The project directly addresses a high priority research topic for 2006-7: developing thresholds for indicators. For biological and practical reasons, we need to identify major structural features of forests that support biological diversity. Biologically, the changes in species composition through time are primarily in response to changing forest structure rather than to time itself. Trees are particularly long-lived and tall, so forests can generate diverse structure. Practically, forest managers are limited to simple approaches (stands are removed in whole or in part; trees are regenerated naturally or by humans) and to simple indices to guide harvest planning (simple enough to cover the large areas necessary in planning or monitoring). We need a way of selecting a short list of habitat features that connect directly with biodiversity, are manipulated by forest practices and are readily monitored. This project collates relevant data and analyses these to extract apparent thresholds for within-stand biodiversity indicators (medium to coarse filter). Data from British Columbia are emphasized, but will be drawn from the entire Pacific Northwest (Alaska, Alberta, British Columbia, Washington, Oregon, Idaho, Montana, and northern California). Relevant data from the SFM network, Europe, and Asia will be incorporated, particularly for boreal forests. The broad base ensures scientific credibility. Within-stand indicators of biodiversity include: large live trees, canopy cover, snags, downed wood (CWD), hardwoods, and shrub and ground cover layers. These are well established as useful indicators in North America, Europe and Australia, but few apparent thresholds have been extracted. For British Columbia, most data are available for vertebrates, but by drawing on a broader literature analyses can be extended to bryophytes, lichens, fungi, vascular plants and, more tenuously, invertebrates. Such breadth ensures that apparent thresholds are not dominated by a few well-known species, and encompass a variety of scales. The approach is to first design a framework that collates the large amount of scattered but relevant material in a fashion connecting directly with forest planning and practice. That framework will focus analyses on extracting thresholds for habitat elements directly influenced by forest practice. Wherever possible, data will be aggregated by Biogeoclimatic Ecological Classification (BEC) unit or stand type (broadly cross-walking data from outside BC). Analyses will attempt to statistically estimate uncertainty associated with thresholds for specific stand types or for BEC-based or broader regional thresholds. That approach helps to estimate natural flexibility in thresholds, focus future research and guide effectiveness monitoring to areas that are most informative. Extension documents will focus on simple graphical or tabular summaries of findings and associated uncertainty or flexibility. The potentially relevant literature base including gray literature is vast, so initial effort will be focused on those within-stand elements with which most forest-dwelling species listed by the Species at Risk Act are strongly associated. These include: cavity sites (large trees and snags), downed wood and hardwoods. It is expected that work on these three can be completed for vertebrates within the first year, and at least partially extended to other components of biodiversity (e.g., bryophytes and lichens).