Spatio-temporal variability of snow cover in the south slopes of central Alborz

Snow cover is one of the most important components of the Earth's surface which has an important role in the global hydroclimate. Snow acts as a temporary reservoir of water and frequently, to maintain the flow of rivers and recharge underground aquifers provide water during the dry season caused billions consumer. Study of spatial and temporal variability of snow cover in arid and semiarid region such as Iran that temporal and spatial variations of precipitation is high and has always suffered from lack of water is important. So that about 60% of surface water and groundwater 57 percent of the country is underway in snowy areas. Therefore, this study aimed to evaluate the accuracy of MODIS snow products and combines remote sensing and terrestrial data to investigate the spatiotemporal changes of snow cover in South Central Alborz slopes and ultimately the relevance of this change is done with climatic elements.
Data and In the present study, we used data from 16 synoptic stations located in the study area and MODIS data. At first, to extract snow cover, the products MOD10A1 and MOD10A2 were used. Then, to evaluate the accuracy of MODIS and relationship between snow cover changes with climatic elements of the data of snow depth, precipitation and temperature (on a scale of hourly and monthly) were selected stations. MODIS images for the detection of snow cover in the index NDSI = (band4-band 6) / (band4 band6) is used. Band4 the spectral reflectance in the visible band (0.555 micrometers) and band6 is the spectral reflectance in the intermediate-infrared band (1.64 micrometers). In this sensor products addition to snow other phenomena is indivisible. Therefore, in order to separate and identify different phenomena pixels, images were processed in GIS software. For evaluation of pictures, daily product images (MOD10A1) to winter (December to February), three years (2007-2009) and precision processing station were evaluated by snow depth data. If one centimeter snow depth at the station and more, pixel located at the station as snow cover and otherwise were considered as no snow. The agreement between the image and the correct classification of stations coefficient that as the number of days (without snow- snow-snow and no snow) divided the total number of days per month in percent, were defined. Finally, the variability of snow cover was evaluated by Mann-Kendall test. To examine the relationships anomalies snow cover and climatic conditions Z index was used.
The percentage of agreement between the earth data and satellite images for three months of winter are respectively from December to February, 81, 67 and 75 percent. However, the mean of all stations is used in snow-prone areas even mean reduced image accuracy up to 40 percent. Studies show that errors are often caused by clouds in the pixel location. Thus the second stage evaluation was performed by removing the cloudy days. The results show that at this stage much increased image accuracy and agreed percentage for each quarter to more than 95 percent. Therefore, according to the movement of the clouds, the product of eight days this sensor was used to monitor and evaluate changes in snow cover. Snowfall in the region since October began with a decrease in air temperature and snow accumulation increased in the period leading up to January 9, the maximum is 34%. January and December respectively 31.5% and 24.8% accounted for the highest monthly snow cover. The highest and lowest snow cover these two months, with respectively 76.2 percent and 5.8 percent is owned by the January 9, 2008 and December 27, 2010. Study shows the monthly change in snow cover, in October, November and March of this phenomenon is on the rise, although not significant. While in December, January, April and especially May in most classes, snow cover has decreased over the last 15 years. This negative trend in the months -2.18 score is significant. Compare snow cover with temperature and precipitation shows, snow cover mostly positive anomalies with positive precipitation anomalies and negative temperature anomalies are consistent and vice versa.
The results of evaluating the accuracy of satellite images showed snow produce of MODIS has the ability to appropriate estimate an area of snow cover. But the cloud is one of the main limitations of MOD10A1. So that in present study after removing the cloudy days, the average accuracy of this images from 67% to 95% and even in snow-prone stations reached to 100 percent. Because the clouds are moving quickly changing daily, but change snow is gradual so for long-term monitoring of snow cover the product (MOD10A2) this sensor is used. The monitoring results showed January and December are the greatest area of snow cover. In terms of spatial continuity and extent of snow cover decreases from West to East in the study area. So that, percentage of snow covers in the Shahrood and Karaj river basins are more than the Semnan and Hablehrood basins. The results of the trend reveal, though for most months and zones of elevation changes of snow cover has a negative trend, but rarely this trend is significantly. In timescale highest positive trend in October with a score of 1.68 and the severest decline in May with a score of -2.18 was observed. In terms of spatial the greatest decline was observed in the elevation 2500 to 3750 meters. In most years positive anomalies of snow cover coincides with positive (negative) anomalies of precipitation (temperature).