Spatial Scale Dependence of Hydrological Climate Change Signals

Precipitation is the major component of the hydrological cycle and a commonly used variable for climate change studies. The simulation of precipitation is, nevertheless, challenging as it is influenced by different meso- and macro-scale processes such as convection and large-scale atmospheric circulations. Global climate model (GCM) is the primary tool for precipitation simulations. However, the spatial resolution of the current generation of GCMs is still coarse and unable to capture sub-grid scale processes. The processes that cannot be resolved in the horizontal grid spacing of GCMs are parameterized, which is a source of large bias and uncertainty in the simulations. A more trustworthy representation of these processes and features is provided at finer spatial resolutions of regional climate models (RCMs) and limited area models (LAMs). This research investigates the role of climate model’s spatial resolution ranging between 2.8 and 312 km on the simulations and projections of precipitation-based indicators in different parts of the world. The considered indicators are extreme precipitation (P_ex), annual (P_a) and seasonal (P_s) precipitation totals, dry day frequency (DDF) and dry spell length (DSL). The results show closer sub-daily P_ex simulations of Belgian LAMs to observations compared to the driving GCMs, whereas no clear improvement in the LAM simulations is found for daily P_ex. Similar results are observed for climate change signals where the LAMs project a larger change for sub-daily P_ex. The comparison of P_s and P_ex changes between the CORDEX RCMs and the driving CMIP5 GCMs reveal the largest difference for summer, with intensification of the signals by the RCMs for 60%-83% of the land area in the Middle East and North Africa (MENA). A larger P_a change is also projected by the RCMs compared to the driving GCMs in the MENA region, while the RCMs project a smaller change for DDF and DSL. Generally, the influence of the model spatial resolution varies with indicator, location and season and must be taken into account for hydrological impact assessments of climate change.