Simulation of the diurnal cycle in a climate model and its evaluation using data from Meteosat 7
Slingo, A., Hodges, K. I. and Robinson, G. J. (2004) Simulation of the diurnal cycle in a climate model and its evaluation using data from Meteosat 7. Quarterly Journal of the Royal Meteorological Society, 130 (599). pp. 1449-1467. ISSN 1477-870X
Full text not archived in this repository.
To link to this article DOI: 10.1256/qj.03.165
The representation of the diurnal cycle in the Hadley Centre climate model is evaluated using simulations of the infrared radiances observed by Meteosat 7. In both the window and water vapour channels, the standard version of the model with 19 levels produces a good simulation of the geographical distributions of the mean radiances and of the amplitude of the diurnal cycle. Increasing the vertical resolution to 30 levels leads to further improvements in the mean fields. The timing of the maximum and minimum radiances reveals significant model errors, however, which are sensitive to the frequency with which the radiation scheme is called. In most regions, these errors are consistent with well documented errors in the timing of convective precipitation, which peaks before noon in the model, in contrast to the observed peak in the late afternoon or evening. When the radiation scheme is called every model time step (half an hour), as opposed to every three hours in the standard version, the timing of the minimum radiance is improved for convective regions over central Africa, due to the creation of upper-level layer-cloud by detrainment from the convection scheme, which persists well after the convection itself has dissipated. However, this produces a decoupling between the timing of the diurnal cycles of precipitation and window channel radiance. The possibility is raised that a similar decoupling may occur in reality and the implications of this for the retrieval of the diurnal cycle of precipitation from infrared radiances are discussed.