Oceanic Chlorophyll Retrieval By Remote Sensing And Ocean Primary Production Estimate

Oceanic chlorophyll concentration is the important indicator of phytoplankton and water quality. So chlorophyll concentration retrieval is meaningful for carbon cycle in ocean-atmosphere system, as well as for oceanic primary production estimate. Ocean primary production is the key factor in ocean ecological system. To some extent, it controls CO₂ exchange in sea-atmosphere surface. So it is the important object for global climate change study. With the support of several programs, we have done lots of work on oceanic chlorophyll and oceanic primary production by remote sensing. As a result, more than 10 papers have been published. This dissertation is on basis of these researches.Presently most of chlorophyll researches have been performed in short-term local sea areas. Long-time series of satellite data can reveal the whole environment variety, which is the significant advantage over the field measurement. First, based on both SeaWiFS and MODIS data, chlorophyll variety in the China Shelf Sea was analyzed on basis of long-term series data. The comparison and impact mechanism were also discussed.The result shows that chlorophyll in China Shelf Sea changes with seasons. It varies regularly between different years. The chlorophyll is the highest in East China Sea, where the inter-annual variety is significant. The annual mean chlorophyll in South China Sea is the lowest. The chlorophyll concentration in the whole China Shelf Sea decreases from coast to open sea. The retrieved chlorophyll well indicates the environment feature in China Shelf Sea.In general, the retrieved results of SeaWiFS and MODIS agree well. In East China Sea they agree the worst, while in South China Sea they do the best. In three different sea areas, SST is significantly correlated to chlorophyll. It shows that SST is one of the major factors which contribute to chlorophyll distribution.Second, in order to improve the algorithm for China Sea, based on SeaWiFS and in situ data, we developed chlorophyll retrieval model using BP ANN (Artificial Neural Network) algorithm for Chinese sea. It demonstrates that for chlorophyll retrieval, ANN has the advantage of precise simulation of non-linear transmission over traditional methods, especially for case 2 waters in which traditional models usually fail.In addition to ANN model, we also constructed new spectral index to retrieve...