How does rising atmospheric CO2 affect marine organisms?

Click to locate material archived on our website by topic


A Multi-Regional Climate Model Hindcast for Africa
Reference
Kim, J., Waliser, D.E., Mattmann, C.A., Goodale, C.E., Hart, A.F., Zimdars, P.A., Crichton, D.J., Jones, C., Nikulin, G., Hewitson, B., Jack, C., Lennard, C. and Favre, A. 2014. Evaluation of the CORDEX-Africa multi-RCM hindcast: systematic model errors. Climate Dynamics 42: 1189-1202.

Background
The authors write that "regional climate model (RCM) data are essential for assessing the impact of climate change on water resources, agriculture, security and natural ecosystems," and they say that "the primary tools for projecting climate are global climate models (GCMs)." But noting that typical impact assessment models require inputs at much finer spatial resolution than what is provided by GCMs, they say that "GCM data are often downscaled using RCMs."

What was done
In a study designed to learn how well this downscaling performs in the case of Africa, Kim et al. conducted an RCM hindcast for the period 1989-2008 that employed boundary conditions derived from the ERA-Interim reanalysis, focusing on 21 sub-regions of the continent and utilizing gridded surface observational data sets obtained from the University of East Anglia's Climatic Research Unit.

What was learned
The thirteen researchers report there were significant systematic biases in all ten of the RCMs they tested. In the case of precipitation, they found that (1) "most RCMs overestimate the magnitude of its spatial variability," that (2) a "large ENS [multi-model ensemble] precipitation bias occurs in the northern Sahara, South Africa, and Arabia Peninsula," that (3) "most RCMs suffer difficulties in simulating the annual cycle in these dry/semi-dry regions," that in the case of temperature, the ENS (4) "overestimates TMIN," which (5) "results in underestimation of diurnal temperature range," while the most noticeable ENS TAVG errors are (6) "the cold biases in the tropical west coast, east coast, and much of the Sahara regions" and (7) "warm biases in the eastern Arabia Peninsula and the subtropical west coast regions," that in the case of clouds, (8) "a majority of RCMs underestimate the overland-mean cloudiness," as well as (9) cloud "spatial variability."

What it means
In their final comment on their findings, Kim et al. write that "the systematic variations in RCM skill may indicate common weaknesses in physics parameterizations used in these models." One can only hope that they - or someone - will get these models fixed ... and soon!

Reviewed 28 May 2014