Are datasets from commercial agricultural analyses sufficiently robust for national-scale monitoring of topsoil indicators?

Soil is a vital natural resource which performs a range of functions including acting as a growth medium for food, fuel and fibre, filtering water, storing carbon (climate regulation) and as a repository for biodiversity. Agricultural soils in Great Britain are subject to a range of pressures which lead to changes in their properties over time that can (despite farmers' best efforts) exacerbate degradation threats such as erosion (loss of the soil resource), compaction and nutrient losses to freshwater. Agricultural soils must also have sufficient nutrients in available forms to ensure they can maintain large crop yields to feed an increasing population. It is therefore essential that we carefully monitor (over time) key properties of British soils to ensure they can continue to function effectively and also that these properties of the soil resource mean it is less likely to be degraded. These key properties include the organic matter content of the soil, the available quantity of two key nutrients for plant growth (phosphorus and potassium) and soil pH (its level of acidity - important in determining nutrient availability and plant growth in general). Government administrations across Great Britain (Defra, Welsh and Scottish Governments) have funded national-scale research to monitor the properties of agricultural topsoil since 1971 and various data have been analysed statistically to report on changes and trends of key soil indicators up to 2007. With declining research budgets, government departments are considering alternative, cost-effective ways to monitor these indicators and one approach is to utilise the soil measurements which farmers pay for, undertaken by commercial laboratories. Although there are many more of these farmer-initiated measurements in comparison to those undertaken using government research funds, the former are not collected to a statistical design and so need to be analysed in a different way to the latter. The farmer-initiated soil measurements may not be representative of soils more generally which could lead to different (biased) results in comparison with the measurements led by government-funded researchers. The aim of our project is to undertake comparisons between the data from farmer-initiated soil measurements and research led measurements to determine the size of any differences in the average values of the key indicators at regional and national scale. We will use statistical procedures which mean we can make consistent comparisons between the two types of data. If the differences we observe are small, it may be possible to use the farmer initiated surveys to monitor soil properties, thereby avoiding the costs of future national-scale, government funded surveys. Rather than approaching individual farmers, we have secured permission to use those data generated by the largest commercial agricultural soil laboratory in Great Britain. We will compare the values of key soil indicators from farmers' data and government-funded research data for the two main agricultural land use types (arable production and grassland) over the last ten years. We will work with, and present our findings to, those in government which have responsibility for soil policy at national scales.