Ocean wind and wave parameter estimation from ship-borne x-band marine radar data
暂无分享,去创建一个
Ocean wind and wave parameters are important for the study of oceanography, on- and
off-shore activities, and the safety of ship navigation. Conventionally, such parameters
have been measured by in-situ sensors such as anemometers and buoys. During the
last three decades, sea surface observation using X-band marine radar has drawn wide
attention since marine radars can image both temporal and spatial variations of the sea
surface. In this thesis, novel algorithms for wind and wave parameter retrieval from
X-band marine radar data are developed and tested using radar, anemometer, and buoy
data collected in a sea trial off the east coast of Canada in the North Atlantic Ocean.
Rain affects radar backscatter and leads to less reliable wind parameters measurements.
In this thesis, algorithms are developed to enable reliable wind parameters
measurements under rain conditions. Firstly, wind directions are extracted from raincontaminated
radar data using either a 1D or 2D ensemble empirical mode decomposition
(EEMD) technique and are seen to compare favourably with an anemometer reference.
Secondly, an algorithm based on EEMD and amplitude modulation (AM) analysis to
retrieve wind direction and speed from both rain-free and rain-contaminated X-band
marine radar images is developed and is shown to be an improvement over an earlier 1D
spectral analysis-based method.
For wave parameter measurements, an empirical modulation transfer function (MTF)
is required for traditional spectral analysis-based techniques. Moreover, the widely used
signal-to-noise ratio (SNR)-based method for significant wave height (HS) estimation
may not always work well for a ship-borne X-band radar, and it requires external sensors
for calibration. In this thesis, two methods are first presented for HS estimation from
X-band marine radar data. One is an EEMD-based method, which enables satisfactory
HS measurements obtained from a ship-borne radar. The other is a modified shadowingbased
method, which enables HS measurements without the inclusion of external sensors.
Furthermore, neither method requires the MTF. Finally, an algorithm based on the Radon transform is proposed to estimate wave direction and periods from X-band marine radar
images with satisfactory results.