A regional GIS-based model to predict long-term responses of soil and soil water chemistry to atmospheric deposition:: Initial results

The aim of this study was to develop GISSMART, a GIS-based model, for estimating relative regional changes in soil and soil water chemistry for given atmospheric deposition and nutrient uptake scenarios. The regional application was performed by dividing the study area into grid cells, the size and the form of the pixels being determined on the basis of the available GIS data bases of Finnish soil and vegetation types. GISSMART is the biogeochemical model SMART, extended for handling GIS data in space and time. The Kangasvaara catchment (56 ha) in eastern Finland was used as the first test site. Data from four other nearby catchments were also used, together with Multisource National Forest Inventory data (25 x 25 m(2)) from an area of 10 x 40 km(2) for estimating soil chemistry. Multisource forest inventory data from the Finnish Forest Research Institute, including tree species, age and growth by tree species was based on Landsat TM-images and field measurements. The same satellite based estimation method was applied to obtain regional soil chemistry input data for GISSMART. Base saturation was the key state variable and nutrient uptake the key driving variable in this application of GISSMART. The results of the model test runs were reasonable. A slight decrease in base saturation was predicted for the assumed deposition and nutrient uptake scenarios. In future work, the aim is to improve the description of hydrological processes and land use management in the model. This would also enable assessment of the effects of climate change and land use management scenarios.