Soil organic carbon sequestration: scoping study

4 Datasets for Scotland with measurements of soil carbon

Appendix 1 lists some of the datasets that have measurements of SOC in Scottish soils and these will be used throughout to provide examples of the types of data that can be used to answer the questions regarding how much SOC there is and where, has it changed over time and what are the potential for loses or gains. Those datasets marked * in Appendix 1 are component parts of the ‘Scottish Soils Database’ held by the James Hutton Institute.

A total of 37 datasets have been identified. Some of these datasets (Appendix 1) cover a wide range of soil types and habitats such as the Representative Soil Profile dataset, the National Soil Inventories of Scotland (NSIS), held by the James Hutton Institute and the Countryside Survey (CEH- Centre for Ecology and Hydrology). Other data sets are more restricted to specific land uses, for example, the BioSoil dataset for Woodlands or the Trends in pollution of Scottish Soils for upland habitats.

Appendix 1 also gives information on the extent (National, Regional or Local) of the data and an indication if the soils have been resampled so that change over time can be assessed (for example, NSIS 1978-88 and NSIS 2007-9, Countryside Survey). It also gives information about whether SOC stocks can be calculated, an indication of the depth over which the soils were sampled and the strengths and weaknesses of each dataset. While there are numerous datasets in Scotland where SOC concentrations have been measured there are fewer where SOC stocks have been calculated.

Other data sets that were collected for specific purposes, such as an assessment of aggregate stability or to develop soil health testing (Soilbio) represent specific land use types. These can be used to explore relationships between carbon and other factors such as fertility, erodibility or water retention but also, they hold valuable information on SOC contents for different land uses and land management.

As can be seen from the list of datasets in Appendix 1, Scotland has a wealth of soil data due its long mapping and soil characterisation programme and the long history of soil research, much of it primarily funded by Government.

4.1 Limitations in the datasets

In general terms, datasets that limit the sampling depth to less than the thickness of most topsoils, for example, to 0-15 cm are less useful than those that sample over the full thickness of the soil (for example, to 100 cm) or at least to the full thickness of the topsoil where fixed depth sampling to 15 or 30 cm depth would have resulted in a false result. This is also the case in soils with organic surface layers that comprise almost entirely of organic material, where changes in depth are often the main way that carbon is lost or gained.

A disadvantage in some of the datasets, for example, the Representative Soil Profile database, is the widespread age of the data (from 1940s to present). Changes in analytical methods over time can affect the measured values. Chapman et al. (2013) illustrated this when assessing change over time using the National Soil Inventory data. They found that reported SOC concentrations in the soils sampled in the late 1970s and early 1980s were 11.5% greater when reanalysed. Subsequent analyses of early samples going back to the 1960s, also showed similar differences between old and new analyses. In an internal report, Chapman (2017) suggested multiplying the older SOC concentration data by around 0.9 to align the values with current measurements. Interestingly, Chapman et al. (2013) did not see a similar change in the Loss on Ignition (LoI) values when the soils were reanalysed indicating that the difference was due to changes in methods and/or the instruments used to measure SOC.

The limitations of differences caused by analytical methods or equipment can also be overcome by reanalysing the original sample alongside new samples to ensure consistency in the method. This was the approach taken by Chapman et al. (2013) and by Lilly et al., 2019 and highlights the benefit of archiving sampled soils for future analyses.