Abstract

The Tropical Rainfall Measuring Mission (TRMM), launched in late November 1997 into a low earth orbit, produced the longest microwave radiometric data time series of 17-plus years from the TRMM Microwave Imager (TMI). The Global Precipitation Measuring (GPM) mission is the follow-on to TRMM, designed to provide data continuity and advance precipitation measurement capabilities. The GPM Microwave Imager (GMI) performs as a brightness temperature (Tb) calibration standard for the intersatellite radiometric calibration (XCAL) for the other constellation members; and before GPM was launched, TMI was the XCAL standard. This dissertation aims at creating a consistent oceanic multi-decadal Tb data record that ensures an undeviating long-term precipitation record covering TRMM-GPM eras. As TMI and GMI share only a 13-month common operational period, the U.S. Naval Research Laboratory's WindSat radiometer, launched in 2003 and continuing today provides the calibration bridge between the two. TMI/WindSat XCAL for their > 9 years' period, and WindSat/GMI XCAL for one year are performed using a robust technique developed by the Central Florida Remote Sensing Lab, named CFRSL XCAL Algorithm, to estimate the Tb bias of one relative to the other. The 3-way XCAL of GMI/TMI/WindSat for their joint overlap period is performed using an extended CFRSL XCAL algorithm. Thus, a multi-decadal oceanic Tb dataset is created. Moreover, an important feature of this dataset is a quantitative estimate of the Tb uncertainty derived from a generic Uncertainty Quantification Model (UQM). In the UQM, various sources contributing to the Tb bias are identified systematically. Next, methods for quantifying uncertainties from these sources are developed and applied individually. Finally, the resulting independent uncertainties are combined into a single overall uncertainty to be associated with the Tb bias on a channel basis. This dissertation work is remarkably important because it provides the science community with a consistent oceanic multi-decadal Tb data record, and also allows the science community to better understand the uncertainty in precipitation products based upon the Tb uncertainties provided.

Graduation Date

2018

Semester

Spring

Advisor

Jones, W Linwood

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Electrical Engineering and Computer Engineering

Degree Program

Electrical Engineering

Format

application/pdf

Identifier

CFE0006987

URL

http://purl.fcla.edu/fcla/etd/CFE0006987

Language

English

Release Date

May 2018

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Share

COinS