A Multi-Sensor Water Vapor Climate Data Record Using Cloud Classification
- Visit Multi-Sensor Water Vapor Climate
We are assembling a record of atmospheric water vapor using a variety of recent and historic data sets. This work is coordinated with the MEaSUREs water vapor record effort titled "Improvement of the NVAP Global Water Vapor Data Set for Climate, Hydrological, and Weather Studies." The goal of the combined effort is a long term, credible record of atmospheric water vapor scanning several decades.
We hypothesize that both clouds and water vapor are changing, so both must be tracked carefully to understand climate variability. Most of our effort will utilize the state-of-the-art humidity and temperature observations from space made by AIRS/AMSU/HSB, AMSR-E, MLS, and MODIS in the A-Train satellite constellation. We are currently assembling a multi-year record of A-Train water vapor and temperature observations. We will use the multi-sensor A-Train observations to classify individual water vapor scenes by cloud type, using well-established cloud classification methods. Initially, our cloud classes are the standard cloud class product created by the CloudSat team, and these will be applied to AIRS and AMSR-E water vapor observations. We will next apply CloudSat cloud classes to MODIS and MLS, where appropriate. We will also examine the feasibility of extending similar classes to other cloud observations, especially those from MODIS. If feasible, the MODIS classes will be applied to the entire A-Train record. The resulting data sets are designed to separate trends due to changes in water vapor from trends due to in viewing conditions induced by clouds or precipitation.
This record will be merged with the historic record in several ways. Radiances from two MLS instruments - one currently on Aura and an earlier one on the UARS satellite - will be processed using a consistent algorithm to provide relative humidity observations in the upper troposphere dating back to 1991. Microwave-only observations from operational satellite sensors are less affected by clouds than are infrared observations so are less susceptible to cloud-induced changes in sampling. We will use microwave water vapor observations from recent operational microwave instruments, co-located with A-Train observations, to constrain trends in the A-Train data. Because operational microwave data are fundamental to the NVAP effort, this comparison will be coordinated with the NVAP team as they update their record into the A-Train era starting in 2002. We will also examine the feasibility of classifying the NVAP observations using well-established cloud classes provided by the ISCCP climatology. The NVAP record begins in 1987 with the SSM/I instruments. We will also work with the NVAP team to examine the possibility of extending the atmospheric water vapor record back in time to 1979 when the TOVS instruments from began operating.
Dr Eric Fetzer - PI, JPL
Distributed by GES DISC
Page Last Updated: May 2, 2019 at 11:41 AM EDT