Daily precipitation, particularly intense and extreme rainfall events, plays a critical role in the occurrence of hydrological hazards, especially in mountainous regions. However, relying solely on mean-based indices or fixed thresholds is insufficient to fully capture the actual behavior of such events. The present study aims to investigate the spatial patterns of daily precipitation in West Azerbaijan Province using percentile classification and geostatistical methods.
For this purpose, daily precipitation records from 17 meteorological stations during the period 2004–2024 were analyzed after excluding precipitation amounts below 1 mm. The precipitation data were classified into four categories: minimum to mean, mean to the 75th percentile, 75th to 90th percentile, and 90th to 100th percentile. Subsequently, geostatistical tools including histograms, quantile–quantile (Q–Q) plots, trend analysis, Moran’s spatial autocorrelation index, and various kriging interpolation methods were employed to evaluate the spatial structure of precipitation.
The results indicated that all percentile classes exhibited non-normal distributions and nonlinear spatial trends. Furthermore, Moran’s index confirmed the presence of significant clustered spatial patterns across all classes. Analysis of the interpolated maps revealed that spatial dependence decreased with increasing precipitation intensity, and the precipitation distribution pattern gradually shifted from relatively continuous structures to more heterogeneous and localized patterns. In particular, precipitation events within the 90th–100th percentile class exhibited the highest degree of spatial heterogeneity.
The proposed approach provides a more comprehensive understanding of the spatial behavior of daily precipitation and can contribute to improved flood management and water resources planning.