Evaluation of GFDL SCM Cloud Fractions and Surface Radiation Fields with Those from the Ground-based Remote Sensing at SGP
| Kim, Byung-Gon | Princeton University |
| Klein, Stephen | Lawrence Livermore National Laboratory |
| Mace, Gerald | University of Utah |
| Benson, Sally | University of Utah |
The various kinds of approaches to obtain cloud fraction in the model still have the limitations, which accordingly result in the consequent errors in the model radiative fluxes. Because of the different schemes to estimate the cloud fraction by the model and also observation artifacts, the prudent comparisons should be made in order to reach the right answer. This study demonstrates the more revised approaches of comparisons to improve the evaluation of cloud fraction. To do this, a relatively inexpensive and computationally efficient model-evaluation method is the use of so-called single column model (SCM), which contains all of the physical parameterizations of the GCM and can be forced with observations. In terms of the observation, a technique to derive cloud faction as a function of height using ground-based radar and lidar is applied to the SCM evaluation. The full-year measurements of 2000 over the SGP site are used to evaluate treatment of clouds in the single column model which uses the same parameterization as are in the GFDL climate model (GFDL Global Atmospheric Model Development Team, 2004). To reduce comparison issues of single point and represent the temporal scale more comparable to the size of the SCM domain, the observations and the SCM outputs are averaged in time to 3 hr. Several comparisons demonstrate the considerable potential of active instruments for validating the representation of clouds in models. Excluding times when either the SCM precipitation or the observed precipitation is greater than 0.5 mm hr-1 appears to significantly improve the agreement of SCM cloud fractions with observations. In addition, the cloud field is not counted as cloud when the SCM LWC is less than 10 mg m-3 and IWC less than 1 mg m-3, for which no cloud should be present especially in the upper layer. Furthermore, the contribution from the SCM precipitation area by snow and rain is included because the radar cannot distinguish precipitating snowflakes and raindrop from nonprecipitating clouds. Lastly we investigate how this modification is associated with the evaluations of the surface radiation fields.
This poster will be displayed at the ARM Science Team Meeting.