Dee catchment, NE Scotland

REFRESH Catchment case studies - responding to future change: Dee catchment, NE ScotlandDee Catchment Cows

The River Dee catchment is a large (ca. 2100 km2), relatively unspoilt area, famed for its salmon fishing, shooting and hill walking. The catchment is subject to significant pressures, including morphological alterations and nutrient inputs from sewage and agriculture, and the area remains a top conservation priority.

Two sub-catchments of the Dee were selected for more in-depth studies: (1) the Tarland Burn sub-catchment, an area of mixed land use in the middle reaches of the Dee was classified as being at ‘Moderate’ ecological status, primarily due to morphological alterations, namely channel straightening and resultant loss or degradation of habitat. Water quality is also of concern, due to diffuse inputs of nutrients and sediments from agriculture; (2) the Loch of Skene sub-catchment in the lower reaches of the Dee, a predominantly agricultural catchment, the main tributary being the Loch of Skene is the Corskie Burn. Both the stream and the lake were classified as having ‘Poor’ ecological status in 2008, due to phosphorus loading from sewage treatment works and diffuse agricultural inputs and barriers to fish migration. High nutrient inputs lead to annual cyanobacterial blooms within the lake.

Modelling future change

A suite of climate, land use and atmospheric nitrogen deposition scenarios were developed, to allow us to investigate the impact of potential environmental change on water quality and ecological status. Three ecological indicators were identified: freshwater pearl mussel (Margaritifera margaritifera), macroinvertebrates and macrophytes. To link hydrochemical model output to ecological impact, relationships were developed between nutrients and these ecological indicators. Empirical relationships were used to link nutrient status and macrophyte and invertebrate response.

Key findings from the scenario analysis are that:

  1. Between now and 2050, climate change alone is not projected to be large enough in this region to bring about significant changes to water quality. Land use change driven by socio-economic factors and climate could have a more significant impact.

  2. In the Tarland Burn, simulated changes in suspended sediment, total phosphorus and nitrate concentrations were large enough to be significant despite uncertainty in model output; phosphate concentrations showed less response. However, the direction of these changes varied, with the potential for water quality to improve or worsen, depending on the land use scenario.

  3. In the Skene sub-catchment, projected changes in climate and land use are very small, leading to little change in simulated nutrient status in either the stream or the loch between now and 2050. As such the loch is predicted to remain at ‘Poor’ ecological status (according to the WFD classification).

  4. Predicted ecological response is highly uncertain, but preliminary findings are that: (i) neither the Tarland nor the Skene sub-catchments are likely to meet water quality targets for the freshwater pearl mussel; (ii) macroinvertebrates in the Tarland Burn are likely to remain in the ‘Clean’ category, with a shift towards the ‘Clean’/’Doubtful’ boundary under some scenarios.

Managing future change

A literature review was carried out to identify a suite of potential mitigation measures that could be used in the two study sub-catchments to reduce nutrient levels and improve water quality. Workshops were then organised, attended by local farmers, land managers, agency representatives and other interested parties. The aim was to discuss key local sources of pollution, the feasibility of implementing a range of mitigation measures and their perceived cost-effectiveness. This resulted in a set of mitigation measures considered relevant for the two sub-catchments.

The hydrochemical models were used to predict the catchment-scale effectiveness of measures to reduce stream nitrate concentrations in the Tarland Burn and stream phosphate concentrations in the Corskie Burn. Estimates of the costs of measures were based on losses to farm income, capital investment (if applicable to a particular measure) and any additional operating costs resulting from implementing management options.

Cost and effectiveness data were combined. The result was a set of the best measures for achieving target nutrient reductions at minimum economic cost. In the Tarland sub-catchment, results suggest that the target is most cost-effectively achieved by: (1) a 20% reduction in applied nitrogen fertilizer to cropland, (2) constructed farm wetland areas, and (3) managed 10 m buffer strips adjacent to arable land. In the Corskie Burn, the target was best achieved by: (1) a reduction in sewage treatment work effluent phosphate concentration to 1 mg SRP-P l-1, (2) a 50% decrease in fertilizer phosphorus application to improved grassland, and (3) a 20% reduction in fertilizer phosphorus application to arable land. Of these, sewage-related measures appear to be the most effective.

The effectiveness of these mitigation measures was then re-tested under scenarios of future climate and land use, to see whether they are ‘future-proof’. Under the worst case scenario of climate and land use change, the measures to reduce nitrate concentration in the Tarland Burn resulted in similar nitrate concentrations to the baseline, as the reduction in fertilizer usage is balanced by the increased area of arable land. However, under the best case scenario nitrate concentrations may decrease by up to 30%. In the Corskie Burn, the projected changes in climate and land use are small enough that the effectiveness of measures is unlikely to change.

Where the cost of meeting environmental objectives is more than the resulting benefits, the WFD may allow less stringent targets to be imposed, or exemption altogether from having to meet targets. To investigate whether this applied in the study sub-catchments, the costs of implementing mitigation measures were compared against non-market benefits resulting from the achievement of good ecological status. The latter were derived through the use of stakeholder participatory techniques and benefit transfer.

Societal benefits were found to outweigh the costs of meeting WFD targets. However, some of the measures may not be affordable for farmers, for example the conversion of arable land to grassland and the creation of 10 m buffer strips. In addition, distributional effects need considering. Often farmers or water companies bear the greatest burden in terms of paying for improvements to water quality, whilst the benefits are enjoyed by the general public, often in a separate location to where the measures were put in place.

Dee Water graphs

Above: A combination of fertilizer reduction and investment/upgrading of waste water treatment works (WWTW) can achieve water quality target with respect to P in the Loch of Skene catchment. Investment in WWTWs appears to be the most effective and economically feasible strategy to deal with P loading problems. Water quality standards for P can be achieved with modest total cost by implementing the combined measures identified in this study (fertilizer input reduction and investment in WWTW).

For further information see REFRESH reports below

Workshop proceedings on collaborative scoping of solutions, Dee catchment, UK

River Dee modelling, final report 

Cost-effectiveness Analysis reports for the Dee catchment, UK, including analysis of proportionality/disproportionality

Synthesis of work at River Dee catchment, UK 

Proceedings of stakeholder workshop for improving water quality in the Loch of Skene and Leuchar Burn sub-catchments, UK.

Waylen, K.A., Blackstock, K., Cooksley, S. (2011).  Land-manager contributions to protecting the Dee Water Environment.  REFRESH deliverable 1.14, The James Hutton Institute, Aberdeen, UK.