Retrieval Of Suspended Sediment Concentration In Deep Bay Using HJ CCD Images Based On MODIS Data

Coastal estuarine bay water is turbid, belonging to case two water. SSC is important water quality parameter of water body, which is really vital in the change process of erosion and deposition coastal zone and changes in the aquatic ecological environment. The traditional sampling method needs lots of time and money, and unable to meet the demand of high dynamic monitoring of water. Remote sensing for its advantages of large area high efficiency long time-series data, more and more widely applied in the field of water environmental research. In this paper, based on the validated retrieval of MODIS of Suspended Sediment Concentration, we established HJ-1A/1B satellites of CCD image retrieval model. The research of Suspended Sediment Concentration of quantitative remote sensing retrieval in Deep Bay was conducted, verifying the parameter retrieval feasibility under the condition imprecise atmospheric correction. This method has positive significance to further promote China’s domestic satellite application. In the article, here is the majority content:(1) SSC retrieval model based on MODISTo MODIS/Terra data which spatial resolution is 250 meters, using SeaDAS version 6.4 official software package for atmospheric correction for remote sensing reflectance, combining with the synchronization in situ SSC of 2007 and 2008 two years two stationary established red/green band ratio retrieval model in Deep Bay. The R2 is 0.80, the root mean square error 5.06, relative error 15%, The results showed that this model can satisfy the retrieval accuracy requirement.(2)HJ-1A/1B satellite CCD RC and SSCI methods retrieval of SSC based on MODISUsing the remote sensing reflectance from exact Rayleigh correction of HJ-1A/1B satellite CCD, and the 2009-2012 years the MODIS data by model extracting quasi-synchronous with the HJ-1A/1B satellite CCD data as the foundation, the suspended sediment concentration was retrieved in the RC method. The remote sensing reflectance of HJ-1A/1B satellite CCD is obtained from exact Rayleigh correction, the Suspended Sediment Concentration Index method was used by analysing the theory of baseline in which analyses the change of the baseline height of the index of CCD red channel in different turbidity degree of SSC in case two water to effectively solve the problem of the difficulty of eliminating aerosol multiscattering signal by water color remote sensing of traditional atmospheric correction in turbid water. This method was used to calculate remote sensing reflectance of HJ-1A/1B satellite CCD. With the 2009-2012 years the MODIS data by model extracting quasi-synchronous with the HJ-1A/1B satellite CCD data on base, SSCI retrieval experiment was carried out.(3) Precision test and verify of SSC retrieval based on RC and SSCI of HJ-1A/1B imageryEstablished RC method retrieval models of single band, band ratio, single CCD and four CCD together in Deep Bay of MODIS data in auxiliary HJ-1A/1B image, and red band SSCI method retrieval models of single CCD and four CCD together and verified the accuracy and evaluation. Model test shows that, for any CCD sensor data, the RC models and SSCI models both have high determination coefficient R2 and low root mean square error (R2≥0.76, RMSE≥11.78), the models can satisfy the requirements of retrieval. Among them, based on the MODIS auxiliary HJ-1A/1B satellite CCD sensor SSCI red band the exponential model is most suitable for SSC retrieval of Deep Bay waters. The SSC products retrieved from the model showed that the method has a positive meaning that HJ-1A/1B satellite CCD is data resource for of observation and continuation of SSC after MODIS retired, and can also used for other sensor parameters retrieval of water environment lacking of synchronization measured data.