TITLE:
Changing Characteristics of Extreme Precipitation during Tanzania’s Long Rain Season
AUTHORS:
Elisia Hamisi Zobanya, Yi Fan, Baraka Charles Bunini
KEYWORDS:
Precipitation Extreme Indices, Long Rain Season, Spatial and Temporal Variability
JOURNAL NAME:
Journal of Geoscience and Environment Protection,
Vol.14 No.1,
January
23,
2026
ABSTRACT: The long rain season from March to May (MAM) is a critical contributor to water resources, agriculture, and ecosystem sustainability in Tanzania, yet it is increasingly characterized by high interannual variability, alternating between severe droughts and damaging floods. This study aims to examine the spatial and temporal variability of extreme precipitation during Tanzania’s long rainfall season over the period 1981-2024 and to identify dominant modes of variability and large-scale climate drivers influencing these extremes. Daily rainfall data from the Climate Hazards Group InfraRed Precipitation with Stations dataset (CHIRPS) at 0.25˚ resolution were used to compute six precipitation extreme indices describing both wet and dry conditions. Climatological results show that consecutive dry days exceed 50 days in the western and southwestern highlands, while values below 40 days dominate the northern, eastern, and southern highlands. Total seasonal rainfall exceeds 200 mm in the northern regions, coastal areas, and southwestern highlands, but remains near 50 mm over the central and northeastern highlands, with maximum five-day rainfall totals exceeding 80 mm along the coast and in the southwestern highlands. Trend analysis reveals increasing consecutive dry days at rates of 0.2 - 0.6 in parts of central and northeastern Tanzania, while decreasing trends of −0.2 to −0.8 occur south of Lake Victoria. Total rainfall on wet days increases by more than 0.5 mm per year over most of the country, accompanied by increases in rainfall intensity and very heavy rainfall frequency. The leading mode of total rainfall explains 37.5 percent of the variance and shows a significant increase after 2000, while wet-spell persistence is significantly linked to the Indian Ocean Dipole with a correlation coefficient of 0.34. Overall, the results indicate a shift towards wetter but more unpredictable long rain seasons emphasizing the need for improved climate adaptation and disaster risk management strategies.