Variants of the Budkyo hydrological model describe the effects of land use change on annual water yield. A recent modification using a simple process-based ecohydrological model provides insight into the partitioning of rainfall between runoff and evapotranspiration. In particular, the 'effective vegetation rooting depth' becomes the single free parameter in the model and can be related to land use and climate. We applied this approach to investigate the relations between mean annual runoff from 2000 to 2011, catchment average effective rooting depth and the proportion of forest cover across eleven catchments in South-west Western Australia. The proportion of forested and cleared land was partitioned using MODIS minimum annual average LAI values from 1 km2 pixels over 2000-2011, with forest clearing ranging from 1% to 98% across the 11 catchments. Estimated mean annual runoff using catchment averaged effective rooting depths for forest and cleared (grassland) land obtained using an independent physiologically-based model gave better estimates than a widely used 'default' Budkyo-based model. If effective rooting depth declines with aridity, as described by the model then runoff decline may be considerably less (about 50%) than predicted if the effective root depth remains unchanged (70%-92% decline). This highlights the importance of understanding ecohydrological feedbacks between vegetation and climate in projecting scenarios of water yield response to climate change. © 2014 by the authors; licensee MDPI, Basel, Switzerland.
|Publication status||Published - 2014|