Total water vapor retrieval data in the Arctic between 2007-2009 using microwave humidity sounders

• Main Author: Scarlat, Raul Cristian
• Format: Article in Journal/Newspaper
• Language: English
• Published: PANGAEA - Data Publisher for Earth & Environmental Science 2018
• Subjects: Arctic; Sea ice
• Online Access: https://dx.doi.org/10.1594/pangaea.888100; https://doi.pangaea.de/10.1594/PANGAEA.888100

Quantitative retrievals of atmospheric water vapour in the Arctic present numerous challenges because of the particular climate characteristics of this area. Here, we attempt to build upon the work of Melsheimer and Heygster (2008) to retrieve total atmospheric water vapour (TWV) in the Arctic from satellite microwave radiometers. While the above-mentioned algorithm deals primarily with the ice-covered central Arctic, with this work we aim to extend the coverage to partially ice covered and ice-free areas. By using modelled values for the microwave emissivity of the ice-free sea surface, we develop two sub-algorithms using different sets of channels that deal solely with open-ocean areas. The new algorithm extends the spatial coverage of the retrieval throughout the year but especially in the warmer months when higher TWV values are frequent. The high TWV measurements over both sea-ice and open-water surfaces are, however, connected to larger uncertainties as the retrieval values are close to the instrument saturation limits. This approach allows us to apply the algorithm to regions where previously no data were available and ensures a more consistent physical analysis of the satellite measurements by taking into account the contribution of the surface emissivity to the measured signal.The retrieved fields of TWV using two separate versions of the AMSU-B retrieval algorithm, an original version with higher accuracy but low coverage and the new version with greatly increased Arctic coverage but higher retrieval uncertainty over ice free areas. The data covers 4 representative months, March, June, September and December in 2007, 2008 and 2009.This data collection is split by year, month and retrieval method.