Contract EVK1-1999-00087 - RECOVER:2010

Part of the 'Sustainable Management and Quality of Water'

Ecosystem Functioning

Directorate General Research

Macaulay Land Use Research Institute logo CEH Logo BIOTEK Logo NIVA Logo CNR Logo Czech Logo IIASA Logo University of Virginia IVL Logo



An assessment of the potential impact of the Gothenburg Protocol on surface water chemistry using the dynamic MAGIC model at acid sensitive sites in the UK

A. Jenkins and J.M. Cullen
Centre for Ecology and Hydrology, Wallingford, Oxon OX10 8BB,UK

Full Reference

Jenkins, A. and Cullen J. (2001). An assessment of the potential impact of the Gothenburg Protocol on surface water chemistry using the dynamic MAGIC model at acid sensitive sites in the UK. Hydrology and Earth System Sciences Vol. 5, No. 3, 529-541.

Summary of Research

The MAGIC model has been systematically calibrated to 12 sites in the UK, which form part of the UK Acid Waters Monitoring Network, using best available data. Predictions for the future are made assuming no further emission reductions from present day (constant deposition at current level) compared to reduced sulphur and nitrogen emission agreed under the Gothenburg Protocol (reduced sulphur dioxide emission by c.80%, nitrogen oxides by c.45% and ammonia by 20% by 2010). In addition, uncertainty in our understanding of future nitrogen dynamics is assessed using 'best' and 'worst' cases of nitrogen leaching in the model. The results clearly indicate the need to achieve further emission reductions in sulphur and nitrogen beyond present day levels to prevent continued surface water acidification. The predictions further indicate that if the emission reductions agreed under the Gothenburg Protocol are achieved by 2010 this will promote a recovery in acid neutralising capacity by 2020 at all sites. Differences between 'best' and 'worst' case nitrate leaching are relatively small, emphasising the need to achieve the sulphur reductions in the shorter term. In the longer term, beyond 2020, increased nitrogen leaching under the 'worst case' leading to further acidification is likely indicating a need for further reduction of nitrogen emissions.

Table 1 Observed and simulated water chemistry

The model successfully simulates observed changes in major ions and acid neutralising capacity over the period 1988 to 2000.
Table 1 shows the calibrated model successfully matches mean 1988-1992 major ion chemistry at all sites and there is no systematic bias in the calibrated determinands.

Site names and descriptions can be accessed from Evans and Monteith 2001

The historical simulations from background (1848) to 1970, representing approximately the peak of S deposition across the UK, show the degree of acidification that is predicted to occurred (Figure 1).Sulphate concentrations in surface waters have increased over the periodalthough not uniformily, but reflecting the location of the sites in relation to emission sources. The historical acidification is demonstrated by the simulated changes in surface water pH and ANC. In general, the biggest decrease in ANC and pH is predicted to have occurred at those sites with the lowest initial pH and ANC and, hence the biggest sensitivity to acidic inputs.

Fig.1.pH 1970 v 1848
Fig.1.ANC 1970 v 1848Fig.1.So4 and NO3 in 1970 v 1848

Figure 2 Simulated water SO4 under constant and Gothenburg protocols

Future predictions to 2020 assume two different deposition scenarios for S; constant deposition at the current level and reduced deposition as predicted by HARM in response to the Gothenburg Protocol (Figure 2). The reductions are assumed to be acheived by the year 2010. These changes in S deposition are reflected in reduced SO4 concentrations in surface waters at all sites. Under constant future S deposition, changes in SO4 concentrations are the result of the assumptions in the model regarding re-afforestation.

Fig 3 Changes in NO3 under worst case scenario

The two different models of N dynamics lead to different surface water NO3 concentrations (Figure 3). The deviation in ANC between 'best' and 'worst' cases is entirely the result of N leaching (Figure 4). For constant deposition at current levels, only the 'worst case' N leaching scenario is presented on the basis that this represents the highest concentrations that can occur. Assuming that N emission reductions are achieved in line with those agreed under the Gothenburg Protocol, however, future NO3 concentrations are predicted to increase by 2020 at all sites, relative to present day, but to remain lower than under constant deposition. Only under the 'best case' of N retention are NO3 concentrations predicted to be lower in future than currently observed, at all sites.

Fig 4 Water ANC under different scenariosThe assumptions made regarding N leaching in the model and the constant and reduced deposition scenarios for N and S promote a wide range of ANC responses both within and between sites (Figure 4). At the higher ANC, less sensitive sites, the predicted ANC concentration shows little change under all three scenarios. At these sites, soil buffering processes (ion exchange) dominate and lower concentrations of acid anions are accompanied by decreased concentrations of base cations and, hence, little change in ANC or pH (high F-factor). At the lower ANC, acid sensitive sites, a substantial further decrease in ANC is predicted for the constant deposition scenario whereas a modest recovery is predicted under the Gothenburg Protocol scenario. In all cases, the predicted ANC in 2020 under 'best', and 'worst' case N leaching is rather similar.