Landscape trajectory analysis: spatio-temporal dynamics from image time series

Landscape dynamics can be revealed through analysis of image times series. The temporal development of spatial patterns holds significant information about ecosystem processes. Yet, the tools for the characterization of spatio-temporal structure are few. We describe some of the challenges facing spatio-temporal analysis and demonstrate a method for identifying and measuring pattern in image time series using a standard data product, the USGS maximum AVHRR NDVI biweekly composites for 1990-2000. We selected six distinct and diverse ecoregions as delineated and revised by Omernik and colleagues. The analysis extracts expectations of landscape spatial pattern during the growing season. The resulting landscape trajectories can serve as building blocks for the development of operational environmental monitoring and ecological forecasting systems.