Open AccessBioinformatic approaches for objective detection of water masses on continental shelves
Author(s) -
Oliver Matthew J.,
Glenn Scott,
Kohut Josh T.,
Irwin Andrew J.,
Schofield Oscar M.,
Moline Mark A.,
Bissett W. Paul
Publication year - 2004
Publication title -
journal of geophysical research: oceans
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2003jc002072
Subject(s) - water mass , continental shelf , ocean color , satellite , boundary (topology) , geology , sea surface temperature , seawater , oceanography , surface water , remote sensing , environmental science , mathematics , mathematical analysis , environmental engineering , engineering , aerospace engineering
As part of the 2001 Hyper Spectral Coupled Ocean Dynamics Experiment, sea surface temperature and ocean color satellite imagery were collected for the continental shelf of the Mid‐Atlantic Bight. These images were used to develop a water mass analysis and classification scheme that objectively describes the locations of water masses and their boundary locations. This technique combines multivariate cluster analysis with a newly developed genetic expression algorithm to objectively determine the number of water types in the region on the basis of ocean color and sea surface temperature measurements. Then, through boundary analysis of the water types identified, the boundaries of the major water types were mapped and the differences between them were quantified using predictor space distances. Results suggest that this approach can track the development and transport of water masses. Because the analysis combines the information of multiple predictors to describe water masses, it is an effective tool in detecting water masses not readily recognizable with temperature or chlorophyll alone.