Improved visualization for trend analysis by comparing with classical Mann-Kendall test and ITA


Guclu Y. S.

JOURNAL OF HYDROLOGY, vol.584, 2020 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 584
  • Publication Date: 2020
  • Doi Number: 10.1016/j.jhydrol.2020.124674
  • Title of Journal : JOURNAL OF HYDROLOGY
  • Keywords: Mann-Kendall trend test, Pettitt change point test, Rainfall, Sen's innovative method (ITA), Time series, Trend analysis, IDENTIFICATION, TEMPERATURE, FLOW

Abstract

This study compares classical Mann-Kendall (MK) test and simple Sen-innovative trend analysis (ITA) method by proposing a new trend analysis methodology with a special graphical representation. Another attempt is to classify sub-categories as "low" and "high" objectively by means of the Pettit change point test. The applications are based on 50-year rainfall records at various meteorology stations in Turkey. The proposed methodology provides different trend conditions than classical MK trend test and ITA approach on the Mediterranean, Black Sea, and continental climate regions of Turkey. Mediterranean climate reflects completely increasing trend especially on "high" values and the Black Sea region stations have non-monotonic decreasing trends at "low" and "high" values with increasing and decreasing trends, respectively on the proposed graphical representation. Continental climate region stations have monotonic increasing trend without sub-categorization, because of no change point via Pettit test on new visualization graphs.