Shallow Water Depth Inversion Based on the Multi-Spectral Data of GF-1 Satellite— Case Study on the Ganquan Island
- Lu Wenhu (National Marine Data and Information Service) | Liu Shuming (National Marine Data and Information Service) | Cai Renhan (National Marine Data and Information Service, Tianjin University) | Yin Yue (National Marine Data and Information Service) | Yu Ting (National Marine Data and Information Service)
- Document ID
- International Society of Offshore and Polar Engineers
- The 28th International Ocean and Polar Engineering Conference, 10-15 June, Sapporo, Japan
- Publication Date
- Document Type
- Conference Paper
- 2018. International Society of Offshore and Polar Engineers
- remote sensing, GF-1 satellite, water depth inversion, islands and reefs
- 0 in the last 30 days
- 13 since 2007
- Show more detail
Remote sensing technology is an efficient and accurate tool for realtime large-scale inversion of numerous and complex islands and reefs in South China Sea. The satellite multi-spectral data over Ganquan Island acquired by China GF-1 satellite on May 12, 2016 was converted to land surface reflection by radiometric correction with absolute calibration coefficient, geometric correction and atmospheric correction. And the Pearson correlation analysis of the single-band reflectance and dual-band reflectance ratios of 93 water depth measurements and their corresponding pixels in the study area were analyzed. The correlation analysis showed that the negative correlation of reflectance value with the single band (P <0.01), and the ratio of double band was significantly correlated with water depth (P <0.01). The Band2, Band3, Band2/Band1 and Band4/Band2, which have the highest correlations with the water depth, are selected to construct the single band model, double band model and band-ratio model, respectively. The parameters of the inversion model are obtained by the regression analysis. The results show that band ratio model of Band2/Band1 and Band4/Band2 fit the best of water depth inversion in the study area. Together with the in-situ observation of 30 stations in the study area, the RMSE, MAE and MRE methods are used to analyze the precision of the water depth inversion. The comprehensive accuracy of water depth inversion derived from the model on the selected band is 1.44m, which is better than that of traditional blue-green band, suggesting a more suitable remote sensing inversion in Ganquan Island area.
The South China Sea (SCS) is a northeastward marginal sea with very complex topography. It is one of the world's busiest shipping areas (Zhang et al. 2015). However, the islands and seamounts scattered in continental shelf put potential threats to the safe navigation there. In order to ensure the safety of navigation, the shallow water depth measurement has always been an important part of the marine surveying and mapping. Currently, shallow water depth measurement is mainly based on traditional ship borne equipment, including Single Beam and Multi-beam Echo-soundings (Huang et al. 2013). These methods have high measuring accuracy, as well as high cost and long observation cycle (Shi et al. 2008). By using multi-spectral remote sensing technology, extraction of water depth information from shallow sea could be obtained in a relatively short time (Dang and Ding, 2001). This method is characterized by large coverage and real-time. It is possible to dynamically obtain the water depth information in the large and shallow seas, and has an important implication for measuring the shallow water depth around the islands in SCS (Zhou, Zhang and Zou, 2012; Liu, Hu, and He, 2014).
|File Size||2 MB||Number of Pages||6|