A Long-Term Precipitation Dataset with Uncertainty Information
Christian Kummerow - PI, Colorado State University
The broad social and economic relevance of rainfall combined with the development of new satellite sensors/techniques for global monitoring has led to multiple long-term global products that rarely agree despite significant efforts towards calibration and validation. These discrepancies are particularly problematic for climate studies because uncertainties in spatially and temporally averaged values are the result of a number of factors including differences between meteorological regimes, sensor calibration, and satellite orbit characteristics.
As a result, these uncertainties cannot be computed from errors in the instantaneous rainfall estimates, but require further insights into the sources of bias errors. Building on algorithm development work funded through NASA's Precipitation Measurement Program, we propose to create a new rainfall and cloud water dataset using the latest version of the Goddard Profiling (GPROF) algorithm and based on a multi-decadal constellation of satellite radiometer data including Special Sensor Microwave Imager (SSM/I), Tropical Rainfall Measuring Mission’s (TRMM) Microwave Imager (TMI), Advanced Microwave Scanning Radiometer - EOS (AMSR-E), WINDSAT and Special Sensor Microwave Imager/Sounder (SSMIS).
The algorithm will be run on the Level 1C brightness temperature dataset, which is part of a currently funded Research, Education and Applications Solutions Network (REASoN) effort to produce a consistent calibrated brightness temperature dataset from this diverse set of radiometers. While this proposal aims to fold in prior data beginning with SSM/I F8 launched in 1987, and has a radiometer intercalibration component that goes beyond the algorithm development work, the bulk of this proposal is not the generation of a new long-term homogeneous data set, but the characterization of the intrinsic uncertainties, including those related to the partitioning between cloud water and rainfall, the sensitivity of rainfall products to meteorological regimes and the relation between this product and previously established ones.
By evaluating and interpreting differences in light of the intrinsic uncertainties, algorithm formulations and input brightness temperatures, we hope to add information about the product that is as essential as the product itself for use in studying climate variability and trends related to the global water and energy processes.
Distributed by NASA's Precipitation Processing System
Last Updated: Nov 14, 2018 at 5:00 PM EST