中文摘要：

积雪是地表最活跃的自然要素之一,其动态变化对气候、环境以及人类生活都产生了重要影响。本文利用MODIS积雪产品和IMS雪冰产品,首先通过Terra、Aqua双星合成和临近日合成去除MODIS积雪产品中的部分云像元,再与IMS融合,获取了青藏高原2002-2012年逐日无云积雪覆盖产品,并逐像元计算每个水文年的积雪覆盖日数（SCD）、积雪开始期（SCS）和积雪结束期（SCE）,分析了不同生态分区积雪的时空变化特征,以及积雪开始期和结束期与温度、降水的关系。结果表明：青藏高原积雪分布存在明显的空间差异,南部喜马拉雅山脉和念青唐古拉山地区以及西部帕米尔高原和喀喇昆仑山脉为SCD的2个高值区,年均积雪日数在200 d以上。18.1%的区域SCS表现出明显的提前趋势,主要集中在青藏高原中东部;羌塘高原南部、念青唐古拉山西段以及川西地区有显著推迟趋势,占高原面积的8.5%。23.2%的区域SCE显著推迟,主要集中在果洛那曲高寒区、昆仑山区和念青唐古拉山地区;而仅有6.9%的区域表现出提前趋势,主要分布在高原西南部。总体上,不同生态单元内积雪开始与结束期受温度、降水的影响差异很大,表现出不同的空间格局与演变趋势。

英文摘要：

Snow cover is one of the most active natural components on Earth＇s surface. The variability of snow phenology has a major impact on water cycle, climate change, environment and human activities. The Qinghai-Tibetan Plateau has a wide range of seasonal snow cover, and its accumulation and rapid meltdown can affect the regional and global climate change. Studying the snow variability in the Qinghai-Tibetan Plateau is therefore important. In this study, the MODIS snow product and IMS snow-ice product were used. Firstly, the Terra and Aqua satellite images were combined to reduce the proportion of cloud pixels. Secondly, the temporal combinations were employed to further reduce the cloud pixels. Finally, the processed MODIS snow product and IMS were fused to produce the daily cloud-free snow product of the Qinghai-Tibetan Plateau from 2002 to 2012. Then, the snow-covered days （SCD）, snow cover start （SCS） and snow cover end （SCE） dates were calculated for each hydrological year, and their spatial and temporal variations in different eco-geographical regions were analyzed. The correlations among the SCS, SCE and climate fac- tors were also investigated. The results show that the distribution of snow cover over the Qinghai-Tibetan Plateau was very uneven. The longest SCD, totalized to be more than 200 days, occurred in the Himalayas, Karakoram, Nyainqentanglha Mountains and the Pamirs Plateau. Up to 18.1% of the area of SCS showed a significantly advanced trend, which mainly occurred in the Golog-Nagqu high-cold region and the southern Qinghai high-cold region; while 8.5% of the area showed a slightly delayed trend. Up to 23.2% of the area of SCE was delayed, occurring mainly in the central and eastern Tibetan Plateau; while only 6.9% of the area showed an advanced trend. The SCS and SCE were greatly affected by temperature and precipitation, but showed different spatial patterns and evolution trends in different ecological zones. Generally, the higher temperature delayed the SCS and advanced the SCE, but