Journal of Geography, Environment and Earth Science International

Aim: Farmer perceptions of climate change vary across spatial scales, and these perceptions can be validated using synoptic weather-station data. However, in data-poor countries such as Sierra Leone, weather stations are sparsely distributed and are also affected by substantial inconsistencies and errors in data capture. Against this backdrop, we present the first validation of farmer perceptions of climate change-related weather patterns in southern Sierra Leone using Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) and the fifth-generation European ReAnalysis for the land component (ERA5-Land) reanalysis data.

Study Design: The study employed a mixed-methods research design involving farmer perceptions of climate change-related weather patterns and corresponding data from the CHIRPS and ERA5-Land datasets.

Place and Duration of Study: The study was conducted in southern Sierra Leone. Field data on farmer perceptions were collected in June 2025, and CHIRPS/ERA5-Land data were acquired for the period 2004–2024.

Methodology: To assess farmer perceptions of climate change-related weather patterns, 613 farmers were interviewed; 56.6% were male and 43.4% were female, and the participants were aged 18–65 years. Qualitative data obtained from focus group discussions were analysed using content analysis, while quantitative data obtained from household surveys were analysed using descriptive and inferential statistics, including percentages, the coefficient of variation, Spearman's rank correlation, and the Kruskal–Wallis test. CHIRPS and ERA5-Land data were analysed using the Hamed–Rao modified Mann–Kendall test and Sen's slope estimator. In addition, Kendall's rank correlation coefficient was used to measure agreement between farmer perceptions and observations from the CHIRPS and ERA5-Land datasets.

Results: Farmers reported rising temperatures, which were corroborated by increases of C to C in annual mean temperature recorded by ERA5-Land across the four districts; all increases were statistically significant at p < 0.005, based on the Hamed–Rao modified Mann–Kendall test. Farmers also perceived irregular rainfall patterns. Based on the CHIRPS data, the coefficient of variation in annual rainfall ranged from 7.1% in Bo to 9.1% in Bonthe, indicating substantial year-to-year fluctuations without a directional trend. Moreover, the 12-month Standardized Precipitation Index (SPI-12) showed more droughts after 2015, supporting farmers’ perceptions of increased drought frequency. Furthermore, farmer perceptions of extreme weather events varied significantly across districts, and Moyamba and Bo, which are more economically active and inland, recorded the largest increases in extreme heat based on ERA5-Land reanalysis data.

Conclusion: The study shows a high degree of concordance between farmer perceptions of climate change-related weather patterns and corresponding CHIRPS/ERA5-Land records. This finding suggests that local perceptions can serve as an integral source of data in local climate studies and that integrating them with global climate products could improve understanding of changing weather patterns from local to regional scales.