I felt throughout this research process was how to best maintain positive and productive working relationships with individuals as a research facilitator, even when issues of mistrust led people to refuse to work with one another. I came to understand how widespread corruption within a society has the potential to cause deep mistrust among individuals and a deterioration of social fabric. Although residents of E l Carmen, where I lived, outwardly always treated me with kindness, some residents commented to a project participant that "Shannon must be a spy for a foreign government or multinational corporation - Why else would she decide to move here to live in E l Carmen?!" Although this struck me mostly as comical at the time, it was a lesson in the mistrust which is present within society in this region. Throughout my experiences living and working in Argentina over the past few years, I have often been inspired by the creativity, determination, resiliency and sense of humour people are able to maintain in the face of adversity. I admired university students, who although facing 114 challenges with racism in society, acknowledged their indigenous ancestry, and expressed their desire to return to communities where their grandparents lived to learn traditional crafts, skills . and knowledge. I felt inspired by the dedication of local public school teachers, who despite extremely low salaries, and near non-existent classroom resources, enthusiastically strive to create stimulating and enriching activities for their students. I admire volunteers with the Jujuy Model Forest Association who for years have collaborated to create environmental education and conservation programs to benefit local community members and ecosystems, with few to no financial resources. I wil l always hold a very special place in my heart for so many members of the community of E l Carmen, and the broader Jujuy Model Forest, who so affectionately and kindly welcomed me into their lives, and I look forward to continued learning together. 115 4.6. References Abule, E., H.A. Snyman, and G.N. Srnit. 2005. Comparisons of pastoralists perceptions about rangeland resource utilization in the Middle Awash Valley of Ethiopia. Journal of Environmental Management 75:21-35. Angassa, A. 2005. The ecological impact of bush encroachment on the yield of grasses in Borana rangeland ecosystem. African Journal of Ecology 43: 14-20. Arango, N . , M.E. Chaves, and P. Feinsinger. 2002. Guia metodologica para la ensenanza de la ecologia en el patio de la escuela. New York, N Y : National Audubon Society Programa para America Latina y El Caribe. Ballard, T .M. 2000. Impacts of forest management on northern forest soils. Forest Ecology and Management 133:37-42. Belsky, A.J . and D . M . Blumenthal. 1997. Effects of livestock grazing on stand dynamics and soils in upland forests of the interior west. Conservation Biology 11:315-327. Brady, N.C., and R.R. Weil. 2002. The nature and properties of soils. 13 t h Edition. Upper Saddle River, NJ: Prentice Hall. 960 p. Braun Wilke, R.H., E.E. Santos, L.P.E. Picchetti, M.T. Larran, G.F. Guzman, and C R . Colarich. 2001. Carta de aptitud ambiental de la provincia de Jujuy. San Salvador de Jujuy, Argentina: Departamento de Suelos y Ecologia, Facultad de Ciencias Agrarias, Universidad Nacional de Jujuy. 245 p. Braun Wilke, R.H., B.S. Villafane, and L.P.E. Picchetti. 1995. Plantas de interes ganadero de Jujuy y Salta Noroeste Argentino. San Salvador de Jujuy, Argentina: Secretaria de Ciencia, Tecnica y Estudios Regionales and Universidad Nacional de Jujuy. 309 pp. Brown, A r D . , H.R. Grau, L.R. Malizia and A . Grau. 2001. Argentina. In: M . Kappelle, and A.D. Brown [EDS.]. Bosques nublados del neotropico. Santo Domingo de Heredia, Costa Rica: Instituto Nacional de Biodiversidad. p.623-659. Bull, W.B., and M.T. Brandon. 1998. Lichen dating of earthquake-generated regional rock-fall events, southern Alps, New Zealand. GSA Bulletin 110: 60-84: Bulmer, C.E., M . Krzic, M . G . Schmidt, and V . M . Blouin. In press. Soil mechanical resistance is closely associated with seasonal trends of water content in British Columbia's forest 116 soils. Canadian Journal of Soil Science. Cabrera, A . L . 1976. Enciclopedia argentina de agricultura y jardineria: Fasciculo 1 - Regiones fitogeograficas argentinas. Buenos Aires, Argentina: Editorial Acme S.A.C.I. 82 p. Chanasyk, D.S. and M . A . Naeth. 1995. Grazing impacts on bulk density and soil strength in the foothills fescue grasslands of Alberta, Canada. Canadian Journal of Soil Science 75:551-557. Chapin, F.S. III. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260. Chartier, M.P. and C M . Rostagno. 2006. Soil erosion thresholds and alternative states in northeastern Patagonian rangelands. Rangeland Ecology and Management 59: 616-624. Clark, L.J. , W.R. Whalley, and P.B. Barraclough. 2003. How do roots penetrate strong soil? Plant and Soil 255: 93-104. Demaio, P., U.O. Karlin, and M . Medina. 2002. Arboles nativos del centro de Argentina. Buenos Aires, Argentina: Literature of Latin America. 210 p. Direction de Bosques Nativos. 2003. Atlas de los bosques nativos argentinos. Available at: http://www2.medioambiente.gov.ar/bosques/bosques nativos/2004 atlas/default.htm Accessed 20 May 2007. Doran, J.W., and T.B. Parkin. 1994. Defining and assessing soil quality. In J.W. Doran, D . C Coleman, D.F. Bezdicek, and B.A. Stewart [EDS.]. Defining soil quality for a sustainable environment. SSSA Special Publication Number 35. Madison, WI: SSSA. Greacen, E.L. and R. Sands. 1980. Compaction of forest soils: A review. Australian Journal of Soil Resources 18:163-189. Grossman, R.B. and T.G. Reinsch. 2002. The solid phase. In: J.H. Dane and G.C. Topp. [EDS.]. Methods of soil analysis: Part 4 Physical Methods. Madison, WI: Soil Science Society of America. Gutierrez, J., and I.I. Hernandez. 1996. Runoff and interrill erosion as affected by grass cover in a semi-arid rangeland of northern Mexico. Journal of Arid Environments 34: 287-295. Hakahsson, I., and J. Lipiec. 2000. A review of the usefulness of relative bulk density values in studies of soil structure and compaction. Soil and Tillage Research 53: 71-85. Hensen, I. 2002. Impacts of anthropogenic activity on the vegetation of Polylepis woodlands in 117 the region of Cochabamba, Bolivia. Ecotropica 8:183-203. Herrick, J.E., J.R. Brown, A.J . Tugel, P.L. Shaver, and K . M . Havstad. 2002. Application of soil quality to monitoring and management: Paradigms from rangeland ecology. Agronomy Journal 94:3-11. Holechek, J.L., R.D. Pieper and C H . Herbel. 1998. Range management: Principles and practices. 3 r d ed. Prentice Hall, Upper Saddle River, NJ. 542 p. International Model Forest Network. 2006. About the International Model Forest Network. Available at: http://www.idrc.ca/imfn/ev-22893-201-l-DO TOPIC.html. Accessed 31 May 2007. Karlen, D.L., M.J. Mausbach, J.W. Doran, R.G. Cline, R.F. Harris and G.E. Schuman. 1997. Soil quality: A concept, definition, and framework for evaluation (A guest editorial). Soil Science Society of America Journal 61:4-10. Krzic, M . , R.F. Newman, K. Broersma and A. A. Bomke. 1999. Soil compaction of forest plantations in interior British Columbia. Journal of Range Management 52:671-677'. Lai, R. 1999. Erosion impact on soil quality in the tropics. In: R. Lai [ED.]. Soil quality and soil erosion. New York, N Y : Soil and Water Conservation Society, CRC Press, p. 285-305. Lai, R., D. Mokma, and B. Lowery. 1999. Relation between soil quality and erosion. In: R. Lai [ED.]. Soil quality and soil erosion. New York, N Y : Soil and Water Conservation Society, CRC Press, p. 237-258. Matthews, J.A. 2005. 'Little Ice Age' glacier variations in Jotunheimen, southern Norway: a study in regionally controlled lichenometric dating of recessional moraines with implications for climate and lichen growth rates. The Holocene 15:1-9. Mayer, A . C , C. Huovinen, V . Stoeckli, and M . Kreuzer. 2006. Plant species diversity, forest structure, and tree regeneration in subalpine wood pastures. In: E . M . Spehn, M . Liberman, and C. Korner [EDS.]. Land use change and mountain biodiversity. New York, N Y : Taylor and Francis, p.249-259. Milchunas, D.G. and W.K. Lauenroth. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecological Monographs 63: 327-366. Mclvor, S. and J.G. Mclntyre. 2002. Understanding grassy woodland ecosystems. In: J.G. Mclntyre, S. Mclvor, and K . M . Heard [EDS.]. Managing and conserving grassy 118 woodlands. Australia: CSER.0 Publishing. McNaughton, S J . 1983. Compensatory plant growth as a response to herbivory. Oikos 40: 329-336. Nambiar, E.K.S. 1997, Sustained productivity of forests as a continuing challenge to soil science. Soil Science Society of America Journal 60: 1629-1642. Olson, K . 2003. Building ecological literacy through ornithology-based fieldtrips: Experiences from an educational program in Patagonia [dissertation]. Missoula, MT: The University of Montana. 133 p. Pellant, M . , P. Shaver, D.A. Pyke, and J.E. Herrick. 2005. Interpreting indicators of rangeland Health. Technical Reference 1734-6, Version 4. Denver, CO: United States Department of the Interior, Bureau of Land Management, National Science and Technology Centre. 122 p. Reich, P.B., M.B . Walters, and D.S. Ellsworth. 1992. Leaf life-span in relation to leaf, plant, and stand characteristics among diverse ecosystems. Ecological Monographs 62: 365-392. Renison, D., I. Hensen, R. Suarez, and A . M . Cingolani. 2006. Cover and growth habit of Polylepis woodlands and shrublands in the mountains of central Argentina: Human or environmental influence? Journal of Biogeography 33: 876-887. Rodd A . V . , Y . A . Papadopoulos, L.F. Laflamme, K . B . McRae, S.A.E. Fillmore, and R.W. Wilson. 1999. Effect of rotational grazing on selected physical properties of a Gleyed Brunisolic Gray Luvisol loam in Nova Scotia. Canadian Journal of Soil Science 79:117-125. Schoenholtz, S.H., H. Van Miegroet and J.A. Burger. 2000. A review of chemical and physical properties as indicators of forest soil quality: Challenges and opportunities. Forest Ecology and Management 138:335-356. Schreier, H. 2004. Professor, Institute for Resources and Environment, University of British Columbia. Personal communication. November, 2004. Sonneveld, M.P.W., T .M. Everson, and A. Veldkamp. 2005. Multi-scale analysis of soil erosion dynamics in Kwazulu-Natal, South Africa. Land Degradation and Development 16:287-301. Stringer, E.T. 1999. Action research. Thousand Oaks, CA: Sage Publications. 229 pp. 119 USDA, NRCS. 2003. National range and pasture handbook. Grazing Lands Technology Institute. Available at: http://www.glti.nrcs.usda.gov/techn^ Accessed 15 April 2007. Talbot, L . M . , S.M. Turton, and A.W. Graham. 2003. Trampling resistance of tropical rainforest soils and vegetation in the wet tropics of north east Australia. Journal of Environmental Management 69: 63-69. Van Haveren, B.P. 1983. Soil bulk density as influenced by grazing intensity and soil type on a shortgrass prairie site. Journal of Range Management 36: 586-588. Weil, R.R. and F. Magdoff. 2004. Significance of soil organic matter to soil quality and health. In: F. Magdoff and R.R. Weil [EDS.]. Soil organic matter in sustainable agriculture. New York, N Y : CRC Press, p. 1-43. Wischmeier, W. H . and D. D. Smith. 1978. Predicting rainfall erosion losses - a guide to conservation planning. U.S. Department of Agriculture, Agriculture Handbook No. 537. 58 p. 120 APPENDIX I Forest regions of Argentina, and detailed map of the Yungas-Chaco transition forest zone in northwestern Argentina. This material has been removed due to copyright restrictions. This page contained a map of the distribution of Yungas and Chaco forests in northern Argentina, and the location of the study area in relation to these forest types. Figure A.l Forest regions of Argentina. The Selva Tucumano Boliviana is a synonym for Yungas forest, and Par que Chaqueno is Chaco forest. Map source: Direccion de Bosques, Secretaria de Ambiente y Desarrollo Sustentable (2003). 12 This material has been removed due to copyright restrictions. This page contained a map showing the mosaic of forest types in the Yungas - Chaco forest transition zone of northwestern Argentina. Figure A.2 Detail of the Yungas - Chaco transition zone in northwestern Jujuy. Map Source: Direccion de Bosques, Secretaria de Ambiente y Desarrollo Sustentable (2003). . • 122 APPENDIX II Field site names and locations Table A.1 Field site names and locations Site Ecological Zone Coordinate Coordinate Elevation # Site Name (South) (West) (m) 1 Algarrobal PAnt Anthropogenic pasture 24°27.266 65°17.467 1220 2 Los Naranjos PAnt Anthropogenic pasture 24°26.640 65°17.220 1194 3 Las Lanzas PAnt Anthropogenic pasture 24°26.912 65°17.361 1212 4 Los Cedros Anthropogenic pasture 24°28.512 65°17.292 1223 • 5 Potrerillo Este PAnt Anthropogenic pasture 24°23.662 65°21.118 1419 6 Potrerillo Oeste PAnt Anthropogenic pasture 24°23.620 65°21.238 1437 7 Los Naranjos BC Deciduous forest 24°26.814 65°18.286 1311 8 Las Lanzas BC Deciduous forest 24°27.417 65°18.226 1317 9 Algarrobal BC Ruta Deciduous forest 24°27.701 65°17.666 1398 10 Algarrobal BC Arriba Deciduous forest 24°27.676 65°18.179 1510 11 Las Urracas BC Deciduous forest 24°26.287 65°18.728 1309 12 E l Mogote BC Deciduous forest 24°24.248 65°18.332 1221 13 La Cornisa Este Highland pasture 24°29.990 65°19.414 1664 14 La Cornisa Oeste Highland pasture 24°29.993 65° 19.479 1665 15 Algarrobal PAltura Highland pasture 24°27.903 65°19.061 1774 16 Potrerillo PA1 Highland pasture 24°24.317 65°20.786 1665 17 Potrerillo PA2 Highland pasture 24°24.502 65°20.830 1663 18 Potrerillo PA3 Highland pasture 24°24.644 65°21.044 1654 123 ( APPENDIX III Soil sampling layout at study sites Figure A.3 Soil sampling layout around each grazing exclosure at study sites. Numbers indicate the distance in metres from the centre of the grazing exclosure. The top left-hand transect is labelled to illustrate where soil sampling was carried out along each transect. A l l four transects were sampled in the same manner. (A) Soil cover and L F H horizon depth (B) bulk density (C) organic C, total N , pH, texture (D) mechanical resistance. 124 A P P E N D I X IV Comparison of 2005-2006 precipitation levels at San Salvador de Jujuy with 98-year average precipitation levels O (0 o CD i _ . CL o Monthly Precipitation from June 2005 to May 2006 in Comparison with the 98-Year Average for San Salvador de Jujuy, Argentina ^ 300 E E 250 200 150 100 -I 50 0 4 $—5-June July Aug Sept Oct Nov Dec Jan Feb Mar Apr May Month -•— 2005 - 2006 Precipitation o 1908 - 2006 Average Precipitation Figure A.4 Monthly precipitation from June 2005 to May 2006 in comparison with the 98-year average for San Salvador de Jujuy, Argentina. • These precipitation measurements were collected approximately 30 km north of our study area, within the same biogeoclimatic zone as our study sites. Error bars are standard deviation for the long-term average precipitation. • Total June 2005-May 2006 precipitation at San Salvador de Jujuy was 1099 mm, while the 98-year average yearly precipitation for this station was 863 mm (standard deviation =197 mm). 125 APPENDIX V Schedule of activities for community-based rural tourism pilot activity Community-based Rural Tourism Activity, January 24, 2007 Milking and Cheese Production 8:30-9:30 • Visitors help a family milk a criollo (creole) cow • Demonstration of the first step in the preparation of cuajada (curd) for quesdlo (local specialty cheese) production • Visitors learn about the tradition of mate, and how leaves of local plants can be added to mate for flavour or medicinal purposes Quesillo Production 9:45 - 10:45 • Visitor help a second family to convert cuajada into quesillo Sheep Shearing and Wool Spinning 11:00-12:00 • Visitors assist a family in shearing one of their sheep with hand-scissors • Visitors are taught how to spin sheep wool into yarn with a hand spindle • A family member demonstrates how she weaves her yarn into blankets Preparation of Empanadas and Bread 12:15-14:45 • Two families gather together to teach visitors how to prepare home-made bread and empanadas, and cook them in a mud oven • Family members and visitors share the meal they have prepared together Corn Cultivation . 15:00-15:30 • An individual livestock owner leads visitors on a tour of his corn field, and explains his cultivation techniques Horse-back Riding • 15:30 -17:00 • A community member leads visitors on a horse-back ride to view the irrigation dam, and learn identification and uses of native trees Bottle-feeding Young Goats and Sheep 17:00 - 17:30 Members of two families help visitors bottle-feed baby goats and sheep Community Gathering and Celebration 17:30-20:00 • A l l community members in the valley are invited to come together to share mate, cook traditional tortillas over the fire, and partake in cultural sharing with visitors • One community member serenades us with traditional folk music on his guitar • Community members demonstrate and explain the stories of traditional folk dances; visitors learn steps to the chacarera folk dance • Visitors teach community members the hokey pokey (in English and Catalan) • Visitors and community members play soccer in the pasture 126 APPENDIX VI Field workbook used for 'Living with the Forest' school field trips Field workbook each student uses to record observations: EXplor^ndo Jmfos el ?$$ioreo j OifoS [kos |os £cosis{e^s forestales Afowkre: "Living with the Forest: Exploring livestock grazing and other uses of forest ecosystems together" 128 130 o H i SOSUOJ-I N 0 O s N sqoq.iv :9ju9iqiuy Ejquios wquios V, !°S lyqmos Bjquios soauofj sausnbiq 131 132 133 Translation of "Living with the Forest" student field workbook • Living with the forest: Exploring livestock grazing and other uses of forest ecosystems together o Name, group • Observing with all of our senses.. .(Observando con todos los sentidos) o The forest has a huge amount to teach us... Stop for a moment: • What do you hear? • What do you see? • What do you smell? • Observations • Research tools (Herramientas de investigation) o What is a GPS? What do we use it for? o Why is the slope of a site important? o Notes • Native trees (Arboles nativos) o What is a native tree? o Drawing of the leaf of a native tree o Chart for students to record data for three native tree individuals (species, circumference, height, notes) • Trees and Erosion (Arboles y Erosion) o What is erosion? o What is the relationship between trees and erosion? o How can we reduce erosion? • From rocks to soil (De la piedra al suelo!) o How is soil formed? o What is a lichen? o What is moss? o What role do fungi play in the soil? • In the search for lichens, fungi and moss... (En la busqueda de liquenes, hongos y musgos) o Chart for students to record how many species of lichens, moss and fungi they found in two different environments: • on rocks (in the sun, partial shade and shade) • on trees (north, south, east and west-facing bark) o Warning! - Soils under construction! (iOjo! Construction de suelo en action!) • Leaf litter layer (La Hojarasca) o What is the leaf litter layer? o Site description o Human uses of the site o What do you see in the leaf litter layer? o Depth of the L F H layer: (4 observations) • Herbarium (Herbario) • Alternative forest products (Productos alternatives del bosque) o Observations 134 APPENDIX VII Typical schedule of activities for a "Living with the Forest" school field trip EI Cronograma Detallada - Salida con La Escuela Parroquial, Sabado, 13 mayo 2006 Cronograma Ines Elena Veronica <& Pedro Alejandro & Carina Claudia Shannon De la Piedra al Suelo Medicion de Arboles Rol de Arboles en el Bosque Herbario Usos del Bosque Hojarasca 9:00-9:15 Sentidos Sentidos Sentidos Sentidos Sentidos Sentidos 9:15-9:40 Cebil Colorado Libelulas Picaflores Palo Borrachos Comadrejas Bichos Palos 9:40-10:10 Picaflores Cebil Colorado Libelulas Comadrejas Bichos Palos Palo Borrachos 10:15-10:40 Libelulas Picaflores Cebil Colorado Bichos Palos Palo Borrachos Comadreias 10:45-11:10 Mate en el Bosque... 11:15-11:40 Palo Borrachos Comadrejas Bichos Palos Cebil Colorado Libelulas Picaflores 11:45-12:10 Comadrejas Bichos Palos Palo Borrachos Picaflores Cebil Colorado Libelulas 12:15-12:40 Bichos Palos Palo Borrachos Comadrejas Libelulas Picaflores Cebil Colorado 12:40-1:00 . Ref lexiones y preguntas antes que volver a la escuela Grupos (con 5-6 integrantes cada uno): Los Cebil Colorados Rotaciones: Ale <& Carina Shannon Claudia Los Picaflores Cola Largas . Ines -> Elena Vero <& Pedro Las Libelulas Los Palo Borrachos Las Comadrejas Los Bichos Palos • Students were separated into groups of 4 to 6, and rotated in their groups among the six forest activities. Each group was named after a local tree, bird or insect species (Example: Los bichos palos = Walking stick insects). APPENDIX VIII Analysis of variance (ANOVA) tables for the forage production, soil quality and forest structure study • Treatment (k) = ecological zone = 3 (Ar-l) = (3-1) = 2 • n = 6 (6 field sites / ecological zone) k (n-l) = 3 (5) = 15 Source of Variation df Mean Square F-Value P r > F Ecological Zone 2 9172675.3 14.59 0.0004 Error 14y 628480.9 y Horses ate the forage inside one grazing exclosure in the anthropogenic pasture ecological zone, and therefore forage production was not measured at this field site. Table A.3 ANOVA table for the effect of ecological zone on unpalatable biomass Source of Variation df Mean Square F- Value P r > F Ecological Zone 2 125963.4 6.69 0.0091 Error 14y 18817.4 y Horses ate the biomass inside one grazing exclosure in the anthropogenic pasture ecological zone, and therefore unpalatable biomass production was not measured at this field site. Table A.4 ANOVA table for the effect of ecological zone on the proportion of forage consisting of grasses Source of Variation df Mean Square F- Value P r > F Ecological Zone 2 6024528.9 SU55 0.0036 Error 14 . 696455.3 Table A.5 ANOVA table for the effect of ecological zone on the proportion of forage consisting of forbs Source of Variation df Mean Square F-Value Pr > F Ecological Zone 2 112187.7 5.29 0.0194 Error 14 21206.6 Table A.6 ANOVA table for the effect of ecological zone on the proportion of forage Source of Variation df Mean Square F-Value P r > F Ecological Zone 2 311523.5 1.86 0.1924 Error 14 167614.9 136 Table A.7 ANOVA table for the effect of ecological zone on the number of saplings • Source of Variation df Mean Square F-Value P r > F Ecological Zone 1 52997.35 36.26 0.0002 Error 9 y 1461.47 y There were no trees at field sites in the highland pasture ecological zone, therefore k = 2 for sapling counts; saplings were not counted at one field site in the deciduous forest. Table A.8 ANOVA table for the effect of ecological zone on % bare soil Source of Variation df Mean Square F-Value P r > F Ecological Zone 2 5857.41 31.03 < 0.0001 Error 15 188.764 Table A.9 ANOVA table for the effect of ecological zone on 0 ^ litter cover Source of Variation df Mean Square. F-Value P r > F Ecological Zone 2 8787.85 29.02 < 0.0001 Error 15 302.803 Table A.10 ANOVA table for the effect of ecological zone on % rock cover Source of Variation df Mean Square F Value P r > F Ecological Zone 2 545.81 2.75 0.0959 Error 15 198.31 Table A. 11 ANOVA table for the effect of ecological zone on soil litter depth Source of Variation df Mean Square F Value P r > F Ecological Zone 2 114.62 42.74 < 0.0001 Error 15 2.6815 Table A.12 ANOVA table for the effect of ecological zone on % organic C Source of Variation df Mean Square F Value P r > F Ecological Zone 2 22.968 9.69 0.0020 Error 15 2.370 Table A.13 ANOVA table for the effect of ecological zone on % totalN Source of Variation df Mean Square F Value P r > F Ecological Zone 2 0.2617 11.77 0.0008 Error 15 0.0222 137 Table A.14 ANOVA table for the effect of ecological zone on the ratio of organic C to total N (C:N) Source of Variation df Mean Square F Value Pr > F Ecological Zone 2 24.574 7.77 0.0048 Error 15 3.164 Table A.15 ANOVA table for the effect of ecological zone on soil pH Source of Variation df Mean Square^ ' _F Value P r > ^ Ecological Zone ~ ~ 2 ~ ~ 1 3 1 0 5 ~ . " ~ 5 . 2 7 " ~ 0.0185 Error 15 0.24868 Table A.16 ANOVA table for the effect of ecological zone on bulk density Source of Variation df Mean Square F Value Pr > F Ecological Zone 2 0.1001 1.34 0.2912 Error 15 0.07466 Table A.17 ANOVA table for the effect of ecological zone on soil penetration resistance from 0-5 cm depth in November, 2005 Source of Variation df Mean Square F Value Pr > F Ecological Zone 2 47463707.1 9.72 0.0020 Error 15 4882306.0 Table A.18 ANOVA table for the effect of ecological zone on soil penetration resistance from 5-10 cm depth in November, 2005 Source of Variation df Mean Square F Value Pr > F Ecological Zone 2 39919800.1 7.64 0.0052 Error 15 5226368.0 Table A.19 ANOVA table for the effect of ecological zone on soil penetration resistance from 0-5 cm depth in February-March, 2006 Source of Variation df Mean Square F Value P r > F EcologicafZone 2 36348782.1 4.53 0.0289 Error 15 8022096.3 Table A.20 ANOVA table for the effect of ecological zone on soil penetration resistance from 5-10 cm depth in February-March, 2006 Source of Variation df Mean Square F Value Pr > F Ecological Zone 2 25321488.2 2.76 0.0950 Error 15 9159749.6 138 Table A.21 ANOVA table for the effect of ecological zone on soil penetration resistance from 0-5 cm depth in April, 2006 ' Source of Variation df Mean Square F Value Pr > F Ecological Zone 2 15072592.1 11.57 0.0009. Error 15 1302171.7 Table A.22 ANOVA table for the effect of ecological zone on soil penetration resistance from 5-10 cm depth in April, 2006 • Source of Variation df Mean Square F Value P r > F EcologTcai^o^e lT' IQU^\5A 534 0.0126 Error 15 1708953.19 1 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0101124/manifest

Comment

Related Items