Precipitation Estimation Using Collocated goes Satellite and Surface Data.
Abstract
The separation of precipitation from non-precipitation events using a matrix of GOES-E digitized infrared and visual satellite data was studied. Precipitation verification was conducted with collocated surface observations. The data set consists of 70,623 surface observations of which 29,342 have collocated satellite data. The visual data were normalizd and converted to albedos using the Muench and Keegan (1979) normalization scheme. The data set was separated into four categories (precipitation/no-precipitation and infrared/visual) and after testing for normality, it was determined that none of the categories were normally distributed. Using histograms, a distinct separation between the peaks of precipitating and non-precipitating events was found, but some overlap does exist. Testing of infrared/visual thresholds for precipitation/no-precipitation events used in automated cloud and precipitation research yielded a correct estimation rate of 92% when the infrared and visual thresholds were combined.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1984
- Accession Number
- ADA154248
Entities
People
- D. W. Rust
Organizations
- Naval Postgraduate School