A Surface-Wind Speed Conversion Technique Using WSR-88D Doppler Weather Radar Velocity Data During Tropical Cyclone Landfall Events

For the past seven decades, reconnaissance aircraft have provided the bulk of wind data that have been used to estimate the intensity of tropical cyclones (TC). However, this indispensable in situ information has provided little more than snapshots in time and space of a tropical cyclone’s intensity and wind field structure, and generally at altitudes of approximately 300, 800, 1500, and 3,000 meters above sea level or 975, 925, 850, and 700 hPa, respectively. Various remote sensing platforms such as space-borne scatterometers (e.g., ASCAT) and microwave radiometers (e.g., SSMI/S) can provide much more detailed wind data about a TC’s entire near-surface (i.e., 10-meters altitude above sea level) wind field, but these data are gathered infrequently and cannot resolve peak intensities above approximately 41 ms-1 (80 kt) due the relatively coarse field-of-view resolution (12.5 ─ 50 km) of the various instruments.

By contrast, land-based WSR-88D Doppler weather surveillance radars operated by the U.S. National Weather Service can provide a complete depiction of TC’s wind field at multiple altitudes, albeit at relatively short distances from populated coastlines, by utilizing the Doppler velocity data. The WSR-88D Doppler radar can aid forecasters by filling in the gaps between the narrow aircraft flight legs, along with temporal changes in the structure of a TC’s horizontal and vertical wind fields and eyewall, including the cyclone’s estimated intensity and radius of maximum winds (RMW).

Although Doppler weather radars can only measure winds that contain a component of motion toward away from the radar site (i.e., radial velocities), those velocity values indicate at least the minimum wind speed present at any location within the cyclone. However, when the wind direction is aligned parallel to the radar beam, then the Doppler velocity measured is also the actual wind speed at that altitude.

Reconnaissance aircraft sample winds along a very narrow flight path, whereas Doppler weather radar data are collected within a relatively large and contiguous volume sample that increases in size with increasing range. The surveillance aircraft typically collect peak 10-second wind data along approximately 1-km-long flight track segments; WSR-88D Doppler data are collected in 250-m long radial bins along 360 azimuthal directions. To approximate the 1-km-long flight leg segments, WSR-88D Doppler radar velocity data were averaged along four (4) contiguous 250-m radial bins yielding a 4-bin average velocity (4BAV) value. The 4BAV value was then divided by the cosine of the angular difference between the aircraft-derived 30-second wind direction and the radar radial azimuth, yielding an environmental wind speed or actual tangential velocity (V_actual). The V_actual values were then compared and linearly correlated to the corresponding aircraft-derived 10-second average wind speeds.

WSR-88D radar collection times for U.S. landfalling hurricanes from 2005-2018 were limited to within +/-2.5 minutes of the reconnaissance aircraft wind observation and within +/-250 m (820 ft) of the aircraft flight-level altitude. Despite the non-concurrent aircraft and radar data sampling, a correlation coefficient of 0.95 was obtained, indicating excellent fidelity between the aircraft-derived wind speeds and the radar-derived 4BAV V_actual wind speeds.

In this study, the V_actual wind speeds were reduced to equivalent 10-meter wind speeds using the National Hurricane Center’s standard aircraft flight-level reduction values of 0.80, 0.75, 0.80, and 0.90 corresponding to altitudes of 975, 925, 850, and 700 hPa, respectively, for some past historical hurricanes. In the cases of Hurricane Katrina (2005) and Hurricane Harvey (2017), 4BAV V_actual wind data indicate that Katrina actually contained 10-meter wind speeds of approximately 64.5 ms^-1 (125 kt) or Category 4 strength vice the post-operationally assessed Category 3 status at the landfall locations in southeastern Louisiana and along the Mississippi coast, while Harvey made landfall along the Central Texas coast with an intensity of 56.5 ms^-1 (110 kt) or Category 3 strength vice the operationally and post-operationally assessed Category 4 status.