Comparison of Incoherent and Coherent Wave Field Measurements Using Dual-Polarized Pulse-Doppler X-Band Radar

Abstract

Radar-based remote sensing for measurement of ocean surface waves presents advantages over conventional point sensors such as wave buoys. As its use becomes more widespread, it is important to understand the sensitivity of the extracted wave parameters to the characteristics of the radar and the scatterers. To examine such issues, experiments were performed offshore of the Scripps Institution of Oceanography pier in July 2010. Radar measurements in low wind speeds were performed with a dual-polarized high-resolution X-band pulse-Doppler radar at low grazing angles along with two independent measurements of the surface waves using conventional sensors, a GPS-based buoy, and an ultrasonic array. Comparison between radar cross section (RCS) and Doppler modulations shows peak values occurring nearly in-phase, in contrast with tilt modulation theory. Spectral comparisons between Doppler-based and RCS-based spectra show that Doppler-based spectra demonstrate greater sensitivity to swell-induced modulations, whereas RCS-based spectra showgreater sensitivity to small-scale modulations (or generally have more noise at high frequency), and they equally capture energy at the wind wave peak. Doppler estimates of peak period were consistent with the conventional sensors, whereas the RCS differed in assignment of peak period to wind seas rather than swell in a couple of cases. Higher order period statistics of both RCS and Doppler were consistent with the conventional sensors. Radar-based significant wave heights are lower than buoy-based values and contain nontrivial variability of ∼33%. Comparisons between HH and VV polarization data show that VV data more accurately represent the wave field, particularly as the wind speeds decrease.

Use of Proper Orthogonal Decomposition for Extraction of Ocean Surface Wave Fields from X-Band Radar Measurements of the Sea Surface

“…These low Hs cases are associated with lower wind speeds, with the exception of one dataset that had a similarly low wind speed (~7 m/s), but a larger Hs, presumably due to the second wave (swell) component. The inaccurate Hs for small waves and low wind speeds is typically due to insufficient Bragg scatterers, and is a common problem with radar-based measurements [2,3,5]. The lowest wind speed case (~2 m/s) was not associated with the lowest wave height case, although only one wave system was present.…”

“…Thus, we chose to compute the statistics along the radar identified peak wave direction with the understanding that any secondary wave system will only be encompassed as a projection of their orbital velocity along that (primary system) direction. This procedure is also adopted because in the conversion between orbital velocity and wave height, it is assumed that the radar is looking along the wave propagation direction [5].…”

“…Vrms and λp statistics are computed the same way as described in the previous paragraph for the buoy data, except the conversion from a frequency to a wavenumber spectrum is not needed, because the 1D radar spectrum is already a function of wavenumber. Hs is computed in the same manner as the buoy data, except that the 1D spectrum of vertical velocity is first converted to a sea surface displacement spectrum using the method outlined in Hackett et al [5] and Hwang et al [2]. The resulting statistics are then averaged over all of the radar frames in the time series, which covers 2-3 min in duration depending on the dataset.…”

“…Because the modulation transfer function between radar backscatter and sea surface height is complex [1], Doppler measurements of the sea surface have become an increasingly more common measurement technique because the transfer function is more straightforward [2][3][4][5][6]. Regardless of whether using a coherent or incoherent system, marine radar measurements of the In summary, these results are consistent with the numerical results of Hackett et al [14], and show experimentally that POD is a viable alternative to FFT-based dispersion curve filtering for the separation of wave contributions to radar measurements of the sea surface and the subsequent calculation of wave statistics.…”

Ocean Wind and Wave Measurements Using X-Band Marine Radar: A Comprehensive Review

“…Presently, X-band coherent marine radar is undergoing rapid development for wave field and sea surface current measurements [109][110][111][112][113]. The advantage of coherent radar is that it provides not only the image intensity but also the radial velocity of the sea surface scatterers.…”