Variability in rainfall and temperature results in different impacts on agricultural practices. Assessesment of variability and trend of rainfall and temperature for the district of Musanze in Rwanda was conducted using six meteorological stations for a period of 37 years, ranging from 1981 to 2018, and data were obtained from Rwanda Meteorology Agency. Musanze district is located in highland areas of Rwanda, understanding the variability and trend in rainfall and temperature is paramount to increase the uptake of climate information and support strategic orientation. The Mann–Kendall nonparametric test and modified Mann–Kendall were used to assess the trend in rainfall and temperature, whereas Sen’s slope estimator was used to assess the magnitude of change. The results from both methods showed much similarity and consistency. The assessment of variability and trend in rainfall and temperature in Musanza district indicated that increasing temperature and decreasing rainfall trends gave an indication of changes in variability and trend in rainfall and temperature. The annual pattern revealed a substantial downward tendency of −25.7% for Nyange, the only station with constant decreasing trend over all seasons, DJF, −61.4%, SON, −12.2%, JJA, −40.3%, and MAM, −4.35. Temperature analysis for both maximum and minimum indicated increasing trend which was signal for constant warming up in the area. The results from coefficient of variation indicated a high disparity in rainfall variation from June to August which ranged between 51 and 74%, and other seasons changes were moderate.

Analysis of the trends and variability of climate variables and extreme climate events is important for climate change detection in space and time. In this study, the trends and variabilities of minimum, maximum, and mean temperatures, as well as five extreme temperature indices, are analyzed over Rwanda for the period of 1983 to 2022. The Modified Mann–Kendall test and the Theil–Sen estimator are used for the analysis of, respectively, the trend and the slope. The standard deviation is used for the analysis of the temporal variability. It is found, on average, over the country, a statistically significant (α = 0.05) positive trend of 0.17 °C/decade and 0.20 °C/decade in minimum temperature, respectively, for the long dry season and short rain season. Statistically significant (α = 0.05) positive trends are observed for spatially averaged cold days (0.84 days/decade), warm nights (0.62 days/decade), and warm days (1.28 days/decade). In general, maximum temperature represents higher variability compared to the minimum temperature. In all seasons except the long dry season, statistically significant (α = 0.05) high standard deviations (1.4–1.6 °C) are observed over the eastern and north-western highlands for the maximum temperature. Cold nights show more variability, with a standard deviation ranging between 5 and 7 days, than the cold days, warm nights, and warm days, having, respectively, standard deviations ranging between 2 and 3, 4 and 5 days, and 3 and 4, and, especially in the area covering the central, south-western, south-central, and northwestern parts of Rwanda. Temperature increase and its variability have an impact on agriculture, health, water resources, infrastructure, and energy. The results obtained from this study are important since they can serve as the baseline for future projections. These can help policy decision making take objective measures for mitigation and adaptation to climate change impacts.

The variability, intensity, and distribution of rainfall have drawn a lot of interest globally and especially in nations where rainfed agriculture is the norm. This article uses rainfall data from the Rwanda Meteorology Agency for the years 1981 to 2021 to delineate and analyze rainfall variability and trends in the Kayonza District. The time series were grouped using the K-means clustering technique based on computed Euclidean distance, the total within-cluster sum of squares, and the elbow plot technique to determine the optimal number of clusters. The coefficient of variation measures was employed to analyze rainfall variability, while Sens slope and the MannKendall (MK) test were used, respectively, to find trends and changes in magnitude. The results indicated four near homogeneous zones named region one to four. The dry seasons indicated higher variability compared to rainy seasons and annual rainfall total with a variability of 128142 over the southeastern part during June to August (JJA) season, while a variability of 1648 was observed over most of the district during both annual and rainy seasons. It was further noted that the areas in the central part of the Kayonza District indicated a significant increasing trend at a significance level of 95 and above during January to February (JF), September to December (SOND), and on annual basis, while March to May (MAM) and JJA season exhibited no significant trend. The findings of this study are essential for creating adequate mitigation strategies to lessen climate changes effects on agriculture as well as other socioeconomic sectors.