Lessons learned from North American Regional Climate Model (NRCM) Experiments

Regional climate modeling is one of the priority projects at the National Center for Atmospheric Research. A nested regional climate model (NRCM) has been developed from the Advanced Research WRF Model (ARW) and tested for a series of long-term simulations. North American Regional Climate experiments were conducted to predict changes in North Atlantic tropical cyclone (TC) characteristics due to climate change. During this set of experiments the NRCM was directly forced with Community Climate System Model (CCSM) ICPP-AR4 output.

Serious climate biases were discovered in the regional climate simulations when driven by raw CCSM output. The results showed almost no TC formation in tropical North Atlantic, and East Pacific TC formations were pushed westward. Analysis revealed that this bias was due to anomalously high vertical wind shear across the tropical North Atlantic in the CCSM data, which arose largely from the well-known eastern Pacific Ocean surface temperature anomalies. NRCM results were greatly improved when the CCSM data was first adjusted to account for the biases in the climate model. High vertical shear was substantially reduced and NRCM was able to produce TCs in tropical North Atlantic at a reasonable spatial and temporal scale. This paper presents lessons learned from the NRCM simulations, when regional climate models are driven directly with raw global climate model data.