A revised L-band radio-brightness sensitivity to extreme winds under tropical cyclones: The 5 year SMOS-Storm database
|Author(s)||Reul Nicolas1, Chapron Bertrand1, Zabolotskikh E.2, Donlon C.3, Quilfen Yves1, Guimbard Sebastien1, Piolle Jean-Francois1|
|Affiliation(s)||1 : IFREMER, Ctr Bretagne, ZI Pointe Diable, LOPS, CS 10070, F-29280 Plouzane, France.
2 : RSHU, Satellite Oceanog Lab, Malookhtinsky Prosp 98, St Petersburg 195196, Russia.
3 : European Space Agcy, Estec, Earth Observat Programme Directorate, Mission Sci Div, Keplerlaan 1, NL-2200 AG Noordwijk, Netherlands.
|Source||Remote Sensing Of Environment (0034-4257) (Elsevier Science Inc), 2016-07 , Vol. 180 , P. 274-291|
|WOS© Times Cited||15|
|Note||Special Issue: ESA's Soil Moisture and Ocean Salinity Mission - Achievements and Applications|
|Abstract||Five years of SMOS L-band brightness temperature data intercepting a large number of tropical cyclones (TCs) are analyzed. The storm-induced half-power radio-brightness contrast (ΔI) is defined as the difference between the brightness observed at a specific wind force and that for a smooth water surface with the same physical parameters. ΔI can be related to surface wind speed and has been estimated for ~ 300 TCs that intercept with SMOS measurements. ΔI, expressed in a common storm-centric coordinate system, shows that mean brightness contrast monotonically increases with increased storm intensity ranging from ~ 5 K for strong storms to ~ 24 K for the most intense Category 5 TCs. A remarkable feature of the 2D mean ΔI fields and their variability is that maxima are systematically found on the right quadrants of the storms in the storm-centered coordinate frame, consistent with the reported asymmetric structure of the wind and wave fields in hurricanes. These results highlight the strong potential of SMOS measurements to improve monitoring of TC intensification and evolution. An improved empirical geophysical model function (GMF) was derived using a large ensemble of co-located SMOS ΔI, aircraft and H*WIND (a multi-measurement analysis) surface wind speed data. The GMF reveals a quadratic relationship between ΔI and the surface wind speed at a height of 10 m (U10). ECMWF and NCEP analysis products and SMOS derived wind speed estimates are compared to a large ensemble of H*WIND 2D fields. This analysis confirms that the surface wind speed in TCs can effectively be retrieved from SMOS data with an RMS error on the order of 10 kt up to 100 kt. SMOS wind speed products above hurricane force (64 kt) are found to be more accurate than those derived from NWP analyses products that systematically underestimate the surface wind speed in these extreme conditions. Using co-located estimates of rain rate, we show that the L-band radio-brightness contrasts could be weakly affected by rain or ice-phase clouds and further work is required to refine the GMF in this context.|