Dynamical downscaling of ERA40 reanalysis data over southern Africa: added value in the simulation of the seasonal rainfall characteristics

We investigate the capability of the regional climate model REMO to adequately simulate the mesoscale climate patterns of southern Africa, thereby focusing on South-West Africa (SWA). The simulations are conducted at a horizontal resolution of about 18 km for the period from 1958 to 2007. Boundary forcing is taken from ERA40 reanalysis and subsequent operational analyses data. This study is the first long-term dynamical downscaling study for southern Africa on a very high horizontal resolution. To evaluate the model results, we use a set of different observations to compare with model output data. This analysis indicates that the model results agree rather well with observations for both spatial as well as temporal patterns. Especially along the focus region in SWA, REMO results are very close to observations. For this region we show that dynamical downscaling of the ERA40 reanalysis data with the REMO model leads to an improved description of the seasonal precipitation characteristics. Remaining deficits, which mainly occur in the simulation of the 2-m air temperature, can be explained with the setup of the model's soil routine and will be investigated further. On the basis of the findings of this study we conclude that the model is well suited for the simulation of the regional climate over the southern African region. High-resolution model output implies more detailed climate information and can lead to a significant improvement in the representation of spatial patterns. It is therefore beneficial to use regional climate model data when assessing climate change-induced impacts on a regional scale. Copyright (c) 2010 Royal Meteorological Society