FN Archimer Export Format
PT J
TI Low-Frequency Sea Surface Radar Doppler Echo
BT 
AF YUROVSKY, Yury Yu.
   KUDRYAVTSEV, Vladimir N.
   GRODSKY, Semyon A.
   CHAPRON, Bertrand
AS 1:1,2;2:1,2;3:3;4:2,4;
FF 1:;2:;3:;4:PDG-ODE-LOPS-SIAM;
C1 Russian Acad Sci, Marine Hydrophys Inst, 2 Kapitanskaya, Sevastopol 299011, Russia.
   Russian State Hydrometeorol Univ, Satellite Oceanog Lab, 98 Malookhotinskiy, St Petersburg 195196, Russia.
   Univ Maryland, Dept Atmospher & Ocean Sci, College Pk, MD 20742 USA.
   Inst Francais Rech Exploitat Mer, F-29280 Plouzane, France.
C2 RUSSIAN ACAD SCI, RUSSIA
   UNIV RUSSIAN STATE HYDROMETEOROL, RUSSIA
   UNIV MARYLAND, USA
   IFREMER, FRANCE
SI BREST
SE PDG-ODE-LOPS-SIAM
UM LOPS
IN WOS Ifremer jusqu'en 2018
   DOAJ
   copubli-int-hors-europe
IF 4.118
TC 9
UR https://archimer.ifremer.fr/doc/00445/55646/57298.pdf
LA English
DT Article
DE ;radar;ocean;backscatter;Doppler shift;wave groups;non-linearity;modulation
AB The sea surface normalized radar backscatter cross-section (NRCS) and Doppler velocity (DV) exhibit energy at low frequencies (LF) below the surface wave peak. These NRCS and DV variations are coherent and thus may produce a bias in the DV averaged over large footprints, which is important for interpretation of Doppler scatterometer measurements. To understand the origin of LF variations, the platform-borne Ka-band radar measurements with well-pronounced LF variations at frequencies below wave peak (0.19 Hz) are analyzed. These data show that the LF NRCS is coherent with wind speed at 21 m height while the LF DV is not. The NRCS-wind correlation is significant only at frequencies below 0.01 Hz indicating either differences between near-surface wind (affecting radar signal) and 21-m height wind (actually measured) or contributions of other mechanisms of LF radar signal variations. It is shown that non-linearity in NRCS-wave slope Modulation Transfer Function (MTF) and inherent averaging within radar footprint account for NRCS and DV LF variance, with the exception of VV NRCS for which almost half of the LF variance is unexplainable by these mechanisms and perhaps attributable to wind fluctuations. Although the distribution of radar DV is quasi-Gaussian, suggesting virtually little impact of non-linearity, the LF DV variations arise due to footprint averaging of correlated local DV and non-linear NRCS. Numerical simulations demonstrate that MTF non-linearity weakly affects traditional linear MTF estimate (less than 10% for typical MTF magnitudes less than 20). Thus the linear MTF is a good approximation to evaluate the DV averaged over large footprints typical of satellite observations 
PY 2018
PD JUL
SO Remote Sensing
SN 2072-4292
PU Mdpi
VL 10
IS 6
UT 000436561800064
DI 10.3390/rs10060870
ID 55646
ER

EF