Internal Solitary Wave Retrieval Model From Sun Glint Of Optical Remote Sensing Images

Internal solitary waves are ubiquitous at the marginal sea and the continental shelf. Generated and propagating along the pycnocline, internal waves greatly theaten the submarine navigation and the operation of the offshore oil-drilling rigs with the large pycnocline influctuation and the strong speed shear. Furthermore, internal waves can also cause the sound field undulation and reduce the sonar efficiency. As a way of ocean engergy transport, internal waves propagation leads to onflow and sea water mixing and has an effect on the local water exchange, sediment and resuspending. So the research on internal waves has both realistic application aquirements and significant scientific meanings.Due to the generation and propagation under the sea surface, regular marine observations cost a lot and are usually difficult to carry out because of the environmental limitations. SAR imagery are usually available at the present research, however SAR are deficient for the long revisit period, high expense and strong noise. Although affected by the cloud, the moderate resolution spectral sensors such as MODIS are able to offset the disadvantage for the wide swath, high temporal resolution and the free data support and become an important data source for internal waves'observation.The thesis focuses on the internal waves optical remote sensing based on the successful SAR detetion of internal waves. It first discusses the internal waves features in the optical images and the comparison between the MODIS and SAR images suggests the internal waves in the sun glint has the same features as that in the SAR images, but the dark and bright bands are ranked inversely which are also related to the polarity. Then, it presents two image equalization methods to solve the internal waves identification problem, which enhance the internal waves characteristics in the whole image and facilitate the informateion extraction. The second section also decomposes the internal waves and surface waves using the two-dimensional EMD methods, providing the internal waves information retrieval from high resolution optical images. The critical section, the simulating of internal waves optical remote sensing dection, introduces the strong nonlinear KdV3N model and numerical arithmetic and similates the large-amplitude internal waves in the northern South China Sea accurately with the wave form consistent with the in-situ measurment. What's more, it establishes the experical model of sun glint internal waves and gets results similar to the real image features. On the basis of these models, the thesis further discusses the experimential model to estimate the amplitude from the distance between the dark and bright bands in optical images. The application of the model in the small area near the Dongsha corral reef produces satisfying estimation. At last, it makes full use of the advantages of optical images in internal waves research to present the statistical characteristics of internal waves in the northern South China Sea, including the internal waves speed distribution and the spatio-temporal distribution. The thesis realizes the application of sun glint information of the optical images in internal waves study, and presents the preliminary use of optical remote sensing in internal waves research.