Multi-dimensional Soil Erosion and Associated Chemical Transport: Experiments, Mathematical Modelling and Field Evaluation

The EU Water Framework Directive includes specific requirements to control diffuse pollution and also places requirements on national governments to set water quality objectives based on good ecological status. The EU Habitats Directive and the UK government's Quality of Life indicators set targets for achieving favourable conditions on conservation sites, and the UK Biodiversity Action Plan aims to reverse and restore populations for key species and habitats. Given the pivotal role phosphorous, nitrate and sediment play in influencing water quality and biodiversity, and that agriculture is thought to be responsible for 50% of the inputs to surface waters, predicting the movement of these diffuse pollutants from land to water is becoming increasingly important. Eroded agricultural soil is carrier of pollutants such as pesticides, herbicides, fertilizers and microbes. The transport of eroded agricultural soil and associated pollutants from farmlands, results in the eutrophication of surface waters, damage to freshwater ecosystems and the microbial contamination of surface water sources. It is now also seen as a major pollutant source responsible for the reduction of water quality in UK beaches and coastal environments. Pollutants are either carried in dissolved form in the runoff water, or through attachment to the soil particles. Since pollutants bind preferentially to different sized soil particles, determining the pollutant loading to waterways requires predicting or calculating the sediment size distribution of the eroded sediment. We propose that this can be achieved by solving the multi-particle size class, multi-dimensional soil erosion model of Hairsine and Rose in conjunction with a transport equation for the dissolved pollutants. Validation and calibration of the model will be done through a hierachical programme, moving from well controlled laboratory experiments, to natural hillslopes and finally a small catchment. At each scale we will determine model performance both at the outlet and spatially. For the flume (1.5 m x 4 m) measured outflow data will include time varying particle size and pollutant distributions as a function of soil type, surface geometry, rainfall rate and initial soil moisture conditions. Addition high resolution spatial data will be obtained through the use of digital photogrammetry where local aspect and local slope measurements of the surface topography in the flume can be measured for comparison against predictions from the erosion model. At the hillslope scale we will link with the Defra funded Mitigation of Phosphorus and Sediment Project, using a supplementing data collection from 52 instrumented hillsides on three contrasting soil types. We will also utilise rare earth element oxides as tracers at this scale to determine where the sources of sediment are and compare this spatial data with our model predictions. At the catchment scale we will use existing rare earth catchment data obtained from Coschocton, Ohio to test the spatial predictions of the model. This research offers significant potential for achieving major advances in the physical understanding of the transfer of diffuse pollutants from soils to waters, providing scientific underpinning for the more applied research currently being funded by Defra and for the implementation of policy measures such as the Water framework directive and Catchment Sensitive Farming.