Characteristics of satellite chlorophyll-a concentration speckles and a removal method in a composite process in the East/Japan Sea

• Published in: International journal of remote sensing 2013, Vol.34 (13), p.4610-4635
• Main Authors: Park, Kyung-Ae, Chae, Hwa-Jeong, Park, Ji-Eun
• Format: Article
• Language: English
• Subjects: Animal, plant and microbial ecology; Applied geophysics; Biological and medical sciences; chlorophyll; Climatology; Concentration gradient; Deviation; Earth sciences; Earth, ocean, space; Error analysis; Errors; Exact sciences and technology; Fundamental and applied biological sciences. Psychology; General aspects. Techniques; Internal geophysics; normal values; Post-processing; Sea-viewing Wide Field-of-view Sensor; spring; Teledetection and vegetation maps; Winter
• ISSN: 1366-5901; 0143-1161; 1366-5901
• DOI: 10.1080/01431161.2013.779397

Anomalously high values of Sea-viewing Wide Field-of-view Sensor (SeaWiFS) chlorophyll-a concentration have been observed in the sea, which have had limited scientific application research. Ten-year SeaWiFS data from 1998 to 2007 indicate that speckles with anomalously high concentration were randomly scattered throughout the entire study area. In order to eliminate the speckles from SeaWiFS chlorophyll-a concentration images, a methodology was developed by considering the statistical characteristics of the speckle distributions. Since the speckles had strong seasonality in terms of occurrence frequency and magnitude, dynamic thresholds were applied to each pixel value by normalizing it to chlorophyll-a concentration climatology. Non-speckle normal values, eliminated due to high concentration in the frontal zone during the spring bloom, were survived by a spatial deviation normalized to the magnitude of chlorophyll-a concentration gradient in the frontal zone. The chlorophyll-a concentration values, passing through the post-processing of the speckle errors, were composited to generate each monthly map. As a result, all the speckles were satisfactorily eliminated, resulting in a maximum of about 14% of abnormal variability being reduced. The errors were most significant in winter, particularly in December, for the past decade. This study addressed the importance of speckle errors in chlorophyll-a concentration from SeaWiFS data and presented an optimized speckle-free composite method for more reliable chlorophyll-a data in scientific application research.