Plant specific biomarkers help to identify erosion and soil carbon loss hot spots in river catchments

Identification of soil erosion “hotspots” is critical for sustainable land and water management. Soils are the largest carbon pool on Earth and provide vital ecosystem services, including biomass production, grazing land, forestry and water filtering. The ability to store carbon and absorb water makes soil an indispensable part of climate change mitigation and adaptation. However, soil erosion compromises soil function, leading to loss of fertility and damaging off-site problems due to sedimentation, including water pollution and damage to infrastructure and aquatic environments.

As soil erosion reflects the combined effect of land management intensity and soil properties, identifying erosion hotspots and the sources of soil organic carbon (SOC) in waterways is challenging. Hence, a PhD study by Dr Catherine Wiltshire, funded by the NERC STARS Doctoral Training Partnership and supported by scientists from the James Hutton Institute (Miriam Glendell, Barry Thornton, Steve Addy, Nikki Baggaley), Cranfield University (Toby Waine, Bob Grabowski) and University of Liege (Jeroen Meersmans) tested a novel sediment tracing technique to improve identification of dominant terrestrial land-use sources of organic carbon in freshwater sediments in river catchments. Three interlinked studies were conducted in two catchments, including Loch Davan in Aberdeenshire.

The first study combined sediment fingerprinting using land use specific plant biomarkers with soil erosion modelling to understand the origin and fate of eroded SOC and the role of catchment connectivity, such as riparian buffer zones, in intercepting sediment delivery to streams. Sediment fingerprinting and modelling using n-alkane biomarkers identified that riparian woodland disconnected the upslope delivery of eroded soil organic matter to water while also providing an input of woodland-derived organic matter to the streams.

A second study tested novel combinations of biomarkers (i.e. n-alkanes, n-alkane isotopes and short chain neutral lipid fatty acids – the latter characteristic of soil microbial and fungal communities) to compare their ability to discriminate sediment sources.  Acombination of n-alkanes and their isotope fingerprints improved discrimination between arable and pasture sources while the combination of n-alkanes and neutral lipid fatty acids improved discrimination between all land uses (incl. arable, pasture, forest and moorland).

Looking towards Loch Davan (top right) from the slopes of Roar Hill, Aberdeenshire.

Finally, using sediment fingerprinting as a benchmark, the ability of two soil erosion models to identify sources of streambed sediments was evaluated. The Scottish erosion risk maps developed at the James Hutton Institute were confirmed to be the most reliable in identifying likely land use sources of sediments and organic matter in the Scottish Loch Davan catchment. The (Scottish) soil erosion risk map shows the risk of a bare soil being eroded by water under intense or prolonged rainfall. The susceptibility to erosion is based on soil texture, with coarser textured soils being most susceptible to erosion, and capacity to absorb rainfall combined with slope to determine how erosive overland flow could be. Soils with mineral topsoils have been classified separately from those with organic (peaty) surface layers, while organic soils (peats) are considered to be highly erodible so are always considered to be at a high risk of erosion.

These cutting-edge sediment and organic matter tracing techniques are now available for wider application to evaluate the effectiveness of river restoration projects and to inform catchment management practices to improve soil health, reduce soil erosion and mitigate its off-site impacts.

This article was first published in Soil Sentinel in March 2023. 

For further information please contact Dr Catherine Wiltshire Katy.Wiltshire@cranfield.ac.uk and Dr Miriam Glendell miriam.glendell@hutton.ac.uk

Disclaimer: The views expressed in this blog post are the views of the author(s), and not an official position of the institute or funder.