Developing a method to estimate the costs of soil erosion in high-risk Scottish catchments: final report

3. The case study catchments

An existing soil erosion risk map (Lilly et al., 2002; Lilly and Baggaley, 2014), covering about 95% of cultivated land in Scotland was used to identify representative catchments to take forward as case studies (Figure 1). The inherent erosion risk is related to slope gradient, soil texture and the ability of the soil to absorb rainfall (thus reducing the likelihood of runoff generation). The latter is based on the HOST (Hydrology of Soil Types) classification (Boorman et al. 1995).

A description of the erosion risk classes (and the map) can be found on Scotland’s soils website (https://soils.environment.gov.scot/maps/risk-maps/map-of-soil-erosion-risk-partialcover/). For soils with mineral topsoils, the risk of soil erosion is classified into 3 main classes: Low, Moderate or High. Each main class is then divided into 3 subclasses (L1-3, M1-3 and H1-3). The higher the number in each class indicates a greater risk of erosion within that class. This is dependent on slope gradient, soil texture and the ability of the soil to absorb rainfall (thus reducing the risk of generating potentially erosive runoff). Organo-mineral and peat soils are also divided into 3 classes: Low, Moderate or High. The Low and Moderate classes are then subdivided into subclasses (Li-iii and Mi-iv), again with the higher the number in each risk class indicating a greater risk of erosion within that class (Figure 2).

If available, data on erosion rates, impacts, mitigation and costs were sourced from the literature review. Examples of the datasets available are shown in Figure 3. These observations were captured in a GIS and plotted onto the map (Figure 3). The aim was to relate the observed rates to soil type and land use (and other characteristics if known) to allow ‘extrapolation’ to other areas with similar site characteristics. However, it should be noted that only a very few of these sites have quantified data.

A long list of 55 potential catchments with information regarding their size, soil types, land uses (based on the Land Cover of Scotland (1988) classes (i.e. arable, moorland, improved grassland etc.) and soil erosion risks (classified as high, moderate or low) was presented to the Steering Group (Table 2). The location, land cover, Soil Series and soil erosion risk were sourced from relevant datasets (Figure 3).

A selection of 10 catchments (Table 2; Figure 1) was made to reflect a broad and representative range of soil types, land covers (including different proportions of different land covers in each catchment) and erosion risks, ideally accompanied with observations and surveys. For example, the Coyle (NS395214) in Ayrshire and the East Pow (NO069256) in Perthshire are contrasting catchments, representative of differing land use, soils and water quality issues. They were used in a pilot study to develop and demonstrate the soil erosion risk classes based on existing 1:25,000 scale soil maps (see https://soils.environment.gov.scot/maps/). They are both sub-catchments of SEPA’s priority catchments (http://www.sepa.org.uk/water/river_basin_ planning/dp_priority catchments.aspx).

Geographical spread was also considered in the selection process, to reflect climate variability. The locations of the catchments are shown in Figure 1. Considering the resolution of data available in these catchments (Table 2), the area covered by the current 1:25,000 soil erosion risk maps is shown on the top left map (in darker grey) in Figure 1. Six of the 10 catchments lie wholly within the detailed Soil Map of Scotland (partial cover), while the others have some areas currently not mapped at that scale. Even so, it is still possible to apply the methodology (see Section 6) to these areas, with the proviso that this would be at a lower resolution and with increased uncertainty, as it would be based on the National Soil Map of Scotland soils data.

Table 2. Long list of potential catchments

Catchment (short listed)

• Coyle
• Pow
• Lunan Water
• River Ugie
• River Earn
• River Forth
• Water of Girvan
• River Tweed
• River Irvine
• River South Esk (Tayside)

Catchment

• Moray Coastal
• River Nairn
• River Ythan
• Wick River
• River Esk (Lothian)
• Cromarty Coastal
• River Doon
• Dumfries Coastal
• Galloway Coastal
• River Lossie
• River Leven (Fife)
• Dundee Coastal
• Forth Estuary (South) Coastal
• Thurso Coastal
• Stewartry Coastal
• River North Esk (Tayside)
• North Ayrshire Coastal
• East Lothian Coastal
• River Dee (Grampian)
• Stirling Coastal
• River Don
• Bervie Water
• Kincardine and Angus Coastal

• Buchan Coastal
• Whiteadder Water
• River Esk (Solway)
• Inverness Coastal
• South Fife Coastal
• River Clyde
• River Almond
• Earn Coastal
• Banff Coastal
• River Nith
• Eye Water
• River Garnock
• Urr Water
• River Eden
• Lochar Water
• River Dee (Solway)
• River Tyne
• River Ayr
• Dighty Water
• River Deveron
• River Tay
• North Fife Coastal
• River Annan

The consensus from the 1st Steering Group meeting was that around 4 - 5 catchments would be selected for the economic analysis of soil erosion rates and associated impacts. The following criteria were considered in selecting the final 5 case study catchments for the project:

• geographical spread (to reflect climate variability);
• representative land uses;
• availability of soil erosion observations;
• range of ecosystems goods and services affected by erosion (e.g. cultural impact, archaeology, historical sites, etc.)
• designated areas of special interest.

Some coastal catchments were not selected because of the small size of catchments included. In consultation with the Steering Group, 5 catchments were selected (Table 3). Maps of land use, Erosion Risk Classes and IACS data for each of the case study catchments are given in Figure 4, Figure 5 and Figure 6 below.

Table 3. Selected catchments

North / North East
Ugie (agricultural runoff, peatland areas)

Central
South Esk (selected for impact on cultural services e.g. bathing water, tourism, fishing, shellfish)
Pow

South East
Lower Tweed (large upland areas; few observations available, although some SEPA data and from Spiers and Frost). Selected for the impact on cultural services.

South West
Girvan (20% forested). There was some debate as to whether soil erosion was an issue in forested areas. The National Forest Inventory (NFI) can identify areas recently felled and newly planted, which are likely to generate erosion. Of these areas, it is assumed that only a small % erode in any one year. However, the NFI does not identify other ‘erosion generating activities’ i.e. haul roads, drainage ditches etc.