Ecological factors governing existing and future patterns of biodiversit
An overarching requirement in biodiversity conservation research is understanding the relationship between ecosystem function (e.g., plant succession, trophic structure, hydrologic and nutrient cycles) and patterns of species richness and abundance (Schultze and Mooney, 1994). The importance of diversity to ecosystem function, and vice versa, led one recent reviewer to call 2002 the "year of the diversity - ecosystem function debate" (Cameron, 2002) . In NW Yunnan, with its heavily humanized landscapes, one also needs to understand how habitat area and fragmentation affect population and ecosystem processes, as well as how to predict the impact of future land use changes on ecosystem health.
Issue 1. Land use systems and sustaining native plant and animal species . A first step towards designing strategies for biodiversity preservation is to survey existing patterns of biodiversity and correlate these patterns with potential factors that impact biodiversity. To identify the effect of human development on biodiversity, surveys must distinguish human impacts from coincidental and confounding factors (Underwood, 1994) , and this requires comparisons among sites that differ in human impacts but are similar in other ways. The highly dissected terrain of Yunnan creates numerous replicate valley systems, and the heterogeneous advance of infrastructure (particularly roads) creates an ideal situation for comparative studies (Mensing et al ., 1998) . Fruitful specific avenues of research would be to investigate patterns of plant and animal diversity in valleys with contrasting adoption of new agricultural technologies, forestry practices, sewage treatment, industrial development, or other features of infrastructure. With our Chinese counterparts and international collaborators, we are selecting flagship organisms that can be monitored for lake and wetlands health, and wild lands conservation. The longitudinal and comparative surveys will be designed and performed by faculty trainers with extensive relevant experience in collaboration with Chinese colleagues and international collaborators Missouri Botanical Garden, International Crane Foundation, The Nature Conservancy).
Issue 2. Identifying the causes of loss of biodiversity. Whereas surveys provide broad hypotheses about the consequences of land-use changes on collections of species, numerous problems can only be addressed with detailed investigations of single species or groups of related species. Students will have a wide range of opportunities for focal species studies, and the faculty trainers (both from the UW-Madison and from China) have collectively a wide range of expertise that can guide students. For example, IGERT faculty, Drs. Karasov and Reed, already have a student (of Dr. Ji's) studying how land use changes effect the nutritional ecology of Yunnan snub-nosed monkeys, a species of sufficient concern that it is the center of a bioreserve program in Yunnan. Another faculty, Dr. Porter, has developed the mathematical tools to model animal landscapes (Porter et al., 2002). These microclimate-and-animal models, based on first-principles using GIS based topography, climate, and vegetation data, and simple animal properties, can identify where species habitat exists, and predict the impact of changes in land cover or climate on population densities.
Dr. Zhu, based on the knowledge of habitat requirements of some key endemic species is combining this information with GIS/Remote Sensing techniques under fuzzy logic to map the current spatial extents of these habitats and identify habitat "hotspots" that need to be maintained. The analysis will also indicate impending gaps that must be filled to meet the requirements of these endemic species as well as areas currently protected but that are of low efficacy in biodiversity and thus could be used for economic development (Zhu et al, 2001; Scott et al., 1993). Other areas of expertise represented by IGERT trainers include toxicology, metapopulation dynamics, biological control, forest management, and wetlands restoration.
Issue 3. Land-use and water resources. A critically important ecosystem function in NW Yunnan is the maintenance of water resources. Managing water resources involves clean water for drinking, agriculture and recreation (Wilson & Carpenter, 1999) ; water for hydroelectric power generation, irrigation and navigation; and flood prevention. Locally our focus will be on the ecohydrology (Zalewski, 1997) of lakes and wetlands, as well as both nonpoint sources (e.g. crop and livestock production) and point sources (e.g., tourist facilities, highway construction) of pollution. Our research scope will be at the watershed level (Cedfeldt et al . , 2000) , building on the experience of our trainers in wetland, agricultural, watershed and basin systems. Nationally, the impacts of changing land-use patterns in Yunnan have been blamed in the flooding of the Yangtze River (Abramovitz, 2001) and internationally, dams along the Upper Mekong are affecting fish catches throughout SE Asia.
Livelihood strategies and human population dynamics that drive resource exploitation
The evolution of biodiversity and ecological systems in regions such as NW Yunnan hinge on the natural resource use decisions that humans make. The task of the livelihoods researchers is not only to describe the patterns of natural resource use among the local people but also to understand the factors that shape them (Ives and Messerli, 1989). Generally, in diverse and remote ecological settings, the livelihood strategies (activities and investments) of rural people tend to be highly heterogeneous within and across villages (Coomes, Barham, and Takasaki, 2003). These livelihood strategies depend on a number of factors including natural resource endowment, household wealth, family labor resources, social capital, community institutions, and social and cultural norms (Shouying, Carter and Yao, 1998;). While most economic activities leave a mark of some kind on the landscape, the severity of these impacts can vary dramatically in terms of their implications for key conservation objectives. Thus, the challenge in areas of conservation interest, where exclusion is not viable, is to analyze:
(i) the underlying logic of livelihood strategies especially those that make heavy use of land and natural resources;
(ii) the link between these activities and investment choices, and key human and environmental outcomes; and,
(iii) the types of technologies, policies, programs, and incentives that might assist in the adoption of improved livelihood and resource management strategies, ones that address both conservation and development objectives.
Work on local livelihoods will provide two types of benchmarks for the IGERT team, one related to the lessons about what types of practices and approaches have proven to be sustainable in diverse ecological settings of the region and the other related to the underlying biology of these systems. This second issue is a crucial one given the tremendous ecological diversity present in this mountainous region and the concomitant disparate farming systems.
Issue 2. Sustainable development strategies and policy interventions . Household livelihood strategies are inextricably linked to wealth, natural resource endowment, family labor resources, social capital, and social and cultural norms. The tight linkages among these forces result in heterogeneous behavior both in terms of how households exploit their economic and natural environments at any given point in time, and how they accumulate assets to exploit them in the future (Zimmerman and Carter, 2003). Understanding such heterogeneous behavior requires a theoretically informed vision of the constraints that underlie human decisions, and empirical methods that are in turn consistent with the theoretical vision (Carter and Yao, 2002). Matching this analytical challenge is the policy challenge of designing incentives and mechanisms that will enhance the preservation of biodiversity across a range of households that are pursuing strategies that degrade the resource base to a greater or lesser extent in response to a variety of constraints.
Issue 3. The effects of the regional economy on local livelihood strategies. Major changes are beginning to take place in NW Yunnan with the establishment of more bioreserves, construction of hydroelectric dams, and the promotion of tourism. As well as affecting local biodiversity, all the changes can fundamentally transform livelihood strategies of the local people, as the opportunities and constraints on traditional activities change and the incentives to pursue new ones increase (Buck et al, 2001). The regional economic changes could also lead to fundamental shifts in population, significant changes in land tenure relations and local management regimes, as government policies supplant traditional arrangements (Meinzen-Dick, et al., 2002). Among the key issues that the UW-CAS team want to investigate include: 1) the key influences on demographic changes that affect the way lands and resources are used; 2) the role changing land tenure relations play in the land and natural resource management strategies of households; and 3) how might current economic activities be made more environmentally sustainable given the increased attention to and investment in eco-tourism?
Governance structures and policies that impact biodiversity and economic development
The management decisions of diverse actors are shaped by the complex, dynamic, and often place-specific interactions of regulatory constraints and imperatives (e.g., law, policy etc.), and indigenous cultural orientations and economic strategies. In other words, human use and action on the landscape is mediated through formal and informal institutions (Carney and Farrington, 1998; Healey 1997). Understanding the institutional dimensions of landscape change in a region also characterized by cultural diversity is therefore a crucial area of research for (i) understanding the 'drivers' of land use change, (ii) understanding the efficacy of various policy measures, and (iii) identifying effective ways of halting or altering the trajectory of environmental change in a given locale by fashioning strategies for improved equity, political stability, environmental management and restoration (Wood, Stedman-Edwards, and Mang, 2000).
NW Yunnan provides a superb laboratory for pursuing research questions in environmental governance. While these questions have universal, academic salience, they also are compelling in terms of the trajectory of change occurring in Yunnan.
Issue 1. Decentralization, democratization, and natural resource governance : China, like many biodiversity-rich nations, is a transitional political and economic system. Processes of decentralization of governance, village-level democratization, and changing land tenure rules (among others) are having important consequences for the way in which natural resources of NW Yunnan are used and managed. While the decentralization of natural resource governance is said to simultaneously promote efficient management and democratize policy and resource use, (Buck et al.2001) some recent empirical work highlights the dysfunctional dimensions of localized control and the potential for enhancing inequality and tension among diverse local actors (Leach et al . 1999, Ribot, 2002). Our IGERT students, under the leadership of Friedman, Jiang, Lane and Streiffer, will have the opportunity to study how decentralization is influencing the implementation of environmental policy in NW Yunnan and whether the goals of efficiency and equity in environmental management are being achieved as a result of decentralization.
Issue 2. Reconciling national interests with diverse local interests and imperatives : The utilization of the landscapes and resources of NW Yunnan for broad national purposes and interests potentially conflicts with the imperatives of diverse local groups. This reflects a universal problem in environmental governance: fashioning environmentally just outcomes (Harvey, 1996). These issues are accentuated by processes of globalization, which might simultaneously serve to intensify tribalism (or difference), particularly in terms of competition for access to productive natural resources, and weaken the mediating effect of the central state (Diamond 1999). Given the differing value and importance of NW Yunnan for diverse local actors and broader, national interests, a crucial question is how national and local interests are reconciled. Subsidiary questions include: How do local people interpret and respond to formal institutional and policy prescriptions? How equitable is environmental policy and management? What are the main sites of contestation and interaction?
Issue 3. Institutional 'bricolage': Governance arrangements for a region such as NW Yunnan are so complex that they appear chaotic. The principal actors include: (i) the agencies of central, provincial, and local government, (ii) diverse local peoples and informal institutions, (iii) transnational actors including international NGO's and private corporations, and (iv) global actors and regulators (e.g. UNDP etc.). These multiple institutional arrangements in environmental management are characterized by intersecting points and a lack of distinction between informal and formal or local and state. We need therefore to understand this institutional bricolage, and understand the complex ways in which practices and priorities unfold in a given locality, and identify the points or moments for effective intervention and reform (Mehta et al . 1999). Discerning the patterns of this complexity is a crucial precondition for considered improvements in governance.
Issue 4. Governance and valuing diverse landscapes and resources. Environments and natural resources are not merely physical, material, or tangible products. They are also viewed in symbolic terms, having historically and culturally constructed meanings and values. The ways in which different actors - at multiple scales -- understand and assign value to different environments and resources is a crucial dimension of policy and management (Berkes, 1999). There are two principal issues here. First, failure to account for indigenous knowledge and values in landscape has been directly linked to failures in environmental plans and policys (Scott 1998). Second, failure to recognize local meanings and values is linked to unjust environmental practices and policies (Harvey 1996). IGERT Fellows, will seek to understand these diverse meanings and values and will investigate whether they are recognized in policy.