Comparing High Resolution Lidar and Conventional Digital Elevation Models for Identifying Potential Phosphorus Transfer Pathways.

Pathways of phosphorus (P) losses from agricultural landscapes to surface waters are influenced by fine-scale topographic features that effectively re-route surface runoff. Such features can now be captured at sub-metre resolution in digital elevation models (DEMs) derived from Light Detection and Ranging (LiDAR), which could improve critical source area (CSA) identification. In two agricultural catchments in south-east Ireland, areas prone to overland flow were modelled using the highest values of Topographic Wetness Index (TWI) grids at four different resolutions, derived from 0.25 m, 1 m and 2 m LiDAR DEMs and a conventional 5 m DEM. Grid resolution had a critical effect on predicted locations of overland flow-prone areas, resulting from differences in topographic information within DEMs which affected upslope contributing areas. Flow pathways derived from LiDAR DEMs were re-routed away from the general hillslope scale flow direction by tramlines, hedgerows and roads, and tended to accumulate at downslope field boundaries and ‘breakthrough’ at gateways and other hedgerow openings. As such features were not captured by 5 m DEMs, modelled flow pathways at this resolution conformed to larger scale topography. Predicted locations of overland flow-prone areas were compared to field observations at target sites during storm events. LiDAR-derived TWI maps showed high spatial precision, accurately identifying 67-100% of observed overland flow-prone areas, compared to 52-60% at 5 m resolution. Locations where overland flow would accumulate, breakthrough field boundaries or be delivered to the drainage network were predicted with 80-100% accuracy by LiDAR-derived maps and 43-56% accuracy by 5 m maps. This indicates the potential for targeted, sub-field scale mitigation measures to intercept and trap P in overland flow close to the source using LiDAR data. If CSAs and subsequent mitigation measures are to be accurately located, the choice of DEM resolution should reflect the scale of the topographic features that influence surface hydrology.