The success of global models in estimating future terrestrial impacts on climate depends on our ability to recognize and include those processes that have large effects on the ocean-land-atmosphere system. For example, it is hypothesized that climate system models which allow transient movement of vegetation in response to changing climate, land use and wildfire regimes would be an improvement over earlier models which specify constant or step-changing vegetation. An efficient way to test such a hypothesis to first order is to examine the immediate, local response to details of vegetation characteristics in a column version of a climate system model. This study will present such tests for the Alaskan region using the column version of the Arctic Region Climate System Model (ARCSyM), and will also report on the climate modelling implications of the first field seasons of the ATLAS (Arctic Transitions in the Land-Atmosphere System) and FROSTFIRE projects.