Detection of sediment area in MODIS imagery utilizing atmospheric channels

Abstract

The maximum reflectance for Moderate Resolution Imaging Spectroradiometer (MODIS) ocean color channels are significantly smaller than those for the land channels at similar wavelengths. Over bright coastal water areas, ocean color channels (0.488, 0.53 and 0.551 urn) are often saturated due to the contribution of suspended sediment reflectance in the turbid water in the MODIS imagery. These phenomenon leads to the lost of geophysical and biological activities in the data. In order to overcome this problem, it is reasonable to use MODIS land and atmosphere channels for the detection of sediment influences water areas. In this paper, a simple algorithm to detect the sediment influences pixels in the MODIS imagery is proposed. The algorithm is based on the gradient difference of the line connecting the 0.47 with 2.13 μm channels and 0.47 with 0.66 μm channels of a log-log graph of the apparent reflectance values against MODIS wavelengths. It is shown that, over clear water area, the 0.47, 0.66 and 2.13 μm fit very well in the corresponding line, with a correlation value of R>0.99. Over turbid water, a substantial increase in the reflectance of 0.66 μm lead to a low value of correlation. By subtracting the gradient of the line connecting 0.47 and 0.66 μm with the gradient of the line connecting 0.47 and 2.13 μm, the threshold value to discriminate turbid and shallow coastal water from clear water pixel can be obtained. If the gradient difference is greater than zero, the pixels are then masked as sediment influenced pixels. The results of this algorithm are then evaluated by comparing each masked area with the corresponding masked generated using other established technique.