Trend Assessment by the Mann-Kendall Test and the Innovative Trend Analysis Method (North-West Algeria)

Abstract

In this article, a trend study, in two watersheds Coastal-Oran and Macta (North-West of Algeria), is carried out using monthly and seasonal rainfall during the period 1975–2015 at 38 stations. Mann-Kendall (MK) and Innovative Trend Analysis (ITA) methods are applied comparatively. The aim of this study is to verify if the trend detection ability is the same in both methods. Monthly rainfall analysis by the MK test shows that 82% of the stations have no significant trends. In contrast, the ITA method detects positive (negative) trends in 56% (33%) of stations in the low rainfall component, 53% (40%) in the average and 41% (58%) in the high rainfall. On a seasonal scale, the MK test still not aware of trends in 75% of the studied stations, while the ITA shows that 63% (31%) of stations with increasing (decreasing) trends. Studying the rainfall trend, using the ITA and Mann-Kendall methods, in the same area can help to understand past variations and therefore extrapolate the results to other areas with the same climate. This is a crucial step in managing climate-related issues. In addition, the ITA detects hidden sub-trends in the low, medium and high components that the MK test does not, which shows the superiority of the ITA method over MK method.

Keywords: Hidden trend; ITA; Mann-Kendall; North-West Algeria; Rainfall.