Snow cover changes and affecting hydro-climate variables in Sabalan Mountainous region in Northwest Iran

Abstract

Introduction

Mountain snow cover is a crucial component of the hydrological cycle, particularly in semi-arid regions, where snowmelt serves as a significant water source for river flow, agriculture, and ecosystems. Snowmelt runoff significantly contributes to total river flow during spring and summer in mountainous regions, which, although limited in area, are hydrologically important. Mountain snow cover serves as a major source of surface and groundwater, supporting agriculture, domestic use, and aquifers. It also helps reduce natural hazards like floods and avalanches. Snow patterns are influenced by regional and global climatic factors, and changes in climate directly impact snow cover, snowmelt, and water availability. Key variables affecting snow distribution include elevation, slope, aspect, solar radiation, temperature, wind, humidity, and precipitation. Climate change, especially global warming, has altered snow dynamics, causing reduced snow cover, earlier melting, and changes in runoff. Monitoring snow through ground stations is often difficult, so remote sensing has become a valuable method for tracking snow characteristics on larger spatial scales. In recent years, the effects of climate change, coupled with anthropogenic pressures, have posed serious challenges to the sustainability of water resources in mountainous regions globally. Snow cover dynamics, influenced by variations in temperature, precipitation, and streamflow, provide key insights into climate-driven changes affecting hydrological systems. Snow accumulation in watersheds serves as a crucial water reserve, making the study of snow amount, quality, and melt processes important for hydrology and natural resource management. Unlike rainfall, snow provides a more predictable water source during dry seasons due to its gradual release. Sabalan Mountain in Ardabil Province, Iran, is a key source of water, with its snow cover playing a major role in maintaining water supply for drinking, agriculture, and industry. However, recent climate changes are altering snow patterns on the mountain, affecting river flows and water availability. As a protected and ecologically valuable area, Sabalan's hydrology, primarily driven by snowmelt, is increasingly influenced by shifts in climate and land use. Sabalan Mountain in Ardabil Province, northwestern Iran, is one of the most ecologically and hydrologically significant areas in the region. It functions as a water reservoir due to its prolonged snow accumulation and gradual melt patterns, feeding the headwaters of many important rivers. However, recent studies point to a transformation in the hydro-climatic regime of this area, linked to climate change and possibly to land use changes. This study investigates the temporal evolution of snow cover in the Sabalan Mountain region over a 24-year period (1991–2015), and examines how hydro-climatic factors, including temperature, precipitation, and discharge, affect snow cover variability. The results contribute to a better understanding of climate-snow-water interactions in mountainous regions and offer insights into future water management strategies.



Materials and Methods

Sabalan Mountain, with elevations exceeding 4,800 meters, is a volcanic massif located in Ardabil Province. The region experiences a semi-arid and cold to freezing climate, classified based on the Emberger method. The average annual precipitation is about 360 mm, largely falling during autumn and winter months, predominantly in the form of snow. Temperature decreases and precipitation increases with altitude. The mountainous terrain feeds multiple river basins, and several hydrometric and climatological stations, including Meshgin-Shahr, Nir, Lay, Atashgah, and Pole-Soltani, are established in and around the area. Snow cover changes were assessed using Landsat satellite imagery acquired from the US Geological Survey for five key years: 1991, 1998, 2003, 2009, and 2015. These dates correspond to similar seasonal conditions in May to minimize inconsistencies caused by intra-annual snowmelt variations. Landsat 4 and 7 images were used, covering paths 33 and 34. The snow detection process was conducted using IMAGINE ERDAS 2014 software. Pre-processing involved gap-filling, band merging (to include parts of East Azerbaijan), and visual interpretation of snow-covered pixels. Classification accuracy was evaluated using the Kappa index and total accuracy metrics, with values ranging from 0.7073 (1991) to 0.9155 (2015), confirming the reliability of the snow cover detection process. Monthly data for precipitation (P), temperature (T), and river discharge (Q) were collected from multiple gauging stations for the period between 1991 and 2015. To evaluate long-term trends, the non-parametric Mann-Kendall (M-K) test was used, which is robust against non-normality and missing data. The M-K test was applied to detect trends in monthly precipitation, discharge, and temperature values for November and December—critical months for early snow accumulation. To assess the relationship between snow cover area and hydro-climatic variables, Pearson correlation coefficients were calculated using R software. This enabled the examination of linear associations between snow cover and temperature, precipitation, and discharge for corresponding timeframes.



Results and Discussion

Results showed a consistent decrease in snow cover from 1991 to 2015. The spatial extent of snow-covered areas declined by 12.7% between 1991 and 2003, and by 65.1% across the entire study period. While intermediate years (1998 and 2003) exhibited slight increases in snow cover compared to 1991, the overall trend points to a fragmented and contracting snowpack. Results confirms classification reliability through high Kappa indices, especially for 2009 and 2015. These values indicate that the image-based analysis accurately captured snow distribution patterns and that observed trends are likely reflective of true environmental changes. The decreasing snow trend may result from increased winter temperatures and changing precipitation patterns, which alter the phase of precipitation (from snow to rain) and accelerate snowmelt. At the Pole-Soltani station, a significant decreasing trend in discharge was observed in both November and December (p < 0.01). The Nir station also showed a significant decline in November discharge (p < 0.05). Interestingly, the Lay station exhibited an increasing trend in December discharge, although this trend was not statistically significant. Meanwhile, the Meshgin-Shahr station displayed a weak upward trend in December discharge. At the Meshgin-Shahr station, November precipitation increased significantly (p < 0.05). January precipitation at the Nir station also showed a significant increase (p < 0.05). However, overall precipitation trends fluctuated over time without a consistent long-term increase or decrease. Although temperature increased at both the Meshgin-Shahr and Nir stations, the trends observed in the analyzed months were not statistically significant. The M-K test findings align with the broader literature, indicating that climate change is affecting mountain hydro-climatology. Decreasing discharge in some stations, despite rising precipitation at others, could reflect reduced snow accumulation and earlier melt, resulting in diminished late-autumn and winter runoff. According to the results, a positive relationship is observed between snow cover and river discharge, particularly in snowmelt-dominated stations such as Nir and Atashgah. As snow cover declined, discharge patterns also changed, often decreasing in downstream stations due to reduced snowmelt contributions. Despite increasing precipitation at certain stations, discharge did not always increase, suggesting that precipitation falling as rain (rather than snow) and increased evapotranspiration may reduce effective runoff. The results showed precipitation trends, particularly a decline at the Nir station in recent years. Meshgin-Shahr also shows reduced precipitation in later years. Combined with warming temperatures, this suggests reduced snow accumulation and earlier melting. This warming trend, particularly from 2009 to 2015, likely contributed to the sharp snow cover loss observed during the same period. Temperature is a critical factor in determining the form of precipitation and the persistence of snow. Even minor increases can shift precipitation from snow to rain, significantly affecting snowpack volume and duration. The warming trend observed in the Sabalan region supports global findings on mountain warming and snow loss.

Pearson correlation coefficients calculated between snow cover area and hydro-climatic variables showed weak and statistically insignificant relationships overall. This may be due to several factors: There exists a spatial and temporal mismatch between the point-based climate data collected at meteorological stations and the areal representation of snow cover derived from satellite imagery, which may affect the accuracy of correlation analyses. The study does not account for other potentially influential variables such as wind speed, solar radiation, and land use changes, which could significantly impact snow cover dynamics but remain unmeasured in the current analysis. The snow cover data, limited to only five discrete time points, lacks the temporal continuity of the hydro-climatic records, thereby constraining the ability to detect finer-scale variations and long-term trends. Although no strong statistical correlations were identified, the visual and trend-based analysis indicates that hydro-climatic changes, especially rising temperature and declining precipitation, are likely contributing to observed snow cover decline.



Conclusion

A significant reduction in snow cover extent, with a total decrease of over 65% across the study period. Statistically significant decreasing trends in river discharge at key stations, particularly Pole-Soltani and Nir, during early winter. Increasing temperature trends at all studied stations, though not always statistically significant. Limited but suggestive evidence of decreasing precipitation, especially in the later years. Weak statistical correlation between snow cover and hydro-climatic variables, possibly due to the complex interplay of multiple climatic and environmental factors. These findings underscore the vulnerability of mountain snowpacks to climate change and highlight the importance of continued monitoring using satellite remote sensing and ground-based observations. Accurate snow cover data, when integrated with hydro-climatic analysis, provide a vital basis for sustainable water resource planning, especially in semi-arid mountainous regions. Future research should incorporate additional variables such as land use change, snow depth measurements, and higher temporal resolution satellite data to improve the accuracy and explanatory power of snow-hydrology relationships. Additionally, hydrological modeling can help predict future water availability scenarios under projected climate change. By identifying the patterns and drivers of snow cover changes in the Sabalan Mountains, this study offers valuable guidance for regional water management authorities seeking to adapt to changing hydro-climatic conditions.