High-resolution climate surfaces are widely used in climate classification, agricultural zoning, ecological and hydrological applications. However, their reliability and temporal representativeness remain a concern, particularly due to the need for updated climate normals under ongoing climate change. This study aimed to validate an updated 30-year climatological precipitation surface, generated from monthly time series rasters downscaled and provided by WorldClim 2.1 based on the CRU-TS 4.06 dataset. Data from 335 stations were used, and raster values were extracted at station locations. Statistical agreement was assessed using Spearman’s correlation ( \(\rho \) ), mean absolute error (MAE), root mean square error (RMSE), and percentage errors, computed for monthly and annual data. In addition, a methodological framework was developed to evaluate potential geometric errors by identifying spatial misalignments between stations and surfaces. Local Indicators of Spatial Association (LISA) were applied to detect significant spatial clusters of error. Monthly MAE ranged from 3.2 mm (July; 38.8%) to 21.6 mm (December; 8.3%), while monthly \(\rho \) varied from 0.66 (April) to 0.90 (February), reaching 0.84 annually. Annual MAE averaged 87.5 mm (6.9%), with higher errors concentrated in mountainous regions in the south and east of the state. LISA clusters revealed spatially structured patterns of systematic over and underestimation, and a northwestward geometric misalignment was identified. Overall, the evaluated climatological precipitation surface showed good performance, but the identified spatial errors highlight the importance of regional validation. The proposed methodology provides a robust framework for evaluating climatological surfaces for climate-related applications and can be extended to other regions and datasets.
