| Summary: | Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, February 2007. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Includes bibliographical references (p. 229-234). This thesis develops and validates the MM5/TBSCAT/F([lambda]) model, composed of a mesoscale numerical weather prediction (NWP) model (MM5), a two-stream radiative transfer model (TBSCAT), and electromagnetic models for icy hydrometeors (F([lambda])), to be used as a global precipitation ground-truth for evaluating alternative millimeter-wave satellite designs and for developing methods for millimeter-wave precipitation retrieval and assimilation. The model's predicted millimeter-wave atmospheric radiances were found to statistically agree with those observed by satellite instruments [Advanced Microwave Sounding Unit-A/B (AMSU-A/B)] on the United States National Ocean and Atmospheric Administration NOAA-15, -16, and -17 satellites over 122 global representative storms. Whereas such radiance agreement was found to be sensitive to assumptions in MM5 and the radiative transfer model, precipitation retrieval accuracies predicted using the MM5/TBSCAT/F([lambda]) model were found to be robust to the assumptions. (cont.) Appropriate specifications for geostationary microwave sounders and their precipitation retrieval accuracies were studied. It was found that a 1.2-m micro-scanned filled-aperture antenna operating at 118/166/183/380/425 GHz, which is relatively inexpensive, simple to build, technologically mature, and readily installed on a geostationary satellite, could provide useful observation of important global precipitation with ~20-km resolution every 15 minutes. AMSU global precipitation retrieval algorithms for retrieving surface precipitation rate, peak vertical wind, and water-paths for rainwater, snow, graupel, cloud water, cloud ice, and the sum of rainwater, snow, and graupel, over non-icy surfaces were developed separately using a statistical ensemble of global precipitation predicted by the MM5/TBSCAT/F([lambda]) model. Different algorithms were used for land and sea, where principal component analysis was used to attenuate unwanted noises, such as surface effects and angle dependence. The algorithms were found to perform reasonably well for all types of precipitation as evaluated against MM5 ground-truth. The algorithms also work over land with snow and sea ice, but with a strong risk of false detections. AMSU surface precipitation rates retrieved using the algorithm developed in this thesis reasonably agree with those retrieved for the Advanced Microwave Scanning Radiometer for the Earth Observing System (AMSR-E) aboard the Aqua satellite over both land and sea. (cont.) Surface precipitation rates retrieved using the Advanced Microwave Sounding Unit (AMSU) aboard NOAA-15 and -16 satellites were further compared with four similar products derived from other systems that also observed the United States Great Plains (USGP) during the summer of 2004. These systems include AMSR-E aboard the Aqua satellite, the Special Sensor Microwave/Imager (SSM/I) aboard the Defense Meteorological Satellite Program (DMSP) F-13, -14, and -15 satellites, the passive Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) aboard the TRMM satellite, and a surface precipitation rate product (NOWRAD), produced and marketed by Weather Services International Corporation (WSI) using observations from the Weather Surveillance Radar-1988 Doppler (WSR-88D) systems of the Next-Generation Weather Radar (NEXRAD) program. The results show the reasonable agreement among these surface precipitation rate products where the difference is mostly in the retrieval resolution, which depends on instruments' characteristics. A technique for assimilating precipitation information from observed millimeter-wave radiances to MM5 model was proposed. Preliminary study shows that wind and other correction techniques could help align observations at different times so that information from observed radiances is used at appropriate locations. by Chinnawat Surussavadee. Ph.D.
|
|---|