GEO-ECOLOGICAL TRANSECT STUDIES IN NORTHEAST TIBET (QINGHAI, CHINA) REVEAL HUMAN-MADE MID-HOLOCENE ENVIRONMENTAL CHANGES IN THE UPPER YELLOW RIVER CATCHMENT CHANGING FOREST TO GRASSLAND

Geo-ecological transect studies in the pastures of the upper catchment of the HuangHe (99 degrees 30'-100 degrees 00'E/35 degrees 30'-35 degrees 40'N'; 3,000-4,000 in a.s.l., Qinghai province, China) revealed evidence that pastures replace forests. Plot-based vegetation records and fenced grazing exclosure experiments enabled the identification of grazing indicator plants for the first time. The mapping of vegetation patterns of pastures with isolated juniper and Spruce forests raise questions as to the origin of the grasslands, which arc widely classified as "natural" at present. Soil investigations and charcoal fragments of Juniperus (8,153 +/- 63 uncal BP) and Picea (6,665 +/- 59 uncal BP) provide evidence of the wider presence of forests. As temperatures and rainfall records undoubtedly represent a forest climate, it is assumed that the present pastures have replaced forests. Circumstantial evidence arising from investigations into the environmental history of the Holocene effectively substantiates this theory.

Effects of vegetation control on ecosystem water use efficiency within and among four grassland ecosystems in China

Through 2-3-year (2003-2005) continuous eddy covariance measurements of carbon dioxide and water vapor fluxes, we examined the seasonal, inter-annual, and inter-ecosystem variations in the ecosystem-level water use efficiency (WUE, defined as the ratio of gross primary production, GPP, to evapotranspiration, ET) at four Chinese grassland ecosystems in the Qinghai-Tibet Plateau and North China. Representing the most prevalent grassland types in China, the four ecosystems are an alpine swamp meadow ecosystem, an alpine shrub-meadow ecosystem, an alpine meadow-steppe ecosystem, and a temperate steppe ecosystem, which illustrate a water availability gradient and thus provide us an opportunity to quantify environmental and biological controls on ecosystem WUE at different spatiotemporal scales. Seasonally, WUE tracked closely with GPP at the four ecosystems, being low at the beginning and the end of the growing seasons and high during the active periods of plant growth. Such consistent correspondence between WUE and GPP suggested that photosynthetic processes were the dominant regulator of the seasonal variations in WUE. Further investigation indicated that the regulations were mainly due to the effect of leaf area index (LAI) on carbon assimilation and on the ratio of transpiration to ET (T/ET). Besides, except for the swamp meadow, LAI also controlled the year-to-year and site-to-site variations in WUE in the same way, resulting in the years or sites with high productivity being accompanied by high WUE. The general good correlation between LAI and ecosystem WUE indicates that it may be possible to predict grassland ecosystem WUE simply with LAI. Our results also imply that climate change-induced shifts in vegetation structure, and consequently LAI may have a significant impact on the relationship between ecosystem carbon and water cycles in grasslands.

Short-term effect of targeted placements of sheep excrement on grassland in inner mongolia on soil and plant parameters

Grazing animal excrement plays an important role in nutrient cycling and redistribution in grazing ecosystems, due to grazing in large areas and return in small areas. To elucidate the changes to the soil and pasture caused by sheep urine, fresh dung, and compost patches, a short- term field experiment using artificially placed pats was set up in the autumn of 2003 in the Inner Mongolian steppe. Urine application significantly increased soil pH during the first 32 days in soil layers at depths of both 0 - 5 cm and 5 - 15 cm. Rapid hydrolysis of urea gave large amounts of urine- nitrogen ( N) as ammonium ( NH4+) in soil extracts and was followed by apparent nitrification from day 2. Higher inorganic N content in the urine- treated soil was found throughout the experiment compared with the control. No significant effects of sheep excrement on soil microbial carbon ( C) and soil microbial N was found, but microbial activities significantly increased compared with the control after application of sheep excrement. Forty- six percent of dung- N and 27% of compost- N were transferred into vegetation after the experiment. The results from this study suggest that large amounts of nutrients have been lost from the returned excrement patches in the degraded grassland of Inner Mongolia, especially from sheep urine- N.

Short-term effects of sheep excrement on carbon dioxide, nitrous oxide and methane fluxes in typical grassland of Inner Mongolia

Excrement patches of grazing animals play an important role in greenhouse gas (GHG) fluxes due to the high nitrogen (N) and available carbon (C) deposited in small areas, but little information is available for the effect of excrement in the Inner Mongolian grassland (43 26 degrees N, 116 degrees 40'E). To elucidate the effect of grazing sheep urine, fresh dung and compost on fluxes of methane (CH4), carbon dioxide (CO2), and nitrous oxide (N2O), a short-term field study (65 days) was carried out in the typical grassland of Inner Mongolia with the optimised closed chamber/GC technique. Compared with the control, cumulative net CH4 consumption decreased 36, 31, and 18% from urine, fresh dung, and compost plots, respectively; net CO2-C output increased by 6.5, 1.5, and 1.2% from urine, fresh dung, and compost treated soil, respectively; about three times as much N2O-N was emitted from urine and the fresh dung treatments during 65 days. Nitrous oxide emission was positively correlated with CO, emission (R = 0.691, P < 0.01) and water-filled pore space (R = 0.698, P < 0.01). The percentages of N2O-N loss of applied-N were 0.44 and 1.05% for urine and fresh dung, respectively. Our results suggest that in autumn in the degraded grassland of Inner Mongolia, the effect of sheep excrement may be ignored when evaluating the total GHG emissions.

Suitability of extractants for predicting available sulfur in natural grassland in the Inner Mongolia steppe of China

It was the objective of this study to compare the suitability of different extractants for predicting the availability of sulfur (S) in natural grassland in a sulfur response trial on three different soil types in the Inner Mongolia steppe of China. For soil analysis, seven different extractants have been employed. The inorganic SO4-S concentration was determined by ion chromatography. Additionally, in the Ca(H-2-PO4)(2) extract the total soluble S was determined employing turbidimetry. Weak salt solutions (0.15% CaCl2, Ca(H2PO4)(2), and KH2PO4) extracted similar amounts Of SO4-S. Extraction with 0.025 M KCl provided the lowest SO4-S values. Deionized water dissolved significantly more SO4-S in the control plots than most weak salt extractants. The concentration of soluble organic S decreased in the control plots after 100 days of plant growth, indicating that the organic S pool contributed significantly to the S nutrition of the forage crops. Significant relationships among the SO4-S in the soil determined in different extracts and crop yield, sulfur content in the forage, and total sulfur uptake were only found for the Ca(H2PO4)(2) extract. In general, the correlation coefficients proved to be unsatisfactory for field experimentation.

Seasonal changes in weight and body composition of yak grazing on alpine-meadow grassland in the Qinghai-Tibetan plateau of China

Forty-five male yaks (born April 2001) were studied to determine how seasonal changes on the Qinghai-Tibetan plateau affected BW and body composition. Thirty yaks were weighed monthly from birth to 26 mo of age to determine seasonal changes in BW. The remaining 15 yaks were allocated randomly to five groups (three yaks per group), designated for slaughter at 13, 15, 18, 22, and 25 mo to measure seasonal effects on body chemical composition. All yaks were grazed on the alpine-meadow grassland of the plateau without any supplementation. All BW and body composition data were calculated on an individual basis. Body weight and body composition data were both compared across seven growth periods spanning 2 yr and defined by season. From April (birth) to December 2001 of the first growing season, yak BW increased (P < 0.01); however, during the subsequent cold season (December 2001 to May 2002), BW decreased (P < 0.01). The second growing season ran from May 2002 (13 mo of age) to October 2002 (18 mo of age), and the second live weight-loss season ran from October 2002 until May 2003. The weight loss experienced by yaks during the first weight loss season was 25.64% of the total weight gain in the first growing season. The weight loss experienced by yaks during the second weight-loss season was 29.73% of the total weight gain in the second growing season. Energy retention in the second growing season was 291.07 MJ, 50.8% of which was consumed during the subsequent cold season. Energy accumulation in the summer (from May to July) and fall (from July to October) of the second growing season did not differ (5.01 and 6.30 MJ/kg of EBW gain, respectively; P = 0.63). The energy mobilized during the second winter (from October 2002 to February 2003) was 16.49 MJ/kg of EBW, and in the second spring (from February to May 2003), it was 9.06 MJ/kg of EBW. These data suggest that the decrease in grazing yak BW during the first cold season is much less than during the second cold season, and that the energy content per unit of BW mobilized is greater (P = 0.02) in winter than in spring. Results from this study demonstrate highly efficient compensatory growth in grazing yaks following the first weight loss period during the first cold season. This benefit could be exploited by herders to improve yak production. Yaks may have developed a type of self-protection mechanism to overcome the long cold seasons in the Qinghai-Tibetan plateau.

基于土壤水分平衡原理的毛乌素沙地草地种植咨询系统

本研究初步建立了基于土壤水分平衡原理的毛乌素沙地草地种植咨询系统“ASSG”，该系统用于进行毛乌素地区沙地草地利用和种植管理咨询，并可进行更新知识库和存储有关咨询结果的操作。该系统适用于田间工作者、农场管理者、有关科研人员以及任何对生态科学和自然资源管理感兴趣者。 在文中描述了该咨询系统的开发过程，即系统的需求分析、设计、实现和运行，该系统实现了专家系统和仿真模型的结合，该模型即基于水分平衡的最佳植被覆盖率模型，有如下初步结论产生：土壤水分的平衡随植被覆盖率变化的情况（上升或下降）取决于立地条件的差异，笼统地讲“植被覆盖率越高越好”或“植被覆盖率一概不宜太高”未免失之偏颇。 在解决复杂的生态学问题时人工智能／专家系统技术的应用以及在定性模型和方法（如专家系统）的应用中使用定量模型（如仿真模型）正处于发展之中，本系统的实现是应用专家系统与仿真模型结合的方法解决实际生态学问题的又一个例子。

草地生态系统碳循环及其对全球变化的响应_锡林河流域羊草草原生态系统循环研究

本研究从三个方面研究了草地生态系统的碳循环及其对全球变化的响应： 1)系统整理了世界范围内各种类型草原群落碳循环主要过程和影响因素的有关资料，在此基础上，对草地生态系统巾的碳素贮量和主要流量进行了全球估计;在CENTURY模型碳循环业模型的基础上，对草地生态系统碳循环的分室模型进行了说明;阐述了土壤呼吸作用在草地碳循环中的作用，综述了草地土壤呼吸的测定方法、影响草地呼吸的主要环境因素，对世界范围内草地土壤呼吸量和土壤中的有机碳周转速度进行了初步估计;讨论了全球气候变化、 C02浓度增加和土地利用变化对草地生态系统的可能影响。 2)对内蒙古锡林河流域一个固定样地内的羊草草原群落的碳循环特点进行了初步研究;采用实测数据和CENTURY模型模拟相结合，对该群落中碳素的输入与输出量，碳素贮量和土壤中有机碳的周转口寸间进行了首次估计，得出了“未利用状态下羊草草原群落基本处于碳素平衡状态”的结论。 3)对世界范围内草地利用现状的资料分析表明，全世界70%的草地出现了不同程度的退化， 过度放牧是导致草地退化的主要原因，在退化草地总面积中，约有35%是由于过度放牧造成的。过度放牧是锡林河流域羊草草原退化的主要入为因素，40年来，过度放牧共导致了羊草草地O - 20cm土层中23%的土壤有机碳损失。结合CENTURY模型的预测结果，土地利用变化对羊草草原土壤有机碳贮量的影响将比气候变化和C02浓度增加的影响更加重大。

松嫩平原碱化草地景观空间格局及环境因子的综合分析

本文对松嫩平原羊草地的群落结构及空间格局进行了分析，主要内容包含以下三个方面： 1、本文第一部分是对碱化草地植物群落分布的空间、环境因子分离。决定碱化草地植物物种在空间分布的因素可以大致分解成4个部分：（1）通过物种的生态位起作用的环境因子;（2）决定物种、群落间竞争激烈程度的空间距离因子;（3）环境和空间因子的交叉或耦合作用;（4）其它未知因子（如生物和随机因子）。本研究在1996年野外实地采样的基础上，运用DCA消势对应分析（Detrended Correspondence Analysis）对群落主要变化趋势及其与土壤环境因子之间的关系进行了分析。运用DCCA消势典范对应分析（Detrended Canonical Correspondence Analysis）对影响松嫩平原碱化草地群落结构季节动态的空间和环境因素进行了定量的分解。结果表明： 在影响群落分布的各因子中，环境因子独立约占40％，而环境一空间耦合因子占35％，空间因子独立约占3％，其它因子约占20％。在诸多因子中，土壤的盐渍化程度在整个生长季起着决定性作用，但土壤水分和氮素的作用因季节而变化。在干旱季节，植物生长的主要制约因子表现为土壤碱化度和土壤水份，而土壤氮素的作用处于次要地位。但在降雨较多、土壤湿润度较大的季节，土壤氮素的影响明显增强，成为仅次于土经度的决定性因素。 2、第二部分对东北松嫩平原碱化草地植物群落空间格局的分形性质进行了分析，分别用边长－面积指数和Korcak指数估计了斑块边界复杂性和斑块面积分布的分形维数。结果表明： 随放牧强度增加，占优势的羊草斑块的相对斑块化加剧，而斑块的边界在中度放牧时最不规则： 在水淹地，占优势的羊草斑块和次优势的碱茅斑块、獐茅斑块的斑块边界复杂性和斑块化程度比未水淹地都低，说明水降低了群落复合体内的异质性： 水淹地优势植物斑块的边长－面积指数和Korcak指数均低于次优势种，但在重度放牧地结果好相反，说明两种样地处于不同的演替阶段; 斑块边界复杂性符合同一尺度规律，在现有的面积范围以内没有尺度转换，而斑块面积分布则存在尺度转换点。对重牧地和水淹地的尺度分析表明，较小尺度上， 群落空间格局的相对斑块化程度较低，说明较小尺度上的空间格局要相对稳定一些，而较大尺度上则相反。可能原因是放牧干扰对大尺度的斑块影响更大。 3、对1989年开始围栏封育的样地10年来的空间格局变化进行时间序列分析，研究恢复演替过程中各个主要斑块类型的分布特征及动态，以及整个样地总体格局在演替系列中的变化。得出如下结果： 羊草是该区植被的优势种，其格局动态为：1989－1993年，羊草斑块总面积增加，斑块数目减少，相对斑块化指数降低，1994年后羊草的空间格局基本稳定。 羊草的斑块面积分布曲线在20平方米左右发生转折，即空间格局在此尺度发生尺度转换。不同尺度上斑块的相对斑块化指数及其在恢复演替中的变化趋势不同：小斑块的相对斑块化指数低，在1989－1993年期间增加，而后降低;大斑块的情况正好相反。说明羊草空间格局的主要变化之一是中等大小斑块的合并和数目减少，羊草斑块生长对格局的影响比斑块合并的影响要小。 獐茅斑块的格局动态为：1989－1994，斑块总面积、斑块数目、最大斑块面积增大，1995年开始降低。相对斑块化程度则先降后增。碱蓬斑块在1989－1993年间的格局动态与羊草正好相反：总面积减少，斑块数目增加，斑块面积大小的变化范围变窄，说明1993年之前，控制碱蓬斑块的主要生态学过程是其他物种的侵入，隔离原来较大的斑块，同时占据某些小斑块。1994年之后，碱蓬斑块的变化比较随机，其格局变化受到其他物种如杂草对策种虎尾草等的影响很大。 样地总体格局的变化为：1991年前，斑块种类和总斑块数目增加，斑块化程度随着增大;1993年后，斑块种类基本稳定下来;1991－1995，斑块化指数降低。在此期间，斑块数目增加，所以斑块化程度降低意味着斑块大小频率分布趋于均匀，大斑块被分裂，小斑块在长大，群落总体空间格局逐渐稳定。总体格局在三个尺度上有不同的自相似规律，这三个尺度分别为：La(a)<=1; 15.5。小尺度上相对斑块化程度低，但年际间变化较中尺度剧烈;中尺度相对斑块化程度变化较为缓，大尺度上，斑块分布一直趋于减少最大的斑块数目。总的变化趋势为，随恢复演替的深入，最大和最小的斑块数目都减少，处于中间尺度的斑块最多。

植物群落复杂性与群落结构组建的描述与分析

时间、空间格局的保持和发展及其对种群、群落和生态系统动态的影响是生态学中两个相互关联的基本课题。围绕这个中心议题作者在植物个体、种群、群落各个层次上开展了研究。 1)研究了克隆植物构型及其对资源异质性的响应，给出了描述该连续体的一个指数v：In1/ln刀．（士

放牧对内蒙古冷蒿小禾草草原植物群落特征的影响

了解天然草原放牧生态系统中，不同管理措施对草－畜系统的影响及系统中草－畜互作方式，是实现草原生态系统持续管理的基础。在内蒙古典型草原区，我们建立了24hm2围栏放牧实验样地。本文通过对试验样地5个放牧率（1.33，2.67，4.00，5.33和6.67羊／hm2）和两种放牧方式下（连续放牧与轮牧）一年排除放牧后植被变化的研究，试图揭示8年的放牧对草地群落的作用结果，探寻放牧率与植物群落生产力、种类组成和演替方向的关系，以求为建立合理的放牧制度提供佐证。 研究发现：①不同的放牧率下8年的放牧使地上生物量出现显著差异，中度放牧（SR3）下的草原群落达到了最大地上生物量，在较轻的放牧率（SR1，SR2）下有超补偿生长出现，地下生物量随着放牧率的升高而下降。②放牧使群落植物组成发生重大变化，随着放牧率的升高，表现出明显的向星毛委陵菜退化草原演替的趋势，无牧的群落表现出恢复演替的趋势。③在轻牧（SR1）处理下植物群落达到了最高的多样性，在多样性与生物量之间，不能发现较具规律性的关系。④轮牧群落的生产力潜力显著高于连续放牧，连续放牧方式造成植物斑块状分布，降低了草场的利用效率。⑤放牧造成了草原植物生态位分化。

内蒙古草原地耳和发菜的固氮特性及在氮素循环中的作用

以内蒙古干旱半干旱草原上的固氮蓝藻－普通念珠藻和发状念珠藻为研究对象，对其分布、生态特点、形态结构、生物量及不同条件下的固氮活性进行了研究，并根据降水与气温状况对其年固氮量进行了估算。结果表明：同发状念珠藻相比，普通念珠藻的分布更为广泛，其形态也有较大的差异。普通念珠藻的生物量在以禾本科为主的退化草场和沙地上较大，在pH值偏碱、含水量300％和温度30℃的情况下，其固氮活性较高，达2500nmolC2H2/g•h（干重）以上，年固氮量也高达4.19kg/h.m2，占草原氮素总输出的10％左右。由此可见，在维持干旱半干旱草原与荒漠草原生态系统的氮素平衡中，固氮蓝藻这类低等固氮植物可能起着重要作用。

探索一种监测典型草原草场退化的新方法--结合群落样方调查资料和卫星遥感数据（TM）

大范围、实时、准确地监测典型草原地区草场退化或健康状况对于草原生态系统的保育、农牧业的可持续性发展具有非常重要的意义。比起传统的群落学研究方法，遥感技术对于监测大尺度的植被状况具有无可争议的优越性，并且已经被广泛引入监测植被覆盖变化的研究中。本项研究系统地分析、综述了过去用非遥感手段对放牧和草场退化的关注和研究，介绍了遥感技术应用于植被研究的理论基础、主要途径（植被指数）、有关领域的研究进展。特别是本文提出草场退化状况或整体健康状况可以由基干相互独立的层面表示，而过去监测植被变化主要依赖的NDVI等植被指数只能监测草原植被的个别层面（总量层）。 本文以草场放牧退化比较典型的内蒙古锡林河流域为研究对象，在进行了大量的野外样方调查的基础上，提出一种结合群落样方调查和遥感技术的监测草场健康状况的新方法。本文引入主成份分析方法（PCA），从包含12个反映群落各方面信息的变量中提取出3个有特定生态学意义的主成份，并进一步对其进行分析组合，得出一个能比较敏感、全面反映群落健康状况的新指标-草场健康指数（GHI）。 从6波段的植被光谱反射数据中比较理想地提取出2个主成份：可见光因子和红外光因子。表征群落总量、放牧退化的主成份和GHI与样方光谱反射值有相当的相关性，由此得到GHI与可光、红外光因子的回归模型。 应用此模型到卫星遥感数据（TM），得到GHI影像，并与同一数据的NDVI影像作对比研究，发现GHI在反映放牧等人为干扰对草原植被的影响效应方面比NDVI有明显的优点。此外，GHI影像对植被分布格局，特别是斑块结构有更好的显示效果。应用GHI到历史TM数据，对所研究地域的植被覆盖变化、农牧业的变迁模式等进行了定性研究。研究还发现有较长放牧史的过度放牧区的植被类型没有沿牧压梯度的规律性分布，而是呈随机斑块分布模式。

内蒙古温带草原生态系统过程对气候变化响应的实验研究

羊草草原、大针茅草原和草甸草原是内蒙古温带典型草原地区的三种主要植被类型，本文以此为研究对象，应用时空替代、温室、网室、自然降水接移、养分添加等多种方法模拟未来可能发生的气候变化，研究了这些变化对上述草原植物群落和生态系统过程的直接和间接影响。主要研究结果如下： 1. 气候变暖及其诱导的土壤养分释放速率增加对植物群落有明显的影响，表现在种种植物群落在高度、盖度、密度和生物量等都有不同程度的变化。说明我国温带草原生态系统对气候变化反应敏感。 2. 在群落说平上，气候变暖直接或间接使地上生物量提高17%～90%，但在种群水平，不同种群对模拟气候变化的响应具有很强的个性，而不具普遍性，不同种群反应的大小、方向不同。因此由在不同群落进行研究得出的推论只有部分是可能的，并且即使观测到相类似的反应，它们内在的机制也很可能是不同的。 3. 气候变化显著地影响着凋落物的分解特征。但不同的气候变化情景下，凋落物分解过程的响应不同。在温度升高降水变化不大或升高的情景下，凋落物的分解速率将加快;在气温上升，降水明显下降的情景下，凋落物的分解速率将降低。 4. 草甸草原土壤碳素释放过程和氮素矿化过程对气候变暖有明显的响应。气候变暖将促进土壤有机碳的释放，使草甸草原土壤成为碳源，同时加速氮素矿化速率，在一定程度上提高土壤有效养分的浓度。 5. 在中小尺度上，海拔样带所是应用时空替代原理，研究陆地生态系统与气候变化的理想平台，生物量是综合反应气候变化对土壤-植物系统影响的敏感指标。 6. 草原土壤有机碳和氮在2m～3m的小尺度上存在空间自相关性，其空间自相关尺度受人类活动的影响。

静态箱-气相色谱法研究人类活动对内蒙古草原主要温室气体交换通量的影响

本研究全面分析了陆地生态系统主要温室气体N2O、CO2和CH4通量研究的主要方法，系统阐述了改进后的静态箱-气相色谱法观测内蒙古半干旱原N2O、CO2和CH4交换通量的具体应用。就用此观测系统，于1998年5月至2000年2月长期定位观测研究了内蒙古锡林河流域主要人类活动（放牧、家垦）对N2O、CO2和CH4交换通量强度及季节变化的影响，同时对水分、地温等相关因素的影响作了初步探讨。 通过对主要温室气体N2O、CO2和CH4通量观测主要方法的比较分析认为：静态箱-气相色谱法是观测内蒙古半干旱草原N2O、CO2和CH4交换通量行之有效的方法，改进后的观测系统其准确性和再现性基本可以满足研究要求，但箱法的局限性也导致对CO2及不同牧压下草原N2O通量的观测很难得到客观准确的结果。 农垦对内蒙古草甸草原N2O、CO2和CH4通量有着重要的影响：农垦显著增大了草甸草原N2O排放通量，但对N2O排放通量的季节变化无显著影响;在植物生长旺期之前，农垦对草甸草原CO2排放通量无显著影响，在植物生长旺期及之后的小麦成熟期，农垦明显增大了CO2的排放。农垦还使得CO2排放通量的季节变化由单峰型变为双峰型;农垦对CH4通量的强度及季节变化均无显著影响。 连续两年的观测结果表明，研究放牧对内蒙古典型草原CO2和CH4通量的影响应结合水分条件来考虑。不同牧压下冷蒿-小禾草草原CO2排放能量受水分条件的显著影响，水分条件改变了CO2通量的季节变化。丰水年CO2排放强度随着放牧强度的增大而增大，正常年份放牧强度对CO2排放通量强度无显著影响。两年的观测均发现不同的放牧强度对CO2通量的季节变化没有影响;土壤-植被系统吸收CH4的能力在不同放牧强度处理之间没有显著差异，而同一放牧率下年际间CH4通量变化很大，正常年份比丰水年吸收通量大。同时分析了温度与不同放牧强度草原CH4通量的关系，认为影响放牧条件下冷蒿-不禾草草原CH4吸收通量的主要因子是地表以下10cm土壤温度。