A simple method for distinguishing global Case-1 and Case-2 waters using SeaWiFS measurements

Since the combinations of water constituents are different between Case-1 and Case-2 waters, bio-optical models, retrieval algorithms for water constituent concentrations and other applications in water-color remote sensing are also very dissimilar between these waters. Use of the algorithms specifically developed for Case-1 waters returns inaccurate results in Case-2 waters, and vice versa. To select an appropriate algorithm for a given water pixel, it is important to first determine whether it is a Case-1 or Case-2 water and to clarify its temporal variations. This paper presents a simple method based on the inherent optical properties (IOPs) of water bodies for discriminating global Case-1 and Case-2 waters based on satellite data. Compared with the previous methods, the newly proposed method only requires two remote-sensing reflectances at 412 and 443 nm for relative comparisons, and thus it not only can easily be implemented using satellite data but also is robust even for satellite data with imperfect atmospheric correction, unpredictable noise pixels in the images, and so on. The new method was then applied to seasonal SeaWiFS 9-km data to map the global distribution of Case-1 and Case-2 waters for each season in 2003. The results showed that more than 80% of global waters belong to the Case-1 category throughout the year, and the Case-2 waters are mainly concentrated in the Northern Hemisphere along the coasts. Both the area and distribution of Case-1 and Case-2 waters changed seasonally. By using a sub-dataset from NOMAD, it was found that when the ratio of [a(ph)(443) + a(w)(443)]/a(443) was larger (smaller) than 50%, about 70%(75%) of the samples were identified as Case-1 (Case-2) waters by the new method. Moreover, the semi-analytical algorithm GSM01 was more accurate for distinguishing Case-1 than Case-2 waters, which implies that use of the proposed method to select the appropriate remote-sensing algorithm would be important. (C) 2012 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS) Published by Elsevier B.V. All rights reserved.