Study On Wind Retrieval Algorithm For Altimeter At High Wind Speeds

Study on Wind Retrieval Algorithm for Altimeter at High Wind SpeedsStrong winds accompanying typhoons and hurricanes may bring not only catastrophic destruction to marine activities, but also have dramatically influence on physical and ecological processes of the upper ocean. So it is significant to improve altimeter wind retrieval algorithms at high wind speeds. A better algorithm is not only effective for better understanding and forecasting of the cyclone's path and intensity, but also useful to improve the retrieval of sea surface height under cyclone. The wind speeds retrieved with higher accuracy are related to better estimates of oceanic heat storage and flux, which are directly influenced by the passing typhoons or hurricanes. In words, strong winds have a close relation with oceanic engineering, physical oceanography and marine biology. The study on strong winds has scientific significance and practical importance.The study on altimeter wind retrieval algorithm has always been a hot topic. However, the altimeter wind speed algorithm at high wind speeds remains unsolved due to lack of observed data. This study aims at developing a wind retrieval algorithm for altimeter at high wind speeds.First, some proposed wind-speed algorithms and typhoon models were introduced and analyzed. We follow Young's (1993) idea, but do not use the Holland typhoon wind field model. The wind speeds in our study are generated by Yin's wind field model (Yin et al., 2007) with parameters provided by the Joint Typhoon Warning Center (JTWC). The accuracy of Yin's typhoon model has been validated by comparing it with recorded data from a weather station. By comparing the normalized radar backscatter cross sections detected by the Jason-1 altimeter with wind speed data inferred by Yin's typhoon model, an empirical algorithm valid for a range of wind speeds between 10 and 40ms-1 is developed and proposed. A comparison between the proposed algorithms and present algorithm is carried out. The result indicated when wind speed is lower than 27 ms-1, the proposed algorithm and Jason-1 operational algorithm have similar accuracy. For wind speeds higher than 20 ms-1, altimeter wind speeds retrieved from the proposed algorithm fit the actual typhoon wind speed better; the RMS and bias of the new algorithm are lower. Hence, the new algorithm can retrieve wind speed more accurately and can be treated as a supplement for the Jason-1 operational algorithm at condition of high wind speeds. Remote sensing reflectance has a direct relationship with backscattering coefficient of water. However,backscattering data is quite deficient due to the limitation of survey technology and funding, therefore study on the water body optical properties through the survey data is unfeasible.This text introduced how to measure and dispose parameters related to inherent optical properties in Bohai Sea in brief. Base on the data of AC9 and HS6 at the spectral bands of 442nm,488nm,532nm,676nm, statistical relations in the Bohai sea between beam scattering coefficient b_p(λ) and backscattering coefficient b_bp(λ) are obtained. The author find there is high correlation between the beam scattering coefficient b_p(λ) and the backscattering coefficient b_bp(λ) for particulates. Regression analysis shows that there are two distinct trends between b_bp(λ) and b_p(λ). In the cross sections of HHK, statistical relation between b_bp(λ) and b_p(λ) can be described by a polynomial function; in the cross sections of BHW, LDW and QHD, statistical relation between b_bp(λ) and b_p(λ) can be described by a power function. Thus this statistical relation can be used to retrieve the backscattering coefficient from the beam scattering coefficient measured by AC9. By comparing the backscattering coefficient measured by HS6 and those retrieved from beam scattering coefficient measured by AC9 using the statistical relation at the residual wave band, we found that the correlative coefficients (R2) between the retrieval data and the in-situ data are 0.97 and 0.96, the average relative errors are 22% and 18%, the root mean square(RMS)errors are 0.031 and 0.055, respectively for two cases. The results indicate the statistical relation between b_bp(λ)and b_p(λ) proposed by this study is usable. Therefore, the backscattering coefficient can be obtained from the AC9 measured data. This is significant for development of the ocean color remote sensing analysis model. The reasons that the statistical relation between b_bp(λ) and b_p(λ) is different along with the sea areas are: (1) The content of inorganic suspended matter determines the relationship between b_bp(λ) and b_p(λ) when inorganic suspended matter is higher; (2) When the inorganic suspended matter is lower, the shape of superiority algae has more effect on the relationship between b_bp(λ) and b_p(λ).