【作者】 冯琦胜；

【导师】 梁天刚；

【作者基本信息】 兰州大学 ， 草业科学， 2012， 博士

【摘要】 基于生物气候的植被分类模型在全球气候变化与生态系统响应研究方面扮演重要角色。本研究利用全球共享气象数据库以及GIS空间分析等方法,在验证了CSCS (Comprehensive and Sequential Classification System)模型的模拟全球潜在自然植被(Potential Natural Vegetation, PNV)分类精度的基础上,分析了全球及区域尺度上当前及过去100年PNV类型的时空分布格局及变化规律；研究了未来50年全球及区域PNV演替趋势；以中国为例探索研究了残存PNV及其典型区的分布范围；研究了CSCS理论在现存自然植被生长状况监测与评价方面的应用；最后以本研究取得的核心成果为基础,开发了基于CSCS的全球潜在自然植被分类管理信息系统。主要研究结果有：1)在全球尺度上,CSCS PNV同RF PNV (Ramankutty&Foley PNV). HLZ (Holdridge Life Zones)、BIOME4模型之间具有较好的一致性。CSCS模型不仅能够成功地预测冻原、荒漠和森林的分布,而且同HLZ模型相比,对草地植被具有更好的分类能力。2)基于CSCS模型,除了南极洲以外,全球PNV可划分为10个类组和42个类型。除永久冰雪和内陆水体外,全球PNV总面积为1.289335×108km2,占陆地总面积的96.07%。中国PNV可划分为10个类组,除亚热极干亚热带荒漠类和炎热极干热带荒漠类之外,全国有40类PNV。由于受人类活动的影响,现存植被同潜在植被之间存在显著差异。PNV受人类影响程度由小至大依次为冻原、荒漠、林地、灌丛和草地。3)过去100年间(1901-2000),除热荒漠和萨王纳植被类组外,其余8种植被类组的分布中心在南北半球的移动方向。热荒漠植被类组在南半球的移动距离是10个类组中最大的。除了冻原与高山草甸、亚热带森林和温带湿润草地植被类组外,其他7种植被类组在南半球的移动距离均大于北半球。4)在全球范围内未来50年(2001-2050)陆地温度将呈上升趋势,尤其是北半球地区温度上升幅度较大,降水的变化没有明显的地域规律。总体而言,全球气候大致向着暖干化的趋势发展,但局部地区存在向暖湿化变化的趋势。中国未来50年温度上升幅度在1.5-4.5℃之间,降水表现出明显的南增北减的趋势。未来50年全球潜在冻原植被的面积急剧减少,潜在荒漠和潜在草地植被的面积增大,潜在森林植被的面积略有增加,但增加幅度不大；中国潜在自然植被面积的变化趋势同世界总趋势类似,但面积变化幅度有所不同。5)中国残存PNV面积约为624.7089万km2,占国土面积的65.07%。残存PNV典型区占残存PNV面积的16.13%,占国土面积的9.65%。典型区主要分布在人口较少、海拔较高、地形复杂的西南和西部地区,而地势较平坦的东北、华北、华中地区没有残存PNV典型区分布,东南及沿海区域有零星的分布。6)青藏高原未来气候在西部地区将呈现暖干化趋势,而在中东部绝大部分地区则呈暖湿化趋势。受此影响,未来青藏高原将出现热荒漠类组。分布于高原南部的潜在森林植被将向高原北部扩展,面积大幅增加,增加率达125%；而潜在草地植被(冻原与高山草甸、冷荒漠、半荒漠、温带湿润草地和斯太普)面积将大幅减少,减少率为25%。其中,明显受温度升高和降水再分配影响的潜在冻原与高山草甸植被类组变化最为剧烈,其面积有大幅萎缩的态势,减少率达56.82%。7)在以上研究的基础上,从系统体系结构、数据组织、系统功能设计等方面出发,设计并开发了基于ArcGIS Server和Flex等技术的全球潜在自然植被分类管理信息系统网站。更多还原

【Abstract】 Bioclimatology based vegetation classification models play an important role in studying the response of terrestrial ecosystems to global warming. On the basis of validation for classification accuracy of Potential Natural Vegetation (PNV) types simulated by the Comprehensive and Sequential Classification System (CSCS) model, using the global sharing meteorological databases and GIS spatial analysis approach, this study analyzed the spatio-temporal distribution patterns, change rules and the succession tendency of PNV types in the global and regional scales in the past100years (1901-2000) and the future50years (2001-2050); Taking China as an example, the distribution of remained PNV types and its typical areas, as well as the application of CSCS theory in monitoring and evaluating the growth condition of existing vegetation were studied; Ultimately, on the basis of the main results of this study, the global PNV management system based on CSCS model was developed.The results are as follows:1) On a global scale there are good agreements among maps produced by the CSCS method and the globally well-accepted Ramankutty&Foley PNV (RF PNV), Holdridge Life Zone (HLZ) and BIOME4PNV models. Comparing with HLZ, the CSCS model can not only predict the spatial distribution extent of tundra, desert and forest, but also identify grassland vegetation types preferably.2) Based on CSCS model,10broad vegetation categories and42classes of the potential vegetation can be identified at global scale excluding Antarctica, the area of global PNV excluding regions of permanent snow/ice cover and inland water is1.289335×108km2, which covers96.07%of the total land area of Earth. In China there are10broad vegetation categories and40classes excluding subtropical extrarid subtropical desert and tropical extrarid tropical desert. There is significant difference between the existing vegetation and PNV due to the influence of human activities. According to the magnitude of human influence in ascending order, the PNV types are tundra, desert, forest, shrubland and grassland, respectively.3) During the period of past100years, excluding the warm desert and savanna, the shift direction of the rest of8broad vegetation categories are different in the northern and southern hemispheres. The shift distance of the warm desert in the southern hemisphere was the largest among the10broad vegetation categories. Except for tundra&alpine steppe, subtropical forest and temperate humid grassland, the shift distances in the southern hemisphere for the rest of7broad PNV categories were larger than that in the northern hemisphere.4) The global land temperature has an increasing trend in future50years, especially in the northern hemisphere. The variation of global precipitation has no distinct regional regulation. Overall, the global climate would be warming and drying, but warming and humid in local area. In China, the temperature rising range would be during1.5-4.5℃and precipitation would be distinctly increasing in south and decreasing in north over the next50years. In future (i.e., by the end of the year2050), the area of global potential tundra would be decreasing rapidly, while the area of potential desert and grassland would be increasing, and the area of potential forest would be slightly increasing; the change tendency of the potential vegetation area in China is similar to that in global scale, but different in area change magnitude.5) The area of remained PNV types is624.7089×104km2in China, accounting for65.07%of total land area. The area of remained PNV typical area account for16.13%and9.65%of total remained PNV area and total land area, respectively. The typical area is mainly distributed in the southwestern and the western regions with the sparsely populated, high altitude and complex terrain. There is no typical area of remained PNV types in northeastern, northern and central areas of China, and rare distributed in southern and coastal area.6) It would be warming and drying in the west, and warming and humid in the most of eastern areas in the Qinghai-Tibet Plateau (QTP) in the future (i.e., by the end of the year2050), and the warm desert would appear in the period2001-2050. In the future, the area of forest in the QTP region would be increasing significantly, and the increasing rate reaches125%. However, the area of grassland (including tundra&alpine steppe, frigid desert, semi-desert, temperate humid grassland and steppe) would have an obvious decreasing trend, and the decreasing rate is25%. Because of affected significantly by temperature rise and precipitation redistribution, the area of tundra&alpine steppe would be reducing severely, almost56.82%area would be decreasing in the future.7) On the basis of existed studies, the system structure, data organization and system function design, the global PNV management system website was designed and developed by use of ArcGIS Server and Flex framework.更多还原