7. The Land Cover Map of Scotland

Why a land cover map of Scotland?

Over 50 percent of the land area of Scotland is covered by semi-natural vegetation of which heather moorland and peatland are the predominant types. Less than 3 percent is urban or rural development. This is a distinctive characteristic of Scotland in relation to the other parts of the UK. Nowhere else has such a high proportion of semi-natural land cover or so little built-up land. Many parts of Scotland are of high nature conservation value, however, Scotland also has a very low proportion of native woodlands. Broadleaved, mixed and native conifer woodlands account for only around 2 percent, whereas (mainly exotic) coniferous plantations make up 12 percent.

These statistics raise questions about what we expect the countryside of Scotland to look like, and how it is changing. The only realistic way we can address these questions is to make a commitment to routinely measuring and monitoring the stock and health of the land resources of Scotland. This commitment is a key element of State of the Environment Reporting, which is central to the ethos of sustainable development.

In 1987, the Scottish Office made the commitment to provide a national baseline survey of the land cover of Scotland. This came about as a result of concern about the nature and rate of change in the Scottish countryside, particularly in relation to urban expansion, and the intensification of agricultural practices which had produced visible changes in the character of agricultural landscapes through loss of stone dykes and hedgerows. Also, changes in the economics of hill and upland land use systems had also meant a decline in traditional management practices like muirburn (controlled burning of heather moorland as a red grouse habitat) and a progressive increase in stocking densities. These were being blamed for the decline in the area of heather moor. Most dramatic was a rapid loss of low intensity agricultural land to commercial forestry.

A backdrop for the 1980s debate about the Flow Country of Caithness and Sutherland was the growing appreciation of the global significance of peatland, mainly in terms of scenic and nature conservation importance, in contrast to its low productive worth for agriculture. Tax incentives for private forestry (withdrawn in the 1988 Budget) on land of marginal value for agriculture meant that large tracts of peatland were drained and planted to conifers20.

What was the nature and extent of these land use changes? Where were they most intense? The uncomfortable answer was that we did not know! The land cover of Scotland survey was born out of ignorance about the state and condition of the land cover of Scotland. This level of ignorance was politically unacceptable given the concern being voiced by the conservation bodies, the government agencies themselves, the public and the media.

In the mid-1980s we did not know much about the land cover of Scotland, particularly how it was changing. This ignorance was heightened by the concern about major changes, such as afforestation.

How the land cover of Scotland was mapped

The LCS survey was the first national census of land cover in Scotland. It was based upon the manual interpretation of specially flown 1:24,000 aerial photography. Interpretation was undertaken by an experienced team of photo-interpreters at the Macaulay land Use Research Institute (MLURI) in Aberdeen. They used mirror stereoscopes to view photo pairs in stereo. Lines were drawn around land cover features on acetate overlays. These boundaries were then transferred to a 1:25,000 OS Pathfinder Series base map.

The advantage of this approach was that each team member was able to use their field experience of the areas in the interpretation process. This assured a higher level of accuracy than would be obtained from interpreters who had no field experience.

Air photo interpretation is dependent upon good knowledge of the area. Try mapping somewhere familiar then somewhere unfamiliar.

The classification system used in the LCS survey is hierarchical, with four levels of feature. There are 6 principal features and a total of 126 land cover features altogether. But below the minimum mapping resolution of the LCS project (10ha semi-natural; 5ha built-up; 2ha woodland) there is additional complexity. So the concept of a mosaic was introduced, allowing the interpreters to map mixtures of features.

Mosaics are defined as mixtures of two land cover features where patches of both are below the minimum mapping area criterion for separate identification. The recognition of mosaics in LCS expanded the number of features from 126 to 1323. Several of these mosaics proved to be extensive, particularly those including heather moorland. For example, mosaics of heather moorland and peatland account for nearly 16% of the land area. Whilst 99% of the land area of Scotland is accounted for by 354 single or mosaic features, the remaining 1% has no less than 969 different features!

Mosaics of land cover features can be used to more accurately depict local heterogeneity in land cover. They do raise problems for calculating areas. Are thematic maps of this type ever “right”?

Once a map sheet had been completed it was digitised (converted into electronic form) using a high resolution digitising tablet. The 1000 OS sheets were digitally ‘fused’ to create a single land cover database. The whole process from the acquisition of the first aerial photo to the completion of the digital land cover dataset took over 5 years and cost over £1M. The benefits have, however, been considerable and the data are being used in a wide range of environmental issues. These include:

The land cover data can be used to produce single feature maps or may be used in combination with other datasets in a GIS to explore resource planning issues. For example, options for disposal of sewage sludge requires information on land suitability, current land cover, distance from major urban centres, proximity to dwellings etc.

Historical look back studies have been possible using a comparison of the 1988 photos with 1946/47 photos for part of the Cairngorms and in part of the Central Valley. These have shown real and significant changes in these areas principally:

Cairngorms

  1. 900% increase in coniferous plantations
  2. 25% decline in native pinewoods
  3. progressive loss of heather moorland

Central Valley

  1. 300% increase in built-up land
  2. 400% increase in opencast mineral extraction
  3. 300% increase in coniferous plantation
  4. 22% decline in broadleaved woodland

What do we now know about the land cover of Scotland?

The LCS88 survey indicates that the land area of Scotland is 78,828 km2. Of this, over 50 percent is covered by semi-natural vegetation of which heather moorland and peatland, and their mosaics, are the most extensive features.

The figures also reveal how little is classed as urban (1.8%) or rural development (0.6%).

Woodlands occur predominantly as single features, with coniferous woodland (7.8%) and recent planting (3.9%) accounting for the greatest area. Broadleaved, mixed and semi-natural coniferous woodland account for 2.5%, which is a remarkably small area given that these represent the natural climax vegetation over much of the lowlands and uplands of Scotland.

Note the very low percentages of semi-natural woodlands. This reflects our long history of woodland clearance and also grazing which often suppresses natural regeneration. Is coniferous plantation a substitute?

One significant concept in landscape ecology is that of heterogeneity. A measure of heterogeneity can be gained from the area of features within mosaics, covering 30% of Scotland. From this it can be appreciated that semi-natural cover types tend to be highly heterogeneous. On the other hand, intensively managed areas, like arable, and coniferous plantation, are very homogeneous. Whilst it must be remembered that heterogeneity is very scale-dependent (e.g. most of the heterogeneity in arable areas is associated with fencelines, which were not mapped in LCS88), it can be taken as a general principle that intensively managed land cover types have little heterogeneity at the landscape scale.

LCS88 data reveal that when semi-natural areas become more intensively managed (e.g. heather moorland planted to coniferous woodland) there is a loss in heterogeneity at the landscape scale. This has potential effects on the nature and range of niches available for wildlife and plants.

Lessons from LCS88

The LCS88 has provided us with new insights into the state and condition of Scotland’s land cover resources. It has allowed us to confirm and qualify many of the beliefs that we might have held intuitively. However, there are also surprises, particularly the very low stock of semi-natural woodlands. These data provide us with a firm basis for debate about Scotland’s land use futures.

  1. See for instance: Stroud, D.A., Reed, T.M., Pienkowski, M.W. & Lindsay, R.A. (1987). Birds, Bogs and Forestry: The Peatlands of Caithness and Sutherland. Nature Conservancy Council, Peterborough.