Study on direction dependence of the fully polarimetric wind-induced ocean emissivity at L-band using a semi-theoretical approach for Aquarius and SMAP observations

REMOTE SENSING OF ENVIRONMENT

卷:265

Article

Ma, Wentao^1,3  Du, Yanlei^1,2,3  Liu, Guihong¹  Yu, Yang¹  Yang, Xiaofeng^1,3  Yang, Jian²  Chen, Kun-Shan^1,4 

[1] Chinese Acad Sci, Aerosp Informat Res Inst, State Key Lab Remote Sensing Sci, Beijing 100101, Peoples R China

[2] Tsinghua Univ, Dept Elect Engn, Beijing 100084, Peoples R China

[3] Key Lab Earth Observat Hainan Prov, Sanya 572029, Peoples R China

[4] Guilin Univ Technol, Coll Geomat & Geoinformat, Guilin 541004, Peoples R China

关键词: Fully polarimetric ocean emissivity;    L-band;    Passive remote sensing;    Directional wave spectrum;    Foam;    Aquarius;    SMAP;   

DOI  :  10.1016/j.rse.2021.112661

来源: Elsevier

【 摘 要 】

The wind direction dependence of the fully polarimetric ocean emissivity at L-band is studied using a semitheoretical approach. The model, which is tuned and validated against Aquarius and Soil Moisture ActivePassive (SMAP) mission observations, incorporates the two-scale polarimetric ocean emission model, an improved directional wave spectrum, the Anguelova and Gaiser, 2013 (AG13) foam emissivity model and a novel foam coverage model. Wind speeds up to 35 m/s are considered. Comparisons of model simulations to satellite data show that the overall accuracy of model could reach 0.2 K, which is slightly worse than the Aquarius geophysical mode function (GMF) error of about 0.14 K. The negative and positive upwind-crosswind asymmetries of L-band ocean emission and the transition of the phase signatures versus wind speed are well reproduced for all polarizations by the proposed model. The average standard deviations of the differences of polarimetric brightness temperature (TB) between observations and model are less than about 0.16 K for wind speeds up to 15 m/s. Simulations also indicate the anisotropy of L-band ocean emissivities for V- and H- pol would be saturated at high wind speeds. The directional signals at the third Stokes component begin to converge at around the wind speed of 24 m/s. The L-band TB variations due to wind direction for the third Stokes parameter show less sensitivity to wind speed at larger observation angles, while the direction-induced TB variations for the fourth Stokes are more sensitive to wind speed with the increase of observation angle.

【 授权许可】

Free