2000-2016年青海湖湖冰物候特征变化

湖冰物候特征是气候变化的灵敏指示器。基于2000-2016年青海湖边界矢量数据,结合Terra MODIS和Landsat TM/ETM+遥感影像及气象数据,利用RS和GIS技术综合分析青海湖湖冰物候特征变化及其对气候变化的响应。结果表明：① 青海湖开始冻结、完全冻结、开始消融和完全消融的时间分别为12月中旬、1月上旬、3月中下旬和3月下旬至4月上旬,平均封冻期和平均完全封冻期为88 d和77 d,平均湖冰存在期和平均消融期为108 d和10 d。② 近16年间青海湖湖冰物候特征各时间节点变化呈现较大的差异性。湖泊开始冻结日期相对变化较小,完全冻结日期呈先提前后推迟的波动趋势,开始消融日期呈先推迟后提前的波动趋势,完全消融日期在2012-2016年呈明显提前趋势。青海湖封冻期在2000-2005年和2010-2016年呈缩短趋势,但减少速率慢于青藏高原腹地的湖泊。③ 青海湖冻结和消融的空间模式相同,即湖冰形成较早的区域则消融较早,且前者持续时间（18~31 d）整体上大于后者（7~20 d）,二者相差约10 d。④ 冬半年负积温大小是影响青海湖封冻期的关键要素,但风速和降水对青海湖湖冰的形成和消融亦发挥着重要作用。

Spatial-temporal characteristics of ice phenology of Qinghai Lake from 2000 to 2016

Lake ice phenology is considered a sensitive indicator of regional climate change. We utilized time series information of this kind extracted from a series of multi-source remote sensing (RS) datasets including the MOD09GQ surface reflectance product, Landsat TM/ETM+ images, and meteorological records to analyze spatiotemporal variations of ice phenology of Qinghai Lake between 2000 and 2016 by applying both RS and GIS technology. We also identified the climatic factors that have influenced lake ice phenology over time and draw some conclusions. First, data show that freeze-up start (FUS), freeze-up end (FUE), break-up start (BUS), and break-up end (BUE) on Qinghai Lake usually occurred in mid-December, early January, mid-to-late March, and early April, respectively. The average freezing duration (FD, between FUE and BUE), complete freezing duration (CFD, between FUE and BUS), ice coverage duration (ICD, between FUS and BUE), and ablation duration (AD, between BUS and BUE) were 88 days, 77 days, 108 days and 10 days, respectively. Second, while the results of this analysis reveal considerable differences in ice phenology on Qinghai Lake between 2000 and 2016, there has been relatively little variation in FUS times. Data show that FUE dates had also tended to fluctuate over time, initially advancing and then being delayed, while the opposite was the case for BUS dates as these advanced between 2012 and 2016. Overall, there was a shortening trend of Qinghai Lake's FD in two periods, 2000-2005 and 2010-2016, which was shorter than those seen on other lakes within the hinterland of the Tibetan Plateau. Third, Qinghai Lake can be characterized by similar spatial patterns in both freeze-up (FU) and break-up (BU) processes, as parts of the surface which freeze earlier also start to melt first, distinctly different from some other lakes on the Tibetan Plateau. A further feature of Qinghai Lake ice phenology is that FU duration (between 18 days and 31 days) is about 10 days longer than BU duration (between 7 days and 20 days). Fourth, data show that negative temperature accumulated during the winter half year (between October and the following April) also plays a dominant role in ice phenology variations of Qinghai Lake. Precipitation and wind speed both also exert direct influences on the formation and melting of lake ice cover and also cannot be neglected.