Frank H. Ruggiero and Artie Jackson Air Force Research Laboratory Hanscom AFB, Massachusetts
Robert P. d'Entremont Atmospheric and Environmental Research Inc. Cambridge, Massachusetts ABSTRACT
In order to properly incorporate cloud water and ice fields along with diabatic effects into a numerical weather prediction model at initialization time it is necessary to accurately identify cloud boundaries. While advanced cloud water analysis techniques use all available information (e.g. radar, pireps, etc), for many military applications satellite data is the primary (and usually only) source of reliable cloud data. In this study, satellite data was combined with conventional three-dimensional weather analysis fields to estimate cloud base and top heights. This is part of larger effort that will eventually result in the production of three-dimensional fields of cloud water for mesoscale model initialization.
The particular approach examined in this study was adapted from the part of the ARPS Data Analysis System's cloud analysis method that uses satellite infrared brightness temperatures. The satellite retrieved fields are cloud top temperature and fractional cloud cover both of which are retrieved from GOES IR data using algorithms from the Support of Environmental Requirements for Cloud Analysis and Archive (SERCAA) system. This information was combined with temperature and relative humidity profiles extracted from analyses used as input to mesoscale numerical weather prediction models. The cloud base estimates are made by computing the lifted condensation level from the thermodynamic profiles. Depending on the cloud top temperature the cloud top height is computed by matching the retrieved satellite cloud top temperature with either the analysis temperature profile or the temperature profile computed by lifting a parcel moist adiabatically from the estimated cloud base. The resulting estimations of cloud base and top height were then compared against subjective interpretations of radiosonde observations. Preliminary results show that for approximately half of the 48 cases used in the study, the SERCAA algorithm correctly identified the category of fractional cloud cover. For those correctly identified cases, the cloud base height estimations showed no significant bias while the cloud top height estimations were generally too low. In addition to complete presentation of the results from the above, we also hope to have for the extended abstract and at the conference results using updated SERCAA algorithms that include improved thin cirrus detection and optical depth estimations.
We believe that the focus of this proposed paper fits under the planned theme of operational applications since vertical cloud boundaries are important to a number of operational users and the work here is part of a developing satellite data assimilation system.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner