Using earth observation for constraining uncertainty in sediment loss predictions.

Current Earth Observation(EO) satellites have limited ability to discriminate variables necessary for predicting the impact of nutrient loss on water quality. This project investigates whether new hyperspectral EO sensors provide enhanced accuracy and more detailed spatial context to aid the characterization of crop type and fractional ground cover. These are key input parameters in models of field and catchment scale nitrogen, phosphorus and sediment loss to streams. Such contemporary information, when coupled with supporting (e.g. digital elevation model) data, will help identify vulnerable areas and provide a framework for developing pollution risk assessment tools based on the spatial co-locationof risk factors (slope, erodable soils, crop type).