Gannan meadow lies in the south west of Gansu Province, on the northeast border region of Tibet Plateau. Gannan meadow is the core region of the important water source supply area of the Yellow River,which provides a natural defense for the upper reaches, playing a unique role in the water source and eco environmental security of the Yellow River Basin.Owing to the climate change and the artificial sabotage, the production performance and ecology function of Gannan meadow shows an obvious declining trend in recent years,the grassland degeneration and deserting trend are accelerated, and the phenological of herbage shows an obvious earlier trend. The ecology and environment of Gannan meadow has obviously changed. The Holdridge life zone system is a common used method to study the relation of the vegetative cover and climate, which has an obvious significance of physics and biology. It is widely used in the study of ecology environment, the dry and wet climate changes, the climate regionalization and so on.Most of the meadows in Gannan Plateau are distributed in Hezuo, Xiahe, Luqu and Maqu, so 4 places are selected as representative area to study the moving tendency of Holdridge life zone and dry and wet climate changes of Gannan meadow. Based on the meteorological data, the method of Holdridge life zone is used to analyze the moving tendency of Holdridge life zone and the dry and wet climate changes of Gannan meadow during 1971—2010. The results show that life zone of Gannan meadow still belongs to the subalpine meadow of Tibet Plateau alpine vegetation system. But impacted by the global climate change, the temperature of Gannan meadow increases obviously with the tendency ratio of 0.44—0.46℃ every decade, meanwhile the bio temperature has increased by 0.23—0.27℃ every decade, too. The variety of precipitation shows a different trend between the northern and southern in Gannan meadow, the moving tendency of Holdridge life zone is diverging from the mean center, and the stability of the Gannan meadow ecosystem is attenuating.The Holdridge REP of Gannan meadow shows an obvious ascending tendency in the rate of 0.02—0.03 per decade. After the 1990s, the climate of Gannan meadow shows a warm drying trend, which is more obvious in the southern of Gannan meadow. The humidity degree of Luqu has changed to humid from the perhumid, and Maqu is in the same transition, too. The most important climatic factor for the Holdridge REP increasing is the temperature, and then is the precipitation and the air humidity.The temperature rising plays a very important role in the warm drying trend of Gannan meadow.Considering the complexity of the environment and the fragility of the ecology system in the border area of Tibet Plateau, the impact of this change on ecology and environment of Gannan meadow still needs further study.

利用Holdridge生命地带系统对1971—2010年甘南草原的Holdridge生命地带偏移趋势及干湿变化进行分析，发现甘南草原目前仍属于青藏高原高寒植被地区的亚高山高寒草甸生命地带，但由于甘南草原生物温度明显升高，甘南草原南部和北部降水量呈现不同的变化趋势，位于青藏高原边坡地带的甘南草原的Holdridge生命地带距平均中心的偏移趋势逐年增大，甘南草原生态系统的稳定性在减弱；甘南草原潜在蒸散率以0.02/10 a～0.03/10 a趋势上升，其中以玛曲上升最明显，达0.03/10 a； 20世纪90年代后，甘南草原呈明显的暖干化趋势，其中以位于南部的碌曲、玛曲变化最为明显，碌曲已由极湿润区转变为湿润区；玛曲有从极湿润区向湿润区过渡的趋势。影响甘南草原潜在蒸散率上升的主要气候因子是温度，其次为降水和空气湿度，温度上升是甘南草原暖干化的主要原因。

摘　要： 应用甘南高原8个气象观测站的月平均气温、月降水量、月蒸发量及水文资料,分析甘南高原气候 历史演变趋势及其对水资源的影响。结果表明：1973～2010年研究区年降水量呈减少趋势,平均每10 a减少11.9 mm,尤以20世纪90年代减少最为明显。年平均气温呈上升趋势,气温每10 a升高0.46℃。年蒸发量总体呈增加趋势,平均每10 a增加21.3 mm,尤以近10 a增加最为明显。年降水量为450 mm的等值线南移,年蒸发量为1 250 mm的等值线和年平均气温为3℃等值线西伸,暖干化面积增大。研究区暖干化导致水资源减少,湖泊水位下降,干旱草地面积扩大,草地涵养水源功能降低,产流 量减少,黄河＂蓄水池＂的作用明显下降。

以甘南牧区2001-2009年的土地利用变更数据为基础,采用 数量分析的方法对全区土地利用结构的时空变化规律进行研究.结果表明:全区主要用地类型为牧草地和林地,分别占总面积的67.30％和24.68％.牧草 地、未利用地、交通运输用地、水域在各县(市)的分布比较均匀,而园地、水利设施用地、其他农用地、耕地在各县(市)的分布不均匀.全区耕地、其他农用地 呈减少趋势,牧草地和林地呈增加趋势；耕地强度指数的绝对值最大,水利设施用地动态度的绝对值最大,分别为0.497 6％和45.570 5％.耕地、牧草地、林地变化贡献率最为显著,分别为0.481 2,0.323 9,0.143 1.空间上临潭县、合作市、迭部县对全区土地利用结构变化的贡献较大.研究表明,退耕还林还草政策在甘南牧区得到了较好落实,取得了显著成效.随着投资力 度加大,甘南牧区的水利设施用地、交通运输用地、居民点及工矿用地面积增加较大.

Alpine pasture is a unique plant community with an ecological set of conditions and environmental pressures which are significantly different from other areas. This simple ecosystem has a naturally slow rhythm. Its functional instability is caused by the cold, alpine pasture environment which makes it vulnerable to disturbance and difficult to repair once damages occurred. To a certain extent, external forces and influences are limited in this cold climate because of its isolation from human activities. This leads to alpine pasture having a highly localized and sparse human population, so that human conflicts only tend to arise over grass-livestock conflicts. Based on characteristics of alpine pastures, the authors have revised the ecological footprint model and ecological carrying capacity model of this unique habitat placing strong emphasis on the key factors controlling ecological carrying capacity. These factors include the numbers and types of livestock, human population, local habitat conditions and regional characteristics. Following the methods of Hezuo, the authors then measured and analyzed the ecological footprint and ecological carrying capacity of 41 administrative units. The results show: 1) The distribution patterns and ecological footprints of the human population and their livestock are very different and roughly the opposite of each other, which indicates there are significant spatial differences between the environmental pressure on alpine pastures coming from the human population and pressures from livestock. 2) The combined ecological footprint of humans and livestock is mainly determined by the location of the livestock ecological footprint, so we can conclude the impact of livestock is the main factor in the formation of the combined ecological footprint. 3) Looking at the distribution of the ecological carrying capacity per unit area, the areas with a low carrying capacity are mainly located in the northern expansive pastoral areas at higher altitudes where the structure of the ecosystem is fairly simple, and main ecological types are alpine meadows and hydric grassland. Also, in these high altitude areas, frost weathering is quite strong, and the effective growing season is very short. Unlike these low carrying capacity and high altitude areas, the areas with high carrying capacity are mainly located around urban areas and along the Taohe River, with both grassland ecosystems and forested ecosystems present.In summary, we can conclude the main factor determining the ecological carrying capacity is our ability to improve anthropogenic factors, while dealing with the complex structure of natural ecological systems and harsh natural environment. 4) When we consider the nature of the spatial distribution of ecological surpluses and deficits, ecological deficit areas are mainly concentrated in pure pastoral areas, and overgrazing is the key cause for the ecological damage. So, the key to increasing the ecological carrying capacity and reducing ecological impact is focusing our attention on the relationships between grasses and livestock. We should take several measures to improve the situation. First, maintaining the balance between forage and livestock and avoiding overgrazing is required. Stocking levels need to be based on the carrying capacity of native pasture. Second, we can improve productivity and carrying capacity of pastoral areas using scientific and reasonable methods, and in this way strengthen protection for grassland ecosystems. Finally, we need to adjust the economic structure, by decreasing human dependence on grassland farming and by promoting the transition from traditional grassland husbandry into modern animal husbandry.

高寒牧区是一类特殊的地域,其生态环境本底和压力明显区别于其它地域类型。从自然本底上讲,"高寒"特征造成生态系统结构单一、节律缓慢、功能稳定性差,因此,脆弱性强,而修复难度大;从压力上讲,"高寒"又一定程度上屏蔽了外界要素的过分介入,导致人口分布的高度离散性和稀疏性,人地矛盾转化为更直观的草畜矛盾。论文结合高寒牧区的特质,对生态足迹模型和生态承载力模型进行了修正,强调了高寒牧区生态承载力的主控因素——牲畜特征、人口特征、生境特征和地域特征,以甘南州合作市为案例,具体测算和分析了41个行政单元的生态足迹和生态承载力。结果表明:(1)人口生态足迹和牲畜生态足迹分布格局大致呈反向态势,说明在高寒牧区,人口与牲畜对环境造成的压力存在着显著的空间差异。(2)从人口生态足迹和牲畜生态足迹对总生态足迹的贡献度上看,总生态足迹的高低主要取决于牲畜生态足迹,说明牲畜是主控因素。(3)从地均生态承载力分布态势看,低承载区主要分布于北部海拔较高的广大牧区,这里生态结构非常单一,主要以高寒草甸草原和沼泽草原生态类型为主,寒冻风化作用异常强烈,生物有效生长期非常短。而高承载区分布在市区周围和南部洮河谷地沿岸,生态系统除草地生态系统外,还有森林生态系统。从而说明人为条件的改善、自然生态系统结构的复杂程度和自然环境的严酷程度是决定高寒牧区承载力高低的主要因素。(4)从生态盈余与赤字的空间分布态势看,赤字区域主要集中分布在广大的纯牧区,超载过牧是生态状况的关键性致因。因此提高生态承载力、减少生态赤字的关键集结于草畜关系上。有鉴于此,需要采取以下措施:一是根据天然草场的承载能力,建立适宜的草畜平衡关系,将超载的牲畜转移出去;二是科学合理的提高草场的生产能力和承载能力,这就需要加强草原生态保护建设;三是调整经济结构,减少社会经济发展对天然放牧业的高度依赖性,推进传统草地畜牧业向现代畜牧业转型。

In order to approach improvement ways of deteriorated grassland on Qinghai-Tibet Plateau, in this paper, the quadrant survey method was used to study the dominant species and photosynthetic characteristics of alpine meadow in Maqu county, Gansu province. The feature index and vegetation cover degree were also analyzed by mean of the grey correlative degree. Results showed that Stipa aliena, Bupleurum chinense, Saussurea nigrescens, Poacrymophila keng and Anemone revelries were main dominant species in this area. The grey correlative degree between the population feature and vegetation cover degree: aboveground biomass>height>density. The grey correlative degree between photosynthetic character index and vegetation cover degree: intercellular CO2 concentration> transpiration rate> net photosynthetic rate> stomatal conductance. The formation of dominant populations depends on the capability to exchange gases between individual plant and outside. To improve the alpine meadow productivity, we should cultivate dominant species basing on the maintenance of biodiversity.

为进一步掌握、了解优势种群，更好的治理我国青藏高原草地退化提供理论依据。采用样方调查法，在甘肃省玛曲县研究了高寒天然草地优势种群及其光合特征，并对特征指标与覆盖度进行了灰色关联度分析。结果表明：异针茅、柴胡、瑞玲草、冷地早熟禾和草玉莓是玛曲高寒草地的主要优势种群。种群特征与覆盖度的灰色关联度依次为：地上生物量>高度>密度。种群光合特性指标与覆盖度的灰色关联度依次为：胞间CO2浓度>蒸腾速率>净光合速率>气孔导度。玛曲高寒草地优势种群的形成主要决定于植株个体与外界气体的交换能力。提高玛曲高寒草地的生产能力，应在维护生物多样性的基础上，加强对优势种群的培育。

The ecological effects of Restoring Grazing to Grassland Project in Maqu County are studied with MODIS data and field investigation data. The results showed that: ①The coverage, the height and the biomass in the project region increased 6.8%, 69.2% and 30.2% in 2005, parallelly 5.6%, 61.5% and 30.6% in 2006, respectively. ②The mean NDVI index in prohibition-grazing and rest-grazing regions increased 3.4% and 3.0% in 2005, and 2.6% and 0.9% in 2006, respectively. ③The NDVI index in enclosure increased 5%~50% in 2005 and 2006 more than 2004, obviously in 2005. In a word, the effects of Restoring Grazing to Grassland Project in Maqu County are obvious, especially in prohibition-grazing region.

为了了解退牧还草工程对玛曲草地生态恢复的效果,利用MODIS卫星遥感资料、地面草地调查资料、退牧还草工程围栏资料和乡行政边界,分别从地面草地监测、不同退牧还草措施的遥感监测对比、玛曲各乡退牧还草效果的遥感监测以及玛曲植被空间变化等方面,对退牧还草工程对玛曲草地恢复的效果进行了研究。结果表明:①退牧还草后的2005年和2006年,工程区内植被盖度较工程区外提高6.8％和5.6％,高度提高69.2％和61.5％,生物量提高30.2％和30.6％。②2005和2006年禁牧区的平均NDVI分别较退牧前提高了3.4%和2.6%,休牧区提高了3.0%和0.9%。③植被的空间分布显示,退牧后2005年和2006年大多数退牧还草围栏内植被指数较2004年高,围栏内NDVI较2004年增加5%~50%,以2005年增加最为显著。④退牧还草工程对玛曲草地生态环境恢复效果均是明显的,其中禁牧的效果较退牧更为显著。

以像元二分法模型为基础,构建了基于改进归-化植被指数 (NDVI)的植被覆盖度定量估算模型,并利用MODIS卫星遥感数据,基于3S技术(GIS,RS,GPS)空间分析功能,分级计算得到了甘南 2002,2004,2006,2008年的草地植被覆盖度,分析了甘南2002- 2008年植被覆盖度变化的空间动态演变过程和趋势.结果表明,2002-2008年,甘南植被覆盖度质量总体呈下降趋势,一级盖度植被的退化比较严重, 草地植被的演变情况主要由优等植被覆盖(一、二级)向低等植被覆盖(四、五级)演变.研究结果揭示了甘南州草地退化状况日趋严重的事实,为相关研究和政府 草地管理提供了有效依据.

In order to predict the actual carrying capacity and theoretical carrying capacity, and to provide the theoretical basis for protecting the ecological environment of grassland and the sustainable development of grassland-animal husbandry, the paper analyzed the real carrying capacity and theoretical carrying capacity of grassland in 4 countries of Gannan in China from 1998 to 2009, based on grey theory and the Markov process, and built a optimized and improved Grey-Markov forecasting model for grassland carrying capacity. The model is equivalent dimensions additional grey model and Markov residual forecasting model. The model was tested by actual stocking capacity in Hezuo country. The prediction by the model had no significant difference with reality (p<0.01, error less than 7.06%). This showed the availability of the model for predicting the carrying capacity of grassland. Furthermore, actual carrying capacity and theoretical carrying capacity in each county of Gannan from 2010 to 2014 were predicted. This study can provide a reference for the relevant departments to determine the reasonable carrying capacities and the grassland management.

为了预测草地实际载畜量与理论载畜量，为草地畜牧业的可持续发展提供理论依据和决策支持基础，通过对甘南州4个县市1998－2009年的草地实际载畜量和理论载畜量数据分析，以灰色GM(1,1)模型和马尔科夫预测模型为基础，建立了优化改进的灰色等维-马尔科夫残差预测模型，以甘南州合作市实际载畜量为实例进行验证，预测结果显著(p<0.01)，预测误差低于7.06%，表明该预测模型精度较高，适用于对草地载畜量的预测研究。最后，利用该模型对2010－2014年甘南州4县市草地实际载畜量和理论载畜量进行预测分析，基本符合草地载畜量实际变化过程，为相关部门合理确定草地载畜量和进行科学草地管理提供参考。

Maqu grassland, belong to Maqu County, Annan Prefecture, Gansu Province, China and located at eastern fringe of Qinhai-Tibet Plateau, is important to the water supply area of Yellow River, but grassland desertification in the region had an vicious effect on this function. This paper aims at the problem of grassland degradation in Maqu pasture, quantitatively analyses reasons for the grassland degradation from natural and artificial factors by applying Cobb-Douglass production function, concludes that the artificial factors are the main reason to cause grassland degradation. The paper further analyzes the artificial factors from three aspects of the raw material model of animal husbandry, operations management system and implementation mechanism, and considers that the institution is the root reason of grassland degradation by applying the method of Chow breakpoint test. The relationship between the institution of grassland property right and overgazing is consanguineous, and the design of the institution of property right is faultiness and institution provided is lagging, which makes utilization and protection of grassland outoforder, tempts herd to overmanage grassland, and leads grassland to be overgrazed over long time. For the existent problems,the paper explores sustainable model of production, operation and management by environmental opportunity, and proposes the sustainable development models of ecocultural tourism, partnerships system and government management and religious propaganda combined. By optimizing development model to realize virtuous circle between the economic development and grassland ecology，it provides references for other plateau pastoral areas of animal husbandry economic development.

针对玛曲草原草地退化问题，运用Cobb-Douglas生产函数和Chow分割点检验定量分析原料型畜牧业和制度对草地沙化的影响，并得出制度是导致草地沙化的根本原因；针对存在的问题，运用环境机会论探索可持续的生产、经营和管理模式。通过优化发展模式，实现经济发展与草原生态的良性循环，为其他高原牧区畜牧业经济发展提供借鉴意义。

In order to guide the production and use local naturally grassland reasonably, an assessment was conducted on grassland primary production and livestock carrying capacity in Maqu County, Gansu. By used the materials from field investigation on Maqu County of Gansu Province and MODIS images of remote sensing, the carried capacities on theory and seasonal ranges were compared and analyzed based on calculated grassland yields and grazing capacities and grassland ranking was conducted in this paper. The general results of study indicated, the main reasons that led to change on grassland productions of Maqu County were excessive grazing livestock and natural climate conditions such as increasing temperature, decreasing annual precipitation which brought about inadequate rainfall on arid grassland. The reverse tend of change on carrying capacities caused the reduction of grassland productions, and natural climate conditions led to the desertion of grassland in some extents. Consequently, the carrying capacities must be strictly controlled and the practically grazing capacities should be set based on the new theoretical and critical carrying capacity in order to turn back the reverse tend of grazing capacity and eliminate the reduction of grassland productions.

为了对玛曲县草地初级生产力及载畜能力进行评价，以指导生产，更加合理利用当地天然草地资源。结合运用甘肃省玛曲县野外调查和MODIS遥感影像资料，计算玛曲县及下属各乡的产草量和载畜量，对比分析了理论载畜量和季节牧场载畜量，并进行草地分级。结果表明：导致玛曲县草地生产力变化的主要原因是超载过牧和自然环境如气温升高，年降水量减少，草地日趋干旱缺水等。载畜量变化的逆向走势使草地生产力下降，而自然因素能在一定程度上引起草地荒漠化。因此，必须严格控制载畜量，并以最新的理论载畜量和临界载畜量来界定实际载畜量，以扭转载畜量的逆向走势，遏制草地生产力的下降趋势。

基于遥感和ＧＩＳ技术，选取１９７５、１９９０、１９９４、２０００、２００５和２００９年６期遥感影像，综合该区地形图、各期沙化监测资料以及土地利用图等资料，采用遥感动态监测、数据解译和野外调查相结合的研究方法，对甘肃省玛曲县近３５ａ来高寒草甸沙化的时空演变过程及其发展趋势进行了解析，并在此基础上，运用灰色关联度分析法，对其沙化驱动因素进行定量辨识和贡献率分析。结果表明，１）近３５ａ来研究区沙化在时间上表现为沙化土地面积处于持续扩张的态势，其扩展经历了缓慢增加－略有减少－快速增加的过程。其中流动沙地（丘）呈现缓慢上升－急速上升－逆转的变化趋势；半固定沙地（丘）呈现缓慢上升－ 急速上升－ 急速逆转的变化趋势；固定沙地（丘）呈现缓慢上升－逆转－急速上升的变化趋势。在空间上表现为沙化土地的分布区域呈沿黄河沿岸不断扩展态势。在沙化程度上表现为波动减弱趋势。２）各驱动因子与沙化土地总面积的关联排序为：国民生产总值＞羊只存栏数＞人口数量＞年均气温＞畜牧业总产值＞年均降水量＞大牲畜存栏数＞ 年均蒸发量＞ 牛羊肉总产量＞ 干燥度指数＞ 黄河径流量＞年均风速；主要自然因素和人为因素对沙化土地总面积的驱动力比重分别为４０．４３％ 和５９．５７％。说明气候变暖的异常扰动导致草场沙化，为诱导和次要因素，人类不合理的活动加速了土地沙化的发展，为强化和主导 因素。研究区草地沙化的扩展是２类因子共同作用和影响的综合过程。

黄河源头区是我国重要的水源涵养地区,是"中华水塔"的重要组成部分.地处黄河源区的玛曲县,自20世纪80年代以来,天然草地沙化面积不断增加,沙化程度日趋严重.本研究以玛曲县为例,对引起草地沙化的主要自然因素和人为因素与草地沙化的相关性及驱动作用进行定量辩识和贡献率分析,结果表明,1)各因子对草地沙化的驱动力大小依次为:牧业产值＞人口数量＞年均气温＞羊只存栏数＞牛羊肉产量＞年均降水量＞大牲畜存栏数＞黄河年径流量＞年大风日数＞年均风速;2)区域年均气温的不断升高,使地面水分蒸发损失和土壤干旱程度加剧,这种变化对土地沙化产生的驱动作用要大于降水量的小幅减少和风力因子变化的影响,是引起该区土地沙化最重要的自然驱动因子;3)对2类因子的相对贡献率进行综合,得到人为因子的驱动力比重为61.55%,自然因子的驱动力比重为38.45%.说明在高寒阴湿的气候条件下,区域气候和水文因子在一定程度上的不良变化只是天然草地沙化的一种诱导力,而人为过度和不合理利用是导致草原植被衰退的根本原因,二者结合导致了黄河源头区草地沙化的加速发展.

20世纪90年代以来,甘南藏族自治州境的生态恶化引起了社会各界广泛的关注.近年来,州政府和人民,对其生存环境恶化的状况逐步有了清醒的认识.采取了一系列保护和建设并重的措施,努力遏制生态环境的继续恶化,取得明显实效.建议社会、国家进一步加强政策调控,运用科技手段,采取措施,优化生态环境.

Based on the 1992～1996 data of 1km monthly NDVI and those of the monthly precipitation and mean temperature collected by 400 standard meteorological stations in China,this paper analyzed the temporal and spatial dynamic changes of the correlation between NDVI and climatic factors in different climate districts of this country.The results showed that there was a significant correlation between monthly precipitations and NDVI.The wave-type time series model could simulate well the temporal dynamic changes of the correlation between NDVI and climatic factors,and the simulated results of the correlation between NDVI and precipitation was better than that between NDVI and temperature.The correlation coefficients (R 2 ) were 0.91 and 0.86,respectively for the whole country.

利用中国大陆1km分辨率的月NDVI数据和相应的400个国家标准气象站点的月降水、月均温数据,依据中国气象局提出的气候分区,探讨了不同气候区NDVI与气候因子之间的时、空变化格局.结果表明,各月份降水与NDVI相关性均显著相关;起伏型时间序列模型能很好地拟合相关性的时间序列的动态变化,NDVI与降水的相关性拟合结果好于与温度相关性的拟合结果,全国范围内,与降水相关性拟合的 R 2 为0.91,与温度相关性拟合的 R 2 为0.86.

Degradation of grassland is a complex and gradual process. In order to make it more capable for monitoring the changes in undergoing-degradation and not-being-completely-deserted grassland with remote sensing data, one of available ways is to identify spectral characteristics in response to the differences in grassland coverage by machine discerning. However, it is challenging since the differences are subtle in a TM / ETM image or other images with similar spatial and spectral resolution. This paper discusses the methods for a learning machine to discern the subtle differences from TM / ETM spectral data. The key technologies can be summarized as: replacing the grass cover ( D ) with the so called "weighted grass cover ( Dw )" to solve the difficulties in the fitting between Normalized Difference Vegetation Index (NDVI) from TM/ETM spectral data and field measured D , designing some mathematical descriptors which are independent of image types and acquisition environment and discerning the indicative differences in response to Dw in a multi-descriptor space so that the error rate in the discerning will be surely reduced. Dw is a weighted sum of the D , the grass height ( H ) and the proportion of the poisonous in all grass ( Dup ). With the use of Dw , The regression equations between Dw , calculated by D , H and Dup measured in the field, and the normalized vegetation indexes, such as NDVI, SAVI, etc., calculated with remote sensing spectral data will fit much better than that with the replacement of D for Dw . In order to improve the sensitivity to the differences in grass coverage by the learning machine, eight new mathematical descriptors, as the supplements of those vegetation indexes, have been proposed and then analyzed one by one in their correlation to Dw or tested in their validity for Dw segmentation. It is verified that some of the combinations of these descriptors are helpful for the learning machine to work out the laws of identifying pixels or cells belonging to different Dw . Otherwise, by comparing the classification accuracy in the use of two kinds of learning machines, i.e. DT and SVM, the types of learning machines are not so crucially to be able to decide discrimination accuracy as mostly desirable and, on the contrary, the construction of search space may be a more decisive factor. Our study shows that in the multi-descriptor spaces, the grasslands belonging to different Dw can reliably be classified with TM / ETM data. Being inspected by field sampling data, it is verified that the classification accuracy is nearer to or better than 80%. Based on the results in Dw classification, then the Dw changes can also be reliably deduced after adjusting the differences of luminance level between two images in different acquisition years.

以加权草被盖度作为草被退化的一种标识值,来标记草场沙化和毒杂草侵蚀的程度,并以机器辨识方法从TM/ETM数据中定量提取加权草被盖度现状及变迁信息。变迁研究的时段从1994年至2008年。为了使图像辨识特征量对高寒草被盖度敏感,提出了植被指数分级密度、植被相对饱和度分级密度等8个新的数学描述符,逐一进行了与加权草被盖度的相关性分析或分割实验;并通过它们的组合训练学习机和实现了对不同盖度草被的划分。通过野外采样数据检核,这种分类的准确率接近或达到80%。在草被盖度正确分类的基础上,通过调整减少不同时相图像照度差异的影响,进一步实现了加权草被盖度变迁信息的自动化提取。

摘　要： 利用2000—2010年16d合成的MODIS／NDVI数据，结合植被异常指数、趋势线分析和Hurst指数等分析方法，对新疆伊犁河谷植被覆盖的时空变化特征进行了分析。结果表明：（1）伊犁河谷内植被覆盖随海拔增高而先增加后减小，最高植被覆盖区位于2000～2500m的高程带；2000—2010年伊犁河谷各高程带内植被覆盖整体下降趋势明显，但海拔低于1000m的高程带除外。（2）受干湿环境影响，伊犁河谷全区平均被植异常指数最高值出现在降水最多的2002年，最低值出现在降水最少的2008年，但不同区域植被异常指数的变化存在较大差异。（3）伊犁河谷内植被覆盖增加和减小的区域分别占总面积的4．09％和19．34％，增加区域主要位于伊犁河两岸的平原区，减小区域主要位于乌孙山两端以及伊犁河谷周围海拔2000m左右的低山区域；变标度极差分析结果表明，伊犁河谷内植被覆盖年际变化呈现很强的持续性，未来一定时间内将保持现有变化趋势不变。

The annual and seasonal change trends of vegetation coverage and their response to climate change in Xilin Gol Grasslands during 2000-2010 were examined by employing MODIS NDVI datasets in growing season from April to October and corresponding climatic variables including monthly temperature and precipitation. The results indicated that the growing season NDVI, on the whole, tended to increase from 2000 to 2010. The correlation coefficient between NDVI and precipitation was 0.89, indicating that the precipitation was a primary factor of vegetation dynamics. For seasonal changes, the NDVI experienced the increasing trend in spring and decreasing trend in summer and autumn. Correlation analyses showed that the vegetation variations were mostly controlled by precipitation in spring and summer, while it was regulated by temperature in autumn. The pixel-based analyses displayed that the regions characterized by increases were distributed in Abbaga county, Xilinhaote city, Taipusiqi and Duolun County, and the regions characterized by decline were mostly located at east Ujimqin county, west Ujimqin county in recent 10 years. The vegetation dynamics and their response characteristic to temperature and precipitation differed from different grassland types.

利用2000-2010年间生长季(4-10月)MODIS NDVI数据和同期气象观测数据,分析锡林郭勒草原植被覆盖度年际和季节变化趋势及其对气温和降水的响应特征.研究表明:生长季平均NDVI总体上呈增加趋势,其变化趋势主要受降水量的控制,年总降水量与NDVI的相关系数达到0.89.从不同季节植被NDVI的变化趋势看,春季NDVI呈上升趋势,夏季和秋季呈下降趋势.与气候因子的相关分析表明,春季和夏季植被覆盖变化趋势主要受降水的控制,而秋季温度的变化对植被生长的促进作用也较明显.在空间分布趋势上,阿巴嘎旗、锡林浩特市、太仆寺旗和多伦县植被覆盖总体上增加趋势较明显,而东乌珠穆沁旗东部、西乌珠穆沁旗的下降趋势最为明显,其气候影响因地区和季节而异.不同草原类型的植被具有不同的变化趋势和受温度和降水量的影响各不相同.

基于MODIS—NDVI遥感 数据,利用像元二分模型估算河东地区2000—2010年的植被覆盖度,并在像元尺度上分析河东地区植被覆盖度的时空变化规律及其驱动因子。结果表明:河 东地区平均植被覆盖度为50.93%,其中,甘南高原覆盖度最高,河东北部最低,植被覆盖度由南向北递减;研究时段内年平均植被覆盖度大于60%(优等) 的区域占23.06%,且90.49%的区域植被覆盖度呈增加趋势,以河东中部植被覆盖度增加速度最快,为33.68%。从空间尺度上来看,降水是其主要 的影响因素;而从时间尺度上,最主要的影响因素可能是植树造林。

根据1982-2006年全国GIMMS NDVI资料,以及全国583个气象站点的气温和降水量数据,以像元为计算单元,分别以植被分区和气象站点为对象分析了植被NDVI、≥10℃积温和降水 量的年际变化及相关关系.结果表明:(1)全国及各分区≥10℃积温均呈显著的增加趋势,降水量和植被NDVI整体上呈减小趋势,植被覆盖面积略有增加, 且≥10 ℃积温增速越大则降水量减速越小.(2)全国植被NDVI与≥10℃积温和降水量分别呈微弱的负相关和正相关.由于地形地貌、地理位置以及人类活动作用的 差异性,不同植被分区相关性有所差异.(3)秦岭-淮河以北地区,降水量是影响植被生在的限制性气候因子.而秦岭-淮河以南地区,植被NDVI与≥10℃ 积温和降水量的分布无显著性规律.

The interaction between vegetation and the atmosphere is important in fields of geography, ecology and so forth, and has become a research focus in recent years. Normalized Difference Vegetation Index (NDVI) is generally taken as a good indicator of terrestrial vegetation productivity. Understanding influences of climatic factors, in particular precipitation and temperature, on NDVI would greatly benefit prediction of changes in terrestrial vegetation productivity under different climatic scenarios. In the present work, variations in vegetation NDVI and its correlation with temperature and precipitation across different climatic zones and land use types in eastern China were examined in detail based on the SPOT VGT-NDVI data, daily temperature and precipitation observations from 205 meteorological stations, maps of climatic zones and land use types. Results indicated that during the period 1999 to 2007, annual and seasonal vegetation NDVI in five climatic zones and four main land use types in eastern China generally showed an increasing trend, with the most increase in vegetation NDVI over the north sub-tropical humid zone and dry field. In summer, vegetation NDVI increased most, but in winter, increases in vegetation NDVI were not obvious statistically. It was found that the maximum correlation coefficient between vegetation NDVI and temperature and precipitation decreased from the northern to southern part of eastern China, with showing stronger correlations between vegetation NDVI and temperature, vegetation NDVI and precipitation in the mid-temperature semi-arid zone, and weaker correlations between vegetation NDVI and temperature, vegetation NDVI and precipitation in the south sub-tropical humid zone. It was also found that in the mid-temperature semi-arid zone, warm temperate semi-humid zone and south sub-tropical humid zone, the correlation coefficients between vegetation NDVI and temperature were greater than that between vegetation NDVI and precipitation for the same period. With respect to the correlations between NDVI and climatic factors for different land cover types, it was suggested that the correlations between vegetation NDVI and temperature, vegetation NDVI and precipitation were strongest in grassland, but weaker in woodland in eastern China. The correlations between paddy field NDVI and temperature, and dry field NDVI and precipitation also seemed stronger. The correlation coefficient between dry field NDVI and temperature in previous 0-90 days was greater than that between dry field NDVI and precipitation for the same period. The variations in vegetation NDVI and its correlation with temperature and precipitation in different climatic zones and land use types in eastern China were basically consistent with other findings in China.

基于SPOT VGT-NDVI数据、气象站点气温和降水量资料、气候区划图及土地利用类型图，分析了中国东部不同气候区和主要土地利用类型植被NDVI的变化趋势及其与同期和前期气温和降水变化的相关性。结果表明，1998年-2007年期间，中国东部5个气候区和4种主要土地利用类型植被NDVI在全年及四季基本上都增加，其中以夏季增加最多，冬季NDVI增加不显著。植被NDVI与气温和降水的最大相关系数在中国东部由北向南逐渐减小，以中温带亚干旱区相关系数最大，南亚热带湿润区最小。草地NDVI与气温和降水变化的相关系数最大，林地NDVI与气温和降水变化的相关系数较小。中国东部不同气候区和土地利用类型植被NDVI的变化及其与气温和降水的相关性与已有研究结果基本一致。

Grassland is the main vegetation type of terrestrial ecosystems, which is an important ecological protective barrier in China. The physical environment (water environment, soil environment, atmospheric environment and other organisms), structure (population, community, landscape) and function (production function, ecological function) of grassland ecosystem are changing under the circumstances of climatic changes. The present paper described the evaluation indices, methods and effects of climatic changes in details, which could provide reference for related researches.

草原是陆地生态系统的主要类型之一,是我国重要的生态屏障。在气候变化的背景下,草原生态系统的物质环境(水环境、土壤环境、大气环境和其他生物)、结构(种群、群落和景观)、功能(生产功能、生态功能)都发生着变化,在总结相关文献的基础上,本文从这3个方面详述气候变化对草原生态系统影响的主要评估指标、方法及其特点,为评估气候变化对草原的影响提供借鉴和依据。

草原资源是大自然的主体部分之一,是人类赖以生存和发展的物质基 础,是维护生态平衡的重要环节.甘南草原是长江、黄河上游重要的水源补给和涵养区,在中国西部生态环境系统中处于十分重要的位置.针对甘南草原面临的严重 草原退化、草原沙化、湿地萎缩、水土流失、生物数量锐减、草场退化、河流流量下降等问题.提出了相应的保护与防治时策.

基于1982—2006年的 AVHRR数据和气象数据,分析了蒙古高原植被时空变化特征及其影响因素。研究发现:(1)近25a来,蒙古高原植被整体呈改善趋势,年增长率为 0.000 6/a。蒙古国植被NDVI年增长率最大(0.000 8/a),其次是中国的内蒙古(0.000 4/a)、甘肃(0.000 3/a)和宁夏(0.000 1/a);(2)植被的变化趋势具有明显的空间差异性和季节性。25a来,蒙古国西北部、内蒙古中部的锡林郭勒盟等区域植被明显改善,而内蒙古东北部呼伦 贝尔市植被呈退化趋势。研究区春、秋季植被明显改善,夏季植被呈退化趋势;(3)植被的变化趋势一定程度上受温度和降雨的影响,春、秋季升高的温度使生长 期延长,植被改善,而夏季温度升高和降水减少共同导致夏季植被退化;(4)草地、农作物面积和产量的增加直接或间接地导致了内蒙古植被NDVI的增加,其 他人为因素,如人口增长、土地利用类型的转换、退耕还林、过度放牧、矿产资源开采等也可直接或间接地影响植被的变化。

Based on the theoretical frame and evaluation index system on assessment of comprehensive benefit of grassland construction engineering project, a case study was conduced in Gannan Tibet Municipality, including Maqu County, Luqu County, Xiahe County and Hezuo City. Gannan Tibet Municipality is one of districts that the recovery and construction projects of Gansu natural grassland were implemented. In this study, 51 assessment indexes were selected, according to the actual conditions of Gannan. Besides, a hierarchy structure and comprehensive assessment model was designed. The total weight of every index was calculated based on the analytic hierarchy process (AHP). The results showed that, the ecological, economic and social benefits were significantly increased with the conducing of projects, and which were increased by 6.135%, 5.433% and 14.042%, respectively. Compared with the corresponding values, which were before the implementation of recovery and construction projects, the landscape benefit was maintained, the starts quo and the environmental benefit decreased from 0.105 to 0.103.

运用草原建设工程项目效益评价理论和指标体系, 以甘肃省甘南藏族自治州天然草原恢复与建设项目为例, 利用多指标综合评价模型对草原建设工程项目的综合效益进行了实证研究。结果表明, 实施天然草原恢复与建设项目后, 项目区的生态效益、经济效益和社会效益都得到了不同程度的提高, 分别提高了6.135%、5.433%和14.042%。项目区的景观效益维持在0.110水平不变, 而环境效益则从0.105下降到0.103, 下降了1.905%。

甘肃甘南天然草地气候生产潜力主要受水分因素和返青期温度因素的 制约，夏河县气候生产潜力受降水影响大，而玛曲和碌曲受返青期温度影响较大。甘南天然草地现实生产力占草地生产潜力的61.5%，还具有一定开发空间。经 过计算分析得出，玛曲县最大载畜量为246×104羊单位，适宜为168×104羊单位；碌曲县最大载畜量为102×104羊单位，适宜为72×104羊 单位；夏河县最大载畜量为208×104羊单位，适宜为121×104羊单位；全州估算最大载畜量为668×104羊单位，适宜为424×104羊单位。 甘南州平均超载率为74.3%，玛曲县为72.0%，碌曲县为71.5%，夏河县最低，为41.3%。甘南天然草地均处在超载状况，一般认为70%以上的 超载率就已经造成草地的过度利用，人为因素在草地退化中占主导地位。甘南高寒草地的开发潜力较大，但因实际载畜过多，超载严重，导致草地退化，成为限制草 地质量提高的主要因素，降低载畜量已成为提高甘南高寒牧业的当务之急。

The effect of restoring grazing to grasslands areas in Maqu County was evaluated in this paper. The 16 days 250m MODIS, annual NDVI and month NDVI data and synchronous meteorological data were used to build some regression models. Using these models we can forecast and evaluate the effect if without the project. Moreover we can use the models to analyze the residual difference change trend of actual NDVI and simulative NDVI from 2000 to 2006, and then to deduce the different project types effect on the grasslands restoring. The result showed that restoring grazing to grasslands was benefit for grassland growing and restoring, especially in the earlier and latter growing seasons. Forbidden grazing measure was the most effective. For the grassland restoring condition, all the grassland areas were in the recovering process, but the effects were various in different areas, and the best condition in Manrima village, the worst in Oula village. As a whole, the project was executed well in most enclosures and the restoring conditions were well, among these enclosures, the 16th enclosure was the best condition. However, some enclosures conditions were bad, such as the 5st enclosure, which can be concluded that there were some absurd grazing behaviors and the project was not well executed in those areas.

利用残差趋势法对玛曲县草地退牧还草工程（禁牧、休牧）效果进行了评价，即利用2000~2006年250m的MODIS16天合成NDVI产品数据及同期气象资料，建立工程区内退牧前2000~2004年两种时间尺度（年、月），3种空间尺度（县、乡镇、围栏）的NDVI值同气象要素的回归模型，并利用模型预测无该工程影响下的模拟NDVI值，分析退牧前后模拟NDVI值与实际NDVI值的残差及变化趋势，从而判断退牧还草工程对草地恢复状况的影响，及分析不同工程措施对草地恢复的作用。结果表明，退牧还草工程有利于玛曲县草地牧草生长和草地恢复，但草地恢复效果存在时空差异，退牧还草工程有助于牧草整个生育期的生长，尤其是牧草生长初期和末期效果最为显著。禁牧措施较休牧措施对于草地恢复效果更显著。整个工程区内草地总体处于恢复中，但恢复状况各有差异，一些区域草地恢复状况不理想，说明这些区域还存在较为严重的放牧或不合理放牧行为，退牧还草工程执行力度差。就各乡草地恢复状况而言，曼日玛乡草地总体恢复趋势最好，欧拉乡草地总体恢复趋势相对较差。各乡下辖各围栏区草地恢复效果显著，但个别围栏区草地仍处于恶化趋势，其中恢复效果最好的是16# 围栏区，恶化最为严重的是5#围栏区草地。

摘　要： 甘南黄河重要水源补给生态功能区退耕还林措施取得到了一定的成效。甘南生态环境不断恶化的趋 势得到了有效遏止,森林保持水土、涵养水源、调节气候、保护牧场的功能开始逐步加强,生态效益更趋于明显,充分发挥了维系国家生态安全、推动经济社会可持 续发展的基础保障作用。甘南牧工商业持续健康发展,人民群生活水平得到了较大改善和提高,林业在促进自治州经济社会协调发展、拉动经济增长中的作用和地位 进一步得到巩固和加强。

基于分布式测算方法，对甘肃省退耕还林工程的生态效益进行了评 估．结果表明，截至2013年，甘肃省退耕还林工程的生态效益总价值量为848.94亿元·a－1，相当于工程自实施到2015年的总投资．6项功能的价 值量由大到小的顺序为涵养水源、保育土壤、固碳释氧、生物多样性保护、净化大气环境和营养物质积累．各市（州）生态效益价值量排序为平凉市＞庆阳市＞陇南 市＞天水市＞定西市＞白银市＞临夏州＞张掖市＞兰州市＞酒泉市＞武威市＞甘南州＞金昌市＞嘉峪关市．说明甘肃省退耕还林工程取得了较好的生态效益，评估结 果真实地反映出甘肃省退耕还林工程所取得的成果．

The status of the “black beach like” degraded grassland in the water replenishment area of upper Yellow River was systematically expounded. It was found that the factors of abnormal climate vibration, overgrazing, rodent damage, anthropogenic influence, etc. have synthetically resulted in this degradation. In order to mitigate grazing pressure, prevent further expansion and spread of the degraded grassland area, and lastly realize the development of animal husbandry, the effective approaches of comprehensively ecological management should be: following natural laws, improving the occasion, associating the models of prevention and treatment in controlling degraded grassland, rationally grazing under grassland carrying capacity, optimizing management, completely implementing the family field policy, increasing investment, and vigorously developing education and strengthen legal system building.

通过对黄河上游水源补给区“黑土滩”型退化草地现状的调查发现,“黑土滩”型退化草地是由于气候的异常扰动、过度放牧、鼠类破坏、人为影响的综合因素引起的。从而提出了生态综合治理有效途径应以遵循自然规律,因势利导为出发点,减轻放牧压力,防止退化草地面积进一步扩张和蔓延;对不同程度退化草地应采用防治结合的模式,以草定畜,优化畜牧业管理模式;同时应及时灭鼠和防治毒杂草,全面贯彻落实家庭牧场政策、增加投入和高效管理,大力发展教育事业,加强法制建设,实现黄河上游水源补给区草与畜的可持续发展。

Understanding the economic value of grassland ecosystem restoration as well as how people perceive its role in providing service can provide insight into the grassland protection and the kinds of policies needed to ensure sustainable uses. We used WTP (maximum willingness to pay for grassland restoration) and WTA (minimum willingness to accept for overloading livestock in grassland) in one questionnaire and Anchored Payment Card (APC) as their eliciting tools to estimate the restoration cost of Maqu grassland, which is one of the important parts in Tibetan Plateau ecosystems. The main results are: ① In 525 families, about 75% (394 families) selected the WTP approach and only 24.6% (131 families) selected the WTA approach. For local people, the reference is that most of them judge the rich or the poor just by the number of livestock that family has. Holding on this traditional opinion, people seldom sell the livestock and wait until they become old enough, which leading to the number of livestock increasing continuously and giving raise to the grassland degradation. ② If we assumed that the restoration of Maqu grassland ecosystem needs 10 years and divided the restoring period into two stages, then the former 4 years cost was 3.4×108RMB, the average cost was 0.85×108RMB per year. While the whole cost of latter 6 years was 0.126×108RMB,average cost was 0.021×108RMB per year. Within the total cost, the payment value (WTP) was only about 6 %, the compensated value (WTA) was 94%. These suggested that the restoration of Maqu Grassland should mainly adopt the measurement of accelerating the superfluous livestock out of the Grassland. Then the pressure of Grassland can be mitigated, and its natural resilience could be achieved. The result also shows that the method of the combined WTP and WTA in one CV (Contingent Valuation) survey is very useful in estimating the cost of environmental improvement projects.

运用目前流行的能同时用于市场物品和非市场物品的条件价值评估法（CVM），对玛曲居民的草地恢复意愿进行了初步调查，并以他们的最大支付意愿（WTP）和最小受偿意愿（WTA）为依据，粗略推算了玛曲草地生态系统的年度恢复成本和总恢复成本。主要结论：① 525户调查样本中，394户选择了出资恢复措施，占总样本的75.4%；131户选择了减少牲畜数量恢复措施，占总样本的24.6%。这一结果表明当地牧民仍然没有摆脱传统数量型畜牧业生产模式，牲畜出栏率低，造成草地持续退化。② 如果玛曲草地恢复以10a计，则总的恢复成本为 3.52×108元，其中前4a每年的恢复成本为0.85×108元，后6a每年的恢复成本为0.021×108元。整个恢复成本中，补偿价值占94%，支付价值只占6%，说明玛曲草地生态系统的恢复要以加速超载牲畜的出栏和淘汰为主，切实减轻草地压力，实现自然修复。

Alpine steppe is the major vegetation community around Qinghai Lake Region. It is regarded as an important ecological barrier of the Qinghai-Tibetan Plateau and a rich site forage base for yak and sheep production. However, the alpine steppe vegetation degraded by overstocking. Therefore, it’s important to study the effects of grazing on community of alpine steppe around Qinghai Lake Region. In this study, the vegetation under different grazing intensity were investigated, based on grazing trial for 3 years in alpine steppe around Qinghai Lake Region. Levins niche breadth and Pianka niche overlap index were used to calculate niche characteristics of major plant populations under different grazing gradients to explain the mechanism of grazing succession of community. Our results showed that dominant species of community had changed after grazing 3 years. The dominance of plants with good palatability ( Stipa purpurea, Poa sp., Medicago ruthenica , etc.) decreased and the dominance of plants with grazing resistance or bad palatability ( Kobresia humilis, Thermopsis lanceolata , etc.) increased with the increase of grazing intensity. S.purpurea had the widest niche breadth, and after then was Thermopsis lanceolata, with the values being 0.994 and 0.959, respectively. Potentilla multifida and Astragalus polycladus had wider niche breadth because of its stronger environmental adaptation ability. The niche breadths of Elymus nutans and K.cristata under different grazing gradients were narrow, indicating that grazing restricted the growth of high grasses. Species with wider niche breadth had higher niche overlap with other species. The narrower niche overlap could be seen as the result of interspecific differentiation on resource utilization. The major plant populations had niche differentiation result from grazing. With increasing grazing intensity, community with S.purpurea and Poa sp. being the dominant populations may be in degradation direction on community with K.humilis, K.pygmaea and T.lanceolata being the dominant populations, indicating that habitat of community had obviously degraded. Therefore, it’s important to control grazing intensity of rangeland.

以环青海湖高寒草原为研究对象,对不同放牧强度下草地植被进行调查,并运用Levins生态位宽度和Pianka生态位重叠指数对不同放牧率梯度下主要植物种群生态位特征进行计算,以期从生态位角度解释植物群落的放牧演替机制。结果表明,经过3年的放牧,群落优势种发生了改变,随着放牧强度的增加,适口性好的植物优势度降低,家畜不喜食或不食的和耐牧的植物优势度增加。紫花针茅生态位宽度最大,披针叶黄华次之,分别为0.994和0.959,多裂委陵菜( Potentilla multifida)和多枝黄芪(Astragalus polycladus)环境适应能力较强,生态位宽度较大。在放牧率梯度上垂穗披碱草(Elymus nutans)和溚草(Koeleria cristata)的生态位宽度较小,说明放牧抑制了高大禾草的层片发育。生态位宽度较大的物种与其他种群间有较大的生态位重叠,生态位重叠度较小是种间对资源需求分化的结果。放牧使主要植物种群生态位发生了分化,随着放牧强度的增加,群落可能从以紫花针茅和早熟禾等为优势种群向以矮嵩草、高山嵩草(K.pygmaea )和披针叶黄华为优势种群的方向发展,表明生境有了明显的退化。因此,控制草地的放牧强度是维持种群健康发展的关键。

近几年，黄河源区高寒草原退化沙化明显。研究了黄河源区紫花针茅高寒草原在不同退化程度下植物群落、生物量和土壤的特征，结果表明，随着高寒草原退化程度加大，植被盖度、草地质量指数和优良牧草地上生物量比例逐渐下降，草地间的相似性指数减小，植物群落多样性指数和均匀度指数呈单峰式曲线变化规律。随着退化加剧，禾草地上生物量减少显著，杂草类植物地上生物量先增加后减少，而莎草科植物地上生物量的变化不受草地退化程度的影响。高寒草原土壤表层０～２０ｃｍ 的植物根系随土壤深度的增加而减少，随草地退化程度的加剧呈倒“Ｖ”型变化。高寒草原物种多样性与生物量之间呈“Ｖ”型变化规律，随着退化程度增加，物种丰富度与生产力的关系由显著正相关转变为负相关。随着高寒草原的退化程度加剧，土壤湿度、土壤有机质、速效磷、硝态氮、速效钾的含量和土壤紧实度都减小。随着高寒草原植被的退化演替，土壤退化越来越严重，贫瘠化不断加剧，到重度退化阶段，旱生沙生植物出现，呈现沙化初始景观。

Effects of different grazing intensities on plant community characteristics of desert steppe were studied through field investigation and data analysis. The results showed that species in communities tended to decrease with the increasing of grazing intensity. With the increasing of grazing intensity, important values of species which livestock grazed decreased, even disappeared, while important values of species which livestock did not graze increased. Plant community biomass, plant community height, plant community coverage and plant community intensity was significantly reduced ( P ＜0.05). Species richness index, homogeneity index and diversity index also showed a downward trend. This is related to the relatively small number of vegetation types and the poor anti-interference ability of the plant community in desert steppe.

本研究通过野外调查与室内分析,探讨了不同放牧强度对荒漠草原植物群落特征的影响。结果表明,随着放牧强度的增加,组成群落的物种数量呈现减少的趋势;家畜喜食植物的重要值降低甚至消失,而家畜不喜食植物的重要值增加;群落生物量、群落高度、群落盖度及群落密度也显著降低( P ＜0.05);随放牧强度增加,物种丰富度指数、均匀度指数与多样性指数也呈现下降趋势。这与荒漠草原的植被种类相对较少、群落整体的抗干扰能力较差有关。

本文从甘南草原鼠害群落类型、 危害区域入手,分析生物毒素、招鹰控鼠、弓箭捕捉、禁牧休牧、补播等技术组装集成的3种鼠害治理模式的方法和效果,结果成功解决了甘南黄河重要水源补给生 态功能区的草原鼠害问题,并且将短期治理、长期防控以及工程措施、生物措施相配套,生态效益显著。

Effect of the population densities of plateau zokor (Myospalax baileyi) in six density grades selected in alpine meadow of Haibei,Qinghai Province on biomass,structure and succession of plant communities was examined quantitatively with data collected in May and August of 1995 and 1996.Species composition,biomass,recovering rate of plants after be covered by mounds among six grades showed strong changes.The species numbers,total biomass of aboveground and monocotyledon biomass showed significantly negative correlation with population density,respectively.However,there was significantly positive correlation between the population density of plateau zokor and the log value of bicotyledons percentage of total biomass.As the population density of plateau zokor increased,the importance value of monocotyledons decreased significantly.However,some bicotyledons which contain second chemical compounds increased rapidly,the importance value of some plants which have strong roots became the biggest in plant community while zokor density was in the middle grade.The important value and biomass of grasses increased significantly,but the biomass of total plants didnt increase after some of zokor population were removed.

研究结果表明，高原鼢鼠种群密度与植物地上生物量呈极显著的负相关关系，与双子叶植物在地上生物量中所占百分比的对数值呈极显著的正相关关系。草地次生植被面积随高原鼢鼠种群密度增加而显著增加，土丘覆盖植物的恢复速率则随高原鼢鼠种群密度增加而显著地降 低。植物群落组成的种数与种群密度呈显著的负相关关系。单子叶植物重要值随高原鼢鼠种群密度的增加而显著下降；双子叶直根类植物与高原鼢鼠种群密度不存在显著的相关性；双子叶植株内含次生化合物的物种重要值则与种群密度呈极显著的正相关关系。捕杀高原鼢鼠后，禾草类、莎草类重要值上升，植物地上生物量增加不显著(P>0.05)，但禾草类增加极显著(P<0.01)。

研究表明,高原鼢鼠的造丘活动在停止放牧扰动的条件下,对于过牧型退化草地而言,非但不会对草地生产力造成破坏,相反,在植被高度、植被组成和产草量等生产力构件上均表现出了显著的促进作用.在相同的休牧条件下,鼢鼠分布地植被性状的各项指标除植物多样性低于无鼢鼠分布地外,其余指标均极显著高于无鼢鼠分布地(P<0.01),其中丘缘地又极显著地高于丘间地(P<0.01);在同为3年的三种休牧方式中,实施绝对禁牧的方式最佳,其次为降低载牧量50%且于返青至6月前禁牧的方式,这两种休牧方式均可较好地满足鼠丘在退化草地植被恢复过程中充分发挥其生态功能所必需的外部条件.

Global change and terrestrial ecosystem are important content in recent global change researches. One of focuses on global change researches is the impact of climate change on terrestrial ecosystem. The impact of climate change on the ecological environment of Northwest China was reviewed generally. Through overall understanding and analyzing the impact of climate change on vegetation，biodiversity，wetland and water resources respectively, the paper also discussed response of the ecological environment of Northwest China with climate change and the existing problems, and proposed countermeasures against the impact of climate change on the ecological environment of Northwest China，based on recent research trends.

全球变化与陆地生态系统是当前全球变化研究的重要内容，气候变化对陆地生态系统的影响及反馈一直是全球变化研究的焦点问题之一。本研究总结了有关气候变化对西北地区生态环境影响研究的趋势，分别通过气候变化对植被、生物多样性、湿地以及水资源等方面影响研究的动态，阐述了西北地区生态环境对气候变化的响应，对存在的问题进行了探讨,并结合当前研究状况提出西北地区应对气候变化的对策及措施。

Grassland desertification in Maqu County adversely affected the replenishment of the water sources in the upper reaches of the Yellow River. Based on an estimation of NPP and the theoretical livestock capacity of grassland, the effects of natural and artificial factors on grassland desertification were analyzed quantitatively. The results were as follows: the key natural factor that affected grassland desertification was increasing temperature, especially the accumulated temperature of ≥0 ℃, and the key artificial factor was overgrazing. The effect of temperature increase accounted for 34.39% of the grassland desertification and overgrazing for 65.61%. The grassland desertification phenomenon did not appear until the accumulative effect of overgrazing was considerable. An increase of 10 000 sheep units in the accumulated overstocking capacity would cause grassland desertification of an area of 3.14 hm2. Protection of grassland from desertification should mitigate or gradually decrease the accumulated numbers of overstocking on the basis of the forage livestock balance.

玛曲草地沙化对黄河上游水源补给产生了不利影响。基于NPP和理论载畜量的估算，定量分析了自然因素和人为因素对草地沙化的作用程度。结果表明：影响玛曲草地沙化的自然因素主要是温度升高，特别是≥0℃积温的增加；人为因素主要是超载过牧。温度升高对草地沙化的影响占34.39%，超载过牧占65.61%。草地沙化是过牧的累加效应达到一定程度时才出现的，累加超载量每增加1万个羊单位会造成3.14 hm2草地沙化。草地沙化的防治应以草畜平衡为基础，逐渐减缓或减少超载量的累加。

摘　要： 利用2000年和2006年TM影像数据、地面调查资料和气象台站观测数据，对玛曲县沼泽湿地进行了目视解译分析，对比研究了非监督分类、监督分类和专家分类的精度，初步探讨了玛曲沼泽湿地退化的原因。结果表明，该县沼泽湿地面积从2000年的19598．67hm^2减少到2006年的14080．80hm^2，沼泽湿地面积平均年递减率为4．69％。非监督分类结果由于受山体阴影、坡向、云的阴影等影响，多分误差高，总分类精度仅为31％；监督分类的总精度为54．1％；专家分类的总精度高达84．2％。因此，专家分类可以用于对玛曲县沼泽湿地的监测和自动提取。草场不合理利用和过度放牧是导致玛曲县沼泽湿地退化的主要原因，黄河干流、各支流径流量减少是导致流域内沼泽湿地退化的直接原因，当地气候变暖促进了这一过程的发生。