Freshwater Umbrella Research

Acidification of Freshwaters: the role of nitrogen and the prospect for recovery (CLAM)

June 1998 — May 2001

Report Availability

The final report can be downloaded from the resources section of the website as a series of 4 volumes. A separate executive summary of the report and the 3 report volumes are available as PDF files.

Adobe pdf icon Download the CLAM Report volumes

Project Summary

The CLAM project, completed in May 2001, was a large research programme with the remit of tackling three key areas of research:

  • Modelling N retention and leaching
  • Measuring and modelling recovery
  • Metal deposition and cycling at Lochnagar

A brief summary of the main findings of the CLAM project are presented below.

Modelling N retention and leaching

This work package addressed some of the problems associated with modelling potential impacts of N deposition on the acidification status of upland waters. Intensive field measurements and laboratory based experimental studies, in combination with extensive regional survey data, have provided further insights into the processes regulating the storage and leaching of N within upland catchments.

Denitirification was found to cause only a minor loss of deposited nitrogen, and the model recommended by the UNECE Mapping Manual for critical loads produces denitrification rates far higher than those observed. It is therefore recommended that the low, soil specific, literature derived values for dentirification produced under the Terrestrial Umbrella programme for critical loads should continue to be used in freshwater critical loads modelling in the UK.

Low denitrification rates in upland catchments provide further evidence that the high observed rates of N retention must be due to immobilisation processes in soils. Extensive regional survey work and intensive sampling within the Lochnagar catchment showed a high degree of spatial variability in soil C and N content both within and between different soil types. In peaty soils much of the organic matter is chemically inert, which may explain why the whole soil C/N ratio is found to be a poor predictor of N leaching from moorland soils. However, intensive studies down soil profiles at four experimental catchments showed that the C/N ratio of the top 5cm of soil profiles, which generally comprises the active soil organic horizon, may be much more closely linked to soil N processes such as mineralization, nitrification and immobilization and may have some potential as an indicator of N leaching, particularly from non-peat soils.

MAGIC modelling work has shown that changes in whole soil C/N ratio, which were thought to be a prerequisite for the high rates of nitrate leaching predicted by the steady-state FAB model, would take perhaps centuries to attain, but future dynamic modelling work will have to focus on adequate representation of the key factors controlling N processes in the active soil C and N pools, which are likely to be much smaller than the inactive pools. Problems of catchment heterogeneity might be addressed through a novel method of calibrating MAGIC to landscape types rather than whole catchments. If landscape units can be shown to have more homogenous C and N characteristics, this approach could have great potential for more accurate simulation of the dynamics of N leaching. However, the importance of current commitments to reduce S deposition under the Gothenburg Protocol were demonstrated by MAGIC7 modelling which showed that planned S reductions have a greater impact on predicted ANC than even the worst case N leaching scenario over the next 50 years.

Measuring and modelling recovery

Reviews of existing studies showed that biological recovery from the large-scale, long-term impacts of freshwater acidification could require proportionately long time scales, perhaps centuries. Attempts to measure and model recovery over more practical timescales should therefore consider damage reversal towards some target endpoint rather than true repair or restoration. In aquatic systems several factors could cause a time lag in recovery, and experiments at Llyn Brianne demonstrated that hysteresis in recovery after liming reflects limited persistence of organisms rather than restricted dispersal. The continued effects of acid episodes in systems undergoing chemical recovery could be a major factor in this hysteresis. This is demonstrated by the significant improvement of regression models which predict biological status from modal chemistry in WAWS stream sites when indices of pH and aluminium episodicity are included.

Data from the WAWS and the AWMN demonstrated that between-year variation in biological communities (invertebrates and diatoms) may be linked to climatic fluctuations driven by the NAO and this could confound attempts to detect biological recovery. Evidence for slight recovery in primary producers has been found in high resolution, integrated sediment trap and sediment core diatom data from some of the Galloway cluster sites, but the data further support the conclusion that biological recovery is likely to be slow.

Recovery targets and management strategies for acidified freshwaters are likely to be driven increasingly by the WFD, and require knowledge of reference conditions prior to acidification. Since there are very few data on the pre-acidification status of surface waters for either chemistry or biology, other means have been developed for identifying recovery targets. This programme built on the analogue matching technique for diatoms, whereby their fossil remains in the sediments of acidified lakes are matched to modern assemblages in other (analogue) lakes, to link the current chemistry of the analogue site to the historic, pre-acidification chemistry of the damaged site. The technique has been expanded to include cladoceran remains as well as diatoms, and to identify biological as well as chemical analogues. With this method, the biology of analogue sites can be used to identify potential biological targets for recovery in acidified lakes. The method is restricted to lakes because of the need for a fossil sediment record. It was found to be robust and reliable for a number of lakes from the AWMN, with analogues identified in the less impacted areas of north-west Scotland.

In a complementary approach, logistic regression models to predict biological status from modelled chemical status were further developed for different organisms. In this way, MAGIC has been used to make biological predictions for several regions of the UK under the Gothenburg Protocol, and predictions indicate only partial recovery at just half of sites in some regions. Furthermore, the approach has not yet been developed to account for the problems of biological hysteresis following chemical recovery mentioned above.

Metal deposition and cycling at Lochnagar

A scoping study produced under this programme summarised evidence that emissions of metals are generally decreasing but that meteorological conditions exert a major influence and must be taken into account in trend analysis. At Lochnagar, mercury deposition appears to have increased over the last 3 years. There is also evidence that some metals are largely retained in catchment soils there, so that climatic changes could affect lakewater levels of metals, for example through changes in soil erosion. Mosses were found to best reflect trends in deposition of metals, while sediment trap techniques were shown to have great potential for monitoring lakewater metal trends while accounting for catchment inputs.

Project Reports
  • Kernan M. and Curtis C.J. (Eds) (2000) Critical loads of acidity and metals: Interim report to the Department of Environment, Transport and the Regions (Contract No. EPG 1/3/117). ECRC Research Report No. 64, ECRC, University College London, London, UK, 63pp.
  • Curtis C.J., and Simpson G.L. (Eds) (2001) Summary of research under DETR contract "Acidification of freshwaters: the role of nitrogen and the prospects for recovery", EPG 1/3/117.. ECRC Research Report No. 79, ECRC, University College London, London, UK, 4 volumes.