Ocean Fronts detection over the Bay of Bengal using changepoint algorithms – A non-parametric approach

Oceanologia

Ravichandran Muthalagu¹  Venkat Shesu Reddem²  Kumar Nimit²  Venkata Jampana²  Pattabhi Rama Rao Eluri³  Venkateswara Rao Bekkam⁴ 

[1] Corresponding author at: Indian National Centre for Ocean Information Services, Hyderabad, India.;Indian National Centre for Ocean Information Services, Hyderabad, India;Jawaharlal Nehru Technological University, Hyderabad, India;National Centre for Polar and Ocean Research, Vasco-da-Gama, India;

关键词: Ocean fronts;    Satellite imagery;    Contextual median filter;    Changepoint detection;   

DOI  :  

来源: DOAJ

【 摘 要 】

Oceanic fronts are regions over the oceans where a significant change in the characteristics of the water masses is observed. Advanced Very High Resolution Radiometer (AVHRR) satellite imagery over the Bay of Bengal shows regions that are populated by frontal structures. Over the Bay of Bengal, some of the strongest gradients in temperature and salinity are observed. In recent years, there has been a tremendous growth in the availability of satellite imagery and the necessity of automated fast detection of the frontal features is needed for services like potential fishing zones over open oceans. In this article, an algorithm to infer oceanic fronts over the Bay of Bengal is described using changepoint analysis. The changepoint algorithm is combined in a novel way with a contextual median filter to detect frontal features in AVHRR imagery. The changepoint analysis is a non-parametric technique that does not put thresholds on the gradients of brightness temperatures of the satellite imagery. In the open oceans, the gradients of temperature and salinity are not sharp and changepoint analysis is found to be a useful complementary technique to the existing front detecting methods when combined with contextual median filters.

【 授权许可】

Unknown