Dynamics of critical source areas: Does connectivity explain chemistry?

Critical source area approaches to catchment management are increasingly being recognised as effective tools to mitigate sediment and nutrient transfers. These approaches often assume hydrological connectivity as a driver for environmental risk, however this assumption has rarely been tested. Using high resolution monitoring, 14 rainfall events of contrasting intensity were examined in detail for spatial and temporal dynamics of overland flow generation at a hydrologically isolated grassland hillslope in Co. Down, Northern Ireland. Interactions between overland flow connectivity and nutrient transfers were studied to test the critical source area hypothesis. While total and soluble phosphorus loads were found to be representative of the size of the overland flow contributing area (P = < 0.05), the dynamics of concentrations throughout storm hydrographs were found to be complex and storm dependant. Near linear relationships were observed between the contributing area and total overland flow volumes (R2 = 0.86). Export coefficients (kg ha− 1) calculated using plot size were found to under estimate annual losses of total phosphorus by a factor of 17, when compared to those calculated using the contributing area. This study shows that current critical source area definitions for implementing mitigation measures may be overlooking the importance of storm characteristics in determining nutrient transfers and hence may be insufficient in determining catchment scale risk.

► We test the critical source area hypothesis that increased connectivity increases nutrient transfer.
► We examine the spatial and temporal dynamics of overland flow at the hillslope scale.
► Total and soluble P loads were well represented by the size of the connected overland flow network.
► Dynamics of phosphorus concentrations throughout storm hydrographs were complex and storm dependant.
► P export coefficients (kg ha− 1) calculated using plot size under estimate annual losses by a factor of 17.