Evaluation of a WRF ensemble using GCM boundary conditions to quantify mean and extreme climate for the southwest of Western Australia (1970-1999)
Andrys, J., Lyons, T.J. and Kala, J. (2016) Evaluation of a WRF ensemble using GCM boundary conditions to quantify mean and extreme climate for the southwest of Western Australia (1970-1999). International Journal of Climatology, 36 (13). pp. 4406-4424.
*Subscription may be required
A high resolution (5 km), single initialization, 30 year (1970–1999) Weather Research and Forecast regional climate model (RCM) ensemble for southwest Western Australia (SWWA) is evaluated. The article focuses on the ability of the RCM to simulate winter cold fronts, which are the main source of rainfall for the region, and assesses the spatial and temporal characteristics of climate extremes within the region's cereal crop growing season. To explore uncertainty, a four-member ensemble was run, using lateral boundary conditions from general circulation models (GCMs) of the Coupled Model Intercomparison Project Phase 3; ECHAM5, Model for Interdisciplinary Research on Climate 3.2 (MIROC 3.2), Community Climate System Model version 3 (CCSM3) and Commonwealth Scientific and Industrial Research Organisation (CSIRO) mk3.5. Simulations are evaluated against gridded observations of temperature and precipitation and atmospheric conditions are compared to a simulation using ERA-Interim reanalysis boundary conditions, which is used as a surrogate truth. Results show that generally, the RCM simulations were able to represent the climatology of SWWA well, however differences in the positioning of the subtropical high pressure belt were apparent which influenced the number of fronts traversing the region and hence winter precipitation biases. Systematic temperature biases were present in some ensemble members and the RCM was found to be colder than the driving GCM in all simulations. Biases impacted model skill in representing temperature extremes and this was particularly apparent in the MIROC-forced simulation, which was the worst performing RCM for both temperature and precipitation. The dynamical causes of the biases are explored and findings show that nonetheless, the RCM provides added value, particularly in the spatio-temporal representation of wet season rainfall.
|Publication Type:||Journal Article|
|Murdoch Affiliation:||School of Veterinary and Life Sciences|
|Copyright:||© 2016 Royal Meteorological Society|
|Item Control Page|