Edge analysis of seasonal variability in chlorophyll maps of the Black Sea

Abstract

The use of remotely sensed oceanographic data would be greatly benefited by being able to detect circulation features automatically through edge detection. In the Black Sea, the use of edge detection to identify fronts can be used to study the effects of river input on circulation patterns and biological and physical interactions. The use of edge detection on remotely sensed chlorophyll data is limited by noisy data, inaccurate measurements, temporal and spatial gaps in data, and limitations on computational power. The algorithm described in this thesis utilizes image processing techniques to create an edge detection process that shows the effects of the Danube river input on Black Sea circulation patterns with little computational complexity.