Heavy and extreme rainfall events are among the most critical climatic hazards in the arid and semi-arid regions of Iran. This study investigates the spatiotemporal patterns of intensity, frequency, and trends of extreme precipitation at 18 selected stations in eastern and northeastern Iran over the period 1995–2024. The 95th and 99th percentile indices were employed to identify extreme events, representing relatively frequent high intensity rainfall and very rare but exceptionally intense events, respectively. The results indicated that the highest intensities of heavy rainfall occurred at Quchan, Kashmar, and Torbat-Heydarieh stations, while the greatest frequencies were recorded at Neyshabur, Torbat-Heydarieh, and Golmakan, highlighting the spatial concentration of extreme events in the northern highlands and confirming the role of topography in intensifying such phenomena. Conversely, the 99th percentile analysis revealed that in southern areas such as Nehbandan, although event frequency was lower, the intensity of extreme rainfall reached remarkably high levels, indicating the occurrence of rare but hydrologically powerful events. This points to a strong spatial heterogeneity in extreme rainfall behavior across the study area. Trend analysis using the Mann–Kendall test and Sen’s slope estimator showed a statistically significant decrease in daily rainfall at some stations, whereas changes in the intensity of extreme events were mostly limited and exhibited no consistent pattern. Overall, the findings emphasize the necessity of simultaneously considering intensity, frequency, and spatial distribution as key components for accurate assessment of meteorological hazards associated with extreme rainfall in eastern and northeastern Iran.
Introduction
In recent years, extreme precipitation events have received increasing attention as a major manifestation of climate change. Characterized by intense rainfall over short periods, these events can generate substantial hydrological, urban, and environmental impacts. Recent studies have shown that the intensity of extreme precipitation in the Northern Hemisphere has increased, in many cases more rapidly than snowfall. Iran, located within the arid and semi-arid belt of the Iranian Plateau, has experienced rising temperatures and a relative decline in precipitation. Previous research indicates that climate change has increased the number of extremely dry days while also altering the probability of heavy rainfall events. Even in areas where mean precipitation has declined, the intensity and frequency of extreme events, particularly those exceeding the 95th and 99th percentiles, remain highly important because warmer atmospheric conditions can increase moisture availability and enhance convective activity.
At broader spatial scales, studies across Asia have documented substantial spatiotemporal changes in extreme precipitation. Increasing trends in indices such as Rx1day and R95p have been reported in East and Southeast Asia, elevating the risk of flash floods. In the Indian subcontinent, heavy daily precipitation has increased in some coastal areas but declined in parts of the interior. In West Asia and Iran, despite reductions in total annual precipitation, the relative contribution of extreme rainfall has increased in many regions. Europe and North America have generally experienced increases in the intensity and frequency of heavy rainfall, whereas trends in Africa and South America remain spatially heterogeneous and strongly dependent on local and seasonal conditions.
Against this background, the present study investigates the intensity, frequency, and long-term trends of heavy and very heavy cold-season precipitation in northeastern Iran over the 30-year period 1995–2024. The main objective is to identify the spatial and temporal patterns of extreme precipitation and assess their long-term behavior using percentile-based indices and non-parametric trend tests, thereby providing information relevant to hydrological risk management and water-resources planning.
Material and Methods
Daily precipitation data for the five cold months of the year (November to March) during 1995–2024 were obtained from the Iran Meteorological Organization. Eighteen synoptic stations in northeastern Iran were selected: Torbat-e Jam, Bojnord, Boshruyeh, Birjand, Torbat-e Heydariyeh, Zabol, Zahedan, Sabzevar, Sarakhs, Quchan, Kashmar, Gonabad, Mashhad, Nehbandan, Nishapur, Ferdows, Golmakan, and Qaen. The data were processed in Excel, and days with precipitation totals below 1 mm were excluded from the analysis. Extreme precipitation events were identified using the 95th and 99th percentiles of wet-day precipitation. Trend analysis was performed using the non-parametric Mann–Kendall test and Sen’s slope estimator. The Mann–Kendall test is widely used to detect monotonic trends in hydroclimatic time series and is relatively robust to outliers, whereas Sen’s slope estimator is used to quantify the magnitude and direction of trends.
Results and Discussion
Based on the 95th percentile, the highest heavy-precipitation threshold values were observed at Quchan (15.25 mm), Kashmar (14.85 mm), and Torbat-e Heydariyeh (14.69 mm), whereas Zabol (9.17 mm), Zahedan (9.59 mm), and Bojnord (9.90 mm) recorded the lowest values. In terms of frequency, the highest numbers of heavy-rainfall days were recorded in Nishapur (71 days), Torbat-e Heydariyeh (70 days), and Golmakan (62 days), while Zabol (16 days) and Nehbandan (17 days) had the fewest. These patterns suggest that topography and elevation may play an important role in shaping the spatial distribution of heavy precipitation.
For the 99th percentile, the highest very heavy precipitation values were observed in Nehbandan (35.24 mm), Torbat-e Heydariyeh (28.51 mm), and Kashmar (27.45 mm). The highest frequencies were recorded in Bojnord (27 days), Quchan (25 days), and Mashhad (24 days). Despite its high intensity, Nehbandan experienced only five days of very heavy precipitation, indicating the occurrence of rare but potentially hazardous events. By contrast, northern stations such as Quchan and Torbat-e Heydariyeh exhibited both relatively high intensity and high frequency, implying greater cumulative hazard.
More than 70% of heavy precipitation events during the cold season occurred between December and March. Golmakan recorded 30 events in March, representing the highest monthly frequency, whereas Nishapur and Torbat-e Heydariyeh showed a more even monthly distribution. Zabol and Nehbandan experienced notable events mainly in only one or two months, reflecting the influence of the arid, low-precipitation climate of southeastern Iran.
Trend analysis using the Mann–Kendall test and Sen’s slope estimator indicated decreasing tendencies in daily precipitation totals at most stations. Statistically significant declines were identified in Nehbandan (p=0.02p = 0.02p=0.02), Boshruyeh (p=0.01p = 0.01p=0.01), and Mashhad (p=0.04p = 0.04p=0.04). The estimated slope was −0.04-0.04−0.04 mm/day in Boshruyeh, while it was close to zero in Torbat-e Jam. No station showed a statistically significant increasing trend. These findings suggest that declines in mean precipitation do not necessarily imply a reduced risk of extreme rainfall events. Annual variations also revealed marked interannual variability in heavy and very heavy precipitation, with peaks in 1999, 2007, 2017, and 2019, and minima in 2001, 2010, 2016, and 2021. Northern and high-elevation areas such as Nishapur, Quchan, and Bojnord experienced more frequent and persistent heavy precipitation, whereas the drier southeastern areas were characterized by infrequent but highly intense events.
Conclusion
Analysis of the 95th and 99th percentiles at 18 stations in northeastern Iran revealed substantial spatial and temporal heterogeneity in heavy and very heavy cold-season precipitation. Northern and high-elevation areas, characterized by both high intensity and high frequency, are more exposed to recurrent flooding, whereas the arid southeastern regions, despite lower event frequency, face an elevated risk of sudden and localized floods due to the occurrence of rare but intense rainfall. The seasonal concentration of events is broadly consistent with the influence of Mediterranean systems and western atmospheric disturbances. Long-term trends indicate a general decline in daily precipitation, although reductions in mean rainfall do not necessarily correspond to a lower risk of extreme events. These findings highlight the need for resilient infrastructure, improved runoff management, and long-term monitoring systems to support more accurate climate-hazard assessment and adaptation planning.