Performance assessment of gridded climate datasets for precipitation and temperature monitoring in a Mediterranean inland region

Accurate and spatially consistent climate information is essential for environmental monitoring, climate change assessment, and resource management in Mediterranean inland regions characterised by complex orography and ground-based observational networks with low spatial density. This study assesses the suitability of three widely used gridded climate datasets—CHIRPS, ERA5-Land and TerraClimate—for climate monitoring purposes in the Alto Bradano area (southern Italy) over the period 2005–2024, using observations from four meteorological stations as reference. The evaluation focuses on annual and seasonal cumulative precipitation, as well as maximum and minimum air temperatures, combining error-based metrics, efficiency indicators, and trend analysis based on the Mann–Kendall test and Sen’s slope estimator. The results highlight substantial dataset-dependent differences, particularly for precipitation. CHIRPS shows the best overall performance in reproducing observed precipitation variability and identifies a statistically significant decline in annual precipitation across most of the study area, supporting its suitability for long-term pluviometric monitoring. ERA5-Land captures temporal variability but exhibits weak and non-significant precipitation trends, while TerraClimate shows systematic underestimation and limited robustness in trend detection. In contrast, temperature trends are more consistent across datasets. Both ERA5-Land and TerraClimate reveal a strong and spatially coherent warming of maximum temperature, while minimum temperature displays widespread nocturnal warming, more robust in ERA5-Land. Overall, the findings emphasise the importance of dataset-specific validation and support the targeted use of gridded climate products for environmental monitoring and climate impact assessments in Mediterranean inland regions.