Measurement of Correlation Functions and Power Spectra in Clouds Using the NRL WARLOC Radar

Abstract

The Naval Research Laboratory W-band Advanced Radar for Low Observable Control (WARLOC) is a highpower 94-GHz radar, with 3-10-kW average and 80-kW peak power, now set up on the western shore of the Chesapeake Bay. It has three orders of magnitude more power and sensitivity than other W-band radar systems. This enables cloud reflectivity to be measured with high signal-to-noise ratios and a resolution of about 15 m over a two (or three)-dimensional region, which can be as large as tens of kilometers on a side. This allows imaging of the internal structure of clouds over great distances in very great detail. At the shortest scale lengths, the structure has a speckle pattern indicating that it is governed at least in part by stochastic processes. The WARLOC data allow the reflectivity correlation function and its Fourier transform, which is the power spectrum, to be computed for scale lengths ranging from 30 m to 10 km. For measurements taken of cirrus clouds on several occasions as well as one case of a precipitating cloud, it was found that for small correlation distance r, the correlation function usually decreases as r 2/3 , or, equivalently, the wavenumber spectrum scales as k −5/3 , where k is the wavenumber. This suggests that fluid turbulence in the inertial range is playing a role; however, unlike classical fluid turbulence, the results suggest that the turbulence here is generally quite anisotropic. Furthermore, for longer scale lengths (the outer range according to the Kolmogorov theory), the reflectivity fluctuations usually show a wavelike behavior in the vertical direction but only occasionally in the horizontal direction.

Geographic smoothing of solar PV: results from Gujarat

“…The expected diurnal peaks at 24 h and harmonics are present. At high frequencies, the plants exhibit a spectrum similar to cloud processes [38,39]. These processes may be a function of the f 5 23°54′N, 71°12′E), has an f 1.76 spectrum.…”

All-Weather Calibration of Wide-Field Optical and Nir Surveys

“…3). Cloud structure can be intricate with significant spatial variations across a single field of view, and can change in the time interval between exposures (Fliflet & Manheimer 2006;Koren et al 2008). In the worst case, the loss of light will be too severe to allow useful data to be taken.…”

Spatial correlation of hydrometeor occurrence, reflectivity, and rain rate from CloudSat

“…An examination of the power spectra (not shown) indicates there is generally a scale break (below the correlation length scale) where the spectrum behaves as $k À3 to k À1.5 depending on location, while at scales larger than the correlation length the spectrum goes as $k À1.5 to k À1.2 . We noted in section 1 that Fliflet and Manheimer [2006] used a high-power scanning millimeter-wavelength (94 GHz, W band) radar to study the spatial correlation of reflectivity from cirrus clouds and that they found the reflectivity autocorrelation function decreased with scale as r 2/3 (that is, b = 2/3 and equivalent to a wave number spectrum of k À5/3 ) for scale lengths ranging from 30 m to 10 km. Restricting the range from 1 to 10 km, we too find b = $2/3 for midlatitude clouds in the upper troposphere (and b $ 0.5 when fitting over the range 1-200 km).…”

“…However, the spatial structure of hydrometeor (cloud and precipitation) reflectivity and Doppler velocities from centimeter and millimeter wavelength radar have also been investigated [e.g., Smythe and Zrnic , 1983; Heymsfield , 1976; Lothon et al , 2005]. Fliflet and Manheimer [2006] used a high‐power scanning millimeter‐wavelength (94 GHz, W‐band) radar to study the spatial correlation of reflectivity from cirrus clouds. They found the reflectivity autocorrelation function decreased with scale as r 2/3 (equivalent to a wave number spectrum of k −5/3 ) for scale lengths ranging from 30 m to 10 km.…”