FD1-1: Landscape distribution of vegetation is largely predictable from state factors, with the residual variation caused by measurement error and legacies of past events

Aerial photos will be flown of both Bonanza Creek Experimental Forest (BCEF) and Caribou Poker Creek Research Watershed (CPCRW), and high resolution Digital Elevation Models (DEMs) will be created. These photos will then be orthorectified and digitized into polygons of vegetation type. After vegetation type has been determined, we will ground truth random points and the LTER permanent sample plots in both BCEF and CPCRW to make sure we are accurately accessing vegetation type. Vegetation type will be based on a combination of dominant species cover and diagnostic species, and will have community names associated with them. These digitized photos will then become a GIS vegetation layer, which can then be linked to 30 m Landsat images, and the 1 km AVHRR vegetation map. Slope, aspect, temperature, and precipitation can then be derived from a combination of Landsat imagery and our DEM. Parent material will be derived from geologic maps of Alaska, and maps of recent fires and stand reconstruction will be added as other GIS layers. We will use our LTER permanent sample plots as well as other random points to determine the accuracy of our state factor derivations. Finally we will use GIS change analysis to look at how both BCEF and CPCRW have changed in vegetation in the 20 years since the last vegetation maps were created.