【作者】 陈思宇；

【导师】 梁天刚；

【作者基本信息】 兰州大学 ， 草业科学， 2015， 博士

【摘要】 青藏高原上分布着世界上面积最大、类型最为独特的高寒草地生态系统和我国近一半的湖泊,是重要的生态屏障,也是全球气候变化的敏感区域。作为高原生态系统的重要组成部分,高寒草地植被和高原湖泊由于受到气候变暖的影响而发生着显著的变化,从而进一步影响高原生态系统的稳定性。因此,研究植被活动和湖泊的变化及其与气候因子之间的关系,在认识高原生态环境变迁和可持续发展方面具有重要意义。本研究基于多种遥感资料,结合1981-2010年月平均气温、月累积降水的格网插值数据,分别分析1981-2010年青藏高原草地植被物候的时空变化趋势和HY2003-2010年(Hydrological year, HYi)湖泊水位变化特点及其对气候变化的时空响应特征。在此基础上,选取纳木错流域作为典型研究区,结合野外调查资料,对流域的气候、湖泊和草地之间的关系以及积雪对草地植被生长的影响进行了系统的分析,得到以下结论：1)1981-2010年青藏高原的草地物候由东南向西北呈现出明显的区域性差异。高原东部和西北部地区的草地植被返青时间(SOG)较早,枯黄时间(EOG)较晚,相应的生长季长度(LOG)也较其他地区长。分析近30年青藏高原草地物候的变化趋势发现,植被物候参数的变化趋势在东西部地区的差异也十分明显。草地返青提前的区域主要集中在高原的东部地区。高原西部的草地具有返青推迟、生长季长度缩短的趋势。高原西南部和青海东部草地的枯黄时间呈现出推迟的趋势。西部和东部部分地区的草地枯黄时间提前。2)从时间角度分析,青藏高原的春季温度和春季降水对草地植被的返青具有一定的促进作用。在EOG提前和LOG缩短的区域内,夏季气温是限制草地EOG和LOG的决定因素。EOG推迟和LOG延长区域的草地EOG和LOG的变化由气温和降水共同决定。在空间分布上,高原东部地区的草地返青时间、高原东部和西部的草地枯黄时间及高原北部地区的生长季长度受温度和降水的共同影响,而其他地区的返青、枯黄时间则主要受到温度的调控,受降水的影响较小。青藏高原东部草地生长季长度受降水的影响较大。3)青藏高原大部分湖泊水位集中在4500～5000 m,主要分布在青藏高原的中部和西部地区,该地区湖泊水位变化十分剧烈,部分湖泊水位上升十分明显。高原南部的打加错、羊卓雍错和高原西部的班公错的湖泊水位下降趋势显著。分析青藏高原湖泊水位变化与气候变化的关系表明,不同流域的湖泊水位随着气温和降水的变化而呈现出不同的变化趋势,湖泊水量的增加除了依靠降水的直接补给以外,也会受到由温度变暖而引起的积雪融水增加的影响。4)近年来,纳木错流域气温的显著上升是造成盛草时期草地植被恶化和纳木错湖泊面积持续扩张、水位上升的主要原因。纳木错流域草地EVI与湖泊面积之间呈显著的负向相关关系,A区(距湖泊边界25 km以内的区域)草地的EVI与湖泊面积的负向相关关系(R2=0.76,P=0.001)高于B区(R2=0.67,P=0.004)(距湖泊边界25-50km的区域)和C区(R2=0.38,P=0.06)(距湖泊边界50-95 km的区域),说明湖泊面积扩张越剧烈,距离湖泊越近地区的草地植被长势的恶化趋势越明显。究其原因有以下2点：(1)湖泊扩张导致湖泊周边地区(即A区)的草地被湖水掩埋；(2)由于冰雪融水形成的地表径流具有搬运作用,在径流汇入湖泊的过程中使得更多的砾石在A区的草地上堆积导致对A区草地植被的侵蚀较B区和C区严重。5)纳木错流域积雪覆盖率与植被返青期呈正向相关关系,与生长季长度呈负向相关关系。在不考虑积雪对植被物候的影响下,温度与植被物候期之间无明显相关关系,说明流域植被物候对4-5月平均积雪覆盖率的响应比春季温度敏感。流域草地植被的长势随着返青提前和生长季的延长而变差,这可能是由于纳木错流域的积雪多为浅雪,其越早消融可能会导致植物过早的暴露在气温较低的环境下,造成霜冻损害。更多还原

【Abstract】 Tibetan Plateau (TP), owning the largest area and the most unique type of alpine grassland ecosystem, as well as nearly half lakes in China, is an important ecological barrier. Since the plateau is a sensitive region to global climate change, significant changes of grassland vegetation and lake in TP have taken place due to the climate warming in recent years. Therefore, studying the variations of vegetation activity and lake as well as their influence on climate change plays a vital role in understanding ecological environment and regional sustainable development. Based on multisource remote sensing data and grid interpolation data of mean monthly temperature and monthly accumulative precipitation from 1981 to 2010, we analyzed the temporal-spatial distribution and trend of grassland phenology from 1981 to 2010, the variation in lake water level from HY2003-2010, and the response of grassland phenology and lake water level to climate changes, respectively. In addition, taking Nam Co basin as a typical example and combining with data from field compaign, the linkages between climate change, lake expansion and grassland vegetation and the influence of snow cover on grassland vegetation growth were systematically examined. Relevant results are as follows:1) Grassland phenology has shown an obvious regional difference from southeast to northwest in TP. The grassland vegetation in eastern and northwestern part of plateau turned green earlier and brown late, with a relatively longer growth season than other regions. The changes of grassland vegetation phenology from 1981 to 2010 in the east and west regions were also remarkable in TP. The start of growth season (SOG) was in advance in eastern region. Grassland in western region had delayed greening and shortened length of growth season (LOG). The end of growth season (EOG) showed a delaying trend in the southwestern of TP and eastern of Qinghai province. The onset of withering period had an advanced trend in the part of western and eastern area.2) From the perspective of time, temperature and precipitation in spring has a certain promoting effect on grassland vegetation greening. In these regions, with an advanced EOG and shortened LOG, EOG and LOG were limited by summer temperature. However, in the area of delayed EOG and prolonging LOG, EOG and LOG were determined by temperature and precipitation. From spatialdistribution, the onset of grassland vegetation greening in eastern of TP, the onset of withering in eastern and western of TP and length of growth season in northern of TP were affected by temperature and precipitation. Nevertheless, in the other areas, greening and withering were mainly influenced by temperature. The length of growth season was influenced by precipitation much more than temperature.3) Most of lakes mainly distribute in the central and western region of TP, and their lake water levels are mainly between 4500 to 5000 meters, with a severely changes such as Selin Co, Nam Co and Kekexili. Lake water level of Tak kyel and Yamzhog Yumco in the southern of TP as well as Panggong in the western of TP showed a decreasing trend. Analysis of the relationship between lake water level and climate factor indicated that lake water level in different basin showed different trends because of varied temperature and precipitation. The increasing trend of lake volume was not only dependent on direct supply of precipitation, but also effected by melted water from glaciers and snow due to climate warming.4) In recent years, it is found that the increasing temperature was the main reason for both the vegetation deterioration and lake expansion. The enhanced vegetation index (EVI) in Nam Co basin negatively correlated well with the lake’s area, the correlation showed gradual decrease as distance away from the lake’s shoreline. Two main factors might have contributed to this:(a) lake expansion buried grassland vegetation in zone A, and (b) more gravel buildup and soil erosion due to runoff from snow melted water in zone A than in zones B and C.5) Snow cover fraction had positive relationship with start of growth season in Nam Co basin, but negative relationship with the length of growth season. Without considering the influences of snow cover on vegetation phenology, there was no significant relationship between the temperature and vegetation phenology, which illustrated that the influence of the vegetation phenology on mean snow cover fraction rate during April and May, was more sensitive than that of the spring temperature. The grassland vegetation growth condition in the basin has become worse with the growth season prolonging and advanced greening, because the advancement of shallow snow melted may lead to vegetation exposure under a relatively lower temperature environment, and caused frost-damage.更多还原