The crisis in the early 1990s over conservation of biodiversity in the forests of the Pacific Northwest caused an upheaval in forest policies for public and private landowners. These events led to the development of the Coastal Landscape Assessment and Modeling Study (CLAMS) for the Coast Range Physiographic Province of Oregon, a province containing over two million hectares of forest with a complex mixture of public and private ownership. Over a decade, CLAMS scientists developed regional data bases and tools to enable assessments of the implications of current policies for biodiversity and have begun using these data and tools to test ideas for solving policy problems. We summarize here four main lessons from our work: (1) Regional ecosystem perspectives, while rewarding, are difficult to achieve. Helping policy makers and the public understand biodiversity policies for an entire province can assist in developing more reasoned policies. However, this result is difficult to achieve because needed scientific building blocks generally do not exist, few policy institutions address regional cross-ownership issues, people can find it difficult to take a regional view, and the appropriate region for analysis changes with the policy problem. (2) Interest in environmental policy analysis may come as much from a pursuit of power as a pursuit of understanding. Biodiversity policy analyses are often viewed as weapons in an ongoing political battle. Also, results that might destabilize existing policies generally will not be well received by those in power. (3) The relationship of regional analyses to civic processes remains challenging and unsettled. Communication between citizens and scientists takes real effort. Also, collaborative processes both inspire and constrain regional policy analysis, and scientific work often proceeds at a different pace than these processes. In the end, CLAMS's most important effect on the civic dialogue may be to change how people think about the Coast Range. (4) An important role exists for anticipatory assessments done independently by scientists. Independent review will be especially important as policy analyses shift to management of nonfederal forests. Our future efforts in CLAMS will focus on evaluating ideas for fundamental changes in forest management.