Recovery of acidified waters in the UK

WP 2: Assessment of "Stock at risk"

WP2 formed a large body of research based around two tasks, one focusing on lake status (Task 2.1) and one on stream status and episodes (Task 2.2). The research conducted on lakes status is summarised below, with the results of Task 2.2 being available on a separate page.

Task 2.1: Regional assessment of lake status

Identification of under-representated regions, updates to FAB mapping dataset and stock at risk assessment

GIS techniques were used to identify three regions potentially "at risk" from acidification and under-represented in the critical loads database. Sampling programmes were undertaken in these regions, with 21 sites in the Surrey and Sussex Heaths plus 19 in the New Forest, both in southern England, and 30 sites in the Trossachs area of central Scotland. In another GIS exercise using digital boundaries for SACs, 29 lakes in SACs designated for freshwater features were also randomly selected and sampled.

For the 2004 CCE call for data, the addition of extra dynamic modelling regional datasets along with the new data collected under this programme resulted in an increased mapping dataset from 1044 Great British sites (lakes and streams) in 2003 to 1595 in 2004. In Northern Ireland the number increased from 119 to 127. A screening criterion to remove sites where seasalt contamination may have led to incorrect critical load exceedance calculations was applied beforehand, reducing an original total of 1797 UK sites to 1722 in the submitted dataset.

The new ANC critical limit of 20 µ eq L-1 agreed at the stakeholder workshop in February 2004, with 0 µ eq L-1 in exceptional cases, was used in FAB applications using a revised version of the model (see Work Package 1). A preliminary critical loads assessment used deposition datasets for 1970 (worst-case), 1995-97 ("present") and 2024 (NECD - best-case) with the pre-screening database of 1797 sites. Stock at risk was assessed on a regional basis using individual subproject (generally regional) site designations.

The least impacted groups with the lowest proportion of critical load exceedances included the CLAG seasalts sites in NW Scotland - due primarily to very low deposition, plus the Scottish Random Survey of 1995 and the 10km Grid Survey of Northern Ireland in 2000, in both latter cases because of the inclusion of a large proportion of non-sensitive sites in the survey datasets. The greatest proportions of exceeded sites were found in the Pennines, Mournes, WAWS lakes, GANE Snowdonia lakes, Galloway and Southern England. In 1970, 15 of the 20 mapping subprojects showed exceedance in more than half of their sites using ANCcrit = 20 µ eq L-1. By 2024, only six groups of sites showed exceedance in more than half of their population. Dramatic improvements in terms of reduced critical load exceedance are therefore apparent for 2024 from the baseline "worst-case" year of 1970. At the country-wide scale using ANCcrit = 20 µ eq L-1, exceedances in England are reduced from 267 sites (67%) in 1970 to 173 sites (43%) by 2024. In Scotland, 402 exceedances (43%) decline to 137 (15%); in Wales 269 (78%) declines to 112 (33%) and in Northern Ireland, 26 exceedances (20%) declines to 17 (13%). For the UK as a whole, exceedance is reduced from 964 sites (54% of the sampled population) to 439 (24%). Note that these populations of sites are not necessarily representative of the country as a whole (except perhaps in Northern Ireland) as site locations in many datasets are biased towards the most acid-sensitive areas.

In April 2004, revised deposition datasets for 1995-97, 1998-2000 and 2024 were released by CEH Edinburgh, plus a new dataset for 1999-2001. These datasets were used to recalculate critical load exceedances for the screened mapping dataset of 1722 sites submitted to the CCE mapping programme. The new data for 2024 in particular led to some changes in the proportions of exceeded sites. In England, a current (1999-2001) exceedance in 208 sites (53%) is reduced to 186 (47%) by 2024. In Scotland, 244 current exceedances (29%) decline to 182 (21%). In Wales, 176 exceedances (51%) decline to 135 (39%). In Northern Ireland, 20 exceedances (16%) are reduced to 18 (14%). In the UK overall, 648 current exceedances (38%) are reduced to 521 (30%) by 2024. This is a larger proportion of exceedances than found in the previous modelling exercise, and this is due primarily to the revised 2024 deposition dataset used.

Acidification in lakes of conservation interest

Of the 29 conservation sites sampled, 10 exceeded their critical loads in 1970. Six sites are still exceeded at present and are still exceeded by 2024. The NECD therefore appears to be of limited success in protecting conservation sites in the UK, with 20% still exceeding critical loads, if it is assumed that the small subset of conservation sites sampled are representative of the whole UK population.

Palaeolimnological study of recovery within a region

Palaeolimnological studies of the recent sedimentary records of five lakes from the Galloway region of Scotland were undertaken. The work demonstrated that four of the five study sites appear to show consistent fluctuations in diatom inferred pH suggesting that the diatoms are responding to some regional forcing factor. None of the studied sites has seen a significant increase in pH over the period studied and that this is inconsistent with the measured hydrochemistry, which does show a consistent and significant increase in pH since the early 1980s. Further work is required to understand the reasons for the discrepancy between the measured pH recovery and the lack of response shown in the diatom sediment samples. The effect of climatic fluctuations associated with, for example, the North Atlantic Oscillation, and the observed increases in dissolved organic carbon may all act to mask diatom recovery and untangling these effects will require a greater understanding of diatom ecology and responses to these other forcing factors.

Palaeolimnological assessment of change in sites of conservation interest

A total of 42 sites of conservation interest (SACs) were analysed to determine the degree of floristic change between reference conditions and present day in sedimentary diatom assemblages. The results of the core top-bottom analysis show that many acid sensitive lakes in designated SACs have experienced substantial change in their diatom communities over reference conditions. These results indicate that 62 % (26 out of 42 sites) of sites in SACs have undergone substantial change, with a further 28% (12 out of 42 sites) having undergone moderate change. Only 10% of the study sites illustrate only minor or very minor change in their diatom communities over reference conditions. The results indicate that a considerable proportion of sites of conservation importance may have been adversely affected by acid deposition and that substantial biological change has taken place.

Monitoring of the Galloway cluster lochs

Long-term monitoring of the Galloway cluster lochs has now continued for over 25 years, although earlier sampling programmes were irregular. The sites show continued recovery from acidification and are at their healthiest in terms of water quality since monitoring began in 1978, with Loch Enoch reaching a mean annual ANC > 0 µ eq L-1 for the first time. Nitrate concentrations have remained stable, but the rapid decline in non-marine sulphate since 1994 has leveled off in recent years, with unknown consequences for future recovery.

National scale applications of biological predictive models

Of the large number of predictive models derived under Task 1.2, four were selected for application to national and regional datasets, to predict the probability of occurrence of the diatom Achnanthes minutissima, the mayfly Baetis spp., plus the sums of acid sensitive diatom and invertebrate taxa. Predictions were based on steady-state ANC for pre-acidification, 1970 and 2024 deposition levels derived using the SSWC model (F-factor predictions). Using a probability threshold of 0.3, the likely distribution of the diatom Achnanthes minutissima was reduced by more than a third of sampled sites in 1970 relative to pre-industrial times, with only partial recovery by 2024. For Baetis mayfly species, probable distribution was reduced by around 20% of sampled sites in 1970 but predicted recovery by 2024 is more widespread than for Achnanthes minutissima. However, using the sums of both acid-sensitive diatom and invertebrate taxa, greater reductions in probable distribution were found for invertebrate groups, while recovery was also slightly less marked by 2024. The same exercise was also carried out for a probability of occurrence of 0.5. Taken together, modelling results indicate that 10-20% of sites in the mapping dataset will fail to reach the baseline biological targets by 2024. Water quality will not have improved sufficiently to support acid-sensitive diatom and invertebrate taxa in geologically sensitive parts of the country. A schematic, conceptual framework has now been produced linking palaeoecological transfer functions (to hindcast water chemistry from biological sediment records using modern chemical-biological relationships), analogue matching (use of current biological communities in analogue sites that match fossil communities in acidified lakes to indicate restoration targets for impacted sites) and ecological response models (modern chemical-biological relationships used to predict biological response from hindcasts or predictions of chemistry). This framework provides the basis for the assessment of recovery towards biological targets and time to "gap closure" relative to pre-industrial conditions.