Response of alpine lakes and soils to changes in acid deposition: the MAGIC model applied to the Tatra Mountain region, Slovakia-Poland

A dynamic process-based model of surface water acidification, MAGIC, was applied to 31 representative alpine lakes in the Tatra Mountains (~50% of all alpine lakes >0.3 ha in the lake-district). The model was calibrated to observed lake chemistry for the period 1980-2002. Surface water and soil chemistry were reconstructed from 1860 to 2002, given estimates of historical acid deposition, and forecast to 2020 based on the reduction in sulphur and nitrogen emissions presupposed by the Gothenburg Protocol. In the 1860s, all lakes were buffered by the carbonate system and only ~6% of lakes had acid neutralising capacity (ANC) <20 μeq l-1. Lake acidification progressed until 1980s, at which time 23% of lakes had a depleted carbonate buffering system and 33% of lakes had ANC <20 μeq l-1. Reversal of water chemistry from acidification started in the late 1980s as a response to decreasing acid deposition. ANC has increased such that only ~16% of the lake population currently has ANC <20 μeq l-1. The number of low ANC lakes is predicted to decrease to 10% by 2020, but the original carbonate buffering system of these lakes will not be re-established. The patterns in long-term changes of sulphate and chloride primarily reflected trends in atmospheric deposition and were similar for all lakes. Base cations (BC), ANC, nitrate, and pH, however, were significantly influenced by catchment characteristics. The water chemistry response to changes in strong acid anion (SAA) inputs varied among the lakes. The changes in SAA were compensated for (1) by parallel changes in BC concentrations (~75% of the SAA change) in non-sensitive lakes (high weathering rates and abundant soils with high base saturation), (2) by inverse changes in bicarbonate concentrations (>50% of the SAA change) in sensitive lakes with intermediate weathering rates and little soils (low BC exchangeable capacity and elevated terrestrial export of nitrate) and (3) by parallel changes in concentrations of protons and aluminium (each ~20% of the SAA change) in extremely sensitive lakes, with the lowest weathering rates and soil base saturation. The full implementation of the Gothenburg Protocol will not be sufficient to allow recovery of the latter group of lakes, which will remain acidified after 2020.