HYDROLOGICAL DATA TREND ANALYSIS WITH WAVELET TRANSFORM


Yerdelen C. , Abdelkader M.

Comptes Rendus De L Academie Bulgare Des Sciences, vol.74, no.8, pp.1194-1202, 2021 (Journal Indexed in SCI Expanded)

  • Publication Type: Article / Article
  • Volume: 74 Issue: 8
  • Publication Date: 2021
  • Doi Number: 10.7546/crabs.2021.08.11
  • Title of Journal : Comptes Rendus De L Academie Bulgare Des Sciences
  • Page Numbers: pp.1194-1202

Abstract

The analysis of natural events such as floods, drought, and earthquakes cannot be limited to simple mathematical equations. Generally, the events related to climate indicators should be considered a random process studied by statistical and probability methods. In this study, the wavelet transform along with classical trend analysis tests are applied to investigate the possible trends in the annual mean streamflow. The paper presents new approach to determine turning point in time series trend. This data analysis approach is used to determine the trend pattern, periodicity and trend change point in sixteen flow measurement stations of the Turkish Mediterranean Sea region. It is identified that the region is experiencing negative trends in annual streamflow starting from the 7th decade of the last century. The results presented in this work show that a combination of signal analysis methods and statistical tests improves time series analysis in engineering studies.