In this paper, intertidal flat of Sishili Bay was the interested area, and the objective of this study was to retrieve and monitor biomass of microphytobenthos using in situ measurement data and remote sensing data. Firstly, retrieval algorithms for estimating chlorophyll-a concentration(Chl-a) of microphytobenthos were established using field data such as hyperspectral data and Chl-a(proxy of microphytobenthos biomass in this study) data. Results showed that ratio algorithm of 675nm to 700nm were better than other retrieval models. The MRE (Mean Relative Error) and the RMSE (Root Mean Square Error) of the retrieved and measured data on the basis of the ratio model were 4.01µg/cm3(n=10) and 14.6% (n=10). Secondly, retrieval algorithms of Chl-a were established on the basis of simulated bands of Hyperion, HICO and Rapid Eye. Moreover, field data which was collected from Muping tidal flat in August 2013 was used to test the retrieval models. Results showed that the precision of band ratio algorithms were better than other retrieval models. The MRE and the RMSE of the band ratio model of B35/B33 of Hyperion were 8.94% (n=10) and 4.14µg/cm3 (n=10). Furthermore, a remote sensing image of Rapid Eye was used to monitor the Chl-a in the tidal flat of southwest of Korea on the basis of Rapid Eye simulated band ratio model. Thirdly, Chl-a of microphytobenthos of tidal flat sediments core was measured with a resolution of 3 mm and 1 mm in day time at the Sishili Bay; and surface reflectance of tidal flat was simultaneously measured. Chl-a remote sensing retrieval algorithm was established on the regression with the NDI-MPB (Normalized Difference Index of Microphytobenthos) which is based on the reflectance at 675nm and 700nm. In situ data showed that Chl-a decreased in an exponential function with the increase in depth, and the vertical distribution of Chl-a might be characterized by the slop of the exponentially decreasing model of Chl-a. The impacts of variations in the vertical distribution of Chl-a on the remote sensing retrieval algorithms were simulated, and results showed that Chl-a retrieval algorithms changed with its vertical distribution. Finally, simulated band ratio model of Hyperion was developed to estimate Chl-a of microphytobenthos from November 2012 to April 2013 in the tidal flat of Yantai coast. Relationship between Chl-a and fine-grained sediment content as well as pheophytin were analyzed, and results showed that there was a remarkable correlation between Chl-a and sediment components; and the correlation between Chl-a and reflectance could be affected by pheophytin.
修改评论