Extended Abstract
Introduction
Clouds are fundamental components of the Earth's climate system, playing a critical role in precipitation distribution, radiative energy balance, and the hydrological cycle. The physical and microphysical properties of clouds—including cloud-top height, temperature, optical thickness, and cloud fraction—significantly influence precipitation intensity and spatial distribution. This study aims to analyze cloudiness characteristics associated with Heavy Precipitation Events (HPEs) in Northwest Iran. Examining the spatiotemporal variations of clouds and their relationship with extreme precipitation provides valuable insights for weather forecasting, water-resource management, and climate-change impact assessment. The analysis of observational and reanalysis data serves as a robust tool for extracting cloudiness patterns and understanding their linkage to precipitation extremes. Identifying the spatial extent and persistence of cloud cover during HPEs also improves the interpretation of precipitation-generating systems and supports regional early-warning, disaster-risk management, and adaptation planning efforts.

Methodology
The study area covers Northwest Iran, encompassing the provinces of West Azerbaijan, East Azerbaijan, and Ardabil (35°–40° N, 43°–49° E). Daily precipitation data from 19 synoptic stations were obtained from the Iran Meteorological Organization (IRIMO) for the period 2000–2019. Total Cloud Cover (TCC) reanalysis data were extracted from the ERA5 dataset. Hourly Total Cloud Cover data for the identified HPE days were retrieved and converted into daily averages. For validation purposes, three selected events were compared against the MERRA-2 dataset. Heavy Precipitation Events were identified based on a combined criterion of intensity (99th percentile and above) and spatial extent (at least 30% of stations and a minimum of five stations recording >= 20.4 mm). The spatial contribution of stations to these events was mapped using the Inverse Distance Weighting (IDW) interpolation method in a GIS environment. Cloudiness characteristics, including intensity, spatial extent, and Total Cloud Cover patterns for 16 Heavy Precipitation Events, were analyzed using ERA5. For each event, the daily mean Total Cloud Cover, the percentage of area with TCC >= 0.7, and the center of mass of cloud cover were calculated. In addition, spatial maps were produced to illustrate the distribution and extent of cloud cover during each event across the study region. The performance of ERA5 for three selected events was evaluated using Willmott’s Index of Agreement (d), Root Mean Square Error (RMSE), and Bias.

Results and Discussion
A total of 16 Heavy Precipitation Events were identified during the study period, forming the basis for analyzing cloudiness characteristics and evaluating reanalysis-data performance. The analysis detailed the spatial coverage of Heavy Precipitation Events, the stations recording maximum precipitation, and the peak precipitation values for each event. Sardasht station recorded the maximum precipitation in 10 out of 16 events (62.5%), highlighting it as a hotspot for precipitation extremes in the region. The highest recorded precipitation reached 185 mm on 3 February 2006 at this station. Validation of ERA5 against MERRA-2 using Willmott’s d, RMSE, and Bias demonstrated that ERA5 performs reliably in representing cloudiness features associated with Heavy Precipitation Events in Northwest Iran. Pixel-based Total Cloud Cover patterns for the 16 events revealed that, although cloud intensity and extent varied across events, high Total Cloud Cover values were predominantly concentrated in the southwestern half of the region, aligning with areas of high station participation. This spatial correspondence indicates a close association between extensive cloud cover and the occurrence of widespread heavy precipitation across the monitoring network. The gradient from minimum to maximum Total Cloud Cover values indicates variations in cloud-system intensity; more intense events were associated with larger spatial clusters of high Total Cloud Cover. This recurring pattern suggests that the spatial structure of cloud cover during Heavy Precipitation Events in this region follows a relatively stable configuration.

Conclusion
The maximum spatial extent of Heavy Precipitation Events was observed on 18 February 2018, when 10 stations simultaneously recorded heavy precipitation, consistent with the extensive cloud cover shown in the pixel maps. The validation results indicated reduced error and optimal statistical performance of ERA5 during high-intensity Heavy Precipitation Events. In 50% of the events, the center of mass of cloud cover was located southeast of Tabriz station, with a spatial cloud extent (TCC ≥ 0.7) of 1, indicating 100% cloud cover. The maximum spatial cloud coverage among the 16 events occurred on 3 February 2006, whereas the minimum value (0.43) was recorded on 30 March 2018.