Segmentation on Temperature Gradient Microstructure Data

Several studies have suggested that small-scale fluid motion, scales smaller than a centimeter, influence the kinetics of biological and chemical particles in aquatic environments. One of the key intrinsic fluid flow parameters is the turbulent kinetic energy dissipation rate. The turbulent kinetic energy dissipation rate is obtainable through the indirect method of Batchelor curve fitting. The estimation procedure assumes homogeneous and isotropic turbulence at small scales. Therefore, the procedure requires partitioning each profile of non-stationary temperature gradient microstructure into stationary segments. In this study, ten profiles of temperature gradient microstructure measured in three inland lakes are analyzed by using a segmentation algorithm. Spectral changes in the temperature microstructure are compared using methods based on AR models and one based on wavelet analysis.