Modelling the impact of acid deposition on the hydrochemistry of the Loch Dee catchments, S.W. Scotland

Abstract

This work describes an investigation into the impact of acid deposition on the hydrochemistry of the three main tributary catchments at Loch Dee in Southwest Scotland. The research covers two main objectives. First, the hydrochemical processes that determine the observed streamwater chemistry are examined empirically through the determination of the hydrochemical budgets for each sub-catchment. Second, the ability of the Integrated Lake Watershed Acidification Study model (ILWAS) to simulate the observed hydrochemical processes is evaluated. From the hydrochemical budgets two major factors are identified as responsible for the spatial variability in the streamwater chemistry at Loch Dee. These factors are the underlying geology and land-use management techniques. The role of afforestation is difficult to judge due to the immaturity of the forest. However, the budgets indicate that any influence is a result of the pre-afforestation ploughing and drainage rather than the presence of the trees. On a temporal basis the hydrochemical budgets show considerable variability on both a monthly and an annual timescale, with the variability in the streamwater outputs of bases and nutrients primarily related to the variability in precipitation quantity. The budgets also indicate that, particularly on an annual timescale, the dry deposition of sulphate and hence of acidity to the catchment varies considerably. Furthermore, on a monthly timescale, the temporal variation in the budgets indicates that the conservative ion chloride can be physically stored within the catchments. This finding has severe implications for the estimation of dry deposition inputs and for the utilisation of simple hydrochemical models. The ILWAS model is generally able to simulate the hydrological response of both catchments. However, the chemical simulations of both catchments are considerably smoothed and bear little resemblance to the observed data. This failure is ascribed to the large number of input variables for which site-specific data are not available, together with the use of monthly averaged precipitation quality to drive the model. Sensitivity analysis of the model indicates that the most critical input variables are the soil depth and soil solution chemistry, and that the majority of the input variables can be considered as calibration parameters. Given the subsequent large number of calibration parameters it is suggested that a unique calibration of the model will rarely be possible. Consequently, it is concluded that the primary use of the ILWAS model will be as an explanatory tool for examining conceptual ideas about the hydrochemical processes that determine acidification. The model's value as a management tool to help mitigate the effects of anthropogenically derived changes in surface water chemistry will be extremely limited.