Climate change poses severe threats to biodiversity, and conservationists have to adapt their management decisions to a changing climate. The challenge is that the biological response to future climate change is uncertain. Climate change will entail a general warming, but even more importantly may increase the frequency of extreme climate events (e.g., multiyear droughts) and extreme weather events (e.g., shorter droughts, heatwaves, and cold snaps). Our previous work has shown that extreme events, such as heat waves and droughts affect biodiversity strongly, and can cause strong declines in bird abundances. However, the causes of these abundance declines are not well understood, and at least four mechanisms may be at play. The first mechanism is lower recruitment, caused either by higher juvenile mortality, or by fewer attempts at reproduction. The second mechanism is adult mortality, which can occur when physiological limits of a species are exceeded during an extreme weather event. The third and forth mechanism are two forms of dispersal, either long-distance dispersal into other parts of a species range (e.g., ducks responding to droughts in the prairie pothole region by migrating to the Boreal to find more favorable breeding conditions), or the concentration of individuals within habitat refugia. Among such habitat refugia, the most important ones are National Wildlife Refuges for waterfowl, and National Forests for forest birds. However, it is not known if these protected areas function as refugia during extreme events. Further, if they do, it is not clear if management actions to enhance food and cover resources could further enhance their role as refugia. Our proposed study thus has two major goals. The first is a basic science question: we seek to predict the effects of extreme climate and weather events, including droughts, heat waves, and cold snaps during the breeding season, on bird demographics, focusing especially on waterfowl and forest birds. As part of this question we will test four alternative hypotheses to explain observed abundance declines: (1) the lower recruitment hypothesis, (2) the adult mortality hypothesis, (3) the long-distance dispersal hypothesis, and (4) the refugia hypothesis. Our second goal is an applied research question: we seek to quantify the role of National Wildlife Refuges and National Forests as refugia for waterfowl and forest birds respectively during extreme events, and to identify management actions to enhance this function. The scope of our work is the conterminous U.S. Our approach will be based on remotely sensed data and in situ wildlife observations. We will use satellite data to map the habitat of bird species and guilds (Landsat, ASTER, and SRTM), and to identify extreme climate events (AVHRR, MODIS, AMSR-E). As biodiversity data, we will use the Monitoring Avian Productivity and Survivorship (MAPS) program, the Breeding Bird Survey, and waterfowl counts from National Wildlife Refuges. Based on the results of the hypotheses tests, we will simulate population viability, and the role of current and future protected areas for a selection of six species with the metapopulation model RAMAS GIS. The simulation model will quantify the effects of predicted future extreme events on population viability, and help identify optimal management responses. Our proposed work will make a substantial contribution to both biodiversity science and to conservation. In terms of the NASA Earth Science research objectives, our project examines how the earth system is changing, how the earth system responds to natural and human-induced changes, and how the earth system will change in the future. We will make strong use of NASA assets since it will be based on, and revolve around, satellite data. Last but not least, the project will have broad societal relevance given widespread concerns about biodiversity declines, and improve broad-scale conservation efforts.