CIG Datasets

Dynamically Downscaled Hydroclimate Projections: WRF model

Overview

A database of dynamically downscaled climate projections for the Pacific Northwest, produced using the Weather Research and Forecasting (WRF) regional climate model, and implemented at an hourly temporal resolution and a spatial resolution of 12 km. These climate projections have also been used to produce hydrologic change projections at a spatial resolution of 1/16th degree (~30 km2) over Washington State.

Figure 1. Snapshot of a future atmospheric river, as simulated by WRF.

The Weather Research and Forecasting (WRF) model is a state-of-the-science mesoscale weather prediction model serving the needs of both operational forecasts and atmospheric research. Recently, the model has been developed to conduct simulations of regional climate for studies focused on the finer-scale impacts of climate change (see Salathé et al., 2010 and Dulière et al., 2013 for details). This is sometimes referred to as “dynamical downscaling,” in which a physical climate model (in this case, WRF) is run with boundary conditions taken from the global model being downscaled.A key advantage of dynamical downscaling is that it simulates the physics of key weather processes, rather than representing them statistically as in statistical downscaling. For example, Salathé et al. (2014) showed that changes in flooding are more accurately represented using a dynamical downscaling approach.

The Climate Impacts Group (CIG) has recently participated in the development of an ensemble of 13 new WRF projections. This effort was led by Cliff Mass in the UW Department of Atmospheric Sciences, with funding provided by the Amazon Catalyst program and King County. Additional details are provided below.

About the Dataset

Regional climate model overview

The WRF model was implemented over the Pacific Northwest at an hourly temporal resolution and a spatial resolution of 12 km (see Salathé et al., 2010 and Dulière, Salathé et al., 2013 for details; older simulations were implemented at a 6-hourly time step). The model performance over the Northwest region has been extensively evaluated using simulations forced by reanalysis fields and is capable of resolving the fine scale structure of storms and their effects on precipitation in complex terrain (Zhang et al. 2009; Duliere et al. 2011). In particular, the model successfully simulates important large-scale features of Pacific Northwest winter storms, such as atmospheric rivers, which have been shown to be the major cause of the largest floods in rivers that drain the windward (western) slopes of the Cascades (Neiman et al. 2011; Warner et al. 2012).

CIG currently archives WRF simulations for the following global models and time periods:

Reanalysis (12 km, 6-hourly): NCEP-NCAR Reanalysis Project (NNRP; Kalnay et al., 1996). Years: 1950-2010

CMIP3 (12 km, 6-hourly):

  • ECHAM5 global model, A1B greenhouse gas scenario. Years: 1970-2070
  • CCSM3 global model, A1B greenhouse gas scenario. Years: 1970-2070
  • CCSM4 global model, RCP 4.5 greenhouse gas scenario. Years: 1970-2070

CMIP5 (12 km, hourly): 

  • ACCESS 1.0 global model, RCP 4.5 greenhouse gas scenario. Years: 1970-2099
  • ACCESS 1.0 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • ACCESS 1.3 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • BCC-CSM1-1 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • CanESM2 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • CCSM4 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • CSIRO-Mk3-6-0 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • FGOALS-g2 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • GFDL-CM3 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • GISS-E2-H global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • MIROC5 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • MRI-CGCM3 global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099
  • NorESM1-M global model, RCP 8.5 greenhouse gas scenario. Years: 1970-2099

Funding

This work was primarily funded by the Amazon Catalyst Program and the King County Department of Natural Resources and Parks.

Contact

Questions and data requests can be directed to Guillaume Mauger.