<p>The ocean temperature profile presents an S—like shape. Considering this feature, a method for fitting temperature profiles using hyperbolic tangent functions is proposed as an alternative to the traditional sigmoid function method, aimed at identifying the thermocline boundary in the eastern Pacific region from January 2010 to March 2024. Compared with five existing methods, namely the entropy value method, tensor analysis method, inflection point method, variable representative isotherm method, and sigmoid function method, the new method demonstrates superior performance in fitting temperature profile and in estimating the top thermocline depth (TTD) and bottom thermocline depth (BTD). In particular, in terms of the fitting accuracy of the temperature profiles, the hyperbolic tangent function method has an R<sup>2</sup> value of 0.9937, whereas the sigmoid function method has an R<sup>2</sup> of 0.9893, showing the higher accuracy of the former in fitting the depth of the temperature profiles.&#xa0;Furthermore, the monthly climatological distribution features of the four characteristics, TTD, BTD, thermocline thickness (TT), and thermocline strength (TS), which are computed by the new method, is carried out, providing significant scientific evidence for comprehending the dynamic variations of the thermocline in this region and its environmental impacts.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Characterization of the thermocline in the eastern Pacific Ocean based on a hyperbolic tangent function fitting method

  • Xiaoqi Ji,
  • Miaomiao Feng,
  • Jichao Wang

摘要

The ocean temperature profile presents an S—like shape. Considering this feature, a method for fitting temperature profiles using hyperbolic tangent functions is proposed as an alternative to the traditional sigmoid function method, aimed at identifying the thermocline boundary in the eastern Pacific region from January 2010 to March 2024. Compared with five existing methods, namely the entropy value method, tensor analysis method, inflection point method, variable representative isotherm method, and sigmoid function method, the new method demonstrates superior performance in fitting temperature profile and in estimating the top thermocline depth (TTD) and bottom thermocline depth (BTD). In particular, in terms of the fitting accuracy of the temperature profiles, the hyperbolic tangent function method has an R2 value of 0.9937, whereas the sigmoid function method has an R2 of 0.9893, showing the higher accuracy of the former in fitting the depth of the temperature profiles. Furthermore, the monthly climatological distribution features of the four characteristics, TTD, BTD, thermocline thickness (TT), and thermocline strength (TS), which are computed by the new method, is carried out, providing significant scientific evidence for comprehending the dynamic variations of the thermocline in this region and its environmental impacts.