Enhancing PMW Satellite Precipitation Estimation: Detecting Convective Class

A decades-long effort in observing precipitation from space has led to continuous improvements of satellite-derived passive microwave (PMW) large-scale precipitation products. However, due to a limited ability to relate observed radiometric signatures to precipitation type (convective and stratiform...

Full description

Saved in:
Bibliographic Details
Published in:Journal of atmospheric and oceanic technology 2019-12, Vol.36 (12), p.2349-2363
Main Authors: Petković, Veljko, Orescanin, Marko, Kirstetter, Pierre, Kummerow, Christian, Ferraro, Ralph
Format: Article
Language:English
Subjects:
Algorithms
Bias
Deep learning
Estimates
Global precipitation
Inverse problems
Lightning
Machine learning
Measurement techniques
Neural networks
Precipitation
Precipitation estimation
Precipitation rate
Precipitation types
Radar
Rain
Remote sensing
Satellites
Sensors
Systematic errors
Variability
Vectors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A decades-long effort in observing precipitation from space has led to continuous improvements of satellite-derived passive microwave (PMW) large-scale precipitation products. However, due to a limited ability to relate observed radiometric signatures to precipitation type (convective and stratiform) and associated precipitation rate variability, PMW retrievals are prone to large systematic errors at instantaneous scales. The present study explores the use of deep learning approach in extracting the information content from PMW observation vectors to help identify precipitation types. A deep learning neural network model (DNN) is developed to retrieve the convective type in precipitating systems from PMW observations. A 12-month period of Global Precipitation Measurement mission Microwave Imager (GMI) observations is used as a dataset for model development and verification. The proposed DNN model is shown to accurately predict precipitation types for 85% of total precipitation volume. The model reduces precipitation rate bias associated with convective and stratiform precipitation in the GPM operational algorithm by a factor of 2 while preserving the correlation with reference precipitation rates, and is insensitive to surface type variability. Based on comparisons against currently used convective schemes, it is concluded that the neural network approach has the potential to address regime-specific PMW satellite precipitation biases affecting GPM operations.
ISSN:0739-0572
1520-0426
DOI:10.1175/JTECH-D-19-0008.1