Developing Climate Data Records from Microwave Water Vapor Channels

Microwave (MW) satellite data are important for weather forecasting as well as climate monitoring and assessment. MW data are widely used to derive various hydrological products, including precipitation, precipitable water vapor, snow cover, tropospheric humidity, cloud liquid water, deep convective clouds, and land surface emissivity. The primary application of MW data is for weather forecasts, so the data are not calibrated with sufficiently stability for climate applications. Therefore, current MW data must be properly corrected for different errors such as geolocation, calibration, scan asymmetry, and scan bias, prior to using them for developing long-term climate data records.

This presentation first describes results of a comprehensive study on the geolocation correction for the observations from microwave instruments (AMSU-A, AMSU-B, and MHS) onobard NOAA satellites then discusses the bias correction and inter-calibration of the observations from water vapor channels of these instruments onboard NOAA and MetOp satellites. In terms of geolocation, several import errors was found including about one degree mounting offset for the NOAA-15 AMSU-A Channel 1 as well as a large timing error related to NOAA-17 AMSU-A observations. The calibration drift was investigated using tropical and polar averages and noticeable drift was found in NOAA-15 and NOAA-16 AMSU-B water vapor channels. The time series of the observations show significant improvement and consistency between multiple satellites after inter-calibration and bias correction.