Analysis of Relationships between Rainfall, Stream Flow, and Climate Indices Using the Wavelet Method: the Case of Wadi Khemis, Northwestern Algeria
摘要
This study assesses the impact of climate variability, represented by key climate indices, on precipitation and streamflow patterns in the Khemis sub-watershed, a part of the Tafna basin in northwestern Algeria, over the period 1970–2019. It is the first to investigate hydro-climatic variability in this region using the wavelet transform method, highlighting its originality and significance. The Mann-Kendall and Pettitt tests revealed significant decreasing trends in precipitation and streamflow, with a detected breakpoint in 1981. Prior to this year, the series exhibited alternating wet years, whereas after 1981, dry years became predominant. The study explores the relationship between hydro-climatic variables and key climate indices, including the North Atlantic Oscillation (NAO), El Niño–Southern Oscillation (ENSO), and Western Mediterranean Oscillation Index (WeMOI), using a monthly time step. Continuous wavelet transform analysis identified dominant variability modes at infra-annual (3–6 months), annual (1 year), and inter-annual (2–4 years, 4–8 years, 8–12 years, 8–16 years, and 16–32 years) scales. Results indicate that low-frequency variability (infra-annual and annual) explains 89 and 75% of total precipitation and streamflow variability, respectively. Wavelet coherence analysis further demonstrates strong correlations between climate indices and hydro-climatic series, with correlation coefficients of 62, 62, and 69% between NAO, SOI, and WeMOI and precipitation, and 65, 71, and 72% between these indices and streamflow. These findings underscore the crucial role of climate indices in shaping hydro-climatic fluctuations in the region, providing valuable insights into the long-term variability of water resources under changing climate conditions.