An interesting seminar was presented at Colorado State University on June 8 2006 by Dr. Ehrhard Raschke of the University of Hamburg, Germany. The seminar title reads,
“Some Controversial Results on the Radiative Forcing of Our Climate System as Extracted from Model Results for the AMIP-2 and IPCC-FAR and for the Radiation Climatologies of the ISCCP-FD and GEWEX-SRB’
with the abstract
“We analyzed only monthly data as provided for the above-mentioned projects and did no further manipulations. This seminar does not report about sensational speculations or new theories on climate change.
Several modeling groups provided data sets describing the radiation fields at both the top and bottom of the atmosphere. Here we compared only values for the incoming solar radiation and found (a) considerable differences in zonal averages (meridional profiles) of up to 15 Wm-2 over both poles during the transitional seasons, (b) systematic regional anomalies of up to ±1 Wm-2, corresponding to seasonal “pulse” of the TSI of 4 Wm-2, (c) different model results on changes of the TSI due to the leap-year and sunspot activity.
ISCCP-FD and GEWEX-SRB radiation products, as they were available until about December 2005 (!) at TOA and at the surface, were simply compared. We found two major errors which need firstly to be removed completely:
(a) ISCCP time series until the data-year 2002 are primarily based on a skin temperature estimate showing a systematic decrease over the entire zone between about 60N and 60S by up to 16 Wm-2, as the inclusion of more recent data, released in April 2006, demonstrates. There are systematic differences in various results on the cloud effect (CRF ) on such products.
(b) GEWEX-SRB data on incoming solar radiation at TOA were often by more than 1% lower than concurrent ISCCP data. This error propagates through all further results on solar radiation budget. It now has been removed in part after a few consultations with the author making the recent data release (2.6) more useful for further studies. Both data sets need a complete reanalysis over both polar regions.
Further, none of both data sets has been validated in detail against satellite measurements of the ERB at TOA. Validations of surface data need a further re-analysis.
We suggest that in future climate studies all model results and climate data sets must be based on the same (well proven) routine for computing the incoming solar radiation at TOA over regions of the earth. Strategies need still development making effective use of ground-based measurements and observations combining the potentials of different networks: e.g. BSRN, ARM, SURFRAD, national and other regional data sets.
`Positive impacts’ of our studies: The GEWEX modelers have invited to report at their workshop on systematic errors in climate models; SRB has done a reanalysis of their solar data during the past 6 months. ISCCP is planning to re-analyze all their radiation products. Our results have been published in Raschke et al., GRL, 2005 and 2006.”
The titles and abstracts of the two papers read,
“How accurate did GCMs compute the insolation at TOA for AMIP-2?”
“Monthly averages of solar radiation reaching the Top of the Atmosphere (TOA) as simulated by 20 General Circulation Models (GCMs) during the period 1985–1988 are compared. They were part of submissions to AMIP-2 (Atmospheric Model Intercomparison Project). Monthly averages of ISCCP-FD (International Satellite Cloud Climatology Project – Flux Data) are considered as reference. Considerable discrepancies are found: Most models reproduce the prescribed Total Solar Irradiance (TSI) value within ±0.7 Wm−2. Monthly zonal averages disagree between ±2 to ±7 Wm−2, depending on latitude and season. The largest model diversity occurs near polar regions. Some models display a zonally symmetric insolation, while others and ISCCP show longitudinal deviations of the order of ±1 Wm−2. With such differences in meridional gradients impacts in multi-annual simulations cannot be excluded. Sensitivity studies are recommended. ”
“An assessment of radiation budget data provided by the ISCCP and GEWEX-SRB”
“The projects ISCCP and GEWEX-SRB compute global data sets of radiation budget components at the top of the atmosphere and at the surface. Time series range from July 1983 to June 2001, and to October 1995, respectively. Comparing monthly averages over broader zones we find that the SRB underestimates the incident radiation at TOA by more than 2–5 Wm−2 over the tropics and up to 40 Wm−2 over polar regions. The ISCCP infrared radiation fluxes near the surface and at TOA, in particular over both polar zones, are higher than those of the SRB. Clouds in the ISCCP appear optically less effective than in the SRB. Interannual and month-to-month variations are observed indicating serious errors in ancillary data. Complete reprocessing is recommended. End products need validation within this large domain in space and time with correlated radiation budget measurements at TOA and at ground. ”
There is a very important conclusion from the Raschke et al. studies. As is stated in the seminar abstract
“We suggest that in future climate studies all model results and climate data sets must be based on the same (well proven) routine for computing the incoming solar radiation at TOA over regions of the earth.”
Regional comparisons would provide a more appropriate quantitative comparison of the multi-decadal global climate models. The ability (or lack of) of regional skill in the global climate model simulations would be crucial information to provide to policymakers. As currently communicated to them, they are presented with regional multi-decadal climate simulations as if they are skillful forecasts for the coming decades. They are not.