The floodplain (varzea) of the main stem Amazon river extends over nearly 100,000 km2, constituting the world's largest riverine wetland. The complex and dynamic landforms of the floodplain result in a mosaic of permanent and seasonal lakes and channels, levees, and flats supporting economic activity and biodiverse plant and animal communities with distributions largely determined by flood depth, flood duration, and successional stage. Over the last 40 years, two major economic drivers have been transforming the floodplain ecosystem: the growth of commercial fisheries, and the expansion of cattle and water buffalo ranching. The impacts of these activities are exacerbated by climate-induced changes to flooding regimes. The observed trends in land use and climate induced changes in flood regime raise key questions for the future of the varzea and entire Amazonia: 1) How are past and current land cover change impacting varzea communities and biodiversity? 2) How are varzea land use systems impacted by climate-induced variability in flood regimes? 3) How are these impacts predictive for the entire Amazon floodplain? 4) What policies and management strategies are most effective for managing changes and mitigating their impacts? To answer these questions, we propose to carry out integrated remote sensing and modeling studies in order to quantify key drivers of land cover and land use change on the lower Amazon floodplain. We will use archived remote sensing datasets (ALOS PALSAR, Landsat TM, historic aerial photography, videography), historic maps and charts, and socioeconomic datasets (fisheries, ranching, and household surveys) in analyses of the vulnerability of varzea land-cover and land-use systems to multiple stressors.The proposed research is organized to test a set of hypotheses regarding the current state and historic trends of the varzea system and the feedbacks between them: 1. Observed reduction in forest area and degradation of grasslands is due to an increase in cattle densities on the floodplain and associated cattle management practices. 2. Reduction of forests and degradation of remaining forests and aquatic macrophyte communities, combined with an increased frequency of extreme flood events, could lead to reduction in the productivity of floodplain fisheries. 3. Household and rancher economic strategies and associated cattle management practices are the primary drivers of deforestation and habitat degradation on the Lower Amazon floodplain. 4. Effective co-management policies can influence household and rancher economic strategies to reduce pressure on forest and grassland habitat and floodplain fisheries and increase resilience to impacts of climate change. Proposed remote sensing work will 1) map contemporary land cover of the lower Amazon floodplain and adjacent uplands at 30 m scale for input to analyses of fisheries productivity and flood pulse variability 2) estimate what proportion of the currently non-forested floodplain was historically forested 3) create spectral fraction images suitable for assessment of pasture degradation due to grazing and 4) test an approach for generating a bare-earth floodplain DEM at 30 m scale. The project will take advantage of research, community development and policy analysis on varzea ecology and resource management conducted in the region since the early 1990s including a regional fisheries dataset collected since 1992 and extensive socioeconomic data collected since 1995. We will complement these data sets with research on the impacts of government management policies on rancher and smallholder economic strategies.