Participatory soil health assessment: exploring soil colour as a predictor of soil organic carbon

Soil organic carbon (SOC) is widely regarded as a central indicator in most soil health assessment frameworks. Despite advances in SOC measurement and monitoring, these methods are not always well understood or accessible to farmers and other stakeholders, which can hinder the adoption of sustainable practices aimed at increasing SOC and improving soil health.

The aim of this project was to investigate how in situ soil assessment, conducted through a participatory approach with farmers, can complement or partially substitute laboratory analyses in establishing robust and reliable baselines for SOC.

Nineteen farmers across the UK assessed soil colour, texture, earthworm abundance, and water infiltration across five fields each, and collected soil samples for laboratory analysis. Samples were analysed for soil organic matter or carbon by a certified commercial laboratory, as well as at university and secondary school facilities, resulting in four independent SOC assessments per sample. In addition, image-based analysis of dry soil colour, and aggregate stability using the SLAKES app were conducted in the laboratory.

While the different laboratory methods were strongly correlated with one another, they showed substantial differences in absolute SOC values. At the individual farm level, these discrepancies could lead to markedly different interpretations of the effectiveness of soil-improving practices. Results showed high variability in farmer-based colour assessments across the 95 samples and moderate correlation with laboratory-measured SOC and other indicators. We will present the best-performing models for SOC prediction using combined farm and laboratory datasets, highlighting soil colour as a low-cost, rapid indicator with potential utility for farmers and other stakeholders.

Overall, this study highlights the inherent variability in SOC assessment and underscores the importance of repeated sampling to establish robust baselines for evaluating SOC trends over time.