Ocean Vector Wind Measurements from GPM to Study Diurnal Cycles of Tropical Winds

This paper presents a combined active/passive ocean vector wind (OVW) retrieval from the Global Precipitation Measurement Core Observatory (GPM-CO). Although the GPM-CO was designed to observe global precipitation, we demonstrate the potential for providing a climate quality OVW data product for the multi-decade GPM- TRMM era. Development of GMFs for the Ku- and Ka-band dual frequency Precipitation Radar (DPR) and the GPM Microwave Imager (GMI) remote sensors onboard the GPM-CO is presented. The unique GPM active/passive OVW retrieval algorithm is described, which uses the modified 2nd Stokes parameter (linear combination of V & H-pol Tb's) to cancel the atmosphere and to enhances the anisotropic WD signal superimposed on GMI Tbs. The PR and GMI GMF's were then combined in a maximum likelihood estimation (MLE) algorithm to infer the OVW. Finally, this OVW retrieval algorithm was validated by comparing retrievals with collocated NASA Advanced Scatterometer (ASCAT) wind vectors. The statistics of WS and WD retrieval accuracy performance are presented and are shown to be comparable with those of conventional scatterometer and polarimetric radiometer missions.

I. Introduction

Many studies have reported strong diurnal and semi-diurnal variabilities of tropical vector winds that may affect the cloud formation and precipitation in the tropics [1]–[3]. Moreover, these signals create regional bias between scatterometer wind observation during ascending and descending passes and affect satellite cross-calibration, and thus need to be removed in order to create a long-term climate quality intercalibrated dataset [4]. Therefore, a comprehensive understanding of the diurnal variability of ocean surface vector wind is essential to understand the fundamental mechanism of diurnal variability of precipitation and air-sea interaction in the tropics, and other regional and global climate studies. However, accurate assessment of diurnal winds has remained to be a challenge because all the present and past satellite scatterometers, except the RapidScat on the International Space Station, have been in Sun-synchronous polar orbits, and they sample a given geographical position twice daily (separated by 12 h), which is insufficient temporal sampling to determine the diurnal and semi diurnal cycles of tropical winds.