بررسی روند تغییرات سطح پوشش برف در حوضه سفیدرود با استفاده از سنجش از دور

برف خوان ها یکی از مهم ترین منابع تامین آب در ایران به شمار می روند. نوشتار پیش رو بر آن است که تغییرات گستره پوشش برف را در حوضه سفیدرود و زیرحوضه های آن واکاوی کند. برای دستیابی به این هدف از تصاویر سنجنده مادیس محصول mod10cm در یک بازه زمانی نوزده‌ساله (20002019) بهره گرفته شد. پس از تجزیه و تحلیل تصاویر در محیط نرم افزار انوی، سطح برف در هر تصویر محاسبه و بر پایۀ آن، مقادیر ماهانه برای تمام دوره آماری استخراج و به محیط نرم افزار آرک جی.آی.اِس. منتقل شد. برای بررسی روند تغییرات سالانه و فصلی از آزمون منکندال اصلاح شده و برای برآورد نرخ تغییرات از تخمین گر شیب سن استفاده شد. بر پایه یافته‌ها می توان دریافت که زمان شروع و پایان برف در حوضه سفیدرود و زیرحوضه های آن دست‌خوش تغییرات بسیار جزئی بوده است. گستره برف پوش حوضه از ماه اکتبر آغاز و پس از افزایش به طرف دورۀ سرد سال، در ماه آوریل به کمترین حدّ خود می رسد. ماه بیشینۀ پوشش برف در تمام زیرحوضه ها ژانویه است. طی بازه زمانی مورد واکاوی، گستره برف در سال آبی 20072006 بیشینه و در سال 20102009 کمینه بوده است. یافته‌های حاصل از برازش آزمون ناپارامتریک منکندال اصلا ح شده روی سری های زمانی سالانۀ گستره برفی حوضه بیانگر روند کاهشی و البتّه غیر معنی دار در بخش اعظم سطح حوضه است، ولی گاه روندهای افزایشی ملایم نیز دیده می شود. به لحاظ پراکنش مکانی، در مناطق مرکزی و شرقی حوضه روند کاهشی معنی دار گستره برفی دیده می شود و نرخ کاهش سالانه آن در برخی مناطق یادشده 77/17% به ازای هر دهه است. روندهای فصلی نیز کاهشی است. بیشترین نرخ کاهش مربوط به فصل زمستان است که در برخی مناطق به حدود 33% به ازای هر دهه نیز می رسد. برف خوان ها یکی از مهم ترین منابع تامین آب در ایران به شمار می روند. نوشتار پیش رو بر آن است که تغییرات گستره پوشش برف را در حوضه سفیدرود و زیرحوضه های آن واکاوی کند. برای دستیابی به این هدف از تصاویر سنجنده مادیس محصول mod10cm در یک بازه زمانی نوزده‌ساله (20002019) بهره گرفته شد. پس از تجزیه و تحلیل تصاویر در محیط نرم افزار انوی، سطح برف در هر تصویر محاسبه و بر پایۀ آن، مقادیر ماهانه برای تمام دوره آماری استخراج و به محیط نرم افزار آرک جی.آی.اِس. منتقل شد. برای بررسی روند تغییرات سالانه و فصلی از آزمون منکندال اصلاح شده و برای برآورد نرخ تغییرات از تخمین گر شیب سن استفاده شد. بر پایه یافته‌ها می توان دریافت که زمان شروع و پایان برف در حوضه سفیدرود و زیرحوضه های آن دست‌خوش تغییرات بسیار جزئی بوده است. گستره برف پوش حوضه از ماه اکتبر آغاز و پس از افزایش به طرف دورۀ سرد سال، در ماه آوریل به کمترین حدّ خود می رسد. ماه بیشینۀ پوشش برف در تمام زیرحوضه ها ژانویه است. طی بازه زمانی مورد واکاوی، گستره برف در سال آبی 20072006 بیشینه و در سال 20102009 کمینه بوده است. یافته‌های حاصل از برازش آزمون ناپارامتریک منکندال اصلا ح شده روی سری های زمانی سالانۀ گستره برفی حوضه بیانگر روند کاهشی و البتّه غیر معنی دار در بخش اعظم سطح حوضه است، ولی گاه روندهای افزایشی ملایم نیز دیده می شود. به لحاظ پراکنش مکانی، در مناطق مرکزی و شرقی حوضه روند کاهشی معنی دار گستره برفی دیده می شود و نرخ کاهش سالانه آن در برخی مناطق یادشده 77/17% به ازای هر دهه است. روندهای فصلی نیز کاهشی است. بیشترین نرخ کاهش مربوط به فصل زمستان است که در برخی مناطق به حدود 33% به ازای هر دهه نیز می رسد.

Investigating the Changes in Snow Cover in Sefidrood Drainage Basin using Remote Sensing

Snow reservoirs are one of the most important sources of water supply in Iran. This study aims to investigate the changes in snow cover in Sefidrood basin and its subbasins. To achieve this goal, snow cover images of MODIS’s MOD10cm product were used over a period of 19 years (2000 to 2019). After analyzing the images in ENVI software, the snow area of each image was calculated and, consequently, monthly values for all years of the study period were extracted and transferred to ArcGIS software. The modified MannKendall test was applied to examine the annual and seasonal trends of snow cover. Then, the Sen’s slope estimator test was used to determine the rate of change. Based on the findings, it can be seen that the time of beginning and end of snow in Sefidrood basin and its subbasins has undergone very little change. The snow cover of the basin starts in October and reaches its minimum in April after increasing towards the cold period of the year. January is the month with maximum snow cover in all subbasins. During the period under review, the range of snow cover was maximum in the water year 20062007 and minimum in 20092010. Findings from the fit of the modified nonparametric MannKendall test on the annual time series of the snow cover of the basin indicate a decrease and of course insignificant trend in most of the basin area, although sometimes slight increasing trends are also seen. In terms of spatial distribution, in the central and eastern regions of the basin, there is a significant decreasing trend of snow cover; its annual decreasing rate in some areas is 17.77% per decade. Seasonal trends are also decreasing. The highest rate of decrease is related to winter, which in some areas reaches about 33% per decade.Extended Abstract1IntroductionAccording to a report, approximately 17% of the world population relies on seasonal snowmelt and glacier melting to provide the water they need. The temporal and spatial characteristics of snow accumulation and melting affect most of the earth vital and nonvital processes. Snow cover varies spatially as well as temporally during different seasons of the year. The expansion or decline of snow cover over a period of time in an environment indicates the presence of changes in that environment and acts as an indicator for climate change. At high altitudes, due to the rugged terrain and harsh climate, there are restrictions on the establishment of meteorological stations. Under these conditions, remote sensing data have a high potential to provide suitable spatial data to describe the spatial and temporal pattern of snow cover. The current study aims to identify the trend of snow cover in one of the most important drainage basins in the country, Sefidrood Drainage Basin, and its subbasins.2Materials and MethodsTo conduct this research, snow cover images of MODIS’s MOD10cm product were used over a period of 19 years (2000 to 2019). These images were analyzed with ENVI software. Then, the snow area of each image was calculated and, consequently, monthly values for all years of the study period were extracted and transferred to ArcGIS software. The modified MannKendall test was used to examine the annual and seasonal trends of snow cover. Besides, the Sen’s slope estimator test was applied to determine the rate of change. The trend of snow cover changes for all subbasins was analyzed at 90% confidence level. It should be noted that the calculations were performed separately for each pixel and for all months of the year.3Results and DiscussionIn Sefidrood basin during the study period, the highest and lowest amounts of snow cover were related to the water years 20062007 and 20092010, respectively. January was also the month of maximum snow cover in all subbasins. According to the calculations, the annual trends of snow cover in different subbasins were similar during the study period and all had a slight decreasing slope. Therefore, there is no noticeable change in the snowy season of the region. MannKendall test showed changes in snow cover in pixels for all Sefidrood subbasins. Based on the results of this test, the general annual trend of snow cover changes in subbasins during the statistical period is decreasing. The only subbasin whose average slope of change is not negative and decreasing is the Mianeh subbasin, whose average is above zero. The most negative slope for a pixel was 17.77 in the GoltepehZarrinebad subbasin, and the most positive for a pixel was 13.93 in the Mianeh subbasin. Similar to the annual trend, the seasonal trends are decreasing. The highest rate of decrease is related to winter, which in some areas reaches about 33% per decade. Thus, in most areas of Sefidrood basin, especially in the eastern, central and southern subbasins, a decreasing trend of snow cover is seen. In some parts of the northern and western regions of the basin, instead of decreasing, there is a slight increase in snow cover. The existence of a decreasing trend in most areas of the Sefidrood basin is consistent with the results of many internal and external studies. In foreign studies, Mityok et al. (2018), Singh et al. (2018) and Malmros et al. (2018) have shown a decreasing trend in the snow cover of the study area, while Rathore et al. (2018) in India have acknowledged the increasing trend.In Iran, Jahanbakhsh Asl et al. (2020), Azizi et al. (2016), Khosravi et al. (2017) and Fattahi and Moghimi (2019) have reported a significant decreasing trend in their studies in different regions of Iran. In the present study, the decreasing trend of snow cover in Sefidrood basin has been emphasized, but it has been stated that this decreasing trend is not very severe in all areas and even in limited areas of the basin there is a very low increasing trend.4ConclusionSince most of the area of Sefidrood basin is mountainous and continuous monitoring of snow cover in this large basin will not be possible with conventional methods, so satellite images, especially MODIS images that provide information on a daily basis, can be very useful and important for monitoring snow cover. It is true that according to the results of this study, the declining trend of snow resources is not very severe and may not be significant in some cases, but due to the growing population and consequently increasing water needs in the drinking, agricultural and industrial sectors, this slight decrease and even stability in the amount of snowfall can lead to a shortage of water resources and create a water crisis in the study area and the whole country. Therefore, continuous monitoring of snow cover and its trend in the region is very important in future studies.