CAUSES: Attribution of radiation biases in NWP and climate models near the US Southern Great Plains

Submitter:

Van Weverberg, Kwinten — Met Office

Area of research:

General Circulation and Single Column Models/Parameterizations

Journal Reference:

Van Weverberg K, C Morcrette, J Petch, S Klein, H Ma, C Zhang, S Xie, Q Tang, W Gustafson Jr, Y Qian, L Berg, Y Liu, M Huang, M Ahlgrimm, R Forbes, E Bazile, R Roehrig, J Cole, W Merryfield, W Lee, F Cheruy, L Mellul, Y Wang, K Johnson, and M Thieman. 2018. "CAUSES: Attribution of Surface Radiation Biases in NWP and Climate Models near the U.S. Southern Great Plains." Journal of Geophysical Research: Atmospheres, 123(7), 10.1002/2017JD027188.

Science

Scientists at the Met Office in the UK, along with colleagues at Lawrence Livermore National Laboratory, have organized an international multi-model inter-comparison project, called CAUSES (Clouds Above the United States and Errors at the Surface). The project aims to identify the physical processes that lead to the formation of a warm surface air temperature bias present in many weather forecast and climate model simulations over the American Midwest.

Impact

This paper is part of a series of papers discussing various aspects related to the warm bias. The focus of this study is on attribution of the surface radiation biases that coincide with the warm bias. The radiation biases are likely to be an inherent link in the chain of model errors leading to too warm screen temperatures in this region. A better understanding of the origin of radiation imbalances is needed for model developers to know where to focus parameterization development and ultimately eliminate the warm bias from numerical weather prediction (NWP) and climate models.

Summary

Using observational data collected from the U.S. Department of Energy (DOE) Southern Great Plains (SGP) sites, it is shown that clouds are the main culprit for surface radiation biases, although most models have significant contributions from a too small surface albedo as well. Deep clouds contribute most to the shortwave bias, either because they are not present frequently enough in the models, or because they are too transparent to shortwave radiation when they are present. Since most models produce precipitation too frequently during the daytime (when cloud errors are most prominent), it appears that the problem is not with the triggering of deep convection, but rather with too large precipitation efficiencies associated with parameterized deep convection and a lack of detrainment of cloud in the troposphere.