Applicability of a forest simulation model for estimating effects of nitrogen deposition on a forest ecosystem: Test of the validity of a gap-type model

Ecosystem models have been used to compile scattered information on various ecosystem processes and to test the hypotheses about ecosystem responses to various simultaneously changing environmental factors. In spite of the widespread use of models, there have been comparatively few quantitative evaluations of model projections compared to long-term observations under changing environmental conditions (e.g. increased nitrogen deposition). In this study we tested the validity of a gap-type forest simulation model (SIMA) in order to extend the applicability of the model for the prediction of how nitrogen deposition influences the production of a boreal forest ecosystem. The validity of the model was tested by comparing the prediction of the model with independent data from long-term fertilization experiments. The predictions provided by the SIMA model agreed fairly well with the results of long-term fertilization experiments. Both the experiments and the model simulations showed similar increases in stem-wood production and in growing stock as a consequence of repeated nitrogen fertilization over the 30-year study period. The addition of nitrogen increased the total production by 30-53% according to field experiments and by 39-63% according to model computations. In both the model calculations and the field experiments, organic matter accumulated in the soil after the addition of nitrogen. The increase in the amount of soil organic matter can be explained as being due to the increased biomass production and the resulting increase in litterfall. According to the model computations, annual litterfall of needles on the mesic site varied from 970 kg ha(-1) to 3050 kg ha(-1) and this agreed well with measured litterfall of the stand. (C) 1998 Elsevier Science B.V.