/PyREClass

Classification of radar reflectivity data using Wavelet transform.

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

PyREClass

Classification of radar reflectivity data using Wavelet transform.

Description

A new method for separating convective, intermediate, and stratiform rainfall based on the wavelet scale analysis. The method uses à trous wavelet transform (WT) to separate heterogeneous convective features from the relatively smooth stratiform field. The heterogeneous region is further split into a convective class for the large wavelet coefficients and an intermediate class for the small wavelet coefficients. There is an improvement in the estimation of convective-stratiform fraction and frequency distribution of rain rates in comparison to the well established Steiner method. This new method correctly reclassified a significant fraction of Steiner’s convective regions.

Reference

Raut, B. A., Louf, V., Gayatri, K., Murugavel, P., Konwar, M., & Prabhakaran, T. (2020). A Multiresolution Technique for the Classification of Precipitation Echoes in Radar Data. IEEE Transactions on Geoscience and Remote Sensing, 1–7. https://doi.org/10.1109/TGRS.2020.2965649