(1) The history, types, and scales of disturbance in Arctic Alaska are reviewed and disturbances organized according to the spatial and temporal domains of Delcourt, Delcourt and Webb. This system is also used as a framework for a regional hierarchical geographic information system (GIS). (2) Natural disturbances vary from frequent small disturbances, such as needle-ice formation, to infrequent large disturbances, such as major glaciations. Most natural disturbances are either directly or indirectly climatically driven and are affected by climate changes, particularly changes to hydrologic regimes. The latter could be influenced by changes in either summer or winter precipitation patterns; increased temperature, which would melt ground ice; or changes in vegetation, which would affect evapotranspiration and run-off. (3) Most anthropogenic disturbances are microscale (10(-1) to 10(6) m2) phenomena, but cumulative impacts associated with large developments, such as the Prudhoe Bay Oil Field, have affected mesoscale regions (10(6)-10(10) m2), and global warming could affect the tundra ecosystem at the macroscale level (10(10)-10(12) m2). (4) In the Arctic, recovery of the vegetation following disturbance is particularly closely linked to recovery of the physical system because of the presence of ice-rich permafrost. Maps of terrain sensitivity to disturbance must consider the influence of ground ice and heat flux to the system following disturbance. (5) A three-tiered GIS hierarchy with five sublevels is presented, with examples of typical scientific questions being addressed at each level, scales and types of databases, and linking elements between levels. (6) At the regional (macroscale and mesoscale) levels, the primary data sources are statellite-derived digital data. At the site level, integrated geobotanical databases derived from field surveys and photointerpretation are used in combination with digital terrain models. At the most detailed (plot or microsite) level, point sampling is used to portray vegetation structure and species composition in 1-m2 plots. (7) Linking or 'scaling-up' elements that affect landscape patterns at all scales are hydrology, geochemistry, and primary production.
