Comparison of different critical load approaches for assessing streamwater acid-sensitivity to broadleaf woodland expansion

Abstract

Due to its potential adverse effects on freshwater acidification, risk assessments of the impacts of forest expansion on surface waters are required. The critical load methodology is the standard way of assessing these risks and the two most widely used models are the Steady-State Water Chemistry (SSWC) and First-order Acidity Balance (FAB) models. In the UK the recommended risk assessment procedure for assessing the impact of forest expansion on freshwater acidification uses the SSWC model, whilst the FAB model is used for guiding emission policy. This study compared the two models for assessing the sensitivity of streamwater to acidification in 14 catchments with different proportions of broadleaf woodland cover in acid-sensitive areas in the UK. Both models predicted the exceedance of streamwater critical loads in the same catchments, but the magnitudes of exceedance varied due to the different treatment of nitrogen processes. The FAB model failed to account for high nitrogen leaching to streamwater, attributed to nitrogen deposition and/or fixation of nitrogen by alder trees in some study catchments, while both models underestimated the influence of high seasalt deposition. Critical load exceedance in most catchments was not sensitive to the use of different acid neutralising capacity thresholds or runoff estimates, probably due to the large difference between critical load values and acidic deposition loadings. However, the assessments were more sensitive to differences in calculation procedure in catchments where nitrogen deposition was similar to the availability of base cations from weathering and/or where critical load exceedance values were <1keqH(+)ha(-1)yr(-1). Critical load exceedance values from both models agreed with assessments of acid-sensitivity based on indicator macroinvertebrates sampled from the study catchments. Thus the methodology currently used in the UK appears to be robust for assessing the risk of broadleaf woodland expansion on surface water acidification and ecological status.

Assessing the effect of broadleaf woodland expansion on acidic dry deposition and streamwater acidification

“…However negative ANC values were recorded in three of the study catchments on some occasions, suggesting that the fish population remained at risk from acid episodes. This was in line with the results of streamwater macroinvertebrate surveys (Gagkas et al, 2010). Streamwater marine SO 4 , non-marine SO 4 (xSO 4 ) and NO 3 concentrations were relatively low, with marine SO 4 concentrations exceeding xSO 4 concentrations, reflecting a maritime dominance of atmospheric deposition (the nearest coast is 57 km to the west).…”

“…Lydersen et al (2004) indicate that the ANC crit value selected should take account of the organic carbon concentration). However, in a sensitivity analysis of CLs calculated using the FAB model, Gagkas et al (2010) showed that, for 14 acid-sensitive UK upland catchments, the CL values were relatively insensitive to different runoff estimates and different ANC crit values; the use of different values did not shift the status of any of the catchments from protected to exceeded.…”

“…Improved representation of many terms, which are currently formulated rather simply, could be accommodated within the FAB model structure, for example, the influence of dissolved organic carbon (DOC) can be accommodated through reformulating the calculation of ANC (Lydersen et al, 2004). Other FAB model terms whose representation could be improved include: (1) N leaching (Kernan et al, 2010;Gagkas et al, 2010);…”

“…(2) anion and base cation inputs in seasalt events (Cresser, 2007;Gagkas et al, 2010) which result in lower ANC values than expected; (3) DOC, which is widely increasing in concentration and may affect the acidity and recovery of freshwaters (Curtis et al, 2005).…”