a Spectral Approach to the Unification of Rawinsonde and Satellite Temperature Profiles
Abstract
The vertical resolution of temperature profiles derived from VAS temperature retrievals was investigated analytically. Using the data collected during the AVE/VAS Ground Truth Field Experiment, temperature profiles from rawinsonde and VAS observations were represented by Fourier cosine series, then harmonic-by-harmonic comparisons were performed for collocated pairs as well as at the same grid points. It was found that the temperature profiles derived from VAS observations correctly represent variations of the zeroth (mean atmospheric temperature) and first harmonics on the synoptic scale. The synoptic scale temperature variations represented by the second to the fifth harmonics contain true variations, but they also contain some uncertainties. Harmonics higher than the fifth represented random variations or noise. The vertical resolution of the physical-mathematical retrieval for VAS temperature profiles was, therefore, determined to be about 500 mb. Only the zeroth and first harmonics in the VAS temperature soundings were treated as a signal. The portion of signal in the VAS temperature soundings replaced the same portion of the temperature fields from conventional data to form a unified data set. The primary purpose of the unified data is to improve data quality over regions of inadequate conventional data. The quality of the unified data was examined with respect to NMC analyzed data and VAS data by comparing the vertical motion fields derived from a linear balance model. Results for three case studies at 1200 GMT on 22, 23, and 24 August 1983 over the western Atlantic showed, in general, that analyses of the unified data were better than analyses of NMC data over ocean areas. It is suggested that combination of conventional and VAS profiles is at least as good as the quality of the conventional data, and demonstrates the potential for improving the data quality over ocean areas where the conventional data are restricted.
- Publication:
-
Ph.D. Thesis
- Pub Date:
- 1986
- Bibcode:
- 1986PhDT.......167S
- Keywords:
-
- METEOROLOGY;
- VAS;
- Physics: Atmospheric Science; Remote Sensing