WP 2: Regionalisation of lake chemistry

Targets     Lead Contractor     Lead Partners    

Task 2.1     Task 2.2     Task 2.3     Task 2.4    

Targets

The water chemistry of mountain lakes is influenced mainly by catchment bedrock and soils and by atmospheric deposition and varies significantly across Europe. In this workpackage we relate atmospheric deposition and catchment soils to water chemistry, produce a chemical classification of lakes, and generate data that allow the MAGIC model to be used for scenario testing of sulphur and nitrogen reductions at the regional scale.

Task 2.1.: Extrapolation of chemical deposition

Chemical deposition has been recently measured on site for a number of remote lakes in Europe. We will use these data and EMEP data to extrapolate chemical deposition to all Lake Districts. Deposition to individual sites will be obtained by interpolating data from the four closest EMEP surrounding grid cells (50 x 50 km) with inverse square distance weighting models, and correcting the interpolation using records of measured deposition at our Experimental sites.

Task 2.2.: Regionalisation of soil properties relevant to mountain lake water chemistry

Although soils are poorly developed in mountain lake catchments it is necessary to characterise catchment soils in order to calibrate models. We will select a maximum of 3 lake catchments from each Lake District for detailed sampling and analysis. We will define the main lithological and vegetation units that most strongly correlate with the different soil types present and then characterise each soil type according to depth, bulk density, cation exchange capacity (CEC), base saturation (BS) and carbon-nitrogen (C-N) content. A system for classifying soils within catchments will thereby be obtained, and this will in turn enable the principal data required by the models (CEC, BS and C-N) to be generated for each site.

Task 2.3.: Chemical classification and empirical regionalisation

Water chemistry data already available, together with new data will be used to establish a lake classification both for individual Lake Districts and for the whole of Europe. This procedure will allow the continuous range of variability to be reduced to a few classes and will facilitate regionalisation according to catchment attributes (WP 6) and enable the application of biogeochemical models such as MAGIC7 and SMART.

Task 2.4.: Regionalisation of process-based chemical models

Process-based models, such as MAGIC that simulate water runoff chemistry, are particularly suitable for forecasting changes in lake-water chemistry in mountain regions. The MAGIC7 model has already been tested at some mountain lake sites, and we will use these previous calibrations for regionalisation trials in Central and Southern Norway, Scotland, the Pyrenees and the Tatra Mountains. Of the two regionalisation techniques that have been previously developed we will use the site-specific approach which involves the calibration of MAGIC7 to each lake/catchment data set individually, followed by the aggregation of the results to the regional scale. This technique allows for GIS mapping of the predicted results within each region. Validation of the predictive capacity of the models will be tested using palaeolimnological and, where available, instrumental data. Palaeolimnological data covering the last 200 years are available from all Experimental sites. However, new cores (one per Lake District) will be analysed for a site in Greenland and for sites in the Retezat and Rila Mountains.

Lead Contractor

1. NIVA - Norwegian Water Research Institute, Oslo, Norway

Lead Partners

• UB-DE - Ecology Department, University of Barcelona, Spain

• HBI-ASCR - Hydrobiological Institute, Academy of Sciences of the Czech Republic, Ceske Budjovice, Czech Republic