Figure 1a. Temperature profile for 18 May 98. Environmental … · 4B.4 EVALUATION OF GCM RADIATION...
3
4B.4 EVALUATION OF GCM RADIATION CODES USING SHEBA DATA James Pinto, Judith Curry and Dana E. Lane* University of Colorado, Boulder 80309 INTRODUCTION Surface Heat Budget of the Arctic Ocean (SHEBA) data are used to evaluate general circulation model (GCM) radiation codes (both longwave and shortwave). The case studies used include a clear sky case, a liquid cloud case, and an ice cloud case. Emphasis was on selecting fairly stationary and horizontally homogeneous clouds for this evaluation. Observations from aircraft, surface based instrumentation, and satellites are used. BACKGROUND The net radiation at the surface is considered one of the most important components of the Arctic energy budget (Pinto et al. 1997). SHEBA is a coordinated project to investigate the role of the Arctic climate in global change. One of the primary goals of SHEBA is to develop and implement models that improve the simulation of the present day Arctic climate, including cloud-radiation feedbacks. The data for this GCM radiation parameterization evaluation come from Phase II of the project, which involved experiments at the North Pole/SHEBA ice station. Additional data are taken from the First ISCCP Regional Experiment (FIRE) Arctic Clouds Experiment (ACE) for the case study days. One of the goals FIRE ACE is to examine the effects of clouds on radiation exchange between the surface, atmosphere and space (Curry et al. 2000). This study continues work described in Pinto et al. (1997). OBSERVATIONAL DATA Several special cases were selected for evaluation of radiative transfer models that are similar to parameterizations found in current GCMs. These days were selected using several criteria. First, the atmospheric condition must be reasonably stationary and homogeneous. Second, observational data from aircraft measuring in situ atmospheric characteristics should be available for the same time period. Additional information from the ER-2 or satellite was also required for determination of TOA fluxes and to verify atmospheric homogeneity during the time of interest. Finally, extensive surface-based radiation data must also be present. For this preliminary study, three days of interest are presented. The first, 24 May 1998 is a clear day with little aerosol input. The second, 18 May 1998 is an example of a layer cloud composed of cool liquid water. Finally, 21 July 1998 is examined as an example of a stratocumulus cloud with discernible two-dimensional features (Benner et al. 2000). Input files containing information about the atmospheric vertical structure have been generated for testing of radiative transfer codes. The vertical data are gridded using nearest- neighbor averaging. Temperature and moisture data were obtained from GLAS radiosondes. Cloud water content and effective radius have been determined from calibrated data collected with the Gerber Probe aboard the NCAR C-130. Aerosol profiles have been determined from C- 130 aerosol measurements below 6 km and a *Corresponding author address: Dana E. Lane, University of Colorado - Boulder, Cooperative Institute for Research in Environmental Sciences, Campus Box 216, Boulder, CO 80309-0216; e-mail: [email protected]. Figure 1a. Temperature profile for 18 May 98.
Transcript of Figure 1a. Temperature profile for 18 May 98. Environmental … · 4B.4 EVALUATION OF GCM RADIATION...
4B.4
EVALUATION OF GCM RADIATION CODES USING SHEBA DATA
James Pinto, Judith Curry and Dana E. Lane*University of Colorado, Boulder 80309
INTRODUCTION
Surface Heat Budget of the Arctic Ocean(SHEBA) data are used to evaluate generalcirculation model (GCM) radiation codes (bothlongwave and shortwave). The case studies usedinclude a clear sky case, a liquid cloud case, and anice cloud case. Emphasis was on selecting fairlystationary and horizontally homogeneous clouds forthis evaluation. Observations from aircraft, surfacebased instrumentation, and satellites are used.
BACKGROUND
The net radiation at the surface is consideredone of the most important components of the Arcticenergy budget (Pinto et al. 1997). SHEBA is acoordinated project to investigate the role of theArctic climate in global change. One of the primarygoals of SHEBA is to develop and implementmodels that improve the simulation of the presentday Arctic climate, including cloud-radiationfeedbacks. The data for this GCM radiationparameterization evaluation come from Phase II ofthe project, which involved experiments at the NorthPole/SHEBA ice station. Additional data are takenfrom the First ISCCP Regional Experiment (FIRE)Arctic Clouds Experiment (ACE) for the case studydays. One of the goals FIRE ACE is to examine theeffects of clouds on radiation exchange between thesurface, atmosphere and space (Curry et al. 2000).This study continues work described in Pinto et al.(1997).
OBSERVATIONAL DATA
Several special cases were selected forevaluation of radiative transfer models that aresimilar to parameterizations found in current GCMs.These days were selected using several criteria.First, the atmospheric condition must be reasonablystationary and homogeneous. Second,observational data from aircraft measuring in situatmospheric characteristics should be available forthe same time period. Additional information from
the ER-2 or satellite was also required fordetermination of TOA fluxes and to verifyatmospheric homogeneity during the time ofinterest. Finally, extensive surface-basedradiation data must also be present. For thispreliminary study, three days of interest arepresented. The first, 24 May 1998 is a clear daywith little aerosol input. The second, 18 May1998 is an example of a layer cloud composedof cool liquid water. Finally, 21 July 1998 isexamined as an example of a stratocumuluscloud with discernible two-dimensional features(Benner et al. 2000).
Input files containing information about theatmospheric vertical structure have beengenerated for testing of radiative transfer codes.The vertical data are gridded using nearest-neighbor averaging. Temperature and moisturedata were obtained from GLAS radiosondes.Cloud water content and effective radius havebeen determined from calibrated data collectedwith the Gerber Probe aboard the NCAR C-130.Aerosol profiles have been determined from C-130 aerosol measurements below 6 km and a
*Corresponding author address: Dana E. Lane, University ofColorado - Boulder, Cooperative Institute for Research inEnvironmental Sciences, Campus Box 216, Boulder, CO80309-0216; e-mail: [email protected].
Figure 1a. Temperature profile for 18 May 98.
combination of the Multifilter Rotating ShadowbandRadiometer (MFRSR) total column optical depth andclimatology above 6 km. Temperature profiles for 18May 98 and 24 May 98 are shown in Figure 1a and1b, respectively.
Flux data at the surface that can be compared tothe output of the radiative transfer models are takenfrom 1-min tower data that is averaged hourly. All
data for this comparison are located on the webathttp://paos.colorado.edu/~curryja/wg5/home.html.
MODEL DESCRIPTION
Several radiation parameterizations that areused in modern GCMs were evaluated as stand-alone radiative transfer codes for each of thecases. The models employed in this study areindicated in Table 1. Each radiative transfermodel was provided with the same informationabout the vertical distribution of water vapor,temperature and ozone, interpolated to anappropriate pressure-level structure. The codesdiffer most in their spectral band structure and intheir ability to include aerosols and trace gases.Therefore all comparisons will be done oncewith the inclusion of aerosols and once without.
For each day of interest the radiative fluxes(longwave and shortwave) are compared toobservations at the surface and at the top of theatmosphere. In addition, information about theefficiency of the radiative transfer calculation arecompiled. Sensitivity studies will be performedfor each model.
RESULTS AND CONCLUSIONS
For the clear sky case study, 24 May 1998,the longwave radiative transfer models agreewithin 20 W m
-2 of each other and the
observations for the magnitude of longwaveradiation reaching the surface (Figure 2).Results are shown for the model runs that didnot include aerosols. There is a greaterdiscrepancy in the results of the shortwavecodes. In particular, the CCM2 codeunderpredicts the amount of shortwave radiationreaching the surface by ~80 W m
-2. This
disparity in the models is similar to that found byPinto et al. (1997). For both longwave andshortwave radiative transfer, the Streamers code(Key 1995) appears to most closely reproducethe observations.
Additional radiative transfer codes that arenot present in this preliminary study arewelcome to participate in this intercomparisonproject. For information on the study, and forcurrent results, please refer to the web site at:http://paos.colorado.edu/~curryja/wg5/pintodata/rad_intcmp/.
REFERENCES
Figure 1b. Temperature profile for 24 May 1998.
MODELREFERENCE
ShortwaveStreamerKey (1995)BugsRadCCM2Briegleb (1992)CCM3Briegleb (1992)HarshvardanHarshvardan et al. (1987)SUNRAYFouquart and Bonnel (1980)
LongwaveReference
StreamerKey (1995)BugsRadRRTMMlawer et al. (1997)CCM2Ramanathan and Downey (1986)CCM3Acker et al. (1996)CCCLWMocrette and Fouquart (1985)
Table 1. List of radiative transfer codes.
Acker, T. L., L. E. Buja, J. M. Rosinksi, and J. E.Truesdale, 1996: Users’ Guide to NCAR CCM3.NCAR Technical Note NCAR/TN-421+IA,Boulder, Colorado, 210 pp.
Benner, T., J. A. Curry, and J. O. Pinto, 2000:Radiative transfer in the summertime Arctic, J.Geophys. Res., in press.
Briegleb, Bruce P., 1992: Delta-eddingtonapproximation for solar radiation in the NCARCommunity Climate Model. J. Geophys. Res.,97, 7603-7612.
Curry, J. A., P. V. Hobbs, M. D. King, D. A. Randall,P. A. Minnis, G. A. Isaac, J. O. Pinto, T. Uttal, A.Bucholtz, D. G. Cripe, H. Gerber, C. W. Fairall,T. J. Garrett, J. Hudson, J. M. Intrieri, C. Jakob,T. Jensen, P. Lawson, D. Marcotte, L. Nguyen,P. Pilewskie, A. Rangno, D. C. Rogers, K. B.Strawbridge, F. P. J. Valero, A. G. Williams, andD. Wylie, 2000: FIRE Arctic Clouds Experiment.Bull. Amer. Meteor. Soc., 81, 5-29.
Fouquart, Y., and B. Bonnel, 1980: Computation ofsolar heating of the Earth’s atmosphere: a newparameterization. Beitr. Phys. Atmos., 53, 35-62.
Harshvardan, R., R. Davies, D. A. Randall, and T. G.Corsetti, 1987: A fast radiation parameterizationfor atmospheric circulation models. J. Geohpys.Res., 92, 1009-1016.
Key, J. R., 1995: Streamer, Version 2.0p User’sGuide, Coop. Inst. for Res. in Environ. Sci.,University of Colorado, Boulder, 71 pp.
Mlawer, E. J., S. J Taubman, P. D. Brown, M. JIacono, and S. A Clough, 1997: Radiativetransfer for inhomogeneous atmospheres:RRTM, a validated correlated-k model for thelongwave. J. Geophy. Res., 102, 16663-16682.
Morcrette, J.-J., 1990: Impact of changes to theradiation transfer parameterizations plus cloudoptical properties in the ECMWF model. Mon.Wea. Rev., 118, 847-873.
Morcrette, J.-J.; and Fouquart, Y, 1985: Onsystematic errors in parameterized calculationsof longwave radiation transfer. Q. J. R. Met.Soc., 111, 691-708.
Pinto, J. O., J. A. Curry and C. W. Fairall, 1997:Radiative characteristics of the Arctic
atmosphere during spring as inferred fromground-based measurements. J. Geophys.Res., 102, 6941-6952.
Ramanathan, V., and P. Downey, 1986: Anonisothermal emissivity and absorptivityformulation for water vapor. J. Geophys.Res., 91, 8649-8666.
Shortwave Results
520
540
560
580
600
620
640
660
680
Models
Longwave results
100
120
140
160
180
200
220
Models
Figure 2. Model comparison for clear sky day, 24 May 98.
