Shifts Detection in Rainfall Regime as an Indicator of Climate Variability at the Kolda Station
Vieux Boukhaly Traore *
Department of Physics, Faculty of Science and Technology, Cheikh Anta Diop University, Dakar, Senegal.
Omar Ngor Thiam
Department of Physics, Faculty of Science and Technology, Cheikh Anta Diop University, Dakar, Senegal.
Mamadou Lamine Ndiaye
Ecological Monitoring Centre (CSE), Ministry of the Environment and Ecological Transition, Dakar, Senegal.
Mamadou Lamine Sow
Department of Physics, Faculty of Science and Technology, Cheikh Anta Diop University, Dakar, Senegal.
*Author to whom correspondence should be addressed.
Abstract
Rainfall variability is a major concern in the Sahel because it affects rain-fed agriculture, water availability, and natural-resource management. This study evaluates shifts in the rainfall regime at the Kolda station in southern Senegal using a 75-year record covering 1951–2025. Annual and monthly rainfall series obtained from the National Agency for Civil Aviation and Meteorology were analysed with the Pettitt, Buishand, Hubert segmentation, and Lee–Heghinian tests implemented in KHRONOSTAT at a 10% significance level. The annual analyses indicated downward shifts, although the estimated break dates varied among the methods. The Lee–Heghinian test identified 1958, Hubert segmentation identified 1967, and the Pettitt and Buishand tests identified 1971. For the monthly series, the tests indicated an upward shift at the beginning of the rainy season: April according to Pettitt, May according to Buishand and Lee–Heghinian, and June according to Hubert segmentation. The Lee–Heghinian posterior probabilities were 0.1688 for the annual shift and 0.2803 for the monthly shift. The convergence of complementary tests supports the presence of non-stationary behaviour in the Kolda rainfall series, while the variation in estimated break dates reflects differences among the statistical procedures. These findings provide locally relevant evidence for agricultural planning and water-resource management. However, their interpretation is limited to rainfall observations from a single station and does not directly establish the climatic mechanisms underlying the detected shifts.
Keywords: Climate variability, rainfall regime, change-point detection, hydroclimatic time series, KHRONOSTAT, Pettitt test, Buishand test, Hubert segmentation, Lee–Heghinian test.