Improving SFMR Surface Wind Measurements in Heavy Rain Conditions

With the installation of Stepped Frequency Microwave Radiometers (SFMRs) on Air Force Reserve Command WC-130J hurricane reconnaissance aircraft, the SFMR has assumed a prominent role for operational measurement of surface winds, and thus, hurricane intensity estimation. The current SFMR wind retrieval algorithm was developed from GPS dropwindsonde surface wind measurements (Uhlhorn et al. 2007; Uhlhorn and Black 2003), and has been successfully implemented across all SFMR-equipped aircraft. The algorithm improvements were specifically targeted at improving surface wind accuracy at hurricane force conditions (> 65 kts, 33 m/s), especially within the eyewall, although the SFMR surface wind vs. emissivity geophysical model function was developed over a broad range of wind speeds (10-140 kts, 5-70 m/s) with the expectation that the hurricane wind field could be readily measured in general.

Over the past few years, it has been recognized that SFMR surface winds show a tendency to be biased high within heavy precipitation. This bias is particularly evident at weak-to-moderate wind speeds (< 65 kts, 33 m/s), which has important implications for identifying tropical systems at the depression and storm stages, and additionally for observing significant outer wind radii. We believe there are two major reasons for this issue: 1) at weak-to-moderate wind speeds, the current GMF version was tuned to GPS surface winds obtained within areas largely free of heavy precipitation; and 2) a biased rain absorption model was used to develop the current surface emissivity vs. wind speed geophysical model function.

In this work, we have expanded the SFMR wind speed versus GPS dropsonde surface-adjusted wind speed database to include data from 2005-2011. Due to this expansion, an increased number of values within the higher rain rate regimes has been included. Based on this database, bin-averaged wind speed differences are evaluated within several wind speed and rain rate bins. The wind speed errors represented in each bin confirm that the largest differences are at the lowest wind speeds and highest rain rates. A weighted polynomial fit to these error data is created and used as a correction to the current SFMR wind speed measurements, where the effects are minimized at the higher wind speeds and are maximized in the lower wind speeds. These corrections are then transferred to operations as forecasters can use the corrections to assess a more accurate wind speed in the heavy rain conditions.