Richard Bardgett Department of Biological Sciences Lancaster University Lancaster LA1 4YQ UK Homepage "Integrating aboveground-belowground relationships into land-use systems" Throughout history it has been known that soil microbes are integral to soil fertility. However, it is only during the last few decades that ecologists have begun to explore belowground communities and their functional significance for plant communities and ecosystem processes. This interest is due, in part, to technological advances that have enabled scientists to extract and characterize soil microbial communities and to assess their function. It is also due to an increasing recognition by ecologists, who have traditionally focused on aboveground organisms, of the importance of belowground organisms as structuring forces in terrestrial ecosystems. The aim of this talk is to illustrate the often unexpected significance of linkages between plants and soil microbial communities in driving plant community and ecosystem services in natural and managed systems, and to illustrate how some of these concepts might be integrated into sustainable land use systems to maximize the efficiency of soil nutrient cycling and nutrient retention, including carbon storage. Specifically, I will refer to some of our recent studies which illustrate the importance of plant-soil-microbial linkages for: (a) nitrogen cycling and the partitioning of chemical forms of nitrogen by co-existing plant species in agricultural grasslands systems; (b) phosphorus cycling, and the maintenance of plant productivity in forested systems; and (b) soil carbon cycling, and especially the goal of sequestering carbon in soil in agricultural systems, an issue that is high on the political agenda.

Kamaljit S. Bawa 1. Department of Biology, University of Massachusetts-Boston 100 Morrissey Blvd., Boston, MA 02125, USA 2. Ashoka Trust for Research in Ecology and the Environment (ATREE) Bangalore, India Homepage "Institutions for Conserving Biodiversity in an Era of Global Change" Humanity's quest for sustainability will be the dominant theme of the twenty first century. The major challenges to this quest are the inter-related problems of environmental degradation and poverty, exacerbated now with climate change. Progress on technological, social and economic fronts would be necessary to meet these challenges. However, existing institutions in much of the developing world do not have the capacity to respond to complex environmental and social challenges. New institutions are required to cope with poverty, environment, and sustainability. Such institutions should be problem driven with explicit goals of addressing environmental degradation, poverty alleviation and social change. Moreover these institutions should transcend traditional disciplinary boundaries to generate new ideas and technologies, and link new findings with changes in policy and governance. I will provide some examples of such institutions and their innovative sustainability programs and discuss ways to make further progress in meeting environmental, social, and economic challenges in the tropics. References: Bawa, K. S., G. Balachander, and P. Raven. 2008. A case for new institutions. Science (editorial). 319: 136. Keim, B. 2008. From the Ground Up: ATREE crossed sectors to breed a new species of conservation agency. Stanford Social Innovation Review 6: 69-70

Erwin Beck Institute of Plant Physiology Gebäude NW1 Universitätsstr. 30 95447 Bayreuth Germany Homepage "Functional aspects of tropical biodiversity, revealed by gradient analysis" Four types of ecosystem services have been differentiated in the Millenium Ecosystem assessment: 1) Cultural services, 2) Provisioning services, 3) Regulating services, and 4) Preserving services. The services 1) - 3) are mainly services of the ecosystems defined towards human needs and their monetary value can at least partly be computed. In contrast the preserving services are the basic ecosystem-bound outputs of ecosystems which brought them into existence and warrant their persistence. These services are also known as functionality in an ecosystem. Functional traits of an ecosystem are difficult to recognize, analyze, quantify and even more difficult to convert into monetary units. Assessment of these services requires an understanding of the particular ecosystem, which best could be gained by ecological experiments. However, in highly biodiverse tropical ecosystems, experiments are difficult to carry out, for logistic as well as for legal reasons. In this situation analysis of parameters along distinct environmental gradients can be used as a likewise powerful approach to gain insight into structure, dynamics and functionality of an ecosystem. In a study of a tropical mountain rainforest in South Ecuador, gradient analysis has been used as well as a comparison of the natural ecosystem with its anthropogenic replacement system "pastureland", respectively abandoned pastureland. Both ecosystems are located in the perhumid eastern Cordillera of the Andes, where they border each other at the bottom of the San Francisco valley, suggesting that their initial environmental conditions had been the same. The area is an extraordinary hotspot of biodiversity of vascular and non-vascular plants and of several animal groups investigated so far. Altitudinal gradients provide ideal settings to study effects of various ecological factors across small spatial scales. Collectively, the non-linear changes of abiotic factors with elevation are mirrored by changes in the vegetation, e.g. species composition, plant life-forms, root-to-shoot ratio of biomass and physiological parameters. Also, distribution patterns of most animal species reveal altitudinal preferences and limits, which in many cases can be related to their thermal physiology. Food-webs also change with altitude, if the elevational ranges of the involved species are overlapping, but discordant. Resulting from the outstanding biodiversity (resulting in a very scattered occurrence of individual plant species) on the one hand, and from seasonal changes of abiotic factors as such (rainy and dry season) and in particular with reference to altitude, biotic interactions are usually of weak nature. This holds for above- and belowground plant-animal interactions as well as for plant-fungal interactions (mycorrhiza). The fact that strong biotic interactions appear to be almost exceptional, conforms to the lack of keystone plant and animal species which by making better use of resources would suppress competitors. Stability of the extraordinary diverse ecosystem "natural mountain rainforest" appears to result from a multitude of weak biotic interactions, favoured by resource limitation and a high dynamics at least of the vegetation cover. At the level of correlations an influence of the plant diversity on the velocity of the nutrient cycle in the forest can be shown. On the opposite side of the valley the natural forest has been replaced by pastures. As a consequence of a non-sustainable destructive land-use, wide areas have already been abandoned as the pasture grass has been overgrown by the aggressive tropical bracken (Pteridium arachnoideum). On the longer run, a few species of violently propagating bushes share the area with the bracken. Wide areas of the Ecuadorian and Peruvian Andes are covered by this type of a successional vegetation, indicating an apparently stable ecosystem, which by contrast to the forest is dominated by a few keystone plant species characterized by an enormous potential of growth and asexual and/or sexual propagation. Ongoing balanced competition apparently keeps this species-poor system stable. Even from the non-flowering bracken, a multitude of invertebrates have been collected, but none of them appears to produce populations which can endanger the fern. Patchwise infection of bracken by a pathogenic ascomycete fungus is conspicuous, but like the fauna, is not detrimental to the fern. Again, biotic interactions appear to be weak except for competition. Irrespective of its apparent stability, resilience of this species-poor ecosystem against serious disturbances is much smaller than of the species-rich forest, because of the occurrence of keystone species. Disappearance of one of them would change the entire structure of the ecosystem. With regard to the provisioning and regulating ecosystem services of the area, it may be of interest that a hydroelectric power plant has been constructed on the forested side of the valley where the vegetation secures a permanent constant flow of water. Literature Gradients in a Tropical Mountain Ecosystem of Ecuador (2008) Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R, eds. Ecological Studies Vol 198, 525 pp, Springer-Verlag, Berlin, Heidelberg (ISBN 978-3-540-73525-0) Beck E, Richter M (2008) Ecological aspects of a biodiversity hotspot in the Andes of southern Ecuador. in The Tropical Mountain Forest - Patterns and Processes in a Biodiversity Hotspot (Gradstein SR, Homeier J, Gansert D, eds.) Biodiversity and Ecology Series 2: p 195-217, Göttingen Centre for Biodiversity and Ecology, Univ. Verlag Göttingen

Strategic Options for Forest Assistance in Indonesia" Timothy H. Brown Sr. Natural Resource Management Specialist World Bank Office Jakarta Jakarta Stock Exchange Building, Tower 2 12th Floor Jl. Jenderal Sudirman, Kav. 52-53 Jakarta 12190, Indonesia Homepage "Sustaining Economic Growth, Rural Livelihoods, and Environmental Benefits: Strategic Options for Forest Assistance in Indonesia" Increasingly, global changes affect the prospects for Indonesia's forests. Man-made policies and incentives have a greater influence than biophysical processes in the medium term. Global pressures include rising demand for raw materials, shifts in trade and competitiveness, rising commodity prices, rule of law and its effect on investment climate, and now, the potential for forest carbon markets. Indonesia's forest resources are not contributing as they could to poverty reduction, economic and social development, and environmental sustainability. Over-exploitation, inefficiency and weak governance in the forest sector have contributed to illegal logging and trade, forest degradation and loss, underperforming plantation lands, losses in government tax earnings, and indebted firms. Industrial output, employment, and competitiveness are declining. Official statistics do not measure illegal earnings from unsustainable practices, nor the high costs of environmental degradation. Forests produce environmental services that contribute to quality of life, both directly (water supply, soil fertility, pest control) and indirectly (watershed and biodiversity protection). When markets do not exist, forest resources and services are undervalued, and essential services are undersupplied. If also laws are not enforced, the upstream sources of environmental services can become degraded, imposing externality costs on downstream users. Forest degradation contributes to decreasing water quality, soil fertility, and land productivity as well as increasing water shortages, fires and haze, health impacts, downstream siltation and flooding. Short run, private, financial incentives are at odds with long run, public economic and sustainability needs. There is increasing recognition that all of these impose costs on the people, put pressure on state assets and budgets and increase conflict, which also increases uncertainty and costs. Under economic globalization, weak governance impedes investment, raises costs, and hinders international competitiveness and access to key markets (for forestry and other products). Indonesia spends hundreds of millions of dollars on rehabilitation of critical land and degradation problems created by past forest management practices. Increasing investment in land conversion to bio-fuels threatens continued encroachment and degradation of forest and peat lands. Improving policy consistency, reducing distortions, and improving management will help Indonesia to become more efficient and competitive, improve environmental services, and free up budgetary resources for priority needs for investment and job creation. Reducing deforestation and the use of fire in land conversion would have multiple benefits to the economy, the society and the environment - even without considering climate change and potential global financial incentives. Local economic benefits include improvement in forest asset values and local livelihoods; health benefits include the value of cleaner water and the reduction of haze pollution; environmental benefits include biodiversity preservation and tourism potential. A more balanced forest and land management approach would help to achieve this range of benefits, compared to the ongoing extractive approach that degrades Indonesia's natural assets. A more diversified, efficient, competitive and sustainable future is feasible, based on more plantation-grown timber, less degradation and encroachment, alternative timber sources, retooled mills and higher value-added processing, with more small-scale enterprises and employment. In terms of Indonesia's pro-growth, pro-jobs, pro-poor strategy and long term economic sustainability, it makes sense to strive toward these objectives. Yet, progress is slow because incentives are mismatched, rule of law is weak, and ownership and responsibility are poorly defined. As well, those who benefit from the current system have the means to resist change, even at the expense of the local public good. At the same time, the costs of change to a new system are poorly understood - and the benefits of change would accrue to different groups. Now, with the potential for payments for carbon sequestration in forests, there may be a source of finance to offset the cost or provide an incentive to take on needed changes. Beyond the local issues, Indonesia's forests also contribute to the global problem of climate change - and have the potential to produce the global public good of carbon storage and sequestration. Emissions from deforestation and land use change (including conversion by fire, especially on peat lands) are a large source of greenhouse gas emissions. The Government recognizes this and is developing an initiative on Reduced Emissions from Deforestation and Degradation (REDD) in collaboration with donors, NGOs, and scientists. Indonesia has the potential to gain hundreds of millions of dollars through a successful REDD effort. (Not all emissions from land use change and fire use would be eligible for payments under a REDD scheme.) For Indonesia to access these resources, it must have credible policies, programs and institutional arrangements in place. Forestry and land use in Indonesia is a complex story of policy, institutional and governance issues. To address emissions from forestry and land use and achieve REDD payments, Indonesia will have to address fundamental management and governance issues that have existed for some time. The potential for REDD payments may help to create new political will and financial incentives for implementing changes that improve forest sector outcomes. There are still significant legal, political, financial, management and incentive challenges that have to be faced on the road to improving forest management. Improvements in policy and governance are needed if REDD incentive payments are to compete successfully against the private financial incentives that induce land use changes that increase emissions and undermine provision of environmental services for the public good, both locally and globally.

Yann Clough Agroecology Georg-August University of Göttingen Waldweg 26 37073 Göttingen Germany Homepage "Cacao agroforestry: an ecological perspective on synergy potentials between agriculture and conservation" Indonesia has become the third largest cacao producer in the world in a record time. The expansion of cacao agroforests in the highlands of Central Sulawesi improved local livelihoods but fuelled agricultural expansion and landscape change. This is a significant development in terms of nature conservation, as this region hosts a major protected area, Lore Lindu National Park, which acts as a refuge for Wallacean flora and fauna. Conversion of forests to agroforests and subsequent intensification is common at the edge of the forest. When compared to most annual crops, agroforests can be species-rich and diverse, providing habitat for a large range of species of organisms and conserving some of the species from the original natural forests. This is conditional on the way cacao agroforests are managed. We collected two years of data covering cacao yield, pests and disease incidence, as well as extensive surveys of several organism groups including vertebrates, arthropods and plants and fungi in over 40 smallholder cacao plots. Here, we show how different aspects of management including shading, weeding and fertilisation practices affect the biodiversity of smallholder cacao, and explore whether biodiversity-friendly forms of management imply losses or benefits in terms of yield. We will put the results in the context of the current trend towards declining yields and increased threat by pests and diseases and discuss how combining production and conservation could sustain cacao production in the study region.

Rainer Finkeldey Forest Genetics and Forest Tree Breeding Büsgenweg 2 Georg-August University of Göttingen 37077 Göttingen Germany Homepage "Genetic variation of dipterocarps: from molecular phylogenies to the identification of the origin of timber" Dipterocarps are dominant components of lowland tropical rainforests in evergreen and seasonal tropical climates in Southeast-Asia. Due to the exploitation for their valuable timber, dipterocarp forests are heavily affected by deforestation, forest fragmentation, and non-sustainable forest management including selective logging. The development of non-manipulable tools to infer the origin of timber and wood products is an urgent need within the context of forest certification in order to follow the chain-of-custody, and to control the international trade with tropical timber. Phylogenetic analyses in dipterocarps have been based on sequence analyses of several chloroplast DNA regions, the internal transcribed spacer region and single copy genes of the nuclear genome. We extended these analyses to a large number of dipterocarp species from Indonesia, from the Philippines and from Vietnam. We also analysed genetic variation within and among geographically distinct populations of the most common and widespread emergents of lowland dipterocarp forests in Indonesia, Shorea leprosula and Shorea parvifolia. Since no clear variation among geographic regions was detectable at chloroplast DNA markers, we applied universal AFLPs (Amplified Fragment Length Polymorphisms) in order to identify markers with pronounced frequency differences between geographic regions in Indonesia. SCARs (Sequence Characterized Amplified Regions) markers have been developed from AFLP markers that unambiguously differentiate between the islands Sumatra and Borneo. We aim to develop a large and reliable set of DNA markers for the verification of the geographic origin of timber. For that purpose DNA isolation methods have been adapted for wood and evaluated by the amplification and sequencing of DNA fragments of different lengths in processed and unprocessed wood probes. Informative DNA markers will be tested in wood probes of defined origins.

Gerald Haug Geologisches Institut CHN H 70.1 Universitätstrasse 16 8092 Zürich Switzerland Homepage "Links between paleoclimates and human history" A unifying theme in paleoclimate research is well summarized by a piece of advice that I once heard the late Sir Nicolas Shackleton give to an audience of paleoceanographers: "Whatever you do, do it in high resolution." The underlying message, I believe, is that much 'noise' in geologic records is actually composed of meaningful environmental signals. A central goal is to use new approaches and techniques that do justice to the complexity of geologic records, in order to allow previously hidden signals to emerge. On the centennial to subdecadal timescale, climate archives with an appropriate memory are anoxic marine basins and lakes. In the anoxic Cariaco Basin off northern Venezuela, millimeter to micrometer-scale geochemical data in the laminated sediments of the Cariaco Basin have been interpreted to reflect variations in the hydrological cycle and the mean annual position of the Intertropical Convergence Zone (ITCZ) over tropical South America during the past millennia. These data with decadal to (sub)annual resolution show that the Terminal Collapse of the Classic Maya civilization occurred during an extended dry period from 700 to 900 AD. Data of comparable quality and resolution have been extracted from sediments of lake Huguang Maar in coastal southeast China. The record indicates a stronger winter monsoon prior to the Bølling Allerød warming, during the Younger Dryas, and during the middle and late Holocene, when cave stalagmite oxygen isotope data indicate a weaker summer monsoon. A remarkable similarity in the records of ITCZ migration in east Asia and the Americas from 700 to 900 AD raises the possibility that the coincident declines of the Tang Dynasty in China and the Classic Maya in Central America were catalyzed by the same ITCZ migrations. Comparison of our records with the Chinese dynastic history suggests that drought played a role in the terminations of Dynasties during the past 4000 years.

Wiebke Kirleis Environmental Archaeology Human Development in Landscapess Christian-Albrechts-Universität zu Kiel 24098 Kiel Germany Homepage "Human Landscapes in Sulawesi: The impact of people on tropical rainforest since prehistoric times" Although many of us are used to consider landscape and environmental settings as a natural fact, there is reason to take up a different position. In many fields of the humanities and social science there is the concept of human landscapes. This concept denies the existence of an untouched physical landscape after people began to found settlements and produce pottery. The Neolithic farmers were the first humans who have shaped the environment according to their needs. In contrast, the foregoing communities of hunter-gatherers are said to have acted in a sustainable way that hardly affected nature. Vegetation history may support this concept since human impact from hunter-gatherer communities is hardly detectable in pollen records. The Indo-Malaysian Archipelago has a long history of hunter-gatherer communities that extends from the Pleistocene to the mid-Holocene. Seven to six thousand years before the present (BP) agriculture started to spread from the Chinese mainland via Taiwan and the Philippines to the Malaysian archipelago, eventually reaching as far as western Micronesia. Earliest Neolithic agricultural and sedentary assemblages found in Sulawesi date from 4,000 BP. Since that time, agriculturalists have colonised and shaped the tropical rainforest, at least in easily accessible areas. In Sulawesi traces of human-environment interactions may be found in both natural and anthropogenic archives. Three regions in particular allow the reconstruction of human impact on tropical rainforest. These are the northern arm of Sulawesi, the southwestern arm of Sulawesi and central Sulawesi. The earliest evidence of human activity in northern Sulawesi dates from 8,000 BP and comes from a shell midden at the shoreline of Lake Tondano. The midden belongs to a pre-ceramic culture that practised hunting and gathering. However, there is little evidence for the presence of people in the Lake Tondano area in the pollen record until the late Holocene when progressive deforestation of the Tondano uplands and the spread of grasslands is recorded. The southwestern arm of Sulawesi has produced the most complete pre-ceramic sequences of late Pleistocene and Holocene stone tool working in the whole Indo-Malaysian Archipelago. Stone tools that range in age between 29,000 to 17,000 years have been recorded. Additionally, a number of sites indicate hunter-gatherer communities known as the Toalian, dating from 7,000 years ago and overlapping with the appearance of pottery in the region. Again, the activities of these hunter-gatherers were not reflected in pollen records from the region so their affect on the rainforest remains uncertain. The palaeo-ecological record from Lake Tempe indicates saline input along the Cenrana river that, around 2,500 BP, reached inland to Lake Tempe. This finding supports the hypothesis that there was formerly a saline channel that transected the peninsula in the vicinity of the Tempe basin and allowed maritime trade along the Cenrana river. Both the environmental history and archaeological record of the Besoa valley in Central Sulawesi show evidence for long-term human impact in the area, most notably through frequent burning and shifting cultivation. Palaeo-records from sites between 1,000 and 1,200 m asl show past human impact on vegetation development and demonstrate how human activity has altered rainforest composition in the area of the Lore Lindu Biosphere Reserve and National Park for at least 2,000 years. This impact is reflected by high amounts of charred fragments in the pollen record plus low values of arboreal pollen, typically resulting from logging and shifting cultivation. In parts of the biosphere reserve, periodical burning has prevented any forest recovery at all. In the Bada and the Besoa valleys, large stone jars and human statues are associated with iron and carved-paddle-impressed pottery. This indicates that they date to the Early Metal phase in eastern Indonesia. The lack of a defined chronology for Sulawesi's prehistory subsequent to the Toalian industry makes dating of the sites difficult. The absence of Chinese pottery suggests the Bada valley sites are older than 500 years and palaeobotanical investigations and radiocarbon dating of sediments from inside the Besoa valley jars suggest a minimum age of 1,000 years for this sites. Both the archaeological and the palaeobotanical records show the existence of human landscapes in different parts of Sulawesi since colonisation by agriculturalists started around 4,000 BP. This means that, in many parts, the existing rainforest has been shaped by humans over the long term. Although human impact left a considerable mark in valley locations, steep mountain areas were less affected. The records of the Besoa valley in Central Sulawesi in particular offer opportunities to implement some of the aims of the conservation philosophy of the Lore Lindu Biosphere Reserve. Given that the integration of local people and their cultural values into the maintenance of the area's high biodiversity and cultural heritage is one of the more important aims of the biosphere reserve concept, it might be one goal to develop new sources of income by implementing nature and culture-based tourism. Recommendations for conservation policy that may be drawn from the area's environmental history include consideration of the impact of human activity in the more easily accessible areas, particularly in face of recent settlement developments.

Stephan Klasen Chair in Development Economics Platz der Göttinger Sieben 3 Georg-August University of Göttingen 37073 Göttingen Germany Homepage "Institutions, Growth, and Rainforest Conversion" This paper investigates the linkages between institutions, technological change, economic growth and rainforest conversion in a sample of villages at the rainforest margin in Indonesia. The paper first examines to what extent economic growth is driven by institutional changes with a particular emphasis on the role of property rights, which in turn affects the introduction of growth-enhancing agricultural technologies. These institutional changes are themselves a result of immigration and the resulting population pressure. Regarding the impact on rainforest conversion, the paper uses satellite data of these villages in 1981 and 2001 and finds that high population pressure and early access to infrastructure hastened rainforest conversion. Conditional on these findings, higher economic growth appears to be associated with lower rates of deforestation. Stephan Klasen (presenting author, University of Göttingen) Michael Grimm (coauthor, Institute of Social Studies, The Hague) Stefan Schwarze (coauthor, University of Göttingen)

Robin Matthews Climate Change Theme Leader Macaulay Institute Craigiebuckler Aberdeen AB15 8QH UK Homepage "Modelling the dynamics of land use change and greenhouse gas emissions at the tropical forest margins" An estimated 13 million hectares of tropical forest are destroyed each year, resulting in the emission of 5.8 Gt of CO2, about 20% of total anthropogenic emissions of GHGs. The drivers of this destruction are many and varied, including a combination of commercial wood extraction, permanent cultivation, livestock development, and the extension of overland transport infrastructure. To add to this pressure, the increasing demand for biofuels in the US and Europe has led many tropical countries to see the growing of crops such as maize and sugarcane for biofuel feedstock as opportunities to kick-start their economies, which, in addition to diverting existing crop production away from use as food, has resulted in some areas of tropical forest being cleared for these crops. Currently, this 20% of emissions is outside the Clean Development Mechanism of the Kyoto Protocol. There are now discussions within the UN Framework Convention on Climate Change to try and develop mechanisms to reduce emissions from deforestation and degradation (REDD). A suggested mechanism is to reward countries that demonstrate a decrease in deforestation rate below a baseline based on average historical deforestation rates. However, it is unclear how targets that are agreed at the national level under international-level REDD rules will be translated to changes in the behaviour of the indigenous people, farmers, ranchers, and loggers who live on the land, and, indeed, whether the approach would be sufficient to overcome the many pressures at the local level leading to deforestation, including the need to produce food, and the economic opportunities afforded by growing biofuel crops. New ways to link the technical and institutional advances on REDD to local stakeholders are needed so that the various scenarios considered reflect local ambitions and response options. Active 'negotiation support' will be needed to achieve the 'free and prior informed consent' status that is seen as a moral imperative to agreements that potentially affect livelihoods of people outside of the centres of political power. Simulation modelling is one tool that allows analysis by stakeholders of different potential mechanisms before intervening in a real system. Many such deforestation models exist - one example is the FALLOW model (van Noordwijk, 2002), which simulates some of the factors influencing land use change in forest margins and the impact that these changes have on carbon stocks and biodiversity, with particular focus on the spatial interactions between different parts of the landscape containing forests and agriculture. Recently, agent-based simulation modelling has been arousing interest due to its ability to model individual decision-making entities and their interactions, to incorporate social processes and non-monetary influences on decision-making, to conceptually reproduce non-linearities ('tipping points') often observed in space-time processes of innovation and change, and to dynamically link social and environmental processes. Various ways in which agent-based land use models have been applied were reviewed recently by Matthews et al. (2007). One such agent-based model is the People and Landscapes Model (PALM, Matthews, 2006), which simulates a number of decision-making entities (e.g. farm households) located on a landscape made up of a number of heterogeneous land units, each of which contains routines to simulate its water balance and carbon and nitrogen dynamics (including GHG emissions) over time. Decisions made by the household agents result in actions which may influence the fluxes of water, carbon and nitrogen within the landscape. PALM has been used to evaluate economic instruments such as taxes ('sticks') or incentives ('carrots') aimed at reducing GHG emissions and the impact that this may have on farmer livelihoods (Matthews & Bakam, 2007). Like other simulation models, a disadvantage of agent-based models is that they can often require more data than is available, particularly in relation to the degree of 'co-variation' between individual characteristics. However, a number of 'hybrid' approaches are now available, which combine a location-specific analysis of spatial patterns with macro-economic data, broadly 'generic' processes valid beyond local parameterisation, and at least some social/economic stratification among the agents. In this paper, we discuss the potential of such models to understand the complex dynamics of land use transitions at the tropical forest margins, give some examples of their use, and outline some of the challenges facing their general applicability. References Matthews, R., Gilbert, N., Roach, A., Polhill, G. & Gotts, N., 2007. Agent-based land-use models: a review of applications. Landscape Ecology 22(10):1447-1459. Matthews, R.B., 2006. The People and Landscape Model (PALM): towards full integration of human decision-making and biophysical simulation models. Ecol. Modelling 194(4):329-343. Matthews, R.B. & Bakam, I., 2007. A combined agent-based and biophysical modelling approach to address GHG mitigation policy issues. In: D. Kulasiri & L. Oxley (Editors), Proceedings of MODSIM 2007 International Congress on Modelling and Simulation, Dec 10-13, 2007. Modelling and Simulation Society of Australia and New Zealand, Christchurch, New Zealand. van Noordwijk, M., 2002. Scaling trade-offs between crop productivity, carbon stocks and biodiversity in shifting cultivation landscape mosaics: the FALLOW model. Ecol. Modelling 149:113-126.

Gerald Moser Department of Plant Ecology and Ecosystems Research Albrecht von Haller Institute for Plant Sciences Georg-August University of Göttingen Untere Karspüle 2 37073 Göttingen Germany Homepage "ENSO drought effects in Southeast Asia - their past, present and future" The El Nino Southern Oscillation (ENSO) phenomenon has caused severe droughts during the past in Southeast Asia, affecting significantly ecosystems, agriculture and socio-economics. Analyses of the climate history have shown that ENSO already exists for 15000 years. It seems that its intensity was low before 5000 BC and that its frequency and intensity increased since then. During the second millennium only few severe ENSO occurred. Since the end of the 19th century intensification could be observed; this was pronounced since the 1950ies. Climate models predict a still increasing frequency and intensity of ENSO in the future caused by global warming. For natural ecosystems this intensification is assumed to cause a significant selective force leading to a displacement of drought sensitive species. The amounts and predictability of agricultural yields in periods of frequent and severe droughts will decrease, causing an expressed destabilisation of the socio-economic status of farmer´s households and their vulnerability to poverty. This economic insecurity combined with a threatening decrease in ecosystem services could accelerate land-use change and deforestation. Drought tolerant crops and adapted management systems will therefore become of striking importance.

Daniel Murdiyarso Jalan CIFOR, Situ Gede, Sindang Barang, Bogor Barat 16680, Indonesia Homepage "Making forestry works for climate change mitigation and adaptation" Managing forests under sustainable systems would potentially contribute to climate change mitigation. These could be implemented by enhancing sink capacity of forests through afforestation and reforestation. As important source of terrestrial carbon, reducing deforestation and forest degradation (REDD) in the tropics are among the best ways to reduce greenhouse gas emissions. Stern (2006) reported that avoided deforestation is the least expensive option for climate change mitigation. He found that the opportunity cost of forest protection in 8 countries responsible for 70% of emissions from land use could be around $5 B/y initially, although over time marginal costs would rise. The question is, would this benefit be attracting to tropical forest countries regarding the opportunity costs of not utilizing tropical forests? The inclusion of REDD in the new climate region post-2012 merit further evidence for a full implementation. Forest ecosystems are among the terrestrial systems that can be used by the community to adapt to climate change. Ecosystem services should be identified and utilized to increase the resilience of forest-dependant community. Among adaptation strategies discussed here is the services provided by the forested watershed regarding sustainable supply of surface and groundwater recharge. The implementation of project activities should identify the synergies and trade-offs between mitigation and adaptation and consider the type of activities, the scale of project in particular ecosystems, and the level of stakeholder participation. Mitigation pathways may be used as entry point, while narrowing the knowledge gaps of forest-water relationship to better inform decision-making processes. Nevertheless, synergies between mitigation and adaptation measures should be developed whenever ecosystems and societies are considered to be vulnerable to climate change.

Meine van Noordwijk International Centre for Research in Agroforestry ICRAF SE Asia P.O.Box 161, Bogor 16001, Indonesia Homepage "Agroforests on the interface of adaptation to and mitigation of climate change"

Susan Page Senior Lecturer & Deputy Head of Department Department of Geography University of Leicester Leicester LE1 7RH UK Homepage "Turning up the Heat! Carbon-Climate-Human Interactions in Tropical Peatlands - Positive Feedbacks to Global Warming" The status of the world's peatlands is a matter of considerable concern since their degradation can lead to carbon emissions to the atmosphere and loss of carbon sink function. Tropical peatlands, located mostly in SE Asia, make a significant contribution to global terrestrial carbon storage, both in terms of their above-ground biomass (peat swamp forest) and thick deposits of peat; they are estimated to contain at least 50 Gt of the global soil carbon store. The carbon sink role of tropical peatland ecosystems is, however, being impacted by rapid land-use changes that increase their susceptibility to degradation and fire. Recent estimates of carbon emissions arising from drainage and oxidation of tropical peatlands are within the range 100-240 Mt C yr-1. Over the past decade widespread fires have also had a major impact on the carbon balance. Page et al. (2002) calculated 810-2570 Mt C were released into the atmosphere as a result of the ENSO-related 1997/98 fires on Indonesian peatlands; significant emissions also occurred during the 2002 and 2006 fire events. In order to provide accurate spatial and temporal values for these emissions it is necessary to study the impact of peatland fires at a landscape scale and to acquire detailed information on fire history (location and return period), fire severity and the dynamics of post-fire vegetation succession and fire susceptibility. To facilitate this, we have employed a multi-temporal remote sensing approach for a study area of 450,000 ha of peatland in Central Kalimantan (Indonesian Borneo) in order to investigate the impacts of deforestation and drainage on peatland fires and vegetation dynamics. The results show that once disturbed, peat swamp forest, in common with other tropical forests, becomes more susceptible to degradation and fire as a result of the modification and fragmentation of the naturally closed structure of the forest, with repeated fires producing a heavily modified secondary ecosystem dominated by ferns, grasses and sedges, in which both fire risk and fire return period increase and, hence, a return to pre-fire vegetation becomes almost impossible. Over a 16 year period, more than 93% of the study area was burnt on at least one occasion; fires are now a regular feature of every dry season, not just those of ENSO-years. Scaling up the emissions data for the study area indicates that fires, alongside drainage and degradation, are resulting in globally significant carbon emissions from SE Asian peatlands and a rapid loss of their carbon sink function.

Rüdiger Pethig University of Siegen Department of Economics Hoelderlinstr. 3 57068 Siegen Germany Homepage "Land use, biomass harvesting and ecosystem services in a theoretical integrated ecological-economic model" In an integrated dynamic general equilibrium model of the economy and the ecosystem, humans and wildlife species compete for land and prey biomass. Humans rely on ecosystem services that depend, in turn, on biodiversity and on the size of habitat for non-human species. As is well known, conversion of land for economic use as well as harvesting of biomass by humans are among the most important links between the ecosystem and the economy. In our model these activities are endogenous and we capture their pervasive intertemporal repercussions in the ecosystem which, in turn, impact on the quality and quantity of ecosystem services. We introduce a competitive allocation mechanism in both submodels such that economic prices and ecosystem (shadow) prices guide the allocation in the economy and in the ecosystem, respectively. We distinguish the scenarios of a habitat for which property rights are either enforced or not enforced. We show that fully enforced property rights are not enough to secure efficiency in the no-policy case and that each property-rights scenario calls for specific corrective policies. The main purpose of the paper is methodological and conceptual but it aims, at the same time, to demonstrate that the application of a general-equilibrium competitive allocation mechanism to a microfounded integrated ecological-economic system yields specific and new insights with regard to efficient management of the ecosystem and its services.

Christian Schulze Faculty of Life Sciences University of Vienna Althanstraße 14 1090 Wien Austria Homepage "The importance of countryside habitats and wooded corridors for maintaining species richness of tropical birds and butterflies." Global biodiversity is seriously threatened by two key factors: (1) human activities modifying and transforming natural habitats and (2) climate change. Although climate change particularly affects biotas of higher latitudes, at least some effects, such as a shift in the altitudinal distribution of bird species, are also documented for tropical areas. However, the major threat for tropical biodiversity is certainly habitat loss caused by human land-use activities. As shown for birds, continental lowland forest species are especially threatened by habitat loss. Particularly understorey species with narrow ranges suffer from forest disturbance and modification, and anthropogenic habitats may be able to support only a small fraction of the forest species pool. Also for range-restricted understorey forest butterflies most land-use systems appear to be of limited conservation value. However, many studies on effects of forest modification and conversion on butterflies do not provide reliable data for evaluating the importance of land-use systems for butterfly conservation. This is due to an inadequate spatial sampling (ignoring significant differences of beta diversity between habitats) and/or neglecting potential effects of the isolation of land-use systems from remaining forest (many records in land-use systems may represent "tourists" depending on remaining forest). Additionally, the increasing isolation of remaining forest areas negatively affects biodiversity, particularly when the countryside, in which forest patches are embedded, acts as an effective barrier for movements of forest species. Strips of gallery forest along rivers (protected by law e.g. in Costa Rica) may represent corridors or stepping stones facilitating the re-colonization of forest patches by forest species. However, as shown by our studies from the southern Pacific slope of Costa Rica such wooded strips are only of limited conservation value for most forest understorey species. There is evidence for birds that island species appear to be more robust to habitat conversion than continental species. Perhaps a lower extinction risk of island species (compared to continental species) is the result of higher population densities (due to decreased interspecific competition) and expanded niches (due to increased intraspecific competition). The later may facilitate the colonization of human-disturbed habitats by forest species. Further studies testing these hypotheses will be of high relevance for developing conservation priorities for continental and island biotas in the tropics.

Susanne Stoll-Kleemann Sustainability Science and Applied Geography Institute of Geography and Geology Ernst Moritz Arndt Universität Greifswald Friedrich-Ludwig-Jahn-Str. 16 17487 Greifswald Germany Homepage "Success Factors of the Protection and Sustainable Use of Tropical Rainforests in Biosphere Reserves and Protected Areas" Tropical Rainforests face many types of threats to their ecological integrity. This is the reason why they are subject to strong conservation efforts around the world resulting in the establishment of protected areas and biosphere reserves. Having adopted a protected area in legal terms does not automatically lead to the "real" protection of the tropical rainforest included in the protected area. Harmful practices originate from the failure of policies, laws and decision-making processes to provide effective guidance and conservation incentives to managers and others involved. Conservation managers and other stakeholders involved in the protection and sustainable use of tropical rainforests confirm that problems at the operational level are closely linked with broader governance issues. This paper presents results from the interdisciplinary research project GoBi (Governance of Biodiversity), which evaluates the success or failure for implementing protected areas and biosphere reserves. Its main hypothesis is that the ecological outcome of implementing protected areas depends on the appropriateness of the selected governance and management systems with regard to the local context, and on broader economic and political issues. The study is based on a large set of expert interviews (n=177) and a global survey with UNESCO Biosphere Reserves (n=213 ) as well as case studies in nine countries on four continents. One of the case studies is the Lore Lindu Biosphere Reserve (Central Sulawesi) to which special reference will be made in the paper . The research project GoBi connects different kinds of data by means of an integrative model to be used as a decision support tool in biodiversity conservation and sustainable resource use. The results show that typical imperfections of governance and management institutions such as enforcement and monitoring problems, insufficient political support, lack of stakeholder involvement, corruption, lack of capacity and leadership play an important role in determining success or failure for implementing successfully protected areas and biosphere reserves. Adaptable institutional arrangements including responsive leadership and capacity building are necessary to protect and sustainably manage tropical rainforests that have complex social, political, cultural and ecological dimensions.

Manfred Zeller Department of Agricultural Economics and Social Sciences in the Tropics and Subtropics University of Hohenheim Schloß, Osthof-Süd 70599 Stuttgart Germany Homepage "Linkages between poverty and sustainable agricultural and rural development in the uplands of Southeast Asia" Most of the upland areas of Southeast Asia are characterized by insufficient infrastructure, low productivity in smallholder crop and animal production, mounting environmental problems such as soil and forest degradation and loss of biodiversity, increasing population pressure, and widespread poverty. While some upland areas in South East Asia have been experiencing considerable progress during the past twenty years, others have stagnated or even declined with respect to economic, social and environmental objectives of development. The conceptual framework for this paper builds on the critical triangle of sustainable rural development. Here, equity or poverty alleviation, economic growth, and the protection of the environment are the three major political objectives. In the short-run and medium-run, there exist trade-offs and synergies between these three objectives. The purpose of the paper is to describe major trends in the upland areas of selected countries in South East Asia. The paper begins with a review of definitions of sustainability, and proceeds with a conceptual analysis of the two-way linkages between poverty and the environment, and poverty and economic growth in rural areas. This is followed by empirical findings from research on agriculture and forestry as the major land uses in upland areas of selected South East Asian countries. Based on the results of different case studies from Cambodia, Laos, Vietnam and Indonesia, we seek to contrast stories of relative success with those of failure in order to identify priorities for policy design and future research. The paper concludes with policy implications for rural and agricultural development policies, and suggests future areas of pro-poor policy-relevant research.