Quantification and modeling of patch-scale shrub expansion in the Alaskan Arctic

Shrub expansion over the course of the 20th century in the Alaskan Arctic has been documented from experimental plot data and broad-scale satellite remote sensing. There is, however, a lack of knowledge regarding expansion at the intermediate patch-scale. Analysis of repeat oblique aerial photography from two dates (late 1940s and late 1990s) in the Colville River basin identified significant expansion (3-80% increase). While it has been proposed that this expansion follows a simple logistic growth model, its precise nature in the interim period is largely unknown. Building on this work, this study aims to more explicitly quantify and model the manner in which shrub expansion has occurred in the Colville basin and other sites throughout the North Slope. From compilation and georectification of historic vertical aerial photos of these sites, along with associated high-resolution orthorectified QuickBird imagery, we will map shrub patches in a GIS. We will calculate pattern metrics of these maps using FRAGSTATS to pinpoint potential variability in spatial patterns. Additionally, we will develop and implement a stochastic cellular model that simulates shrub expansion and incorporates environmental heterogeneity (e.g., topography and hydrology) and biological processes (e.g., clonal expansion and seed dispersal). Analysis of the model output using FRAGSTATS, MANOVA, and Principal Components Analysis (PCA) will allow us to determine which environmental parameters best explain the observed pattern of expansion. This knowledge and methodology can then be used to refine hypotheses about the processes controlling shrub expansion and can also be applied to predict future expansion in Alaska and in other areas throughout the Arctic.