Fine-scale sea surface temperature fronts in wintertime in the northern South China Sea

Abstract

This study presents finer structures and inter-seasonal evolutions of sea surface temperature (SST) fronts in wintertime in the northern South China Sea (SCS) by applying an entropy-based edge detection method to 7-year (2001-2007) satellitederived SST images with a grid size of 1 km. From monthly mean maps of SST front, six significant SST fronts were defined in the northern SCS. This study not only reveals the earlier defined frontal bands, but also provides finer structures and gradient variability of the fronts. One front alongside Luzon Island (the Philippines), which was undetected by a gradient-based method in previous studies, is well defined by the entropy-based method used in the present study. This result suggests that the gradient-based method might not be appropriate for front detection in coastal waters or marginal seas, but using SST data of lower spatial resolution SST data might not be suitable to describe the detailed structure of short or small fronts through the sophisticated edge detection methods.

A comparison of satellite-derived sea surface temperature fronts using two edge detection algorithms

“…Interestingly, the West Luzon Front detected by CCA in Belkin and Cornillon (2003) and by SEA in Chang et al (2010) was not detected by Wang et al (2001) in their application of a gradient based algorithm to SST fields of the northern South China Sea.…”

“…These algorithms have been applied to thermal fronts in marginal seas (Hickox et al, 2000;Wang et al, 2001; Belkin and Cornillon, 2003) as well as open ocean regions (Ullman et al, 2007;Belkin et al, 2009). Several studies have also presented new views of oceanic fronts in coastal and regional seas, such as Ullman and Cornillon (1999) who applied the CCA to the northeastern coast of the US, and Shimada et al (2005) and Chang et al (2006Chang et al ( , 2010 who applied SEA to the Japanese coast and northern South China Sea.…”

Application of a sea surface temperature front composite algorithm in the Bohai, Yellow, and East China Seas

“…Current methods to detect frontal structures mainly include the histogram-based separation of two water masses Cornillon, 1992, 1995;Diehl et al, 2002;Yao et al, 2012), detection of horizontal gradients (Oram et al, 2008;Belkin and O'Reilly, 2009;Zeng et al, 2014;Zhang et al, 2014), and entropy-based approaches (Vázquez et al, 1999;Shimada et al, 2005;Chang et al, 2008Chang et al, , 2010. Furthermore, some new algorithms have been developed such as the gravitation algorithm (Ping et al, 2013(Ping et al, , 2014 and morphology algorithm (Pi and Hu, 2010).…”

“…In the fi rst step, a traditional front detection algorithm is applied to each SST map within the time series data. In the second step, the long-term mean frontal average gradient and frequency are calculated to reveal the front's variations (Shimada et al, 2005;Chang et al, 2008Chang et al, , 2010Pi and Hu, 2010;Yao et al, 2012).…”

Diurnal Variability of Turbidity Fronts Observed by Geostationary Satellite Ocean Color Remote Sensing

“…Compared with conventional gradient-based and histogram-based front detection algorithms used widely in oceanography [14,16,39,40], the automatic front detection technique based on Jensen-Shannon divergence (entropy-based algorithm) is more robust against noise and avoids previous filtering that would blur frontal features [11]. The entropy-based algorithm has already been used to retrieve the fine-scale SST fronts in different regions, such as the SST fronts near the Japanese coasts [11], the upwelling SST fronts in the Taiwan Strait and its surrounding waters [41,42], the inter-seasonal evolutions of fine-scale SST fronts in the northern South China Sea [43], and the SST front disappearance phenomena in the subtropical North Pacific [44].…”