Validation of a pattern scaling approach for determining the maximum available renewable freshwater resource

A pattern scaling approach allows projection of regional climate changes under a wide range of emission scenarios. A basic assumption of this approach is that the spatial response pattern to global warming (scaling pattern) is the same for all emission scenarios. Precipitation minus evapotranspiration (PME) over land can be considered to be a measure of the maximum available renewable freshwater resource, and estimation of PME is fundamentally important for the assessment of water resources. The authors assessed the basic assumption of pattern scaling for PME by the use of five global climate models. A significant scenario dependency (SD) of the scaling pattern of PME was found over some regions. This SD of the scaling pattern of PME was mainly due to the SD and the nonlinear response of large-scale atmospheric and oceanic changes.When the SD of the scaling pattern of PME is significant in a target area, projections of the impact of climate change need to carefully take into consideration the SD. Although the SD of the anthropogenic aerosol scaling patterns tended to induce SDs of precipitation and evapotranspiration scaling patterns, the SDs of precipitation and evapotranspiration tended to cancel each other out. As a result, the SD of the PME scaling pattern tended to be insignificant over most regions where the SD of anthropogenic aerosol scaling patterns were significant. The authors could not find large impacts of land use change on PME scaling pattern, but the former may influence the latter on different time scales or spatial scales.