Studies on clouds with RACMO in recent years have primarily focused on model evaluation of cloud-related parameters and radiative fluxes.
A first evaluation study was carried out among four models including RACMO with ground-based observations from CLIWA-Net (van Meijgaard and Crewell, 2005). Figure 1 nicely summarizes the observations obtained at Cabauw during the BBC-campaign in Aug-Sep 2001 that were used for the evaluation. Concerning liquid water path there was no generic model finding, but an interesting outcome certainly was that all models overestimate frequency and duration of precipitation.
Figure 1 Observed mean liquid water path and relative occurrence at Cabauw during the BBC observational campaign (Aug-Sep 2001) as a function of aggregation time. From top to bottom: I) all non-precipitating periods, II) non- precipitating water clouds, IIc) non- precipitating water clouds under overcast conditions, and IV) free of water clouds. Uncertainty bars signify the sensitivity to variations in the values of the observed cloud base parameters.
The Spinning Enhanced Visible and Infrared Imager (SEVIRI) onboard the geostationary METEOSAT-8 satellite was the first instrument able to provide accurate information on the day-light cycle of cloud properties over land and ocean surfaces. It’s usefulness for evaluation RACMO predicted condensed water path was assessed by Roebeling and van Meijgaard (2010) for summertime conditions over Europe. Overall, RACMO underestimates European scales summertime cloud amount, while it overestimates condensed water path as Figure 2 shows, but there are strong north-south oriented gradients in the biases.
Figure 2 Mean cloud amount (upper panels) and condense water path (lower panels) retrieved from SEVIRI and predicted by RACMO for Europe during the period 15 May to 15 September 2004 using cloudy and cloud free grid boxes. The images in the right panel represent the difference between RACMO predicted and SEVIRI retrieved values.
| SEVIRI
| RACMO
| RACMO-SEVIRI
|
|
|
|
|
|
|
|
|
Following the study over Europe, we carried out a study over Africa for a single summer month, where we combined radiation measurements from GERB with the cloud parameter retrievals from MSG/SEVIRI (Greuell et al., 2011). An important finding, displayed in Figure 3, is that the model overestimates the albedo of trade wind cumulus fields over the southern tropical Atlantic Ocean. There are potentially a number of explanations for this difference but it is argued in the paper that this discrepancy is owing to the model overestimating liquid water path for this type of cloud fields.
Figure 3 July 2006 monthly albedo (a) and monthly means of outgoing long wave radiation (OLR) (b), cloud cover (c), condensed water path (CWP) (d) and effective droplet radius Reff (e) along an east-west transect at 15 °S. The Angolan coast is situated at 0 km. Averages were computed across a 200 km wide zone. Different curves represent the satellite observations (in black) and results obtained from four RACMO runs (in color). EXP0: control run with ECMWF physics from cycle31r1; EXP3: like EXP0 with reduced mixing in free troposphere; EXP4: like EXP3 with effective radius prescribed from SEVIRI inferred estimates; EXP5: like EXP4 with homogeneous cloud assumption, i.e. no spatial variability in liquid water content. The gray shaded area in c) shows the uncertainty in SEVIRI cloud cover due to quantifying cloud-contaminated pixels. Because SEVIRI CWP is only retrieved for solar zenith angles less than 72°, it was converted to full-time CWP by means of the monthly mean diurnal cycle from the University of Wisconsin (UWisc) LWP climatology (multiplication factors between 1.01 and 1.14).
To facilitate the interpretation of model evaluation, we recently build a numerical simulator of the SEVIRI instrument (Jonkheid et al., in preparation) inspired by already existing simulators of space-borne cloud measuring systems like ISCCP and CloudSat/Calipso. The SEVIRI simulator enables us to quantify the error associated to the cloud retrieval of SEVIRI inferred observations. Preliminary results of retrieval errors are shown in Figure 4. Knowledge of the retrieval error structure helps in assessing a (much) more accurate estimate of the model error.
Figure 4 The cloud water path retrieval errors for various cloud configurations processed by the simulator, averaged over the appropriate solar zenith angles and satellite viewing angles. The top row shows the relative RMS error (indicating the overall retrieval error); the middle row indicates the fraction of geometries where the relative retrieval error is greater than 25% (indicating the prevalence of moderate to high errors); the lower row shows the average relative error (indicating a bias in the retrievals). The left column refers to pure water clouds, while the right column refers to pure ice clouds. The central columns refer to configuration of an ice cloud layer situated over a water cloud layer.
Meijgaard, E. van, and S. Crewell, 2005: Comparison of model predicted liquid water path with ground-bades measurements during CLIW-NET. Atmospheric Res. 75, 201-226
Roebeling, R.A. and E. van Meijgaard, 2009: Evaluation of the daylight cycle of model predicted cloud amount and condensed water path over Europe with observations from MSG-SEVIRI,
J. Climate 22, 1749-1766,
doi:10.1175/2008JCLI2391.1.
Greuell, W., E. van Meijgaard, and N. Clerbeaux, 2011: Evaluation of model predicted top-of-atmosphere radiation and cloud parameters over Africa with observations from GERB and SEVIRI, accepted by J. Climate
Jonkheid, B., R. Roebeling, and E .van Meijgaard, 2011: Quantifying uncertainties of cloud physical properties derived from simulated SEVIRI observations, in preparation
)
)
)
)
)
)
)
)
)
)
)
)
)
