Targeting watershed protection in the Guapiaçu-Macacu region of the Atlantic Forest, Brazil: An environmental and economic assessment of the potential for a payment for ecosystem services scheme

Kölner Schriften zur Ingenieurund Naturwissenschaftlichen Forschung Band 1/2015 Targeting watershed protection in the Guapiaçu-Macacu region of the Atlantic Forest, Brazil: An environmental and economic assessment of the potential for a payment for ecosystem services scheme Vanesa Rodríguez Osuna Kölner Schriften zur Ingenieurund Naturwissenschaftlichen Forschung Band 1/2015 Targeting watershed protection in the Guapiaçu-Macacu region of the Atlantic Forest, Brazil: An environmental and economic assessment of the potential for a payment for ecosystem services scheme Inaugural-Dissertation zur Erlangung des Doktorgrades der Naturwissenschaften in Kooperation mit der Universität Leipzig vorgelegt von Vanesa Rodríguez Osuna im Jahr 2015 Betreut durch: Prof. Dr. Hartmut Gaese Institut für Technologie und Ressourcenmanagement in den Tropen und Subtropen Technische Hochschule Köln Prof. Dr. Jürgen Heinrich Fakultät für Physik und Geowissenschaften Universität Leipzig Prof. Dr. Jan Börner Zentrum für Entwicklungsforschung Universität Bonn This research was funded by International Postgraduate Studies in Water Technologies (IPSWaT) scholarship programme (2010-2013), an initiative of the German Federal Ministry of Education and Research (BMBF) Acknowledgements Acknowledgements This dissertation was accomplished between 2010–2014 with the support of a PhD scholarship from the German Ministry of Education and Science (BMBF), to which I am very grateful (specially Mrs. Parisius), via the IPSWaT programme. In addition, this research was performed within the scope of the DINARIO (Climate Change, Landscape dynamics, Land use and Natural Resources in the Atlantic Forest of Rio de Janeiro) project. I have the deepest gratitude to Prof. Dr. Hartmut Gaese for his invaluable constant moral, scientific, and institutional support throughout the whole dissertation process. I especially want to thank Dr. Rachel Bardy Prado (my co-supervisor of Embrapa Soils in Brazil) for her scientific support, valuable guidance and contributions throughout the entire dissertation process. Prof. Dr. Börner was also highly involved in the most important stages of this research and I am very grateful for all of his very valuable contributions and expertise. I am also very thankful to Prof. Dr. Heinrich who provided accompaniment and support during my dissertation effort. I am very grateful for the support of the German partners of the DINARIO project, especially Udo, Juan Carlos, Dietmar, Jens, Santiago, Annika, and Dara. My work in Brazil would not have been possible without the support of DINARIO´s Brazilian partner, Embrapa Soils, therefore I am especially thankful to their MP2 Team, particularly Rachel, Ricardo Trippia, and Joyce Monteiro. I am also extremely thankful to Lenilson Biazatti (from the Rural Workers Union) for his extraordinary engagement and technical support during field campaigns. In addition, I greatly appreciate the technical support from Nicholas Locke (REGUA Association Director), Thabta Matos de Mata from the City Council of Cachoeiras de Macacu, Jocemir Da Silva (Chief technical extension officer of EMATER in Cachoeiras de Macacu), Demerval Pereira de Sousa (President of ALAF), Eng. Veiga and other technical staff from the state water utility company (CEDAE—Imunana Laranjal) and particularly to all of the farmers who enriched this study with their valuable time and cooperation. I sincerely appreciate the institutional support from ITT, especially from Lars Ribbe, who enabled me to present some of my research results at several international conferences. I am also very thankful to the International Office of the Cologne University of Applied Sciences for granting me a scholarship to participate in additional international conferences. Acknowledgements During the first phase of my dissertation, I was working in the company ‘sequa gGmbH’ and I am truly thankful to my colleagues there for the moral support during this phase. I truly appreciate the support of my mentor Dra. Balcázar, who supported my efforts to cope with the interphases of PhD and work within the framework of the MINTTalente Programme at the Cologne University of Applied Sciences and the Association of German Engineers (VDI). This was a one year institutionalised, external, equal gender, one-to-one mentoring programme with the objective of supporting young women from MINT areas (Mathematics, Informatics, Natural Sciences and Technical Areas) for developing leadership skills and facilitating the transition from academic study to the workforce. Thank you to my colleagues at the Centre for Development Research (ZEF) in Bonn, especially those from the project “Shaping environmental policies for sustainable tropical forest bioeconomies” for your moral and scientific support, especially during the last phase of this dissertation. I am very grateful to many for their love and support, especially Stefan and my mates Gaby, Markus, Till, Juliana, Vero, Zuura, Wifag, Tekalegn, Islam, Siggy, and Meghan. All of this academic and institutional support could only have been achieved due to the constant love and support from my family in Bolivia (especially my mum and dad, Andrea and Manuel Alejandro), whom I deeply love and owe all my dedication to this process. I dedicate this dissertation to the memory of my Mamá Guichi, my grandmother, who left this earth on September 2013 to join my grandfather, Papá Esteban, both of whom I most admire and know they would be proud to see me accomplishing this important step in my life, which kept me long away from home. However, they always gave me all the inspiration, love and support I needed to make this possible. Contents i Contents Acknowledgements ............................................................................................................ i Contents .............................................................................................................................. i Tables ................................................................................................................................ iii Figures ............................................................................................................................... iv Acronyms .......................................................................................................................... vi Abstract ............................................................................................................................ vii Resumo .............................................................................................................................. ix Resumen ............................................................................................................................ xi Zusammenfassung ......................................................................................................... xiii 1. Introduction .................................................................................................................. 1 Background and motivation for the study ................................................................ 1 1.1. Research questions, objectives, and hypothesis .................................................... 3 1.2. Thesis organisation ................................................................................................. 8 1.3. 2. Background and Conceptual Framework .................................................................. 9 Ecosystem services: importance to human well-being and a concept that links 2.1. economic and ecological systems ................................................................................. 10 Watershed services: classification ........................................................................ 14 2.2. Sustainable ecosystem services management ..................................................... 15 2.3. 2.3.1 Payment for ecosystem services as a mechanism for safeguarding sustainable ecosystem management ......................................................................... 18 2.3.2 Economic valuation of ecosystem services ................................................... 22 2.3.3 Examples of the economic valuation of watershed services and activities most affected by their degradation............................................................................. 26 The supply of watershed services: farming systems and their opportunity costs . 32 2.4. The demand for watershed services ..................................................................... 36 2.5. Relevant legislation, policy tools, and land-use based options for the management 2.6. of watershed services in Brazil ...................................................................................... 43 Conceptual framework .......................................................................................... 47 2.7. Contents ii 3. Methods ....................................................................................................................... 50 Study area............................................................................................................. 50 3.1. Watershed service supply assessment methods .................................................. 58 3.2. 3.2.1 Farming system and opportunity cost analyses ............................................. 58 3.2.2 Environmental conditions that influence the provision of watershed services: vulnerability of water resources ................................................................................. 62 3.2.3 Identification of priority areas for improving watershed service provision ...... 63 Methods for the assessment of the potential demand for watershed services ..... 64 3.3. 4. Results and Discussion ............................................................................................. 65 Watershed service supply assessment: farming system analysis ........................ 65 4.1. Watershed service supply assessment: cost-benefit and opportunity costs 4.2. analyses ......................................................................................................................... 76 4.2.1 Perceived land-use and water quality changes in the Batatal, Caboclo, and Manuel Alexandre study sites .................................................................................... 82 4.2.2 Perceptions regarding water quality and quantity in the Batatal, Caboclo, and Manuel Alexandre study sites .................................................................................... 86 The Manuel Alexandre River watershed as a reference site ................................ 89 4.3. Vulnerability of water resources in the Guapiaçu-Macacu watershed .................. 90 4.4. Analysis of environmental and economic criteria for watershed service 4.5. conservation and improvement in the Guapiaçu-Macacu watershed ............................ 93 Demand for watershed services in the Guapiaçu-Macacu watershed: water 4.6. treatment ........................................................................................................................ 96 Matching supply and demand for watershed services: is demand great enough to 4.7. justify a ‘payment for watershed services’ scheme in the Guapiaçu-Macacu watershed?.. ................................................................................................................ 104 5. Conclusions and Outlook ........................................................................................ 108 6. References ................................................................................................................ 113 APPENDICES ...................................................................................................................... 1 Appendix I. Summary of PWS programmes in the Atlantic Forest, Brazil ........................... 1 Appendix II. Initial semi-structured survey ......................................................................... 14 Appendix III. Agricultural calendar, mean yields, and farm prices at the study sites ......... 22  Tables iii Tables Table 1.1: Specific objectives, research questions, and data analysis methods ................ 5 Table 2.1: Classification of watershed services ................................................................ 14 Table 2.2: Monetary valuation methods for ecosystem services ...................................... 24 Table 2.3: Examples of the economic value of watershed services .................................. 28 Table 2.4: Metropolitan areas in the USA that have avoided the costs of water filtration plant construction through watershed protection ............................................. 30 Table 2.5: Examples of economic activities most affected by the loss of watershed services ............................................................................................................ 31 Table 2.6: Selected PWS related schemes worldwide ...................................................... 39 Table 2.7: Management practices and coordination requirements for ecosystem services related to watershed protection ........................................................................ 45 Table 2.8: Sustainable land management practices that enhance watershed services .... 46 Table 4.1: General household characteristics in the Batatal and Caboclo study sites ...... 68 Table 4.2: Household crop production characteristics at the Batatal and Caboclo study sites .................................................................................................................. 72 Table 4.3. Fertiliser application rates for each of the typical crops in the Rio de Janeiro study area ........................................................................................................ 73 Table 4.4: Key characteristics of the study area farming systems in the state of Rio de Janeiro, Brazil .................................................................................................. 74 Table 4.5: Summary of estimated annual profits of farming systems in the study area .... 77 Table 4.6: Water utility intake characteristics in the lower Guapiaçu-Macacu watershed . 97 Table 4.7: Water treatment costs and characteristics at the Laranjal Treatment Unit (CEDAE), Rio de Janeiro, Brazil ...................................................................... 99 Table 4.8: Avoided costs based on turbidity levels and annual chemical treatment costs at the Laranjal Treatment Unit (CEDAE), 1998–2011 ........................................ 100 Table 4.9: Turbidity measurements from eleven sites in the Guapiaçu-Macacu watershed ....................................................................................................................... 102 Table 4.10: Main physical and chemical variables measured at the eleven sampling sites in the Guapiaçu-Macacu watershed ............................................................... 103 Figures iv Figures Figure 1.1: Research framework showing the sequence of specific objectives and research questions in pursuit of the overall research goal ............................. 7 Figure 2.1: Ecosystem service categories and their linkages with human well-being .... 13 Figure 2.2: Total economic value of ecosystem services ............................................... 23 Figure 2.3: Bottom-up approach for the assessment of ecosystem services ................. 33 Figure 3.1: The Guapiaçu-Macacu watershed within the state of Rio de Janeiro .......... 50 Figure 3.2: Land use in the Guapiaçu-Macacu watershed ..................................................................................................................... 53 Figure 3.3: Elevation and soil classes in the Guapiaçu-Macacu watershed ................... 54 Figure 3.4: Characteristic landscape features of the three Guapiaçu-Macacu subwatersheds: Manuel Alexandre (left), Batatal (upper right), and Caboclo (lower right) .................................................................................................. 56 Figure 3.5: Farming system analysis steps .................................................................... 58 Figure 3.6: Preparatory work for the household survey effort that included presentations to local producers on the research sampling goal and strategies with scientists from Embrapa Soils ...................................................................... 59 Figure 3.7: Weighted state and pressure indicators based on consultation with hydrological experts for the assessment of the vulnerability of water resources in the Guapiaçu-Macau watershed .............................................. 63 Figure 4.1: Crop production areas visited during the field survey .................................. 66 Figure 4.2: Banana production at the Batatal study site ................................................. 70 Figure 4.3: Typical landscapes at the Caboclo study site .............................................. 72 Figure 4.4: Spatial opportunity costs in BRL per hectare for the ‘agriculture’ (crop production) and ‘pasture’ land uses within each of the Guapiaçu-Macacu sub-watersheds. The white areas correspond to land uses other than crop or livestock production ...................................................................................... 79 Figure 4.5: Reported perceptions on changes in water quality and quantity in the Batatal (N=29 responses) and Caboclo (N=24 responses) study sites .................... 87 Figure 4.6: Typical riparian landscape in the Manuel Alexander River watershed ......... 89 Figure 4.7: Water resource vulnerability among the sub-watersheds of the GuapiaçuMacacu watershed ....................................................................................... 91 Figures v Figure 4.8: Prioritisation of the sub-watersheds of the Guapiaçu-Macacu watershed for improving or maintaining watershed services .............................................. 94 Figure 4.9: The Imunana channel is the main source for the public water supply in the Guapiaçu-Macacu watershed ....................................................................... 97 Figure 4.10: Distribution of sampling sites in the Guapiaçu‐Macacu watershed (Corrego Alegre/UTM Zone 23S) .............................................................................. 101  Resumen xii Con base en los costos de tratamiento, la disposición a pagar de la compañía estatal de aguas por los servicios ambientales hidrológicos será insuficiente para compensar a los productores rurales por renunciar a sus actividades agrícolas para mejorar la provisión adicional de servicios ambientales hídricos. Los resultados sugieren que los costos de los cambios en la cobertura del suelo a una escala requerida para mejorar la calidad de agua probablemente excederán el costo de inversiones adicionales en el tratamiento de agua para el suministro público. Incentivos monetarios condicionados a ajustes específicos de los sistemas de producción podrían tener un papel complementario en la mejora de los servicios ambientales hídricos. Sin embargo, el análisis de la disposición a pagar solo se concentró en costos de tratamiento con productos químicos y sólo en un servicio ambiental, de una amplia gama de servicios ambientales provistos por la vegetación natural en la cuenca de Guapiaçu-Macacu (i.e. mantenimiento de la calidad del agua para el suministro de agua potable). Otros servicios ambientales provistos por la cobertura forestal incluyen el secuestro y almacenamiento de carbono, moderar el impacto de eventos climáticos extremos, regulación del flujo hídrico, así como de la protección de la biodiversidad y la belleza escénica. Incluir esos servicios ambientales adicionales en la ecuación de la disposición a pagar probablemente cambiará las conclusiones de esta evaluación a favor de acciones conservacionistas adicionales, sea a través de pagos por servicios ambientales (PSA) o a través de otros instrumentos de política. Este esfuerzo contribuye al conocimiento de una creciente literatura científica sobre como la combinación de información económica y ambiental espacialmente explícita puede ser utilizada para ofrecer una visión valiosa sobre la viabilidad de implementar esquemas de PSA a la escala de cuencas hidrográficas. Además, los resultados de este estudio pueden servir a iniciativas sobre conservación de cuencas y a políticas públicas en otras cuencas del bioma del bosque atlántico al facilitar la focalización de incentivos para la conservación a través de una gestión costo-efectiva de cuencas hidrográficas. Zusammenfassung xiii Zusammenfassung Der Schutz von Wasserressourcen in der Guapiaçu-Macacu Region im atlantischen Regenwaldbiom Brasiliens: Eine Bewertung der ökologischökonomischen Potentiale eines Projekts von Zahlungen für ÖkosystemDienstleistungen („Payments for Ecosystem Services“) Die Intensivierung von Landnutzung und Urbanisierungsprozessen sind Ursachen für den Rückgang von Ökosystem-Dienstleistungen im Einzugsgebiet des Guapiaçu-Macacu im Bundesstaat von Rio de Janeiro. Ein sinnvolles Instrument zur langfristigen Wasserversorgung der etwa 2,5 Mio. Wasserverbraucher könnten Kompensationszahlungen für Landwirte dieser Region sein, wenn diese auf ihre landwirtschaftliche Produktion verzichten. So kann die natürliche Vegetation im FlussEinzugsgebiet wiederhergestellt und damit die Ökosystem-Dienstleistung „Lieferung von sauberem Trinkwasser“ erhalten werden. Diese Studie hat die Kosten für Landnutzungsänderungen kalkuliert, die der Verbesserung von Ökosystem-Dienstleistungen bezüglich der Wasserqualität dienen könnten. Diese Kosten wurden verglichen mit der geschätzten Kostenvermeidung aus der Wasseraufbereitung für die öffentliche Trinkwasserversorgung, welche ein Indikator für die Zahlungsbereitschaft für hydrologische Ökosystem-Dienstleistungen (Wasserqualität) darstellt. Opportunitätskosten, die beim Wandel der Landnutzung zugunsten verbesserter Wasserqualität entstehen, sind anhand von Haushaltsdaten landwirtschaftlicher Betriebe kalkuliert worden. Dieser Wandel ist ein Prozess, der mit hoher Wahrscheinlichkeit zur Verbesserung der Wasserqualität (gemessen als Reduktion der Trübung) durch Verringerung der Erosion beiträgt. Die geschätzten Opportunitätskosten wurden auf das Einzugsgebiet übertragen, basierend auf Landnutzungsklassifizierungen und einer Vulnerabilitätsanalyse der Wasserressourcen, um prioritäre Zonen für Maßnahmen im Bereich Wasserressourcenmanagement zu identifizieren. Um die potentielle Nachfrage nach hydrologischen Ökosystem-Dienstleistungen zu ermitteln, wurden Daten der örtlichen Wasseraufbereitungsanlage (Hauptverbraucher von Wasser) sowohl zur Wasserqualität als auch bezüglich der Aufbereitungskosten analysiert. Es wurde gezeigt, dass eine Reduktion der ackerbaulichen Landnutzung zum Erhalt von Ökosystem-Dienstleistungen hohe Opportunitätskosten im Untersuchungsgebiet, nahe der Stadt Rio de Janeiro, mit sich bringt. Im Gegensatz dazu bestehen im Bereich der Viehzucht einige kostengünstige Optionen zur Erhaltung von hydrologischen ÖkosystemDienstleistungen. Andere Optionen beinhalten die Implementierung von Bodenkonservierungstechniken; permanenter und effizienter Schutz erosionsanfälliger Gebiete; Schutz und Wiederherstellung der Galeriewälder sowie die Erhaltung von Vegetation an den Wasserquellen und die Umsetzung noch nachhaltigerer landwirtschaftlicher Praktiken. Solche Maßnahmen haben das Potential, die Sediment- Zusammenfassung xiv und Nährstoffeinträge in den Wasserkörper direkt zu reduzieren und damit gleichzeitig die Kosten der Trinkwasseraufbereitung zu verringern. Davon ausgehend, dass die Zahlungsbereitschaft des staatlichen Wasserbetriebes der Höhe der Wasseraufbereitungskosten entspricht, reichen die Ausgleichszahlungen für den Erhalt von Ökosystem-Dienstleistungen nicht aus, um die landwirtschaftlichen Opportunitätskosten zu decken. Finanzielle Anreize können jedoch weiterhin eine ergänzende Rolle bei der Verbesserung solcher Ökosystem-Dienstleistungen spielen, wenn sie mit spezifischen Regulierungen existierender Produktionssysteme verknüpft werden. Die Analyse der Zahlungsbereitschaft in dieser Arbeit basiert lediglich auf den Aufbereitungskosten von Chemikalien für die Verbesserung der Wasserqualität. Weitere positive Effekte durch die Wiederherstellung des natürlichen Ökosystems wurden nicht explizit untersucht. Diese Ökosystem-Dienstleistungen umfassen die Kohlenstoffsequestrierung durch die Vegetation, die Abflussregulierung durch eine erhöhte Wasserspeicherkapazität der Böden, die Erhaltung der Biodiversität und den Erholungswert der Natur. Diese zusätzlichen Dienstleistungen der Ökosysteme in dem Zahlungsbereitschaftsausgleich zu erfassen, könnten das Verhältnis von Kosten und Nutzen zugunsten zusätzlicher Schutzaktionen verändern. Dies kann durch die Zahlung für Ökosystem-Dienstleistungen oder anderer politischer Maßnahmen erfolgen. Diese Studie leistet einen Beitrag zur wissenschaftlichen Diskussion, wie die Kombination aus räumlich expliziten ökonomischen und ökologischen Informationen verwendet werden können. Ziel ist es, einen nutzbaren Einblick in die Durchführbarkeit von PES-Projekten auf der Skala von Fluss-Einzugsgebieten zu erhalten. Dies ist wichtig, um sachkundige Entscheidungsfindungsprozesse zu unterstützen, unter Berücksichtigung der ökonomischen Spielräume des Wasserressourcenmanagements im Guapiaçu-Macacu Gebiet oder ähnlichen Regionen. Die Ergebnisse dieser Studie dienen zur Unterstützung von Initiativen zur nachhaltigen Nutzung von Wasserressourcen. Weiterhin helfen die Ergebnisse dieser Arbeit der Politik bei der Setzung von gezielten Anreizen zur Bewahrung von Ökosystem-Dienstleistungen in anderen Einzugsgebieten des atlantischen Regenwaldes durch ein kosteneffektives und integriertes Management von Wasserressourcen.  Chapter 1. Introduction 1  1. Introduction Background and motivation for the study 1.1. One of the greatest environmental challenges of our time is to revert the trend of on-going degradation of ecosystems while meeting increasing demands for food and biomass (MEA 2005; THOMAS & CALLAN 2010). Population and economic growth are also increasing the demand for water resources and simultaneously amplifying pressure on the ecosystems that deliver watershed services (PORRAS ET AL. 2008). Furthermore, climate change is increasing the pressure on natural and anthropogenic systems through the impacts of climate extremes, which have fundamental implications for water resource demand and availability (MEA 2005; IPCC 2012; UNEP 2012). The degradation of ecosystem services represents a loss of largely undervalued natural capital assets (MEA 2005; MONTES 2007; TEEB 2009). Essential economic sectors depend on these and other forms of natural capital, which together constitute an essential pillar of sustainable development (TEEB 2009). While there is general agreement among scientists and conservation practitioners that land-use choices influence watershed service provision, the magnitude and nature of the effects are highly context specific and are poorly understood in tropical forest environments (CALDER 2005; VAN NOORDWIJK 2005; PORRAS ET AL. 2008; VEIGA 2008). Land and water linkages are challenging to manage because watershed services have a common good characteristic (PORRAS ET AL. 2008), and therefore are often perceived as ‘free’ and are generally unaccounted for in monetary terms (TEEB 2009). As a result they are often degraded and or not reflected in traditional economic measures such as GDP (COSTANZA ET AL. 1997; FAO 2007; TEEB 2009). Land-use patterns and land management practices by upstream landholders in a watershed largely determine the quality and quantity of available water downstream (PORRAS ET AL. 2008). For example, unsustainable land use and agricultural practices by upstream landholders can result in negative hydrological side-effects or externalities, such as increased sedimentation downstream (MEA 2005; VEIGA 2008). Consequently the design of policy measures to encourage farmers to adopt watershed conservation measures and, at the same time maintain or even increase productivity has become a major issue for researchers, natural resource managers, and conservation practitioners. Chapter 1. Introduction 2  Several policy options are available to enhance incentives for the supply of ecosystem services. A command-and-control regulatory approach, such as the prohibition of forest clearing, has long been the preferred policy strategy of many governments for the control of environmentally harmful land-use changes (FAO 2007; PORRAS ET AL. 2008; BÖRNER & VOSTI 2012). More recently, mechanisms based on economic incentives or markets for ecosystem services are being increasingly implemented as cost-effective and complementary tools for promoting ecosystem service conservation (FAO 2007; ENGEL ET AL. 2008; PORRAS ET AL. 2008). These incentive-based mechanisms include payments for ecosystem services (PES) and payments for watershed services (PWS), water quality trading markets, and reciprocal or in-kind agreements (BENNETT ET AL. 2013). Between 2001 and 2011 Latin America registered a transaction value of 528.9 million USD in such investments corresponding to 3.4 million hectares of forest (BENNETT ET AL. 2013). PWS programmes are considered to be the fastest growing and most mature among all of the PES schemes currently implemented in Latin America (STANTON ET AL. 2010; BALVANERA ET AL. 2012; BENNETT ET AL. 2013). PWS efforts in Latin America typically involve compensating rural upstream agricultural producers for protecting and/or restoring natural forest ecosystems or natural highland grasslands (páramo) (GRIEG-GRAN ET AL. 2005; VEIGA & GALVADÃO 2011; PRIA ET AL. 2013). Targeted ecosystems are typically located in strategic water production areas, such as headwaters, riparian forests, or water intake points for public potable water supplies (VEIGA & GALVADÃO 2011; PRIA ET AL. 2013). There is, however, little evidence in the peer review literature of the effectiveness of such schemes for meeting conservation and development goals (see PATTANAYAK ET AL. 2010; ARRIAGADA ET AL. 2012). In Brazil PWS efforts are expanding and there are already 848 ecosystem service suppliers in the Atlantic Forest region alone, mainly organised or supported by the National Water Agency´s Water Producer Programme (SANTOS ET AL. 2010; VEIGA & GALVADÃO 2011). As of 2012, 41 PWS projects had either been implemented or were in preparation across an area of approximately 40,000 ha. Several PWS projects are strategically located in priority conservation areas of the Atlantic Forest or close to significant urban settlements (VEIGA & GALVADÃO 2011). Among the eight projects that had already been implemented in 2009, a pilot initiative was identified in the Guandu watershed, which supplies potable water to the city of Rio de Janeiro. A second project in the state of Rio de Janeiro is currently under development in the Três Picos State Park in the Municipality of Cachoeiras de Macacu (GUEDES & SEEHUSEN 2011). This state Chapter 1. Introduction 3  park is located in the mountainous area of the Guapiaçu-Macacu watershed (GMW), which is the geographical focus of this study. This watershed supplies potable water to almost 2.5 million inhabitants of five municipalities, including the city of Niteroi (PEDREIRA ET AL. 2009). The main drivers of water resource degradation in this watershed are urbanisation, intensive agriculture, and the elimination of riparian vegetation. Decision makers need reliable data and suitable tools to make more informed, scientifically-based environmental management decisions in order to maintain or improve valuable watershed services. Recognition of the value of ecosystem services (including watershed services) can contribute to a more transparent decision-making process (MEA 2005; CHEN ET AL. 2008; TEEB 2009). Identifying and measuring such values have begun to feed policy development processes in what has become an increasing trend worldwide, especially with regard to watershed services (DAILY & MATSON 2008; STANTON ET AL. 2010; GUEDES & SEEHUSEN 2011). This can result in opportunities to save costs through timely or targeted action (TEEB 2010a). Managing watershed services requires solid knowledge about: (1) the costs of providing an additional unit of water quality or quantity (supply side), and (2) the willingness of water users to pay for an additional unit of water quality or quantity (demand side). The economic aspects of watershed services supply and demand are particularly poorly understood and many PWS initiatives have begun operating without quantitative knowledge of such parameters (MARTIN-ORTEGA ET AL. 2012). Moreover, most widescale PES or PWS programmes lack rigorous evaluations of their conservation effects (PATTANAYAK ET AL. 2010; MITEVA ET AL. 2012) and often fail to account for design features oriented to make PWS schemes efficient (ENGEL ET AL. 2008; PATTANAYAK ET AL. 2010). Research questions, objectives, and hypothesis 1.2. The overall goal of this research was to contribute to the understanding of the economic and environmental aspects that influence the provision of landscape-scale hydrological ecosystem services (hereafter referred to as watershed services) based on circumstances in the GMW of Rio de Janeiro, Brazil. Specifically, this research concentrated on the watershed service of water quality maintenance in terms of turbidity reduction for the potable water supply. Factors that influence water treatment costs (related to the demand for services) based on land-use change were also analysed. Improved knowledge of both Chapter 1. Introduction 4  the costs of providing additional watershed services through land-use and land-cover change (supply) and the willingness-to-pay (demand) for watershed services is necessary to enable an informed decision-making process in the context of the GMW. Main research hypothesis: If the costs of converting current land uses to forest cover across farming systems in the GMW (ecosystem service supply) are greater than what the main watershed service beneficiary (water facility company) is willing to pay (ecosystem service demand) then a PWS scheme will not be viable. To fulfil this goal and test the research hypothesis, specific objectives and questions were formulated that are presented in a research matrix described in Table 1.1 and Figure 1.1. An overview of data sources and assessment methods for each research question and objective are also presented in Table 1.1. Each method is described in greater detail in Chapter 3. Chapter 1. Introduction 5  Table 1.1: Specific objectives, research questions, and data analysis methods Specific objectives Research questions Data sources and methods 1. To calculate the economic costs of converting current land uses to forest cover across farming systems of the GMW RQ1: What are typical farming systems within selected sub-watersheds of the GMW? 1.1 How do their form, intensity, and profitability vary?  Farming system analysis based on semi-structured farmer surveys  Cost-Benefit Analysis/Opportunity Cost analyses  Spatial mapping of the opportunity costs of forest conservation  Extrapolation of the mean opportunity costs at the producer (farming system) level to the watershed level using ArcGIS spatial analysis tools 2. To determine the key environmental conditions that determine the provision of watershed services (water quality) for the public water supply RQ2: Where is land-use change more likely to have a key future role in the improvement of watershed services? 2.1 In which areas of the GMW are watershed services highly vulnerable to land-use change?  Secondary spatial data on the vulnerability of water resources in the GMW Chapter 1. Introduction 6  3. To estimate the avoided costs of improving water quality for the primary downstream user: the public water supply provider (State Water and Sewage Company) RQ3: What are the treatment costs of the water supply company associated with improving water quality parameters such as turbidity? 3.1 How have the water treatment costs and key water quality indicators from the GMW changed over the last decade?  Expert interviews of the State Water and Waste Water Company technicians  Avoided Cost analysis 4. To identify priority areas for watershed service (water quality) protection and improvement RQ4: Is a PWS scheme more cost-effective than other land-use based options for enhancing water quality with respect to turbidity? 4.1 If not, what are other feasible options?  Determine the spatial distribution of priority areas for the provision of watershed services by overlaying land uses, the opportunity costs of land-use conversion, and the vulnerability of water resources in the GMW  Literature review on key criteria for identifying priority areas for watershed service protection and improvement  Literature review on the best management practices to improve watershed services Chapter 1. Introduction 7  ENVIRONMENTAL CONDITIONS MANAGEMENT OPTIONS Figure 1.1: Research framework showing the sequence of specific objectives and research questions in pursuit of the overall research goal 2. To determine the key environmental factors that affect the provision of watershed services Overall research goals: to better understand the economic and environmental aspects that determine watershed service provision with particular focus on water quality in terms of turbidity reduction for the potable water supply. To analyse the factors that influence water treatment costs (related to services demand) based on land-use changes. To improve understanding of both the supply and demand for watershed services in order to better inform the decision-making process with regard to watershed management in the context of the GMW. 2.1 Where is land-use change more likely to have key future impacts on water quality? 4.1. If not, what are other feasible options? DEMAND SUPPLY 4. To identify priority areas for watershed service conservation and improvement 3. To estimate the avoided costs of improving watershed services relevant for the main downstream water user 1. To calculate the economic costs of land-use conversion to forest cover across farming systems in the GMW 1.1 How do their form, intensity, and profitability vary? RQ1. What are typical farming systems in these subwatersheds of the GMW? RQ2. In which areas are water resources highly vulnerable? 3.1. How have water treatment costs in the GMW changed over the last decade? RQ4. Is a PWS scheme for this watershed service more costeffective than other measures for protecting or enhancing water quality? RQ3. Which land-use factors are relevant to water treatment costs? Chapter 2. Background and Conceptual Framework 14  water quality for human consumption is considered a ‘provisioning’ service, is used throughout this document. Table 2.1: Classification of watershed services Watershed service classification Sources Provisioning services Water availability: fresh water purification Involves water quality maintenance, including biological purity as well as sediment load (including both sedimentation and erosion control) DUDLEY & STOLTON 2003; TEEB 2010a Regulating services Regulation of water flows, including flood prevention MEA 2005; DAILY ET AL. 2009; TEEB 2010a Water supply “Provision of water for consumptive use (e.g. potable, irrigation, and industrial use).” DE GROOT ET AL. 2002 Waste treatment Forests filter dust particles from the air, and wetlands and other aquatic ecosystems can remove anthropogenic organic waste from relatively large volumes of water thereby acting as ‘free’ water purification plants. DE GROOT ET AL. 2002 Regardless of the classification type there are several essential watershed services provided by forest ecosystems:  Forests act as natural water filters and storage systems, where tree roots and leaf litter facilitate rainwater infiltration and help retain soil humidity (CWP & USFS 2008; VEIGA 2008).  Healthy forest and wetland ecosystems are considered very effective at regulating water flow and improving water quality (TEEB 2010a,b; RUSSI ET AL. 2013). Maintaining water quality includes the control of sediment, nutrients (particularly phosphorous and nitrogen), chemicals, and salinity (TEEB 2010b). These ecosystems regulate water flows by helping maintain ground water and water tables during dry seasons or drought years and controlling floods through the containment and absorption of water when precipitation is excessive (FAO & Chapter 2. Background and Conceptual Framework 15  CIFOR 2005; CWP & USFS 2008; KFOURI & FAVERO 2011). In addition, forest ecosystems can remove pathogenic microbes, sequester and convert inorganic ions, and transform persistent organic pollutants (TEEB 2010b).  Forests enhance stream quality and watershed health by reducing storm water runoff and pollutants delivered to freshwater systems (CWP & USFS 2008; VEIGA 2008). Specifically, riparian forests control erosion and both sediment and nutrient loads, as well as other diffuse sources of water pollution, thus helping to maintain water quality (FAO 2005; KFOURI & FAVERO 2011). Even though these ecosystem services are of great value to human populations, their degradation has been increasing considerably (MEA 2005). One of the primary reasons for this trend is the fact that most ecosystem services have traditionally been considered public goods and their provision is challenged by the combined effects of population and economic growth and larger global integration (FAO 2007). There are, however, several options for addressing this challenge. Sustainable ecosystem services management 2.3. Considering the importance of ecosystem services to human well-being and environmental health, it is a challenge to manage such services in a sustainable manner, but this can be accomplished through several approaches. MEA (2005) identified five types of actions that contribute to the achievement of sustainable ecosystem services management, including: 1. Institutional and governance—Changes related to environmental governance and institutional frameworks are often necessary to enable the fundamental conditions needed for effective ecosystem management. 2. Economics and incentives—Interventions offer effective instruments for regulating the use of ecosystem goods and services. 3. Social and behavioural—These actions consider aspects related to population policy; public education; community, women, and youth empowerment; and from civil society. 4. Technological—The development and dissemination of technologies that help improve resource use efficiency or decrease the impacts of drivers of ecosystem service degradation. Chapter 2. Background and Conceptual Framework 16  5. Information-based—Appropriate ecosystem management is often limited due to a lack of comprehensive knowledge of ecosystem complexity and by the failure to use existing information adequately. Since the loss of natural capital is partly attributed to the exclusion of ecosystem services from traditional economic measures such as GDP (COSTANZA ET AL. 1997; FAO 2007; TEEB 2009), there is a major challenge that can be met through the application of economic policies and incentives for establishing their value. For example, this can be achieved by establishing markets2 for ecosystem services, and by targeting the monetary and financial interests of selected social actors (MEA 2005). Market development for ecosystem services relies on intensifying the cooperative and hierarchical arrangements between beneficiaries and providers to be able to plan group payment strategies and to address the problem of ‘free riders’3 (LANDMILLS & PORRAS 2002). A legal framework is generally required for the adoption of economic instruments, which are dependent on the particular socio-economic situation (MEA 2005). Promising economic instruments and market-based approaches with the potential to contribute to the sustainable management of ecosystem services include:  Taxes or user fees for activities that have negative external costs (externalities)— For example, taxing excessive application of fertilisers (MEA 2005).  Creation of markets such as cap-and-trade systems—Ecosystem services that have been considered free goods, as is often the case for water, tend to be used inefficiently, however, establishing markets for ecosystem services incentivises their conservation and enhances the economic efficiency of ecosystem service allocation (TEEB 2009). This is more easily achieved when there are existing and effective supporting legal and economic institutions. Nevertheless, although markets can improve the efficiency of resource use, they can negatively affect specific user groups. Combining regulated emission caps with market mechanisms for trading pollution rights has often been shown to provide an efficient way of 2 Market is defined here as “an actual or nominal place where forces of demand and supply operate, and where buyers and sellers interact (directly or through intermediaries) to trade goods, services, contracts or instruments, for money or barter” (WUNDER 2013). 3 Free riders are those that benefit from a service without paying for it (FAO 2007; THOMAS & CALLAN 2010). Chapter 2. Background and Conceptual Framework 17  decreasing emissions that are harmful to ecosystem health (MEA 2005; TEEB 2009).  Payment for ecosystem services (PES). These schemes are established to enable individuals, enterprises, or the public sector to pay resource owners for the specific ecosystem services desired (MEA 2005). The focus of PES schemes or programmes varies according to their purpose. For example, the main focus of PES is often to support and improve ecosystem management, while some PES schemes also attempt to reduce poverty. Within PES, payments for watershed services (PWS) are those schemes that particularly focus on hydrological ecosystem services or watershed services (ASQUITH & WUNDER 2008). PWS are generally linked to water supply, availability, and/or quality (FAO 2004). In this context PWS schemes are intended to encourage landowners to adopt practices that contribute to erosion control or the protection and/or restoration of natural ecosystems that do so naturally, especially forests located in strategic water production areas (FAO 2004; VEIGA & GALVADÃO 2011). Normally, these schemes are initiated when water users recognise the existence of a positive externality. Such an externality often occurs when rural farmers implement forest restoration and conservation measures; thus the scheme seeks to incentivise or provide compensation for carrying out such activities. In this form a monetary incentive is provided to the producers for engaging in practices that secure the provision and conservation of watershed services (VEIGA & GALVADÃO 2011).  Schemes promoting consumer preferences to be expressed through markets. Consumers can motivate producers to engage in more sustainable practices in other ways in the absence of effective regulation by the government. This is the case of many existing certification schemes for sustainable fisheries, agriculture, and forestry practices, including biodiversity-friendly agroforestry practices and products such as bird-friendly shade-grown coffee (MEA 2005; WUNDER 2006; TEEB 2009).  Removing subsidies for environmentally harmful activities. Combined global subsidies to the agricultural, fisheries, energy, transportation, and other sectors total up to almost one trillion USD annually (TEEB 2009). More than a third of these subsidies encourage fossil fuel production and consumption (TEEB 2009). Chapter 2. Background and Conceptual Framework 18  In addition to market-based tools for ecosystem conservation, a common policy approach is imposing command-and-control regulations. This approach consists of the design and implementation of regulatory instruments such as prohibiting environmentally harmful activities, creating protected areas, and other legal measures (WUNDER 2006). In developed countries many watershed protection programmes have focused on effective legal land protection in situations where the command-and-control approach can be implemented satisfactorily. In developing countries, however, the effectiveness of command-and-control tools is frequently limited due to weak institutions and poor governance, especially in agricultural frontier areas (WUNDER 2008). 2.3.1 Payment for ecosystem services as a mechanism for safeguarding sustainable ecosystem management PES schemes are increasingly being proposed as local projects and are often supported by international cooperation as a means of managing the trade-offs between agricultural land uses and watershed services in a cost-effective manner. One perceived benefit of this approach is that it is an economic approach that is relevant to both water resource and biodiversity conservation goals (FAO 2007; WUNDER 2007). Since the last decade this tool has been gaining importance among scientists and practitioners working on ecosystem and/or biodiversity conservation efforts (MURADIAN ET AL. 2013). PES is expected to have the potential to mobilise innovative means of financial support for sustainable ecosystem management and simultaneously contribute to other complementary goals such as poverty reduction and/or agricultural development (FAO 2007). This can be achieved by linking the interests of landowners and external actors (e.g. water users) through the provision of compensation in order to address trade-offs or internalise environmental externalities between those stakeholders (WUNDER 2008). In this way PES schemes might facilitate increases in the quality and quantity of services provided by ecosystems that had not previously been supported by monetary incentives from the beneficiaries (FAO 2004; STROBEL ET AL. 2007). The most important principle in PES is that the transaction is voluntary, whereby a buyer or user (service demand) pays the provider (service supply) for the continuous provision of a well-defined ecosystem service (LANDMILLS & PORRAS 2002; WUNDER 2008; PAGIOLA ET AL. 2010). Additionally, the beneficiary or user’s willingness-to-pay (WTP) is often estimated to assess the demand for a given ecosystem service (FAO 2004). In the case of watershed protection, ecosystem service providers are commonly upstream Chapter 2. Background and Conceptual Framework 19  agricultural producers and potable water users the beneficiaries. Another important feature of PES is that the provision of the desired ecosystem service is secured and monitored (WUNDER 2008). The premise of PES is that the landowner (ecosystem service provider) and the beneficiaries have divergent interests and that unless the latter compensates the former, there is a threat to the continued provision of the service (FAO 2007). A key aspect of PES is that payments can be stopped, reduced, or suspended in the case of non-compliance (conditionality principle) on behalf of the service provider (WUNDER 2008). This can be rather efficient, but might create problems if conservation costs are high (WUNDER 2008). Another critical aspect to consider when implementing a PES scheme is land tenure. In areas where the legal property rights of the owners of the forest that provides the service are not clearly defined a PES scheme is unlikely to be feasible (LANDMILLS & PORRAS 2002; FAO 2007; WUNDER 2013). Due to its adaptive nature it is argued that PES can achieve fair and “flexible conservation outcomes” (WUNDER 2013). Cost-effective PES schemes, however, require a cautious design approach that relies on the particular characteristics of the ecosystem service being provided, as well as the local biophysical and socio-economic contexts (FAO 2007; TEEB 2010a). Furthermore, there are certain preconditions for implementing PES schemes that embrace cultural, informational, economic, competitive, and institutional aspects (WUNDER 2008). Cultural aspects need to be carefully considered because PES schemes can only function if the ecosystem service providers are sufficiently motivated by the payments received and that they do not consider them as socially inappropriate (WUNDER 2008; 2013). Moreover, PES should involve some level of trust between the service users and providers, where mutual contract compliance can be expected and misconceptions about intentions (i.e. unanticipated trespassing by the users on the service providers’ land) do not take place (WUNDER 2008). Normally, a broker is required to deal with conflicting interests between service providers and users (LANDMILLS & PORRAS 2002; WUNDER 2008). PES involves transaction costs and demands a relatively intensive information gathering process. Such costs tend to be high during the initial phase of implementing PES schemes. These entail price negotiation, ecosystem service baseline assessment, system design, etc. (WUNDER 2008). In the case of the Pinampiro watershed in Ecuador start-up Chapter 2. Background and Conceptual Framework 20  costs were 69 USD ha-1, while recurrent transaction costs during the operating phase were only 1.6 USD ha-1 (WUNDER & ALBÁN 2008). When PES schemes are properly organised, it is possible to overcome constraints that could hinder their successful implementation. For example, donor funding could defray high start-up transaction costs in certain cases. This is only realistic when PES schemes prove to be cost-effective and sustainable once they are operational. In the case of watershed protection, rapid biophysical assessment methods can reduce the high costs of establishing baselines (WUNDER 2008). One economic precondition of PES is that there is a positive externality. This means that there is compensation to the landowner for the service provided to the beneficiaries (FAO 2007; KFOURI & FAVERO 2011). In this sense, a PES scheme can become an additional income source for participating landowners who avoid using the areas responsible for the provision of the desired ecosystem service and thereafter forgo commercial production in these areas (KFOURI & FAVERO 2011). A second economic precondition for PES to function is that the benefits must be greater than the costs of providing the ecosystem service (i.e. that the ecosystem service user’s WTP for the given ecosystem service must be greater than the service provider’s OCs) (WUNDER 2013). In this case OCs are understood as the foregone benefits of carrying out one activity in favour of another, or giving up their initial preferred land-use plan (UGALDE 1996; FAO 2004; MEA 2005; FAO 2007; WUNDER 2008; WINPENNY ET AL. 2010). Since OC analyses are an important element for the assessment of ecosystem service provision within PES programmes (WUNDER 2008) special attention is given to OCs in Section 2.4. The supply of watershed services: farming systems and their opportunity costs. ENGEL ET AL. (2008) identified key aspects of assessing the effectiveness and efficiency of PES programmes, which are also valid for PWS schemes. Among these, it is desirable that the PES scheme: performs better than the alternative ‘business-as-usual’ scenario, considers ‘additionality,’ avoids ‘leakage,’ and achieves long-term improvements in ecosystem service provision. In this case the term ‘additionality’ refers to the adoption of practices that would not be implemented in the absence of the PES programme (ENGEL ET AL. 2008). In other words PES schemes lack additionality when they pay for the implementation of practices that would have otherwise been adopted (e.g. due to environmental regulations) without the programme (FERRARO & PATANAYAK 2006). Chapter 2. Background and Conceptual Framework 21  ‘Leakage’ in this sense refers to avoiding undesirable practices in the PES intervention zone, but that can be implemented in areas external to the programme (ENGEL ET AL. 2008). Long-term improvements in ecosystem service provision (permanence) should be achieved as a result of PES, which includes timeframes beyond the PES programme and ‘after payments end’ (ENGEL ET AL. 2008). Some authors argue that PES schemes depend on the economic valuation of ecosystem services (MURADIAN ET AL. 2013). Others argue that “even a service of infinite value can be protected through cost-priced PES, as strictly speaking, we don´t need to know what the service is worth, as long as we know that we want to keep it” (WUNDER 2013). In this dissertation the economic valuation of ecosystem services is viewed as an instrument that can support PES design (e.g. by helping to determine service payments), without being a strict requirement for PES implementation (WUNDER 2013). 2.3.2 Economic valuation of ecosystem services There have been important developments in the economic valuation of ecosystem services over the past twenty years, although there remains much to be accomplished (STROBEL ET AL. 2007; TEEB 2009). In this sense the economic valuation of ecosystem services has become an essential tool for increasing public awareness about the value of environmental and biodiversity protection (MEA 2005; FAO 2007; STROBEL ET AL. 2007). Economic valuation includes the identification and economic quantification of the negative impacts resulting from ecosystem degradation or the loss of ecosystem services (TEEB 2009). Economic values are not a natural property of ecosystems and they are usually closely related to the number of beneficiaries and the specific socio-economic context (TEEB 2009). COSTANZA ET AL. (1997) calculated the economic value of the whole biosphere considering 17 global ecosystem services of 16 biomes at a minimum of 33 trillion USD annually. Economic valuation can help make ecosystem conservation financially sustainable. Valuation can significantly support conservation efforts by identifying, measuring, and demonstrating the most important benefits or avoided losses to the beneficiaries associated with a specific ecosystem service. For this reason valuation is a strategy that can help convince decision makers to allocate the necessary resources for ecosystem conservation (PAGIOLA ET AL. 2004). Economic valuation can contribute to the spatial determination of where ecosystem services could be provided at a lower cost than alternative artificial interventions (e.g. for Chapter 2. Background and Conceptual Framework 22  water purification, potable water provision, flood control, etc.). The valuation process requires an ecological understanding of these services together with appropriate monetary valuation methods (TEEB 2009). The rationale behind this endeavour is to better understand the complexities involved in socio-ecological relationships, to make the manner in which humans influence ecosystem service value explicit, and to express such values in units that can be incorporated into public decision-making procedures (TEEB 2009). There are several approaches for determining the value of benefits and services provided by ecosystems. A very common approach for estimating the value of ecosystem services is the concept of total economic value (Figure 2.2). This approach considers the full range of economic values that society attributes to each land use. Figure 2.2: Total economic value of ecosystem services Source: PAGIOLA ET AL. 2004  Direct use values—are those derived from goods or services with an established market value, which usually imply benefits like commodities, fuel wood, timber, non-timber forest products, recreation, education, and tourism (PAGIOLA ET AL. 2004; FAO 2007). Most of these belong to the MEA (2005) category of provisioning services and their valuation is generally relatively simple (FAO 2007).  Indirect use values—are those that society attributes indirectly from ecological functions such as watershed protection, pollination, pest control, and soil fertility (TEEB 2010a).  Option values—are the benefits of preserving potential future direct or indirect use of the resources or service for oneself (option values) or for others (bequest values) (PAGIOLA ET AL. 2004). Biodiversity conservation is commonly comprised of option values (FAO 2007).  Non-use, existence or passive use values—are those that do not imply the utility of ecosystems. These are intrinsic values attributed to the ecosystem simply because Chapter 2. Background and Conceptual Framework 23  they exist and therefore need to be conserved for future generations, such as particular habitats or endangered species (PAGIOLA ET AL. 2004; FAO 2007). When sustainably managed the total economic value of ecosystems is often higher than the economic returns generated from converting natural vegetation to agriculture, clearcut logging, or other intensive uses (MEA 2005). For example, in 1998 the net present value of protecting water quality in the Catawba River (USA) for five years was estimated at 346 million USD. To illustrate the value of water purification by wetlands, in 1992 it was estimated that about half of the total economic value of the Danube River floodplain could be attributed to its role as a nutrient sink (MEA 2005). Estimating the value of ecosystem services is complicated because there are no markets for many of them. However, there are several ways to estimate their value in monetary and non-monetary terms. There are several approaches to monetary valuation, such as: revealed-preference, state-preference, cost-based, and benefit transfer. Non-monetary valuation can include measures of attitudes, preferences, and intentions; decision science approaches; ecosystem benefit indicators; biophysical ranking methods; and civic valuation (LIU ET AL. 2010). Some of the most relevant methods are presented in Table 2.2. The key differentiating aspect between monetary and non-monetary methods is whether the approach uses direct observation of behaviour (i.e. revealed-preference approaches) or responses to hypothetical questions (state-preference approaches) such as: “How much would you be willing to pay for...?” or “What would you do if…?” (LIU ET AL. 2010). On occasions when the ecosystem service is difficult to value using these approaches, it may be possible to estimate the value using the replacement or avoided cost method. Nevertheless, such cost-based approaches should be used carefully as the resulting values do not measure economic value because they do not rely on preferences (LIU ET AL. 2010). In cases where a more timely and cost-effective valuation process is desired the benefit transfer approach is often applied. Benefit transfer is based on the analysis of one or more previous studies that determined the value of similar services in comparable locations. Subsequently, the values obtained in the original studies are adapted and transferred to another geographical location and/or time. This approach is specific to the service that is being valued. Some services have a high degree of transferability, such as carbon sequestration at certain scales, however, the value of services at the local level such as flood control might have much more limited transferability (LIU ET AL. 2010). Chapter 2. Background and Conceptual Framework 30  elevated concentrations of suspended particles and solutes that drain into local water bodies (BATCHELOR ET AL. 1998). Changes in land cover such as the conversion of forest to agricultural land uses tend to accelerate erosion processes, increase superficial runoff, and increase sediment concentrations in rivers that results in downstream siltation (MEA 2005; FAO 2007). Conversion of forest cover to urban land uses also increases runoff, which in turn raises the concentrations of pollutants and reduces the temporal reliability of freshwater systems (e.g. lower long-term groundwater recharge) (MEA 2005). The conversion of undisturbed natural ecosystems to other land uses can severely reduce the capacity to provide essential ecosystem services (POWER 2010). Watershed service provision in agricultural areas is determined at the parcel level due to land-use decisions made by individual agricultural producers based on the relative benefits or returns that can be obtained from among alternative activities (FAO 2007). This involves a series of dimensions that are illustrated in the concept of farming systems in Figure 2.3. A farming system is considered to be a basic unit that can be comprised of a few dozen or up to millions of households and that includes the resources (land, labour and capital) used within this unit (BEETS 1990; DIXON ET AL. 2001). Such systems typically have comparable resource bases, enterprise patterns, household livelihoods, and limitations (DIXON ET AL. 2001). Each individual household has particular characteristics resulting from their resource endowment and family conditions. Independently of their size, each farming system is oriented towards food production and meeting household objectives through the management of available resources within a given social, economic, and institutional context. Farming systems are normally engaged in mutually dependent processes including the production, harvest, and post-harvest treatment of crops and livestock (DIXON ET AL. 2001). Various natural (geographical), technological, economic, cultural, and institutional components of the specific context influence farming systems (Figure 2.3). In this thesis a farming system is considered a social group (generally a group of households) in which the economic environment has the greatest influence on farmer decisions and thus determines land-use decisions without excluding all other relevant factors that are essential for producing goods and services. This rationale is based on the high degree of integration among farmers and markets in the study area, which has increasingly monetised agricultural income, inputs, and labour (FORERO 2002; GAESE 2009). Chapter 2. Background and Conceptual Framework 31  Figure 2.3: Bottom-up approach for the assessment of ecosystem services Regarding the economic context, access to markets and price ratios influences farmer decisions on input purchases, the timing of product sales, and enterprise patterns. Moreover, the available economic and social infrastructure in rural areas is essential for determining transport costs and the availability of services. Other factors such as information and educational services influence household decisions and strategies in the same manner. Available technologies determine the nature of the production and processing of agricultural products and natural resources (DIXON ET AL. 2001). The factors that determine the ecosystem services provided by farming systems have several dimensions. There are changes that can increase the output of an ecosystem service, however, changes can also affect a number of other services in either a positive or negative way (FAO 2007). Farmers typically choose the combination of production practices that maximises their well-being based on consideration of their existing assets and their access to resources and opportunities (FAO 2007). Farmers seldom adopt changes in land use or agricultural practices that are favourable to the environment in the absence of motivational policy measures, such as incentives or Chapter 2. Background and Conceptual Framework 32  command-and-control measures. For example, setting aside agricultural land to preserve or restore forest cover can result in the protection or enhancement of water quality, biodiversity preservation, and carbon sequestration services; however, such decisions represent lower returns to the farmer’s household (FAO 2007). Farming systems can make major contributions to the mitigation of threats to water quantity and quality. Reciprocally, more sustainable land uses in combination with revenues from water management services could help achieve sustainable and profitable farming systems. More sustainable farming systems can generate a wide variety of ecosystem services. Without economic incentives for ensuring that these services are provided, more sustainable land uses are very likely to remain economically unattractive to land owners. Most PWS schemes in developing countries are guided by a ‘land-based’ approach, whereby service providers are paid to improve their land management practices, which are expected to result in the desired watershed services, rather than being paid for the actual service delivery (PORRAS ET AL. 2008). Regardless of the approach, watershed service provision is inevitably linked to land-use decisions (FAO 2007). Consequently, watershed service provision analyses often require cost-benefit analyses of agricultural production systems in order to derive OC estimates. The concept of OCs4 is considered essential in many modern economic analyses (IPCC 2001). In the case of PWS analyses, OCs are most frequently used to express the costs of providing additional watershed services. In the context of watershed services OCs represent any benefits foregone by a farmer by converting existing land uses to alternatives that contribute to watershed service delivery. The OCs calculated from maintaining forest cover can contribute significantly to the cost effective design of forest conservation schemes because the forest conservation costs to farmers in a specific area can be assessed (BÖRNER ET AL. 2009). Ideally, PWS schemes are designed so that, at minimum, payments compensate landowners for the OCs of additional service provision. The extent to which such payments are quantitatively appropriate or not depends on the alternative land-use options in each given area (PAGIOLA ET AL. 2010). 4 “This cost measures the economic value of outputs, goods, and services that would have been possible to produce elsewhere with the resources used to produce the last unit of good X” (IPCC 2001). Chapter 2. Background and Conceptual Framework 33  In general terms, ecosystem service programmes that involve changes in land use tend to be more effective where the OCs of agricultural land uses are low (FAO 2007). In areas where land is abundant or there are greater off-farm employment opportunities that reduce rural populations, the potential to set aside land for non-agricultural uses is greater (FAO 2007). Where land is scarce and labour is readily available the trade-offs between agricultural and non-agricultural services are higher, normally resulting in higher OCs relative to land-use conversion (FAO 2007). The concept of OCs is highly relevant to common mechanisms of assessing ecosystem services such as REDD+ and PES, and can also be a very useful measure for the abatement of CO2 emissions (e.g. in the context of avoided deforestation). OC analyses provide insights into the drivers of deforestation because higher OCs tend to be associated with higher deforestation levels (WBI 2011). Calculating OCs is also useful for determining reasonable compensation to those who change their land-use practices as part of a given forest conservation programme such as REDD+, and to assess whether such programmes provide sufficient incentives to conserve forests. For example, in areas with high OCs (i.e. converting high value agricultural systems such as soybean, palm oil or cattle) on productive soils it is more difficult for schemes to provide sufficient incentives to restore or conserve forests (WBI 2011). The demand for watershed services 2.5. Demand refers to the current beneficiaries of the delivery of watershed services. The assessment of demand includes accounting for the resources available for protecting and conserving watershed services (GUEDES & SEEHUSEN 2011). Demand for ecosystem services, especially from agriculture, will grow due to increases in their scarcity and value (FAO 2007). A significant source of financial support for PWS schemes in developing countries comes from funding or development assistance from the international public sector (PORRAS ET AL. 2008). A key provider of this funding is the Global Environmental Facility (GEF), which acts as a buyer on behalf of service users for conserving global public services. Approximately 108 million USD have been made available as World Bank (WB) loans and 52 million USD as GEF grants respectively for WB/GEF-supported PWS projects (FAO 2007). For example, the World Bank has provided loans to support the development and implementation of well-known PES programmes in Mexico and Costa Rica (FAO 2007). Chapter 2. Background and Conceptual Framework 34  Involvement of the private sector in paying for ecosystem services is growing, including in the context of corporate social (and environmental) responsibility efforts (TEEB 2010c). Depending on the source of funding PES schemes can be classified into two broad categories: user-financed and government-driven (ARRIAGADA & PERRINGS 2009). Small-scale PES watershed protection projects (user-financed) are increasing, especially in Latin America. While PES was originally conceived to concentrate on achieving environmental outcomes (WUNDER 2013), governmental initiatives tend to have multiple goals (i.e. also incorporating poverty alleviation, regional development or political objectives) (WUNDER 2008). Privately-driven and user-financed PES schemes (or in this case PWS), such as efforts by the French water company Vittel, tend to incentivise voluntary actions to decrease water pollution resulting from upstream land uses such as livestock production. This company has provided incentives to farmers to change their land management practices in order to reduce nitrate concentrations in spring water (PERROT-MAÎTRE 2006). The changes in agricultural practices included the elimination of maize cultivation for animal fodder and the application of associated agrochemicals, the reduction of livestock densities, and the modernization of farm facility construction to minimise nutrient runoff (FAO 2007). A similar example of a privately-driven initiative in Germany is an effort by the Bionade Corporation together with the NGO ‘Trinkwasserwald e.V.’ (Forest Drinking Water Association) to undertake reforestation activities to improve groundwater recharge (BENNETT ET AL. 2013). Examples of governmental PWS efforts in Latin America include the PES programme in Costa Rica, the National Programme for Hydrological Environmental Services in Mexico (PSAH), and the Brazilian Socio-environmental Federal Development Programme of Rural Production (Proambiente)5 programme (FAO 2007; WUNDER 2008). The PES programme in Costa Rica pays upstream landowners to preserve forests for erosion control and includes private sector entities that pay for watershed services, including: hydroelectric companies, a beer brewing company, Florida Ice and Farm, and the World Bank (FAO 2007; ECOSYSTEM MARKETPLACE 2010). The programme in Mexico is financed from water user fees and payments are made according to forest type. The price  Chapter 2. Background and Conceptual Framework 35  paid was based on the OCs of land with the assumption that maize production would be an alternative use to conservation. In this case conserved areas of cloud forest and mesophilous forest receive higher payments than similar areas of temperate forest (ECOSYSTEM MARKETPLACE 2010). Proambiente provides farmers with subsidised credit for farming practices that are believed to enhance ecosystem service delivery (BÖRNER ET AL. 2007). Examples of PWS efforts of variable scales are summarised in Table 2.6. Some government-driven PWS programmes that pay rural producers for conservation activities on their properties in the Brazilian Atlantic Forest region include: ‘Bolsa Verde’ in the state of Minas Gerais, ‘ProdutorES de Água’ in the state of Espírito Santo, and ‘Mina D’água’ in the state of São Paulo (GUEDES & SEEHUSEN 2011). These activities include the protection or restoration of native vegetation with a focus on headwaters and riparian forests (GUEDES & SEEHUSEN 2011; VEIGA & GALVADÃO 2011). In general all PWS schemes in the Atlantic Forest region obtain funding from public sources as well as from Basin Committee charges for water use. Few cases in this region involve private initiatives. A user fee system can be an effective approach to efficient water resource use. Such user fee systems have been implemented in the watersheds of the Piracicaba, Capivari and Jundiaí (PCJ) rivers in the state of São Paulo with approximately 8.8 million beneficiaries (VEIGA & GALVADÃO 2011). In this situation, an Inter-Municipal Basin Committee was formed to manage a watershed protection fund and contributions to the fund come from municipal water utility budgets. Chapter 2. Background and Conceptual Framework 36  Table 2.6: Selected PWS related schemes worldwide Case, Country Environmental Services Buyer Seller Targeting Criteria Payment Scheme Government-Financed Programs Environmental Quality Incentives Programme, USA (CLAASSEN ET AL. 2008) Watershed protection, biodiversity conservation (benign agriculture & agricultural land retirement) US government US farmers Participants are selected based on environmental benefits and cost index Annual cash payment. A reserved price is based on the rental value of land adjusted for its productive capability Conservation Reserve Programme, USA (CLAASSEN ET AL. 2008) Watershed protection, soil conservation, wildlife protection and carbon sequestration (benign agricultural practices and agricultural land retirement) US government US farmers Participants are selected based on environmental benefits and cost index Annual cash payment. A reserved price is based on the rental value of land adjusted for its productive capability PES Programme (Costa Rica)1 (ECOSYSTEM MARKETPLACE 2010) Water-based ecosystem services markets The public sector (state power corporation “Compañía Nacional de Fuerza y Luz” - CNFL) and the public utility of the town of Heredia Land owners Annual cash payments for natural regeneration, forest preservation, and new forest plantations. Contracts can also pay per tree for agroforestry contracts. National Programme for Hydrological Environmental Services (PSAH ) Mexico2 (ECOSYSTEM MARKETPLACE 2010) Watershed / payment for hydrological services provided by forests The funds for this programme are designated by Congress to the National Forestry Commission (CONAFOR) Forest land holders or 'ejidos' (community held land) Applications are qualified according to an evaluation of the risk of deforestation and water scarcity, as well as social criteria in order to enroll areas with higher social and environmental benefits. The price paid has been determined by the government based on the opportunity cost of the land and depending on the forest type Chapter 2. Background and Conceptual Framework 37  Case, Country Environmental Services Buyer Seller Targeting Criteria Payment Scheme Working for Water Programme, South Africa (TURPIE ET AL. 2008) Watershed and wetlands protection (clearing invasive alien plants) Previously unemployed individuals that tender for contracts to restore public or private lands Department of Water Affairs and Forestry, water management agencies The programme prioritise areas using ecological and social rationales Cash payments to contractors staff that have been previously unemployed User-Financed Programmes The Vittel (Nestlé Waters) watershed protection programme, France (ARRIAGADA & PERRINGS 2009) Watershed protection (best practices in dairy farming) Vittel Dairy farmers (27 farmers enrolled) Cash payments are based on new farm investment and the cost of adoption of new farming practices Los Negros, Bolivia (ASQUITH & WUNDER 2008; ARRIAGADA & PERRINGS 2009) Watershed and biodiversity protection (cloud forest and páramo conservation) Pampagrande municipality, US Fish and Wildlife Service Santa Rosa farmers (46 landowners) In kind (beehives supplemented for honey production) plus technical assistance Pimampiro, Ecuador (WUNDER & ALBÁN 2008; ARRIAGADA & PERRINGS 2009) Watershed protection Metered urban users (20% fee) Households in Nueva América Cooperative Participant selection has focused on Nueva América because it is located near the water intake Three differentiated cash payments according with forest type 1Voluntary contracts are channelled by the Nacional Fund for Forest Finance (FONAFIFO), a mixed private-public body under the direction of the Ministry of Environment. 2Mexico is working to develop voluntary payments from water users to supplement the funding given by the government under the World Bank/GEF Environmental Service Project. Chapter 2. Background and Conceptual Framework 38  PWS schemes in Brazil are expanding and there are already 848 ecosystem service providers in the Atlantic Forest region who are receiving annual payments that range from 10 to 577 BRL (4.5 to 257.2 USD) per hectare monthly (VEIGA & GALVADÃO 2011). As of 2010, 40 PWS projects had been identified in the region, that collectively cover an area of around 40,000 ha (VEIGA & GALVADÃO 2011). In 2011 these projects were in different stages of development: 8 were being implemented; 20 were under development, and 12 were in the design stage. An overview of these projects is presented in Appendix I based on consideration of the valuation method used, the value paid, relevant detailed case studies, limitations, and recommendations for PWS schemes in the Atlantic Forest region. These efforts include the programme of payments for restoring riparian areas in São Paulo, the Water Conserver project, the Water Producer project, the Producers of Water and Forests project, and the Economic Valuation for pricing water in Três Picos State Park. The majority of these projects are concentrated in southeast Brazil (28 Projects), followed by seven in the south, and only five in the north and northeast of the country. Several projects are strategically located in priority conservation areas of the Atlantic Forest or close to significant urban settlements (VEIGA & GALVADÃO 2011). Thirteen projects are located in significant water supply systems that provide large urban centres with potable water, such as Guandu in Rio de Janeiro and towns near the cities of São Paulo, Brasilia, Campo Grande, Vitória, and Palmas (VEIGA & GALVADÃO 2011). Fifty seven per cent of all PWS projects in the Atlantic Forest are in areas of semi-deciduous seasonal forests and ombrophilous mixed forests, 13% are in savannah areas, and 30% are in transition areas among mixed ombrophilous forest, steppes, and savannahs (VEIGA & GALVADÃO 2011). Most PWS projects in the Atlantic Forest are led by municipal governments and in some cases by water suppliers, NGOs, and the National Water Agency (ANA). Some PWS projects have been supported through public policies of states such as Espiritu Santo, Minas Gerais, and São Paulo. The state of Rio de Janeiro government has already reached advanced discussions regarding PWS legislation (VEIGA & GALVADÃO 2011). Chapter 2. Background and Conceptual Framework 39  Relevant legislation, policy tools, and land-use based options 2.6. for the management of watershed services in Brazil Relevant legislation for the assessment of watershed services can be found in the Brazilian National Law on Water Resources (Law No. 9433/97), which is potentially a basis for establishing watershed service markets and permitting charges levied on users for water consumption as is already practiced in several states (VEIGA 2008). This approach is based on the ‘polluter pays’ principle to transfer payments from the ‘users/polluters’ to the ‘protectors’ of springs and riparian forests. Another important tool for assessing the relationships between forest cover and watershed services is foreseen in Articles 47 and 48 of the National System of Conservation Units ‘Sistema Nacional de Unidades de Conservação’ (SNUC)6. SNUC is the Brazilian law governing protected areas and it foresees payments to conservation units for ecosystem services. Unfortunately implementation of Articles 47 and 48 of the law has been delayed due to the lack of corresponding regulations (STROBEL ET AL. 2007). The ‘ecological’ or ‘green’ tax for the circulation of merchandise and services ‘Imposto sobre a Circulação de Mercadorias e Serviços’ (ICMS)7 is another relevant economic incentive that is being used to support management costs of the private natural heritage reserves. This policy tool is intended to increase the economic attractiveness of creating new private protected areas as well as to improve the management of existing reserves (VEIGA 2008; MARQUES 2009). The ICMS has already been implemented in 11 Brazilian states, including: Paraná, Minas Gerais, São Paulo, Rio de Janeiro, Mato Grosso do Sul, Amapá, Mato Grosso, Rio Grande do Sul, Pernambuco, Rondônia, and Tocantins (VEIGA 2008). Each state has specific environmental criteria that define the proportion payable of the 25% that the municipalities can obtain from this tax (VEIGA 2008). For example, the state of Paraná pioneered this type of tool and since then the number of protected areas within the state has increased by 165% (MARQUES 2009). 6 The National System of Conservation Units (SNUC) instituted by Law 9985/2000 works under the protector-receiver principle, indicating that those who protect natural resources should receive a financial reward for their stewardship (STROBEL ET AL. 2007). Chapter 3. Methods 46  3. Methods Study area 3.1. The study area lies within the 1,265 km² Guapiaçu-Macacu watershed (GMW) located in the Serra do Mar mountain range in the state of Rio de Janeiro (Figure 3.1). This mountain range is considered a distinct biogeographical region (RIBEIRO ET AL. 2009) within the Atlantic Forest and is considered a high-priority conservation area due to its high levels of biodiversity (CEPF 2001). Historically the Atlantic Forest biome has been characterised by high deforestation rates (DEAN 1997), resulting in numerous isolated and disperse forest fragments in a landscape dominated by agriculture (RIBEIRO ET AL. 2009; NEHREN ET AL. 2013). The Atlantic Forest supplies 135 million people with water (PRIA ET AL. 2013); however, a mere 11% to 16% of the original forest cover remain (RIBEIRO ET AL. 2009). Figure 3.1: The Guapiaçu-Macacu watershed within the state of Rio de Janeiro Source: Modified from RODRÍGUEZ OSUNA ET AL. (2014) Chapter 3. Methods 47  Key geographical characteristics of the state of Rio de Janeiro include steep mountains, isolated inselbergs,8 a varied coastal topography, and an assortment of land-cover types (NEHREN ET AL. 2013). Remaining forest cover in the state of Rio de Janeiro and particularly in the GMW (Figure 3.2) is at relatively high altitudes. Forest cover mostly remains on steep slopes that are inappropriate for agriculture, while the foothills and lowlands are dominated by agricultural land cover types (STROBEL ET AL. 2007; NEHREN ET AL. 2013). In 2008 the land cover/land use composition of the GMW consisted of 48.8% forest cover (in all stages of growth), 41.4% pasture, 4.4% crop production, and 5.4% a combination of urban areas, water bodies, bare soil, rock outcrop, and mangrove or other wetlands (FIDALGO ET AL. 2008). The soils of the GMW are the result of a combination of the lithology (gneisses/granite) of the highly dissected relief, the Holocene climate, and dense forest cover. The main soil types are cambisols, ferralsols, acrisols, fluvisols, and gleysols (NEHREN 2008). Alluvial soils (fluvisols) are especially found in the plains of the GMW, cambisols are found in areas with greater relief, and gleysic soils are found in the downstream reaches that developed due to seasonal flooding (Figure 3.3). There are four principal natural ecosystems in the study area: tropical rainforest (dense ombrophilous forests); mangrove; marsh and other wetlands (pioneer formations); and highland grasslands (UFF & PETROBRAS 2007). 8 A tectonic uplift beginning in the early Cretaceous and ending in the Oligocene resulted in the Serra do Mar and Serra da Mantiqueira mountain ranges. The uplift was associated with various volcanic intrusions and extended periods of intense erosion and weathering that resulting in isolated inselbergs and pediments among the pediplains across different altitudinal gradients (NEHREN ET AL. 2013). Chapter 3. Methods 48  Figure 3.2: Land use in the Guapiaçu-Macacu watershed Chapter 3. Methods 49  Figure 3.3: Elevation and soil classes in the Guapiaçu-Macacu watershed Source: FIDALGO ET AL. (2008) & LUMBRERAS (2010). In: PENEDO ET AL. (2011) Most of the rivers and creeks in the GMW originate in Três Picos State Park, with the remainder originating in the Serra dos Órgãos National Park and the Paraíso State Ecological Station. The Macacu and Guapiaçu rivers originate within the state park and constitute the main components of this watershed (STROBEL ET AL. 2007). The intake point for the public water supply is located in the lower GMW at the Imunana channel and is managed by the state water utility company (CEDAE). The Imunana channel joins the Macacu with the confluence of Guapiaçu and Guapimirim rivers. The channel was built in the 1940s to drain the frequently flooded river basin for agriculture and for malaria control (STROBEL ET AL. 2007). In addition to the Imunana channel, the natural meandering Chapter 3. Methods 50  river channel has been replaced by straightened, deepened, and extended channels that together have contributed to the drastic reduction of wetlands in the watershed (DA COSTA ET AL. 2007). There is regional climate variation within the GMW. The coastal area in the southwest experiences lower precipitation rates and higher temperatures than the mountain range in the northeast (PENEDO ET AL. 2011). This is reflected among distinct locations: in the state of Rio de Janeiro at an altitude of 5 m (asl) the mean annual temperature is 23°C and mean annual precipitation is 1,093 mm; whereas in Nova Friburgo at 856 m (asl) in the mountains the mean annual temperature is 18°C and mean annual precipitation is 1,246 mm (NEHREN 2008). Mean annual precipitation across the GMW ranges from 1,200 mm to 2,750 mm (ANA 2011), with an annual pattern that peaks during summer (December through February) and reaches its minimum values during the winter (June through August) (PENEDO ET AL. 2011). In 2010 there were a total of 54,273 inhabitants in the municipality of Cachoeiras de Macacu in the GMW, giving it a population density of 56.9 people per square kilometre (IBGE 2010). There are various domestic, industrial, and agricultural demands on the water supply in the GMW as a result of the high water quality from the main rivers (STROBEL ET AL. 2007). A new petrochemical complex (COMPERJ) under construction in the lower reaches of the watershed is expected to further increase both water demand and pressure on natural resources in the study area (PEDREIRA ET AL. 2009). Three sub-watersheds of the GMW (Figure 3.4) are featured as study sites in this research: (I) Manuel Alexandre, (II) Batatal, and (III) Caboclo. Each of these subwatersheds contains different combinations of the principal land-cover/land-use types (forest, crop production, and pasture), which together account for approximately 95% of the total watershed. These three GMW sub-watersheds with significantly different compositions of these prevailing land uses were selected for a comparison of the variability of agricultural profitability and the effects of land-cover composition variability on water quality (turbidity). Chapter 3. Methods 51  Figure 3.4: Characteristic landscape features of the three Guapiaçu-Macacu sub-watersheds: Manuel Alexandre (left), Batatal (upper right), and Caboclo (lower right) Source: RODRÍGUEZ OSUNA ET AL. (2014) (I) The Manuel Alexandre sub-watershed is a well-preserved landscape with a high proportion of forest cover (87%) located in the Guapiaçu Ecological Reserve (REGUA) (FIDALGO ET AL. 2008; PEDREIRA ET AL. 2009). This GMW sub-watershed served as a reference site due to the relatively undisturbed natural forest ecosystem. Most of this sub-watershed is protected by an NGO called REGUA, which is financially supported by the Brazilian Atlantic Forest Trust (BART), an organisation whose stated mission is the protection of the Atlantic rainforest of the upper Guapiaçu river watershed. This mission is primarily accomplished by purchasing properties to enlarge existing protected areas. (II) In 2008 the land-cover composition of the Batatal sub-watershed consisted of a mosaic of forest fragments (69%), pastures (28%), and crop production (4%) (FIDALGO ET AL. 2008). Perennial crops such as banana predominate in the uplands, whereas annual crops (cassava, maize, beans and other vegetables) are dominant in the flatter lowlands. Forest fragments at a variety of different successional stages are found in higher elevation areas. (III) The predominant land cover in the Caboclo sub-watershed is forest (81%), followed by pasture (14%), and crop production (3%) (NAEGELI 2010). Agricultural systems in the Caboclo study site are considerably more intensive than in the Batatal sub-watershed and Chapter 3. Methods 52  are mainly located along the floodplain. Most of the cultivated crops are annuals such as maize, cassava, beans, and other vegetables. The higher elevations of the Caboclo study site are within Três Picos State Park where agricultural activities are restricted to the buffer zone. Both the Batatal and Caboclo sub-watersheds have undergone similar historical exploitation cycles that have resulted in widespread deforestation, forest fragmentation, forest degradation, and intense soil erosion (NEHREN ET AL. 2013). Chapter 3. Methods 53  Watershed service supply assessment methods 3.2. 3.2.1 Farming system and opportunity cost analyses The collection of contextual information began with informal conversations with local experts and key informants (Figure 3.5 and Figure 3.6) in order to learn about general local conditions and production patterns, and as a basis for the selection of representative farms. Essential information gathered through this process included: the most commonly cultivated crops, crop seasonality and productivity, livestock characteristics, agricultural management practices, variability of resource availability, and production constraints (STRÖBEL ET AL. 1987). Figure 3.5: Farming system analysis steps Chapter 3. Methods 54  Figure 3.6: Preparatory work for the household survey effort that included presentations to local producers on the research sampling goal and strategies with scientists from Embrapa Soils9 To calculate the OCs associated with the provision of watershed services according to the different land-use types, cost-benefit analyses of the representative farming systems in the region were performed. Detailed individual budgets for all activities for each farming system were developed for this analysis (see BEETS 1990) that summarised the revenue 9 Embrapa is the Brazilian Company for Agricultural Research, a technological innovation company linked to the Ministry of Agriculture, Livestock, and Food Supply that focuses on promoting knowledge and technology within Brazilian agriculture. Within Embrapa there is an Embrapa Soils unit and specifically scientists from the area of Geotechnologies and Environmental Monitoring were involved in this research. Chapter 3. Methods 55  and cost information. The activity budgets were later aggregated to calculate the mean rate of return for each land-use type (WBI 2011). Crop budgets were later compared to official current production costs estimated by the Rural Extension and Technical Assistance Agency (EMATER) of the state of Rio de Janeiro. The target population included households that practiced crop production in the Batatal and Caboclo sub-watersheds of the GMW. Household surveys were conducted during two field visits in 2011 and 2012 with the support of key local producers, EMATER of the municipality of Cachoeiras de Macacu, the city council, and scientists from Embrapa Soils. Expert interviews of the director and other staff members of the REGUA reserve in the Manuel Alexandre sub-watershed were conducted to better understand the land-use history and management practices in the study area. An inventory was created to define the sample population using an indirect census technique following FORERO (2002). This process consisted of a participatory mapping exercise based on recent satellite imagery provided by the city council of Cachoeiras de Macacu. The exercise permitted the assembly of a list of all farms within the GMW subwatersheds of interest that was subsequently validated via field visits and local experts who included a member of the agricultural department of the city council (Cachoeiras de Macacu), the president of the Faraó Farmers Association (ALAF) in Batatal, a member of the Rural Workers Union in Caboclo, and the director of the REGUA reserve in Manuel Alexandre. Thirty-two households were identified in Batatal and 60 households in Caboclo, from which a total of 78 households within the two populated GMW subwatersheds participated in a semi-structured survey. No interviews were conducted in the Manuel Alexandre reference sub-watershed. The sample size was consistent with a minimum sample size of 25–30 households for communities with 100 to 500 families following ANGELSEN ET AL. (2011). Two survey samples were conducted during the course of the fieldwork. The first survey sample included randomly selected households within each of the GMW sub-watersheds to determine representative farming system characteristics. Important selection criteria for these farming systems, as suggested by ZIMMER ET AL. (2009) and local experts, included: farm size, land tenure, production and agricultural management practices, and mean farm altitude. The design of the first semi-structured survey (Appendix II) considered key variables for a subsequent classification process, including: information on farming system form, Chapter 4. Results and Discussion 62  Table 4.1: General household characteristics in the Batatal and Caboclo study sites Batatal (N=32) Caboclo (N=46) Relationship to rural property % % Work purposes only 25.0 56.0 Both work and residence 63.0 44.0 Basis of land tenure or land-use rights % % Inherited or provisional possession certificate 66.0 – Settlers (‘assentados’ or ‘parceiros’) – 84.0 Sharecroppers (‘meeiro’) 19.0 5.0 Property keeper (‘caseiro’) 9.0 – Renter (‘arrendatário’) 3.0 – Other 3.0 11.0 Family size % % ≥4 people 87.5 74.0 ≤5 people 12.5 26.0 HH age distribution % % 15–60 years old 65.0 65.0 >60 years old 23.0 9.0 >15 years old 15.0 26.0 Mean percentage of HH members working in the agricultural sector 46.0 68.0 Mean percentage of HH members receiving income from sources other than agriculture 42.0 19.0 Primary income source Agriculture Government assistance Off-farm employment Other % 45.0 22.0 30.0 3.0 % 68.3 7.3 21.9 2.4 Mean parcel size Upland Lowland 1–14 ha 15–50 ha >50 ha % 81.2 12.5 6.2 % 62.5 31.2 6.2 % 100 – – Mean land use Forest cover Crop production Pasture % 54.0 37.9 4.9 % 34.2 47.3 15.9 % Common reserve 72.6 26.6 Distance to sub-watershed headwaters 234 m 466 m 1,200 m Chapter 4. Results and Discussion 63  Four types of farming systems were identified in the Batatal study site and two types were identified in Caboclo (Table 4.2). In Batatal 80% of the upland farming systems specialised in banana (Musa sp.) production, which was classified as FS1. The remaining 20% were mixed systems of cassava (Manihot sp.) and banana, which were classified as FS2. In the Batatal lowlands where agriculture is more intensive there were two additional farming systems classified as FS3 and FS4 that were equally represented. The FS3 category includes mixed production systems of cassava, green maize (Zea mays), yam (Colocasia sp.), and courgette (Cucurbita sp.). The FS4 category consists of the same mix of cassava, green maize, and yam production, but with banana instead of courgette (Table 4.2). Cassava is the dominant crop at the Caboclo study site, followed by green maize, yam, and common beans (Phaseolus vulgaris). Cassava, which requires eight to nine months from planting to harvest, is popular due to its low input requirements and relatively stable prices. Green maize requires greater care and thus has higher cultivation costs; however, it provides relatively rapid returns because it may be harvested only 90 days after planting. Common beans help improve soil fertility through nitrogen fixation and complement other traditional crops. Other short-cycle crops grown in the study area include okra (Hibiscus esculentus) and gilo (Solanum gilo). The most common farming system in the Caboclo sub-watershed (70%), classified as FS5, includes a combination of cassava, yam, common bean, and green maize production. The other farming system category in Caboclo was FS6 (variant FS6a), which is a combination of cassava, yam, common bean, green maize, and okra or gilo production (Table 4.2). Typically two crops of green maize, courgette, and beans can be cultivated each growing season. Chapter 4. Results and Discussion 64  Figure 4.2: Banana production at the Batatal study site An agricultural calendar for the main crops produced in the study area was generated based on the household survey results. The calendar includes the most relevant local agricultural production characteristics such as crop seasonality, crop productivity, and agrochemical application scheduling (Appendix III). The information integrated into the calendar was highly relevant for the second survey effort in support of the cost-benefit analysis of the farming systems in the study area and was also used as an essential input for hydrological modelling efforts, which are being carried out in the study area within the framework of the DINARIO project. Crop yields and the quantities and timing of agrochemical applications served as parameters for the GMW hydrological model.12 12 Santiago Penedo and Annika Künne are completing their PhD research using the J2000-S model developed by the Friedrich Schiller University of Jena, which simulates hydrological factors such as water quality and quantity, as well as nutrient transport in the GMW as part of the DINARIO project. Chapter 4. Results and Discussion 65  Agriculture in the Batatal uplands is less intensive than the lowlands in terms of fertiliser use (Table 4.2) due to the fact fertilisers are not used for upland banana production. The remoteness of the upland production areas made intensive production less attractive relative to the lowlands. Lowland Batatal vegetable producers applied a mean of 240 kg of fertiliser per hectare annually. Over 70% of the households in Caboclo reported using fertilisers for crop production at a mean application rate of 547 kg per hectare annually. These values are consistent with the findings of EMATER (Table 4.3). Figure 4.3: Typical landscapes at the Caboclo study site Chapter 4. Results and Discussion 66  Table 4.2: Household crop production characteristics at the Batatal and Caboclo study sites Batatal Caboclo Upland Lowland Farming system proportions FS1/Banana (80%) FS2/Banana and cassava (20%) FS3/Cassava, green maize, yam, and courgette (50%) FS4/Cassava, yam, green maize, and banana (50%) FS5/Cassava, yam, beans, and green maize (70%) FS6/Cassava, yam, beans, green maize, and okra or gilo (30%) Fertiliser applications in mean kg ha–1 130 240 547 Pesticide use (% of HH) 59 44 89 Fertiliser use (% of HH) 31 56 68 Lime applications for soil pH treatment (% of HH) 25 44 29 Fertility analysis (% of HH) 16% None Most common pesticides Herbicide: ‘Roundup’ (glyphosate) Herbicide: Roundup. Other products: ‘Decis’ (pyrethroid), ‘calda sulfocálcica’ (homemade product with insecticidalfungicidal-acaricidal function) Herbicides: Roundup, ‘Gramoxone’ (paraquat) Other products: Decis, ‘Kumulus’ (fungicide) Most common fertilisers NPK 20–5–20, NPK 4–14–8, and chicken manure Cattle and poultry manure, bone meal, NPK 10–10–10, orthophosphate, NPK 4–14–8, urea, NPK 12–6–12, and NPK 25–20–15 Bone meal (alone or in combination with the same products used in Batatal lowlands) Primary fertilised crops Cassava (occasionally) Most vegetable production (cassava is the only crop that is not fertilised) All Field preparation Manual labour using hoe (‘enchada’) Mix of manual and mechanical (microtractor or tractor) Mix of manual and mechanical Livestock production 1–2 mules for hauling harvested bananas 13% of HH owned >35 beef cattle 8% of HH owned >27 beef cattle Chapter 4. Results and Discussion 67  Table 4.3. Fertiliser application rates for each of the typical crops in the Rio de Janeiro study area Crop Fertilisation pattern Quantity (kg ha-1) Fertilisation Quantity (kg ha-1) Banana No fertilisation – – – Cassava Application during the planting period 200 – – Green Maize Application during the planting period 250 Applied on soil surface 150 Yam Application during the planting period 400 Applied on soil surface 200 Courgette Application during the planting period 900 Applied on soil surface 900 Gilo Application during the planting period 900 Applied on soil surface 900 Okra Application during the planting period 300 Beans No fertilisation Source: Modified from EMATER-RIO (2011a) All households surveyed in the Batatal uplands reported using herbicides for weed control. In the Batatal lowlands a variety of agrochemicals were used for both weed control and fertilisation. The common use of herbicides could be due to the local scarcity and high cost of manual labour for weed control. The Atlantic Forest Law (Law 11.428/1986) and Brazilian Forest Code (Law 12.651/2012) impose certain limitations on agricultural production. The Atlantic Forest Law prohibits the conversion of secondary forest to other land uses. The collapse of market prices for banana (~1999–2000) led to the neglect of many banana plantations that subsequently developed into secondary forest that was later abandoned entirely in compliance with these regulations. In addition, the Brazilian Forest Code establishes that for the Atlantic Forest Biome, 20% of rural properties need to be maintained under permanent forest cover as a ‘Reserva Legal.’ The Brazilian Forest Code also prohibits clearing primary vegetation on slopes steeper than 45°, and along the margins and headwaters of rivers and streams, which are classified as areas of permanent protection (APPs) (MINISTRY OF ENVIRONMENT 2005). Chapter 4. Results and Discussion 68  Site preparation in the Batatal uplands was exclusively performed using manual labour with hoes (enchada), which can be easily used on steep slopes and is permitted under the Atlantic Forest Law. In lowland areas of both Batatal and Caboclo, however, a combination of manual and mechanised site preparation methods predominated (Table 4.4). Typically mechanical preparation was conventional tilling performed with either a tractor or micro-tractor. Mechanical tillage can either rely on private services or is subsidised by the municipality. The municipality only has a limited ability to provide this service due to the fact that there are only five tractors for the entire municipality, which is insufficient for meeting all of the local demand for this service (pers. comm. DA SILVA, 2011). Table 4.4: Key characteristics of the study area farming systems in the state of Rio de Janeiro, Brazil Farming system Main crops Farm size (ha) Mean crop production area (ha) Land tenure Site preparation methods Batatal uplands FS1 Banana 10.3 10.3 Banana Inheritance or provisional (‘posse’) title Manual FS2 Banana/ Cassava 10.3 8.3 Banana/2.0 Cassava Mixed manual and mechanical Batatal lowlands FS3 Cassava/ Yam/Green Maize/ Banana 6.0 2.0 Cassava/2.0 Green Maize/1.0 Yam/1.0 Banana Inheritance or provisional (‘posse’) title Mechanical FS4 Cassava/ Yam/Green Maize/ Courgette 6.0 2.0 Cassava/2.0 Green Maize/1.0 Yam/1.0 Courgette Mechanical Caboclo (lowland) FS5 Cassava/ Yam/Beans/ Green Maize 4.0 1.0 Cassava/1.5 Green Maize/1.0 Yam/0.5 Bean Settlers Mechanical FS6 Cassava/Gre en Maize/ Yam/ Okra/Gilo/Be an 4.0 1.0 Cassava/1.5 Green Maize/0.5 Yam/0.5 Okra or Gilo/0.5 Bean Mechanical The main difference in livestock production between upland and lowland Batatal was that mules were the dominant livestock in the uplands due to their important role in Chapter 4. Results and Discussion 69  transporting harvested bananas from the plantations, which are mostly located on slopes. In the lowlands of both study sites a minority of households engaged in small-scale (<20) cattle production. In the municipality of Cachoeiras de Macacu 95% of the owners of larger cattle farms or ‘fazendas’ lived off the farms and only 30% of all producers were residents of the livestock farms where they worked (personal communication, Luiz Eutalio DE ALMEIDA, president of local dairy enterprise Cia do Leite Maraporã Coop. Agrop. Ind. Ltda., 2011). Among livestock producers, farms under 20 ha are considered small-scale and farms above 400 ha are considered large-scale (pers. comm. DA SILVA 2011; pers. comm. DE ALMEIDA 2011). According to municipal government cattle vaccination records the entire registered herd in all three districts of the municipality was 27,995 animals in 2011 (5,026 in Cachoeiras, 11,241 in Santa Ana de Japuiba, 11,728 in Subaio) (STATE SECRETARIAT OF AGRICULTURE, CATTLE FARMING, FISHERIES & FOOD SUPPLY 2011). Cattle are usually maintained in large, undivided pastures. Small-scale livestock producers are concentrated in the nearby settlements of Sao José da Boa Morte, Marubaí, and Vecci. Beef cattle production is concentrated among a few large-scale properties. Dairy producers are concentrated in the areas of Papucaia and Marubaí where there was a government dairy project. An important characteristic of beef cattle productivity is the rate of weight increase, which varies according to altitude. Mean annual weight gain per head of cattle is between 120– 150 kg in the lowlands, compared to 40–60 kg in the upland forested areas (personal communication, Nicholas LOCKE, REGUA Association Director, 2012). This difference is primarily due to the difficulties of maintaining pastures and productivity in areas with steep slopes. Several cattle producers mentioned experiencing lower productivity (i.e. lower calf survival and calf weight gain rates due to lower pasture productivity) at the higher altitude pastures. Additional information on the use of agroforestry systems, willingness to produce other crops, and observations about erosion were requested in the survey. The survey results were provided to support other scientific groups within the DINARIO project and especially the local project partner, Embrapa Soils. This information was not included in the analyses presented because it is not highly relevant to the objectives of this research effort. Chapter 4. Results and Discussion 70  Watershed service supply assessment: cost-benefit and 4.2. opportunity costs analyses The estimated mean annual profits per hectare for agricultural land uses were 4,115 BRL (1,702 USD) in Batatal and 5,052 BRL (2,090 USD) in Caboclo (Table 4.5). Relative to the highlands, returns on agriculture tend to be higher in the lowlands where production intensity is higher. There is also greater agricultural input use in the lowlands, especially fertiliser use for commercial crop production. Profits earned by cattle producers were approximately 20, 40 and 100 BRL (8.3, 16.5 and 41.4 USD) per hectare annually depending on slope (Table 4.5). In some cases operations belonging in the FS2 group (cassava production) in the Batatal uplands are not profitable, which is primarily due to the relatively high labour requirements and low yields. Area-weighted OCs ranging from 14 to 1,660 BRL (5.8 to 686.6 USD) per hectare in each of the GMW sub-watersheds were spatially projected over the study area using ArcGis software tools (Figure 4.4). In terms of area, livestock pasture is the dominant land use in the GMW. Livestock production had low per hectare profits relative to other cover types. Chapter 4. Results and Discussion 71  Table 4.5: Summary of estimated annual profits of farming systems in the study area Main Crops Unit Batatal Caboclo Pasture slope categories* Upland Lowland Lowland FS1 FS2 FS3 FS4 FS5 FS6a FS6b P1 P2 P3 Banana [USD] 394.4 394.4 398.8 Cassava –541.2 1,602.6 1,602.6 2,122.3 2,122.7 2,122.7 Green Maize (summer) 961.9 961.9 838.2 838.6 838.6 Green Maize (winter) 1,394.8 1,394.8 961.9 961.9 961.9 Yam 1,160.7 1,160.7 2,515.0 2,515.0 2,515.0 Courgette 1,474.0 Gilo 4,438.9 Okra 4,094.4 Beans (summer) 575.7 575.7 575.7 Beans (winter) 617.5 617.9 617.9 Occurrence [%] 80 20 50 50 70 15 15 Typical area by farming system (FS) [ha] 10.3 10.3 6.0 6.0 4.0 4.0 Per hectare profit [USD ha–1] 394.4 213.4 1,579.8 1,759.2 1,983.8 2,181.2 2,224.3 8.1 16.2 40.6 Mean value extrapolated to the watershed level [USD ha–1] Macacu River lowlands Guapiaçú River 1,669.4 2,049.5 8.1 16.2 40.6 Source: Modified from RODRÍGUEZ OSUNA ET AL. (2014) *According to the expert interviews, cattle weight gains on high slope upland pastures ranged 40–60 kg per animal in contrast to lowland pastures with gains of 120–150 kg; ‡ Upland agriculture cover types in the Batatal study site (FS1 and FS2) were not detectable by the available land-use classifications (FIDALGO ET AL. 2008; PEDREIRA ET AL. 2009; NAEGELI 2010), the mean per hectare annual profits for lowland agriculture in Batatal were applied (FS3 and FS4), resulting in a mean annual per hectare value of 4,114.8 BRL (1,669.4 USD). Chapter 4. Results and Discussion 78  4.2.2 Perceptions regarding water quality and quantity in the Batatal, Caboclo, and Manuel Alexandre study sites All water for household consumption in the study area comes directly from headwater areas or river sources (EMATER-RIO 2011b). It was also found that none of the local households use treated water for drinking and that none of the agricultural production areas are irrigated. All reported water use was for human consumption and leisure. In the study area all sewage is discharged directly into surface water bodies. All of the survey respondents in Batatal considered the quality of this untreated water to be ‘very good.’ All respondents reported that there were no seasonal water availability problems, however, three individuals mentioned that in recent years the amount of available water was reduced from June to August. Some interviewees reported water use for other activities such as watering gardens at home and post harvest washing of yams, which is a market requirement. Other responses regarding water use included selling water as ‘bottled water,’ and use for dairy cattle, fish farming, energy generation, and poultry raising. All of the respondents’ properties were located close to water sources. In the Batatal uplands the mean distance from households to their water sources was 234 m; while in the lowlands the mean distance was 466 m. In Caboclo the mean distance between households and waters sources was 1.2 km2. In response to an inquiry about perceived changes in water quality, the most common response in the Batatal study site (45%) was a perceived decrease in stream volume. Some respondents (21%) reported reductions in the fish populations in the rivers, and others reported changes in the annual rain pattern (14%). A few respondents in Batatal (7%) reported that water quality had improved due to the local increase in forest cover from fallow crop production and pasture areas, while the rest reported that they did not know (14%). Some respondents, however, associated perceived decreases in water availability to the land-cover transition from agriculture (primarily banana production in Batatal) to forest cover over previous decades. Contrary to the perceptions of respondents in Batatal, the leading response in Caboclo was a perceived increase in stream volume (42%), while 38% reported decreased stream volume, and the rest (21%) reported that they did not know. Chapter 4. Results and Discussion 79  Figure 4.5: Reported perceptions on changes in water quality and quantity in the Batatal (N=29 responses) and Caboclo (N=24 responses) study sites Forty-eight respondents at both study sites reported their perceptions regarding the cause of changes in water quality and quantity. In Batatal the perceived reduction of water quantity was most commonly attributed (21%) to the increased number of water bottling companies in the area and to sand extraction along the river channels (8%). Fewer respondents attributed water quality or quantity changes to climate change (4%), deforestation (4%), or water quality improvement due to increased forest cover (4%). The decrease in fish populations was attributed to the construction of a dam downstream from the study area that does not allow locally popular fish species such as the ‘Piaba’ to travel upstream to spawn. The remainder of the respondents could not explain the changes in water quality or chose not to respond (13%). Some individuals mentioned that a positive indicator of water quality was the return of the river otter or ‘lontra’ (Lontra longicaudis) to the GMW. In Caboclo decreased stream volume was most commonly associated with climate change (23%) followed by population growth (13%), and some attributed this to the construction of recreational pools for residential housing in upstream areas. The most commonly (10%) cited factor contributing to the perceived increases in stream volumes was ‘no more cattle in the area.’ According to LOCKE (pers. comm. 2012) the most important changes in the study area that are linked to water availability and quality over the past decades include: deforestation, excessive removal from sand from river channels, increased erosion due 0 10 20 30 40 50 Decrease in stream volume Increase in stream volume Less fish Change in rain patterns Better water quality Do not know % of responses Response categories Batatal Caboclo Chapter 4. Results and Discussion 80  the greater velocity of the river current due to the channelization efforts, and the extraction of mineral water by bottling companies (over 1,000 litres of water daily from the Guapiaçu river). These factors were also mentioned as determinants of water quality and quantity (personal communication, Joseph Edwards MATHEWS, local municipal environmental control authority director, 2012). The high water quality in this region is reflected by multitude of water bottling initiatives that have been launched in recent years. There are currently 17 water-bottling projects in the area, at least three of which are located along the Guapiaçu river (pers. comm. DA SILVA 2011; pers. comm. LOCKE 2012; pers. comm. MATHEWS 2012). KOSOY ET AL. (2007) identified misconceptions among common public perceptions (conventional local wisdom) and scientific evidence regarding the relationships between forest cover and hydrology. In their analyses of local perceptions of water provision based on three case studies in Central America (Honduras, Nicaragua, and Costa Rica) they found that greater forest cover in upstream areas was popularly believed to result in downstream improvements to both water quantity and quality. JOHNSON & BALTODANO (2004) found that the same perception was common among rural residents in other parts of Latin America, and this was also true in tropical regions worldwide (WILK 2000). Nonetheless scientific evidence has shown that the relationship between forest cover and water quantity is complex, and that in some cases increased forest cover can lead to lower downstream water availability (KOSOY ET AL. 2007). The relationship is highly context-specific; water yields from cloud forest headwater areas tend to be greater than those from montane forests that are not influenced by fog and low clouds (BRUIJNZEEL 2005) and empirical cases in which increased forest cover has resulted in decreased water availability downstream (KOSOY ET AL. 2007). However, there is a general consensus on the relationship between water quality and forest cover relationships in terms of both scientific evidence and local perceptions (KOSOY ET AL. 2007), therefore PES schemes intended to improve water quality through increased forest cover have a higher likelihood of being effective than schemes that seek to increase water availability (KOSOY ET AL. 2007). The Manuel Alexandre River watershed as a reference site 4.3. The GMW sub-watershed with the highest percentage of forest cover is Manuel Alexandre (89%) (FIDALGO ET AL. 2008). This site includes a private forest reserve with limited anthropogenic impact, primarily in the form of bird watching related tourism (Figure 4.6). This watershed lies below the Serra dos Orgãos mountain range in the state of Rio de Chapter 4. Results and Discussion 81  Janeiro, which is part of the larger Serra do Mar range that is characterised by stunning landscapes of ridges, peaks, valleys, and lowlands. Part of this area is protected by the REGUA, a conservation non-governmental organisation (NGO) that was formed in 2001. REGUA is financially supported by the Brazilian Atlantic Forest Trust (BART), which enables it to purchase properties from local landowners (REGUA 2011). Figure 4.6: Typical riparian landscape in the Manuel Alexander River watershed REGUA enlarges protected areas by purchasing adjacent properties or by negotiating management agreements with landowners. REGUA generates some income through the Bird Lodge tourism operations that provide services to 1,350 overnight guests and other day visitors each year, but most of its operating costs are covered by BART funding. REGUA’s reforestation activities are financially supported by a British registered charity called the World Land Trust (REGUA 2011). Presently REGUA owns 7,380 ha of forest and another approximately 3,300 ha of forest are subject to management agreements. The most significant management agreement in terms of scale is with the beverage company Schincariol, which owns a large forested property (2,500 ha) in the heart of the reserve (pers. comm. LOCKE 2012). The 449 ha Fazenda Manuel Alexander is located in this sub-watershed of the GMW, approximately 447 ha of which are forested (pers. comm. LOCKE 2012). Chapter 4. Results and Discussion 82  Vulnerability of water resources in the Guapiaçu-Macacu 4.4. watershed Water resource vulnerability is determined by both anthropogenic and environmental factors (Figure 4.7). The assessment of water resource vulnerability must therefore rely on indicators that accurately reflect the status of these factors. With regard to environmental state indicators (i.e. geomorphology, hydrogeology, slope, the area covered by APP forest fragments, soils, drainage density, and watershed circularity) the higher reaches of the GMW typically have greater vulnerability to anthropogenic pressure than the lowlands (FERREIRA 2012). Anthropogenic pressure indicators are highly influenced by human population density, land-use practices, the number and size of urban settlements, road density, and other factors. The GMW sub-watersheds with more settlements and larger human populations therefore have relatively greater anthropogenic impacts. In terms of population the higher altitude GMW sub-watersheds have considerably lower impact values as a result of less human settlement (FERREIRA 2012). All factors (state and pressure) were weighted based on the results of consultations with hydrological experts by FERREIRA (2012) (see Figure 4.7). Figure 4.7 shows the range of water resource vulnerability among the GMW sub-watersheds with the darker values indicating greater vulnerability and the lighter values indicating lower vulnerability. The least vulnerable GMW sub-watersheds are in the lower reaches and one of these is a protected natural area subject to land-use restrictions. Chapter 4. Results and Discussion 83  Figure 4.7: Water resource vulnerability among the sub-watersheds of the Guapiaçu-Macacu watershed Source: RODRÍGUEZ OSUNA ET AL. (2014) Chapter 4. Results and Discussion 84  Analysis of environmental and economic criteria for 4.5. watershed service conservation and improvement in the Guapiaçu-Macacu watershed Results of the analyses of environmental (vulnerability of water resources) and economic criteria (OCs of watershed service provision) in the GMW (Figure 4.8) were used to prioritise sub-watersheds according to their potential for watershed service conservation and improvement. The target areas identified have the greatest potential for improving water quality through land-use conversion (typically pasture or crop production areas to forest) and the lowest OCs (Figure 4.8). Most of the higher priority sub-watersheds have slopes that are steeper than the mean slope values in the study area. Chapter 4. Results and Discussion 85  Figure 4.8: Prioritisation of the sub-watersheds of the Guapiaçu-Macacu watershed for improving or maintaining watershed services Source: RODRÍGUEZ OSUNA ET AL. (2014) Chapter 4. Results and Discussion 86  Demand for watershed services in the Guapiaçu-Macacu 4.6. watershed: water treatment According to micro-economic theory the demand for water treatment chemicals by water utility companies was expected to reflect their WTP for water quality, in this case turbidity values under 5 Nephelometric Turbidity Units (NTU) as the specified threshold for human consumption. Thus the treatment cost of each additional unit of turbidity reflects a company’s potential WTP for measures that reduce turbidity by an equal unit (see description of the avoided cost method in PERMAN ET AL. 2003). The avoided cost method was applied to the local water utility company information based on expert interviews and relevant studies (DEARMONT ET AL. 1988; REIS 2004; MEDEIROS ET AL. 2011). Turbidity was identified as the key water quality parameter with respect to treatment costs. The main water treatment characteristics are presented in Figure 4.9. Figure 4.9: The Imunana channel is the main source for the public water supply in the GuapiaçuMacacu watershed Chapter 4. Results and Discussion 87  Table 4.6: Water utility intake characteristics in the lower Guapiaçu-Macacu watershed Laranjal Treatment Unit (CEDAE) Municipality of São Gonçalo Drainage area 1,263 km2 Forest cover in the watershed 48.8 % Treatment type Conventional General treatment phases* Captivation, sedimentation, coagulation, flocculation, decantation, filtration, disinfection, water fluoridation, and pH correction Treated water flow Average flow in 2011 was 5.35 m3s–1 Population supplied with potable water 2,000,000 General chemical products used Aluminium sulphate Al2(SO4)3; polyelectrolyte, hexafluorosilicic acid H2SIF6; chlorine Cl; calcium oxide CaO Chemical water treatment costs (in 2011) 2.31 x 106 BRL (955,426.00 USD) Pre-treatment water turbidity, mean value 17.10 NTU Treated water characteristics Colour: 2.50 uH Turbidity: 0.34 NTU Total chemical and electricity costs for the water treatment unit For the treatment of 6 m3s – 1:  300,000 BRL (124,081 USD) = monthly chemical expenditures  100,000 BRL (41,360 USD) = monthly electricity expenditures for captivation and production Source: Modified from RODRÍGUEZ OSUNA ET AL. (2014) *For further information on the water treatment process see: http://water.epa.gov/lawsregs/guidance/ sdwa/upload/2009_08_28_sdwa_fs_30ann_treatment_web.pdf The avoided costs for a 1% reduction in turbidity were estimated at 15,510 BRL (6,415 USD) a year based on a mean annual treated water volume of 174,545 m3 and a mean cost of 22.2 BRL (9.2 USD) per 1,000 m3 (Table 4.8) using CEDAE data for the 1998– 2011 period (Table 4.7). In 2001 the mean minimum turbidity value was 17.1 NTU and the maximum turbidity value in 2004 was 32.2 NTU (Table 4.7). The Pearson correlation coefficient for mean annual turbidity levels and chemical treatment costs for the 1998– 2011 period was 0.4. Relative to other studies in Brazil this correlation value is low. The correlation value for seven water treatment units in São Paulo was 0.7 (REIS 2004) and the value based on a comparison of 10 watersheds and their treatment costs in São Paulo was 0.9 (CABRAL DE SOUSA 2011). Chapter 4. Results and Discussion 94  Certain livestock and grazing management practices are more likely to compromise water quality to the point that it is degraded and unable to meet water quality standards (EPA 2013). Excluding livestock from riparian areas and improving grazing management practices have been shown to contribute to reduced turbidity in streams (EPA 2013). Silvopastoral practices can offer other on-site benefits such as improving pasture productivity by cycling nutrients and water from soil horizons that are unreachable to the root systems of grasses (PAGIOLA ET AL. 2007). Silvopastoral practices also offer other direct benefits such as providing fodder, fuel wood, fruit, and timber. Increased shade resulting from silvopastoral practices has been shown to reduce thermal stress on cattle and can increase livestock productivity, particularly dairy production through improved soil fertility and the nutritional content of forage (PAGIOLA ET AL. 2007). Silvopastoral practices also offer other benefits such as increased biological diversity and carbon sequestration (PAGIOLA ET AL. 2007). Despite the fact that the WTP of the water utility company fails to meet the estimated OCs of converting productive land uses to forest cover, it may be possible to procure additional contributions from other water users that could be considered in the design of a PWS scheme for this watershed. For example, one well-known case of providing incentives for the application of practices to improve water quality among cattle farmers by a water bottling company, is Vittel in France, where the company pays producers to implement riparian protection practices in watersheds where the company sources water that is bottled and sold (PERROT-MAÎTRE 2006). After this “Vittel experience,” other French water bottlers such as Evian and Volvic followed this initiative (PERROT-MAÎTRE & DAVIS 2001). One important potential stakeholder is COMPERJ, the Rio de Janeiro petrochemical complex, which will consume large quantities of water. Another important feature of this assessment was the fact that turbidity levels were the sole basis of determining water quality because this service is relevant to water consumers downstream. This approach is supported, however, by other research efforts (LELE 2009) on the basis that it is relevant to estimating watershed service values of specific land uses only when considering the effect on human welfare of downstream water consumers. Consideration of additional ecosystem services provided by forests may increase the potential WTP for implementing PWS schemes in this region. The demand estimates were based on the results of an assessment of the WTP for water quality improvement in terms of turbidity by the water utility company CEDAE. Other research efforts in Latin America have found this approach to be helpful for the design of watershed payment schemes (MARTINEZ DE ANGUITA ET AL. 2011; PAGIOLA ET AL. Chapter 4. Results and Discussion 95  2010; MARTIN-ORTEGA ET AL. 2012; GARCÍA-NIETO ET AL. 2013). The application of a modified version of an existing modelling package, such as SWAT, was practical for the scope of other research efforts (see QUINTERO ET AL. 2009; MARTINEZ DE ANGUITA ET AL. 2011), but was considered to yield unsuitable results for this research based on consultations with local hydrologists. Improving our understanding of the relationships between water quality dynamics and land use is one of the greatest challenges in ecosystem services research (DE GROOT ET AL. 2010). Quantification of the water quality improvements resulting from specific land-use changes was not possible without a more comprehensive hydrological model and must be left for future research efforts. Forest ecosystems generate a number of other services in addition to watershed protection such as carbon sequestration and storage, biodiversity protection, and contributing to the aesthetic quality of landscapes. Such a combination of ecosystem services is also provided in the study area, however, this was not addressed in this study, an omission that has been subject to criticism by some authors (KOSOY & CORBERA 2010). The results obtained in this research were intended to support efforts to demonstrate the value of the service provided and to provide relevant information (in the form of OCs) rather than to determine a definitive ‘price’ for watershed services. Greater understanding of the value of watershed services can provide a basis for the design of other programmes such as REDD+ initiatives that consider the implications of reducing carbon emissions on the basis of land use and determining other co-benefits such as biodiversity conservation (KAROUSAKIS 2009). The OCs of land-use transition derived through this research can facilitate the cost effective allocation of limited financial resources for future efforts to implement watershed protection projects (KAROUSAKIS 2009; WBI 2011). Chapter 5. Conclusions and Outlook 96  5. Conclusions and Outlook This research showed how spatially explicit economic data combined with spatially explicit ecological data can provide valuable insight into the feasibility of implementing PES schemes at the watershed scale based on innovative field research in the GMW in the state of Rio de Janeiro, Brazil. Assessing the economic scope for incentive-based watershed management requires knowledge of both the costs of providing additional watershed services through land-use and land-cover change (service supply), and the WTP (demand) for such services. This study quantified these two prerequisite inputs to enable an informed decision-making process in the context of the GMW. Five primary conclusions emerged from this research. 1) The economic costs of converting current land uses to forest cover across farming systems of the GMW were reflected by estimates of the OCs of land-use transition for the improvement of water quality. The OCs ranged from 4,000–5,000 BRL (1,623–2,028 USD) per hectare for crop production areas and less than 100 BRL (40.6 USD) for pastures (Table 4.5). The actual OCs per hectare in many subwatersheds, however, are likely to be lower, specifically if targeted pasture areas are adjacent to rivers and headwaters (Figure 4.4). The estimation of OCs was based on the identification and analysis of six existing farming systems within the study area in the Batatal and Caboclo sub-watersheds of the GMW with sitespecific characteristics and economic determinants of profitability (Table 4.4 and Table 4.5). 2) In general terms the main differences between the two study sites (Batatal and Caboclo) were related to land tenure and land use. In Batatal most land was either inherited or bestowed to farmers via a provisional land title or sharecropping agreement, whereas in Caboclo most farmers lived within a local settlement that maintains a common forest reserve (Table 4.1). Farming systems in Batatal were divided into upland and lowland systems due to clear existing altitudinal differences in agricultural systems; most upland crop production was focused on banana production, whereas lowland systems included a mix of banana, cassava, green maize, and courgette. All of the farming systems at the Caboclo site were located in the lowlands, where a similar mixture of crops was produced in addition to crops such as beans, gilo, and okra. Intensive use of fertilisers was much more common in the Caboclo lowlands (547 kg ha-1) compared to the Batatal lowlands (240 kg ha-1). Chapter 5. Conclusions and Outlook 97  3) Local perceptions of land-use change over the last decade were mostly associated with the transition from pasture or cropland to forest in Batatal, while in Caboclo the prominent change was from pasture to cropland. In the Manuel Alexandre subwatershed of the GMW, which was used as a reference site, the transition of pasture to forest was the predominant change in land use. Regarding water quality and quantity changes over the last decade, some farmers perceived a decrease in stream volume whereas others perceived an increase in stream volume. Few farmers identified any improvement in water quality. The leading perceived causes for these ‘changes’ varied from the extraction of water by water-bottling companies, the removal of sand from deposits along the river, climate change, and population growth. Considering discrepancies in local perceptions and raising awareness of the environmental implications of land and water resource management and their benefits are important steps towards assuring the success of future watershed protection efforts in the study area. 4) Key environmental conditions that influenced the provision of watershed services (water quality) for the public water supply included water resource vulnerability (FERREIRA 2012), wherein both anthropogenic and environmental indicators were assessed. When only environmental state indicators were considered, subwatersheds of the GMW at higher altitudes tended to be more vulnerable to anthropogenic pressure than those in the lowlands. Sub-watersheds of the GMW with relatively high levels of existing anthropogenic pressure were those with higher densities of urban settlements and rural population nuclei (Figure 4.7). 5) The analysis of ecosystem service demand (water quality) identified turbidity as the key water quality parameter relevant for water treatment costs. The analysis of the costs of turbidity reduction to CEDAE over the 1998–2011 period revealed mean turbidity values of pre-treated water at 25.0 NTU with a minimum of 17.1 NTU and a maximum of 32.2 NTU. The mean water treatment cost of 1,000 m3 water for CEDAE varied from 19.3–29.7 BRL (7.8–12.0 USD) (Table 4.7). Based on these data the avoided costs for CEDAE were estimated at 15,510 BRL (6,415.0 USD) per each additional 1% reduction in turbidity levels at the intake point of the public water utility treatment plant (maximum WTP). The findings confirmed the main research hypothesis that the cost of converting current land uses to forest cover across the farming systems in the GMW (ecosystem service supply) is higher than what the watershed service beneficiary (CEDAE) would be willing to Chapter 5. Conclusions and Outlook 98  pay (ecosystem service demand). With regard to supply, the estimated OCs of land-use conversion ranged 4,000–5,000 BRL (1,654.4 and 2,068.0 USD) per hectare for cropland and <100 BRL (<40.6 USD) per hectare for pastures. Among the GMW sub-watersheds, those with a high proportion of pasture and a low proportion of cropland had relatively low area-weighted mean OCs and vice versa. With regard to demand, the assessment of water treatment costs found that a 1% reduction in water turbidity levels had a mean value of 15,510 BRL (6,415.0 USD) to the water utility company (CEDAE). The highest priority areas for watershed service (water quality) conservation and improvement (intervention) were found to be the land-use options (i.e. conversion of pasture or crop production areas to forest) with the highest potential for improving water quality (highest vulnerability of water resources) and the lowest OCs (Figure 4.8). Only a relatively small proportion of the watershed was cropland (5,560 ha) whereas the areas covered by pasture (52,374 ha) and forest (61,665 ha) were much larger. However the area represented by cropland is considerable and has the potential to affect the desired watershed services and, as a result, the relatively high OCs for this land-use category becomes a critical limitation to strategies for incentivising land-use change for the improvement of water quality. This study estimated the costs involved in both the supply and demand of water quality maintenance and improvement, and identified priority areas for the supply or provision of the desired watershed service in order to target watershed management measures or support the design and implementation of PES or PWS compensation schemes. Based on the research findings PWS is not considered the most cost-effective option for enhancing water quality with respect to turbidity compared to alternative watershed management options, however, some feasible options were identified. Given the spatial distribution of pastures and relatively high intensity crop agriculture in the watershed, it was found that payments for forest recuperation are likely to only be cost-effective in sub-watersheds of the GMW with relatively high proportions of pasture cover. Those sub-watersheds with high proportions of pasture that also exhibit higher levels of water resource vulnerability represent high priority areas for watershed management interventions (Figure 4.8). In these areas management efforts such as the implementation of silvopastoral practices and restricting livestock access to riparian areas are much more likely to provide considerable water quality benefits. In addition, sustainable agricultural and soil conservation practices can bring additional benefits by reducing and avoiding sedimentation and erosion, which in turn improve water quality. Chapter 5. Conclusions and Outlook 99  This study will contribute valuable data for local institutions that provide technical assistance to farmers in the study area due to the paucity of local information available to them. This was made possible due to the involvement of the most important local institutions and their support during the survey phases of this research. Furthermore, farming system analyses delivered key inputs for other scientific groups working with the German-Brazilian DINARIO project, especially those working on water quality and quantity modelling, who are currently parameterising the farming systems analysis results so that they can be incorporated into a comprehensive hydrological model of water quality and quantity that is under development for the GMW. In addition, relevant information was shared with local scientists at Embrapa soils who supported the survey efforts and who are already benefiting from the characterisation of current agricultural management practices. This is a starting point for the delivery of information to local producers on how to improve current production systems and implement more sustainable agricultural management practices that are favourable for watershed protection based on consideration of the relevant costs involved in implementing such changes. The results of this study may also prove useful for long-term watershed conservation initiatives such as Rio Rural, a project funded by the World Bank and executed by the Rio de Janeiro State Secretary for Agriculture and Livestock (GOVERNO DO BRASIL 2013). This and many other initiatives in the Atlantic Forest region reflect the considerable interest in incentive-based watershed management approaches. Furthermore, Brazilian legislation such as the Brazilian National Law on Water Resources (Law No. 9433/97), allow for the establishment of watershed service markets and permit charging polluters or consumers for water use (VEIGA 2008). Equally relevant is the more recent Law 9985/ 2000 or ‘SNUC’ that adopts the protector-receiver principle, which allows for rewards in exchange for responsible natural resource stewardship (STROBEL ET AL. 2007). Funding for incentive-based watershed management may be derived from Brazil’s ecological tax system, which compensates districts for conservation costs based on the value-added tax (ICMS) in addition to other sources (VEIGA 2008; MARQUES 2009). The study area reflects watershed management issues that are typical across many parts of the Atlantic Forest region, which is characterised by diverse agricultural mosaics and thus has highly variable OCs. Wherever intensively used cropland dominates the vulnerable zones of large watersheds, land-use planners will find it difficult to rely solely on the use of PWS schemes. Effective watershed management initiatives will then need to be combined with enhanced monitoring and enforcement activities in order to ensure compliance with the Brazilian Forest Law, particularly with regard to riparian forests. Chapter 5. Conclusions and Outlook 100  As emphasized by Germany’s federal Ministry for the Environment and the European Commission through TEEB, an economic approach to address environmental concerns can support decision makers to define the most appropriate use of limited natural resources at all levels (TEEB 2010b). This research is consistent with the TEEB approach where it is emphasised to concentrate efforts: 1) to provide information about costs and benefits; 2) create a common language for policymakers, businesses, and the public that allow natural capital to be widely accounted for decision making; 3) highlight the chances to “work with nature” (by assessing the cost-effectiveness of providing valuable ecosystem services); 4) emphasise the urgency of action by indicating where and when preventing biodiversity loss is more cost-effective than restoration or replacement; and 5) generating information about value that are relevant for the design of policy incentives including (PWS schemes) (TEEB 2010b). Moreover, this research can serve to strengthen the science-policy interface since it can deliver concrete science-based inputs for local decision makers to support more sustainable management of watershed services (IPBES 2013). Research designed with a bottom-up approach that quantifies the potential compliance costs of land users can provide a clear benefit to policy makers by facilitating the targeting of both incentives and disincentives in a cost-effective manner. Chapter 6. References 101  6. References ANA (AGÊNCIA NACIONAL DE ÁGUAS) (2011): Portal do Sistema Nacional de Informações sobre Recursos Hídricos (SNIRH). Available at: http://www.ana.gov.br/portalsnirh/ Accessed 10 Jun 2014 ANGELSEN, A., H.O. LARSEN & J. LUND (2011): Measuring livelihoods and environmental dependence: Methods for research and fieldwork. Routledge, London, UK ARRIAGADA, R.A. & C. PERRINGS (2009): Making Payments for Ecosystem Services Work: An Evaluation of the Effectiveness of PES Schemes. UNEP Working Papers on Ecosystem Services, Nairobi, Kenya ARRIAGADA, R.A., PJ. FERRARO, E.O. SILLS, S.K. PATTANAYAK & S. CORDEROSANCHO (2012): Do Payments for Environmental Services Affect Forest Cover? A Farm-Level Evaluation from Costa Rica. Land Economics 88(2):382–399 ASQUITH, N. & S. WUNDER (2008): Payments for Watershed Services. The Bellagio Conversations. Fundación Natura Bolivia, Santa Cruz de la Sierra, Bolivia BALVANERA, P., M. URIARTE, L. ALMEIDA-LENERO, A. ALTESOR, F. DECLERCK, T. GARDNER, J. HALL, A. LARA, P. LATERRA, M. PENA-CLAROS, D.M. SILVA MATOS, A.L. VOGL, L.P. ROMERO-DUQUE, L.F. ARREOLA, A.P. CAROBORRERO, F. GALLEGO, M. JAIN, C. LITTLE, X.R. DE OLIVEIRA, J.M. PARUELO, J.E. PEINADO, L. POORTER, N. ASCARRUNZ, F. CORREA, M.B. CUNHA-SANTINO, P.A. HERNANDEZ-SANCHEZ & M. VALLEJOS (2012): Ecosystem services research in Latin America: The state of the art. Ecosystem Services 2:56–70 BATCHELOR, C., J. CAIN, F. FARQUHARSON & J. ROBERTS (eds.) (1998): Improving water utilization from a catchment perspective, Vol. 4. SWIM Paper. International Water Management Institute, Colombo, Sri Lanka BEETS, W.C. (1990): Raising and sustaining productivity of smallholder farming systems in the tropics. A handbook of sustainable agricultural development. A Handbook of sustainable agricultural development. AgBe Publishing, Holland BENNETT. G., N. CARROLL & K. HAMILTON (2013): Charting New Waters. State of Watershed Payments 2012. Forest Trends, Washington, USA BÖRNER, J. & S. VOSTI (2012): Managing tropical forest ecosystem services: an overview of options. In: MURADIAN R. & L. RIVAL (eds.) Governing the provision of ecosystem services. Springer 4:21–46 BÖRNER, J., J. MBURU, P. GUTHIGA & S. WAMBUA (2009): Assessing opportunity costs of conservation: Ingredients for protected area management in the Kakamega Forest, Western Kenya. Forest Policy and Economics 11(7):459–467 Chapter 6. References 102  BÖRNER, J., A. MENDOZA & S. VOSTI (2007): Ecosystem services, agriculture, and rural poverty in the Eastern Brazilian Amazon: Interrelationships and policy prescriptions. Ecological Economics 64(2):356–373 BORJA, A., I. GALPARSORO, O. SOLAUN, I. MUXIKA, E.M. TELLO, A. URIARTE & V. VALENCIA (2006): The European Water Framework Directive and the DPSIR. A methodological approach to assess the risk of failing to achieve good ecological status. Estuarine, Coastal and Shelf Science 66(1):84–96 BRANCA, G., L. LIPPER, B. NEVES, D. LOPA & I. MWANYOKA (2009): New Tools for Old Problems: Can Payments for Watershed Services Support Sustainable Agricultural Development in Africa? ESA Working Paper No. 09–10. Agricultural Development Economics Division. The Food and Agriculture Organization of the United Nations (FAO), Rome, Italy BRUIJNZEEL, L.A. (2005): Tropical montane cloud forest: a unique hydrological case. In: BONELL, M. & L.A. BRUIJNZEEL (eds.) Forests, water, and people in the humid Tropics: past, present, and future hydrological research for integrated land and water management. Cambridge University Press CABRAL DE SOUSA, W.J. (2011): Pagamento por Servicos Ecossistemicos: Mata Ciliar, Erosao, Turbidez e Qualidade de Agua: Produtos Tecnicos 1. Projeto de Recuperacao de Matas Ciliares, São Paulo, Brazil. Available at: http://www.sigam.ambiente.sp.gov.br/sigam2/Default.aspx?idPagina=9662 Accessed 12 Jun 2010 CALDER, I.R. (2004): Forests and water: closing the gap between public and science perceptions. Water Science and Technology 49(7):39–53 CALDER, I.R. (2005): Blue revolution: Integrated land and water resources management. 2nd edition. Earthscan, London, UK CARPENTER, S.R., H.A. MOONEY, J. AGARD, D. CAPISTRANO, R.S. DEFRIESE, S. DÍAZ, T. DIETZ, A.K. DURAIAPPAH, A. OTENG-YEBOAH, H. MIGUEL PEREIRA, C. PERRINGS, W.V. REID, J. SARUKHAN, R.J. SCHOLES & A. WHYTE (2009): Science for managing ecosystem services: Beyond the Millennium Ecosystem Assessment. The Ecological Society of America 106(5):1305–1312 CEPF (CRITICAL ECOSYSTEM PARTNERSHIP FUND) (2001): Perfil do Ecossistema Mata Atlantica. Hotspot de Biodiversidade—Brazil. Conservation International, Airlington, USA CHEN, N., H. LI & L.WANG (2008): A GIS-based approach for mapping direct use value of ecosystem services at a county scale: Management implications 68(11):2768– 2776 Chapter 6. References 103  CLAASSEN, R. CATTANEO, R. & R. JOHANSSON (2008). Cost-effective design of agrienvironmental payment programs: U.S. experience in theory and practice. Ecological Economics 65:738–753 COSTANZA, R., R. D'ARGE, R. DE GROOT, S. FARBER, M. GRASSO, B. HANNON, K. LIMBURG, S. NAEEM, R.V. O'NEILL, J. PARUELO, R.G. RASKIN, P. SUTTON & M. VAN DEN BELT (1997): The value of the world's ecosystem services and natural capital. Nature 387(1):253–260 CWP (CENTER FOR WATERSHED PROTECTION) & USFS (US FORESTRY SERVICE) (2008): Watershed Forestry Research Guide. Available at: www.forestsforwatersheds.org/reduce-stormwater/ Accessed 6 Apr 2014 DA COSTA, D.J.R., J. RIBEIRO DE ALMEIDA & G. ALVEIRO LINS (2007): Série Gestão e Planejamento Ambiental. Artigo Técnico No. 07. CETEM (Centro de Tecnología Mineral). Ministerio de Ciencia e Tecnología, Rio de Janeiro. Available at: http://www.academia.edu/1810619/IMPACTOS_AMBIENTAIS_NA_BACIA_HIDR OGR%C3%81FICA_DE_GUAPI_MACACU_E_SUAS_CONSEQ%C3%9C%C3%8 ANCIAS_PARA_O_ABASTECIMENTO_DE_%C3%81GUA_NOS_ Accessed 12 Oct 2014 DAILY, G.C. & P.A. MATSON (2008): Ecosystem services: From theory to implementation. Proceedings of the National Academy of Sciences 105(28):9455– 9456 DAILY, G.C., S. ALEXANDER, P.R. EHRLICH, L. GOULDER, J. LUBCHENCO, P.A. MATSON, H.A. MOONEY, S. POSTEL, S.H. SCHNEIDER, D.TILMAN & G.M. WOODWELL (1997): Ecosystem Services: Benefits Supplied to Human Societies by Natural Ecosystems. Issues in Ecology 2. Ecological Society of America (ESA), Washington, USA DAILY, G.C., S. POLASKY, J. GOLDSTEIN, P.M. KAREIVA, H. A.MOONEY, L. PEJCHAR, T.H. RICKETTS, J. SALZMAN & R. SHALLENBERGER (2009): Ecosystem services in decision making: time to deliver. Frontiers in Ecology and the Environment 7(1):21–28 DE GROOT, R.S., R. ALKEMADE, L. BRAAT, L. HEIN & L. WILLEMEN (2010): Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecological Complexity 7(3):260–272 DE GROOT, R.S., M.A. WILSON & R.M. BOUMANS (2002): A typology for the classification, description and valuation of ecosystem functions, goods and services. Ecological economics 41(3):393–408 DEAN, W. (1997): With Broadax and Firebrand: The Destruction of the Brazilian Atlantic Coastal Forest. University of California Press Chapter 6. References 110  NAEGELI, F. (2010): Evaluation of Forest Fragmentation and Land Use Change Patterns using Remote Sensing Techniques and Field Methods. Master Thesis. Cologne University of Applied Sciences, Cologne, Germany NATIONAL RESEARCH COUNCIL (1993): Sustainable Agriculture and the Environment in the Humid Tropics. Committee on Sustainable Agriculture and the Environment in the Humid Tropics, Washington, D.C., United States of America. Available at: http://www.nap.edu/catalog.php?record_id=1985 Accessed 12 November 2014 NEHREN, U. (2008): Quartäre Landschaftsgenese und historische -degradation in der Serra dos Órgãos, Rio de Janeiro. Doctoral Thesis. University of Leipzig, Germany NEHREN, U., A. KIRCHNER, D. SATTLER, A.P. TURETTA & J. HEINRICH (2013): Impact of natural climate change and historical land use on landscape development in the Atlantic Forest of Rio de Janeiro, Brazil. Anais da Academia Brasileira de Ciencias 85(2):497–518 PABON-ZAMORA, L., A. FAUZI, A. HALIM, J. BEZAURY-CREEL, E. VEGA-LOPEZ, F. LEON, L. GIL & V. CARTAYA (2008): Protected areas and human well-being: Experiences from Indonesia, Mexico, Peru and Venezuela. CBD Technical Series No. 36, Secretariat of Convention on Biological Diversity, Montreal, Canada PAGIOLA, S., K. VON RITTER & J. BISHOP (2004): Assessing the Economic Value of Ecosystem Conservation. The World Bank Environment Department. Environment Department Paper No. 101, Washington, United States of America. Available at: https://openknowledge.worldbank.org/handle/10986/18391 Accessed 26 February 2013 PAGIOLA, S., E. RAMIREZ, J. GOBBI, C. DE HAAN, M. IBRAHIM, E. MURGUEITIO & J. P. RUÍZ (2007): Paying for the environmental services of silvopastoral practices in Nicaragua. Ecological Economics 64(2):374–385 PAGIOLA, S., W. ZHANG & A. COLOM (2010): Can Payments for Watershed Services Help Finance Biodiversity Conservation? A Spatial Analysis of Highland Guatemala. Journal of Natural Resources Policy Research 2(1):7–24 PAIVA, M., S. PENEDO, A. KUENNE, R.B. PRADO & A.E. SCHULER (2011): Qualidade da agua e exportacao de sedimentos em sub-bacias dos rios Guapi-MacacuBioma Mata AtlanticaRJ. Oral presentation at the XXXIII Congresso Brasileiro de Ciencia do Solo, Solos nos biomas brasileiros: sustentabilidade e mudancas climaticas, Uberlandia/ Minas Gerais, Brazil PATTANAYAK, S.K. (2004): Valuing watershed services: concepts and empirics from southeast Asia. Agriculture, Ecosystems & Environment 104(1):171–184 PATTANAYAK, S.K., S. WUNDER & P.J. FERRARO (2010): Show me the money: Do payments supply environmental services in developing countries? Review of Environmental Economics and Policy 4(2):254–274 Chapter 6. References 111  PEDREIRA, B., E.C.C. FIDALGO, M.B.D. ABREU, J.C.N. EPIPHANIO & L.S. GALVAO (2009): Mapeamento do uso e cobertura da terra da bacia hidrografica do rio Guapi-Macacu, RJ. Anais XIV Simposio Brasileiro de Sensoriamento Remoto, INPE: 2111–2118, Natal, Brazil PENEDO, S., A. KÜNNE, R.B. PRADO, A. SCHULER & J. ROEHRIG (2011): Implementation of a Hydro-climatic Monitoring Network in the Guapi-Macacu River Basin in Rio de Janeiro, Brazil. XIVth IWRA World Water Congress, Porto de Galinhas, Brazil PERMAN, R., M.A. YUE, J. MCGILVRAY & M. COMMON (2003): Natural Resource and Environmental Economics. 3rd Edition, Pearson Education Limited PERROT-MAÎTRE, D. (2006): The Vittel Payments for Ecosystem Services: A “Perfect” PES Case? Project Paper No. 3. IIED, London, UK PERROT-MAÎTRE, D. & P. Davis (2001): Case Studies of Markets and Innovative Financial Mechanisms for Water Services from Forests. Forest Trends Association. Available at: http://www.foresttrends.org/publication_details.php?publicationID=134 Accessed 24 June 2014 PORRAS, I., M. GRIEG-GRAN & N. NEVES (2008): All that glitters: A review of payments for watershed services in developing countries. Natural Resource Issues 11. IIED, London, UK POSTEL, B., H. BARTON & J.R. THOMPSON (2005): Watershed protection: Capturing the benefits of nature´s water supply services. Natural Resources Forum 29:98– 108 POTSCHIN, M. & R. HAINES-YOUNG (2011): Progress in Physical Geography. Introduction to the Special Issue: Ecosystem Services 35:771. Available at: http://ppg.sagepub.com/content/35/5/571.full.pdf+html Accessed 4 November 2013 POWER, A. (2010): Ecosystem services and agriculture: tradeoffs and synergies. Philosophical Transaction of the Royal Society B 365(1554):2959–2971 PRIA, D.A., A. DIEDERICHSEN & C. KLEMZ (2013): Pagamento por Servicos Ambientais. Uma estrategia para a conservacao ambiental nas regioes produtivas do Brasil? Sustentabilidade em Debate 4(1):317–340 QUINTANA, B. (2012): Native Tree Species in Silvopastoral Systems: A Bioeconomic Assessment in Cachoeiras de Macacu, RJ-Brazil. Master Thesis. Master of Science awarded by Universidad Autónoma de San Luis Potosí and Cologne University of Applied Sciences, Cologne, Germany Chapter 6. References 112  QUINTERO, M., S. WUNDER & R. ESTRADA (2009): For services rendered? Modeling hydrology and livelihoods in Andean payments for environmental services schemes. Forest Ecology and Management 258 (9):1871–1880 REGUA (Reserva Ecológica de Guapiaçú) (2011): Governance. Available at: http://www.regua.co.uk/index.html Accessed 12 September 2012 REIS, L.V.S. (2004): Cobertura Florestal e Custo do Tratamento de Águas em Bacias Hidrográficas de Abastecimento Público: Caso do Manancial do Município de Piracicaba. Doctoral Thesis, University of São Paulo, Brazil RIBEIRO, M.C., J.P. METZGER, A.C. MARTENSEN, F.J. PONZONI & M.M. HIROTA (2009): The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation. Biological conservation 142(6):1141–1153 RIORURAL (2013): Rio Rural. Programa de Desenvolvimento Rural Sustentavel em Microbacias Hidrograficas. Governo do Rio de Janeiro. Secretaria de Agricultura e Pecuaria, Rio de Janeiro, Brazil. Available at: http://www.microbacias.rj.gov.br/ Accessed 15 Jul 2012 RODRIGUEZ OSUNA, V. (2013): Smallholder Production and Climate Risk: The Lower Amazon Region, Brazil. LAP Lambert Academic Publishing RODRÍGUEZ OSUNA, V., J. BÖRNER, U. NEHREN, R.B. PRADO, H.GAESE & J. HEINRICH (2014): Priority areas for watershed service conservation in the GuapiMacacu region of Rio de Janeiro, Atlantic Forest, Brazil. Ecological Processes 3:16. Available at: http://link.springer.com/article/10.1186%2Fs13717-014-0016-7 Accessed 21 August 2014 RUSSI, D., P. TEN BRINK, A. FARMER, T. BADURA, D. COATES, J. FORSTER, R. KUMAR & N. DAVIDSON (2013): The Economics of Ecosystems and Biodiversity for Water and Wetlands. Available at: http://www.teebweb.org/publication/theeconomics-of-ecosystems-and-biodiversity-teeb-for-water-and-wetlands/ Accessed 21 August 2014 SANTOS, D.G., A.F. DOMINGUES & C.V.T. GISLER (2010): Gestão de recursos hídricos na agricultura: O Programa Produtor de Água. In: PRADO R.B., A.P. TURETTA A.G. ANDRADE (Eds) Manejo e Conservação do Solo e da Água no Contexto das Mudanças Ambientais, Rio de Janeiro, Brazil, 353–376 SMITH, D.M., D. DE GROOT & G.J.J. BERGKAMP (2006): Pay: Establishing payments for watershed services. IUCN, Gland, Switzerland SOUTHGATE, D. & S. WUNDER (2009): Paying for watershed services in Latin America: a review of current initiatives. Journal of Sustainable Forestry 28(3–5):497–524 Chapter 6. References 113  STATE SECRETARIAT OF AGRICULTURE, CATTLE FARMING, FISHERIES, AND FOOD SUPPLY (2011): Campanha de Vacinação contra a febre aftosa. Government of Rio de Janeiro. Cachoeiras de Macacu Region. Núcleo de Defesa SanitáriaEscritorio Cachoeiras Macacu, Rio de Janeiro, Brazil STANTON, T., M. ECHAVARRIA, K. HAMILTON & C. OTT (2010): State of Watershed Payments. An Emerging Marketplace. Executive Summary. Ecosystem Marketplace: A Forest Trends Initiative. Available at: http://www.foresttrends.org/publication_details.php?publicationID=2438 Accessed 31 Jan 2011 STEFFEN, W. (2009): Interdisciplinary research for managing ecosystem services. PNAS 106(5):1301–1302 STERN, N. (Ed.) (2007): The economics of climate change: the Stern review. Cambridge University press STRÖBEL, H., G. DÜRR & H. MAYR (1987): Betriebswirtschaftliche Planung von bäuerlichen Kleinbetrieben in Entwicklungsländern. 2. Materialband. Bundesministerium für wirtschaftliche Zusammenarbeit und Deutsche Gesellschaft für Technische Zusammenarbeit. Eschborn, Germany STROBEL, J.S., JR. W.C. DE SOUSA, R.S. DA MOTTA, M.R. AMEND & D.A. GONCALVES (2007): Critérios Econômicos para a Aplicação do Princípio do Protetor–Recebedor: Estudo de Caso do Parque Estadual dos Três Picos Serie Técnica. Edição 11. Conservation Strategy Fund, Brazil TEEB (THE ECONOMICS OF ECOSYSTEMS AND BIODIVERSITY) (2009): TEEB for National and International Policy Makers. Summary: Responding to the Value of Nature 2009. TEEB, Wesseling, Germany TEEB (2010a): Ecological and Economic Foundations. Earthscan, London and Washington D.C. TEEB (2010b): Mainstreaming the Economics of Nature: A Synthesis of the Approach, Conclusions and Recommendations of TEEB. The Economics of Ecosystems and Biodiversity (TEEB), Malta TEEB (2010c): TEEB for Business report. Executive Summary. The Economics of Ecosystems and Biodiversity (TEEB), Malta THOMAS, J.M. & S.J. CALLAN (2010): Economia ambiental: aplicacoes, politicas e teoria. Cencage Learning, São Paulo, Brazil TURPIE, J.K. MARAIS, C. & J.N. BLIGNAUT (2008): The Working for Water Programme: evolution of a payments for ecosystem services mechanism that addresses both poverty and ecosystem service delivery in South Africa. Ecological Economics 65: 789-799 Chapter 6. References 114  UFF (UNIVERSIDADE FEDERAL FLUMINENSE) & PETROBRAS (PETRÓLEO BRASILEIRO S.A.) (2007): Diagnóstico Ambiental das Bacias Dos Rios Macacu e Caceribu: Cobertura Vegetal, Uso E Ocupação Do Solo. Relatório Final. Instituto de Geociências, Rio de Janeiro, Brazil UGALDE, E.G. (1986): Administración de Empresas Agropecuarias. Editorial Universidad Estatal a Distancia, San José, Costa Rica UNEP (2012): The UN-Water Status Report on the Application of Integrated Approaches to Water Resources Management. UN Water Report. Nairobi, Kenya. Available at: http://www.unwater.org/publications/status-report-on-integrated-water-resourcesmanagement/en/ Accessed 12 November 2014 VAN NOORDWIJK, M. (2005): RUPES typology of environmental service worthy of reward. RUPES working paper. ICRAF-Southeast Asia, Bogor, Indonesia VEIGA, F.C. (2008): A Construção dos Mercados de Serviços Ambientais e suas Implicações para o Desenvolvimento Sustentável no Brasil. Doctoral Thesis, UFRRJ, Rio de Janeiro, Brazil VEIGA, F.C. & M. GALVADÃO (2011): Iniciativas de PSA de conservacao dos recursos hidricos na Mata Atlantica. In: GUEDES B.F. & E.S. SEEHUSEN (eds) Pagamentos por Serviços Ambientais na Mata Atlântica: liçoes aprendidas e desafios. vol Serie Biodiversidade 42, 2nd edition. Ministério do Meio Ambiente, Secretaria de Biodiversidade e Florestas. Departamento de Conservação da Biodiversidade, Brasilia, Brazil WBI (WORLD BANK INSTITUTE) (2011): Estimating the opportunity costs of REDD+. A training manual. Version 1.3, Washington DC, USA WILK, J. (2000): Local perceptions about forests and water in two tropical catchments. GeoJournal 50:339–347 WINNPENNY, J., I. HEINZ & S. KOO-OSHIMA (2010): The wealth of waste. The economics of wastewater use in agriculture. FAO Water Reports 35. Natural Resources Management and Environment Department, Rome, Italy. Available at: http://www.fao.org/docrep/012/i1629e/i1629e00.htm Accessed 11 October 2013 WRI (WORLD RESOURCES INSTITUTE) (2011): Mainstreaming Ecosystem Services Initiative (MESI). Washington D.C., USA. Available at: http://www.wri.org/project/mainstreaming-ecosystem-services Accessed 20 January 2014 WUNDER, S. (2006): Are direct payments for environmental services spelling doom for sustainable forest management in the tropics. Ecology and Society 11(2):23 WUNDER, S. (2007): The Efficiency of Payments for Environmental Services in Tropical Conservation. Conservation Biology 21(1):48–58 Chapter 6. References 115  WUNDER, S. (2008): Necessary Conditions for Ecosystem Service Payments. Conference Paper. Economics and Conservation in the Tropics: A Strategic Dialogue, San Francisco, January 31–February 1, 2008. Available at: http://www.rff.org/Documents/08_Tropics_Conference/Tropics_Conference_Paper s/Tropics_Conference_Wunder_PES_markets.pdf Accessed 21 October 2014 WUNDER, S. (2013): When payments for environmental services will work for conservation. Conservation Letters 6(4):230-237 WUNDER, S. & M. ALBAN (2008): Decentralized payments for environmental services: The cases of Pimampiro and PROFAFOR in Ecuador. Ecological Economics 65(4):685-698 ZIMMER, Y., C. DEBLITZ & K. SEIFERT (2009): Die globale Landwirtschaft besser verstehen: Agrarokonomen betreiben internationales Netzwerk agri benchmark. Forschung Report. Agribenchmark, Braunschweig, Germany       APPENDICES Appendix I. Summary of PWS programmes in the Atlantic Forest, Brazil  Appendix I. Summary of PWS programmes in the Atlantic Forest, Brazil An overview of current PWS schemes in the Atlantic Forest is presented in the following tables according to the valuation method used, the value paid, relevant detailed case studies, limitations, and recommendations for PWS schemes in the Atlantic Forest region. These efforts include the programme of payments for restoring riparian areas in São Paulo, the Water Conserver project, the Water Producer project, the Producers of Water and Forests project, and the Economic Valuation for water prices in Três Picos State Park. Appendix I. Summary of PWS programmes in the Atlantic Forest, Brazil 1  Overview of current PWS programmes in the Atlantic Forest of Brazil Project Name FU1 Valuation Type PWS Value In Execution: Departamento de Meio Ambiente de Extrema - Conservador de Águas MG Total valuation comprises the local OCs and the total property size 176 BRL ha-1 yr-1 Value based in a UFEX (Unidade Fiscal do Município de Extrema), in March 2010. Ana/TNC - Produtor de Água, Bacia PCJ SP According to model of the Programme “Water Producer of ANA”: BRL 25-125 ha-1 yr-1 PAE = 100 (1Z1 / Zo) Three payment categories: Where:  conservation practices (25-75 BRL ha-1 yr-1); PAE varies depending on soil management  restoration of riparian forest with two classes (83-125 BRL ha-1 yr-1); Z = are reference values for erosion abatement  conservation of riparian forests (42-125 BRL ha-1 yr-1) according the level of engagement of the producer and the successional stage of the forest Instituto Terra - Produtores de Água e Florestas – Bacia Guandu RJ Variables considered for the payment calculation: 10-60 BRL ha-1 yr-1 1) Areas to be restored (APPs and water interception areas in two states) 2) Conservation areas (areas surrounding the conservation units; succession stage of vegetation; level of engagement of producers in the restoration and fitting within the priority areas for the water service Local OC served as basis for the determination of the above mentioned weighed factors. Appendix I. Summary of PWS programmes in the Atlantic Forest, Brazil 2  Project Name FU1 Valuation Type PWS Value In Execution: Instituto BioAtlântica/ IEMA - ProdutorES de Água – Bacia Benevente ES Formula comprises the criteria: slope, regeneration stage of the forest and OCs: 80-340 BRL ha-1 yr-1 (max. values established in 510 VRTEs2 (Article 3° of Law 8.995/2009) VSrh = 200 x VRTE x (1-Z) x Kt where: VSrh = environmental service value of the conservation and improvement of water quality and availability in BRL ha-1 yr-1; VRTE = unit for the reference values of the State Treasury Z = coefficient of erosive potential referred to the stage of development of the forest defined by: the stage of initial regeneration, secondary initial, primary, secondary media advanced; Kt = coefficient of topographic adjustment defined by the slope ranges. IEMA - ProdutorEs de Água – Bacia Guandu ES Formula comprises the criteria: slope, regeneration stage of the forest and OCs 80-340 BRL ha-1 yr-1 Fundação Boticário de Proteção à Natureza - Oásis SP Valuation for the reposition cost considering the conservation of the areas and the index for the springs valuation (IVM: Índice de Valoração de Mananciais). 75-370 BRL ha-1 of conserved natural area Payment with greater value for the preserved areas. Three criteria are used:  Water production and storage (99 BRL ha-1yr-1);  Erosion control (75 BRL ha-1yr-1) and  Water quality maintenance (196 BRL ha-1yr-1)–max. value 370 BRL ha-1yr-1)  9  Produtores de Água e Florestas / Producers of water and forests Category: Execution Intervention type Forest restoration in APPs and areas of water interception; forest conservation & rural sanitation Location Scope Investment costs (BRL) Source of funding Legal Base Time frame Beneficiaries Monitoring Municipality of Rio Claro, State of Rio de Janeiro Watershed: of the Guandu River within the biodiversity corridor of Tinguá-Bocaina Micro-watersheds: Das Pedras River within the district of Lídice: 5,227 ha including main springs of Piraí river (responsible for until 15% of water resources available in the Guandu system). Area to be conserved: 3,342 ha Area to be restored: 3,677 ha Total area: 3,677 ha Forest restoration; forest conservation and rural sanitation: 1.9 million BRL yr-1: 1 million yearly as initial costs + 648,908.40 maintenance costs Coordination = 107.19 BRL ha-1 Restoration= 13,820 BRL ha-1 Conservation = 108.70 BRL ha-1 Water use charge – CBH Guandu Federal Law: 9.433/97; State Law 3.239/99, legal base for the water use charge within the RJ State; Law 5.234/08 that modifies Article 27 of the Law 3.239/99. PL of State based PES in discussion 2009–end is not yet known (min. of 5 years) Around 8 million people in the metropolitan area of Rio de Janeiro Nine water quality parameters in addition to precipitation, flow and discharge, icthofauna and avifauna indicators Detailed description of project V aluation of the service Context: This project is carried on within the framework of state and federal policies for the charge for water use and the incentives for the conservation of watersheds. Institutional arrangement: Contracts between rural producers and FAPUR (Foundation that is responsible for the execution of contracts of the CBH Guandu. Annual contracts that can be renewed for a minimal period of five years. Payments are done every six months. Participants: 18 small and medium size rural producers (mean income: 72% until two minimum wages; mean property size: 79% smaller than 100 ha. Partners:  SEA/INEA–Inputs for restoration  CBH Guandu: Payments to the producers providing the environmental services  Instituto Terra: institutional articulation, coordination and execution of field actions  TNC: institutional, technical and scientific support; financial partial support to the restoration/conservation actions  City Hall of Rio Claro: provision of local headquarters and local support  UGP: general project and decision making coordination (all partners). 10-60 BRL ha-1 yr-1 Variables considered for the payment calculation: 1) Areas to be restored (APPs and water interception areas in two states: well preserved and medium preserved 2) Conservation areas (areas surrounding the conservation units; succession stage of vegetation; level of engagement of producers in the restoration and fitting within the priority areas for the water service Local opportunity cost served as basis for the determination of such values, weighed by the factors mentioned above.  10  Economic valuation for pricing waterTrês Picos State Park (project concept) The Conservation Strategy Fund (CSF) developed a concept for establishing the price of water coming from Três Picos State Park. Besides protecting various endemic species threatened in this biome, this park in the mountain range of the Rio de Janeiro State is responsible for the protection of springs that supply more than 1.7 million inhabitants close to the city of Rio de Janeiro (STROBEL ET AL. 2007). The study has five parts: 1) estimating the cost of guaranteeing the hydrological protection afforded by the park; 2) estimating the park’s contribution to water used by the main consumers; 3) definition of economic criteria relevant to the allocation of protection costs among consumers; 4) proposal of three alternative pricing scenarios and 5) description of an institutional arrangement to govern the payment system (STROBEL ET AL. 2007). In order to estimate the protection costs of the park, only those associated to safeguard water resources were taken into account such as sorting out land tenure for disputed parts of the park, paying guard salaries, specific training, equipment, fuel, administrative costs and certain physical infrastructure. The total cost was estimated at around BRL 635,680 (250,084 USD) (STROBEL ET AL. 2007). The park´s water contribution to each user was calculated using GIS with data on topography, rivers, land use, park boundaries, water intake points and rainfall. The system estimates both surface and aquifer supplies originating in the park. To establish how much each user should pay, it was thought that the sum of all payments should equal the cost of protecting the park. However, the economical principle that the demand for a good goes down if the price goes up was taken into account. Therefore, price was set at a level to cover protection costs, presuming slightly diminished consumption (STROBEL ET AL. 2007). For managing the payment system, it was proposed that a committee composed of the park director, one representative of each consumer group and one representative from each municipality sharing the park’s territory be formed.  11  Limitations and recommendations of PWS projects in the Atlantic Forest (based on: VEIGA & GALVADÃO 2011) Limitations Recommendations Economic:  Uncertainty concerning future continuous resources for the maintenance of PES projects (from the perspective of the project executors and rural producers).  High cost of associated activities, especially forest restoration and their appropriate technical support.  High transaction costs due to the complexity in the project elaboration (mapping, socioenvironmental diagnostic studies); shared management of projects; and the elaboration, negotiation and monitoring of individual contracts with each producer.  Difficulties in the identification of total costs of the projects because of the shared management and involvement of partner institutions.  Implementation case by case (lack of standardization).  Lack of private institutions specialized in the implementation of PES Projects. Technical:  Low technical capacity in the forest restoration process (collection of seeds, producing seedlings, maintenance of plantings).  Missing or inappropriate monitoring processes for the implementation phase related to water, as well as the relation with the conservation practices and forest restoration efforts implemented. Institutional & legal:  In some cases, there is no legal framework providing security to involved stakeholders.  Non-definition of fiscal rules that can be applied to PES  Difficulty in the execution of public funding due to lack of a legal framework or bureaucratic Demand generation:  Required, especially from water use charges within the framework of CBH, governmental funds through specific legislation, and governmental programmes as well as private voluntary funds.  Demand remains small on the part of final service users, probably due to the lack of knowledge about the relationships between forest and water among the general public, including some significant industrial users. Training  Sharing results from projects under implementation and lessons learned need to be shared in order to facilitate replication.  It is advisable to promote partnerships between municipalities in different process stages for the exchange of experiences.  To create and strengthen stakeholder networks. Technical assistance in the project elaboration and execution Project development training, especially where potential funding exists. CBH and public programmes might support local NGOs and municipalities so that they can access different available funding sources. Communication  Sharing project results and the importance of hydrological resources in urban and rural areas.  PES can also facilitate regulating rural properties; territorial mapping; application of the Forest Code; and hydrological, soil, and forest resource conservation. In some cases PES can be accompanied by the implementation of sanitation measures, such as in areas close to springs. Tools for supporting decisions  Can support the process of defining priority areas for the project implementation at the watershed and macro levels.  12  procedures in the contract management.  Producers sometimes ignore their environmental responsibilities (which would increase number of adherence to projects).  It is important to also use existing tools that can be applied at greater scales. Supporting studies  Need for research studies concerning the relationships between forests and water, especially among universities.  Review of studies that correlate the economic impacts of erosion on drinking water treatment costs in different watersheds in SP; studies to determine priority areas for the implementation of future PES projects based on crossing spring areas with priority areas for biodiversity conservation in the Atlantic forest and the Cerrado, and the study of relationship between the project water producers and adaptation measures to climate change.  Efforts to identify lower cost areas for assisted forestry regeneration with in selected watersheds by searching for reducing the cost of forest restoration and increasing the turnover of the environmental services under consideration. Monitoring  It is important to monitor the performance of buyers and sellers of the environmental services in relation to the given payments. It is also essential to monitor the quantity and quality of water over the long term because these are the basis of the product that will be sold.  There is a need for improving monitoring processes so that these can increase the credibility of projects and enhance the strategic importance of PES as a conservation tool. Public policies and PES legislation Specific legislation and related programmes can guarantee legal mechanisms of funding transfer to rural producers and show that the importance of environmental services is valued by society. Appendix II. Initial semi-structured survey 13  Appendix II. Initial semi-structured survey This survey was divided into five parts. The first component was intended for general rural unit characterisation. The second component deals with production aspects and management practices. The third component was for the assessment of land-use changes. The fourth component comprises income sources and their prioritisation. The fifth component addressed perceptions regarding water quality. Data collected in the first survey component include: 1. Name of the interviewee 2. Relation to the rural property (considering three sub-categories: a) the interviewee lives in the farm unit, b) the interviewee lives and works in the farm unit and c) the interviewee only works in the farm unit. 3. Condition of land tenure and land use. This was assessed by considering the following subcategories: share-cropper meeiro, rents land arrendatário, partner parceiro, day worker diarista, and landowner proprietário. Such categories are included in the current National Programme for the Strengthening of Family Farming (PRONAF), the National Programme of Agrarian Credit, and in the Law No. 11718 of 2008 establishing the agrarian contracts and their associated rules, which are compulsory in all Brazilian territories. Partnership contracts may be written or verbal. The category of partner refers to those having a partnership contract with the landlord, carrying out agricultural activities, and sharing profits according to agreed interests. Share-croppers are those having a contract with the landlord, carrying out agricultural activities, and dividing the obtained yields and expenditures. 4. The number of people living on the rural property, differentiating among those with more than 60 years old, those between 15 and 60 years, and those less than 15 years old. Additionally, respondents were asked about the number of household members working in the agricultural sector and the number that depend on an income source outside the farm unit. Appendix II. Initial semi-structured survey 14  The general information from the first component enabled to better understand the respondent’s relationship with the rural property, land tenure conditions, available labour force, and respondent’s dependence on the agricultural sector. The second component includes the number of farm units that the person is associated with in terms of their size, altitudinal level (hilltop, slope, or lowland) and the most important uses (agricultural land, pasture, and forest). The objective of this part was to assess the property condition regarding particular specifications of the Brazilian Forest Code (Federal Law 4771/1965), such as the established percentage of rural properties to be maintained as a permanent forest reserve (Reserva Legal1). The Forest Code also prohibits the clearing of primary vegetation on steep slopes (>45°) and along the margins of rivers and streams, all of which are classified as APPs (Permanent Protection Areas)2. For agricultural areas the survey focused on the main crops, the use of lime for improving soil pH, and the use of fertilisers. The name of the product and the approximate yearly total amount used were requested from respondents reporting fertiliser use. Data was also collected on the use of other agrochemicals for pest control and the distance between the agricultural activity and the nearest water source. Regarding other management practices, information on the types of soil preparation and other activities (including the use of burning, agroforestry, willingness to produce other crops, and fertility analyses) were also requested. Considering the fact that this study focused on a forest ecosystem service, respondents were asked whether there was an estimate on the age of local forests and the reasons to have forest on their property. Information on whether erosion was observed in the production unit and livestock herd sizes was also collected. 1A legal reserve is “an area located in the interior of a private property or land claim, except in areas of permanent preservation (APP), necessary for the sustainable use of natural resources, the conservation and restoration of ecological processes, the conservation of biodiversity and the sheltering and protection of native flora and fauna” (Article 1, III: Brazilian Forest Code)”. 2 An area of permanent protection (APP) is a “protected area covered or not with native vegetation, with the environmental functions of preserving water resources, landscapes, geological stability, biodiversity, and genetic fluxes of flora and fauna, as well as protection of the soil and securing the wellbeing of human populations” (Article 1, II: Brazilian Forest Code). Appendix II. Initial semi-structured survey 15  The third component included questions about whether land use had changed over recent decades and, if so, what kind of changes occurred (amongst the categories: forest, agriculture and pasture). Additionally, in order to assess the willingness of the respondent to clear forest, a hypothetical question was asked about which areas could be converted to agricultural production and the justification for doing so. The fourth component included a prioritisation of income sources from agriculture, cattle ranching (differentiating dairy and beef production), fish farming, remittances or governmental aid, and work outside the farm unit. The fifth component included questions about water related variables, such as: the source of water for domestic use, the perceived water quality, whether water is available year round, and, if not, in which months it is not typically sufficient. Finally, respondents were asked if they perceived changes in the quantity/quality of water for the domestic use and if so, what changes; when these changes occurred; what were the drivers of these changes, and at last, whether the riparian forests adjacent to water sources (headwaters) according the Brazilian Forest Code were taken into account. Appendix II. Initial semi-structured survey 16  Appendix II. Initial semi-structured survey 17  Appendix II. Initial semi-structured survey 18 