THE ARCTIC FLUX STUDY - A REGIONAL VIEW OF TRACE GAS-RELEASE

Fluxes of trace gases from northern ecosystems represent a highly unce rtain and potentially significant component of the arctic land-atmosph ere system, especially in the context of greenhouse-induced climate ch ange. The initial goal of the Arctic Flux Study (a part of NSF's Arcti c System Science Program) is a regional estimate of the present and fu ture movement of materials between the land, atmosphere and ocean in t he Kuparuk River basin in northern Alaska. We are measuring rates and controls of processes along a north-south transect running from the ma rshy coastal plain to mountain valleys. Important vertical fluxes unde r study are the release of CO2 and CH4 from soils and water, lateral f luxes are surface water, nutrients, and organic matter. A hierarchy of measurements allow the rates and understanding of processes to be sca led from plots to the landscape, regional, and circumarctic level. The se include gas flux measurements in small chambers, measurements over larger areas by eddy correlation from small towers, and measurements a t the landscape scale from airplane overflights. Experimental manipula tions of carbon dioxide, soil moisture, nutrients and soil temperature from this and other studies give information on process controls. The distribution of plant communities has been described at several lands cape-scale sites and a hierarchic GIS has been developed for the regio n at three scales (plot, landscape, region). Climate is measured at si x sites and hydrological processes are being studied at each watershed scale. In the soils, measurements are being made of soil organic matt er and active layer thickness and of availability of soil organic matt er for microbial transformation into CO2 and CH4. Fluxes and process u nderstanding have been incorporated into a hierarchy of models at diff erent scales. These include models of regional climate nested in a GCM ; of regional- and continental-scale plant productivity and carbon cyc ling including CO2 release under altered climates; watershed and regio nal models of hydrology; and surface energy budgets. After the first y ear of study the regional climate model has been successfully configur ed to the northern Alaska region We have also measured a large release of carbon dioxide from tundra soils in all but the coldest and wettes t parts of the transect. The rates from eddy correlation towers (lands cape level) agree closely with rates from chambers (plot level). Obser vations, experimental manipulations and modelling analyses result in t he prediction that the combination of warmer and drier soils is respon sible for the large CO2 release.