四川省杂谷脑河流域景观格局与生态脆弱性评价研究
|
摘要
岷江江上游是四川省乃至西部地区重要的生态屏障,天然林保护工程的重点实施区域和长江上游的生态屏障,且长期以来人类不合理的资源开发和利用方式造成人口-资源-环境之间的矛盾日益突出,成为区域经济社会发展的主要限制性因子,造成生态环境脆弱化且已危及到长江流域的生态安全。基于此背景下,本文以岷江上游干旱河谷—杂谷脑河流域为研究对象,借助空间信息技术GIS和RS),以1980-2006年四期遥感数据为主要数据源,结合地形分析,从时间序列上和空间尺度上对研究区域的土地利用及其景观格局进行了深入分析,运用土地利用图谱模型对土地利用类型的数量变化、时空演变过程和空间转移规律进行了剖析,在此基础上,对研究区域开展生态环境脆弱性的评价研究,探索生境脆弱性生态区的生态恢复途径和方法,以期为研究区的生态保护和可持续发展提供理论基础和科学建议,对该区域生态环境的恢复与重建、流域可持续发展等方面具有重要的战略意义。本文得出的结论如下：
‘(1)研究区域内各景观类型面积、斑块数量与优势度分布不均衡,差异明显。区域自然景观(森林景观、草地、灌木林地、岩石裸露地等)占绝对主导地位,面积总和均达到90%以上,而半自然景观、人为景观(采伐迹地、农地、居民地等)占据较小的比例,说明杂谷脑河流域以自然景观主导区域景观格局特征,人为景观则明显处于支配地位。
(2)1980-2006年杂谷脑河流域的土地利用类型变化表现为林地和未利用地面积减少；灌木林地、草地、耕地、居民地和水域面积增加的态势,林地面积减少的最大,未利用地面积减少的最小；灌木林地面积增加的最大,水域面积增加的最小
(3)采用5x5km的格网作为统计单元,并在ArcGIS的支持下,对土地利用综合程按自然断裂法(Natural Break)对土地利用综合度进行空间聚类,获取其空间分布格局,土地利用综合程度较高的区域主要在杂谷脑河流域沿岸的区域,归因于区域气候适宜,适宜人口生存,人类的干扰程度较大,呈现出不同土地利用兼备的格局现状。
(4)林地的重心坐标变化相对较为复杂,变化趋势为高纬度----低纬度----高纬度；草地重心坐标由低纬度向高纬度转变,高海拔梯度的分布频率逐渐增加；灌木林地的重心坐标由低纬度区域向高纬度、高海拔区域转变,变化较为曲折性；未利用地的重心坐标由高纬度低海拔区域向低纬度高海拔区域转变。
(5)河谷区(1980-2001年)形成森林-耕地-水域等交错的区域景观格局；2006年形成森林-灌木林地-耕地等交错的区域景观格局；亚高山区主要形成森林-灌木林地-草地等交错的亚高山针叶林带景观格局；半高山半山区主要形成草地-灌木林地-林地等交错的亚高山灌丛草甸带景观格局,该地带是森林景观向草地景观的过渡地带；高山区主要形成草地-未利用地-林地等交错的高山草甸带景观格局。
(6)居民地在各个坡度分区上都有分布,主要集中分布在平坡上,水域主要集中分布在平坡区域,居民地和水域都是在无坡向区域的面积达到最大；林地和灌木林地都是在陡坡上处于优势地位,林地主要分布于半阳坡和半阴坡区域,灌木林地在阳坡上的分布比阴坡偏多,草地的分布刚好相反,且草地和灌木林地在坡向上的分布差异明显。
(7)1980-2006年稳定型图谱单元在各类变化模式图谱单元中占比例较大,占研究区总面积的92.5245%,面积最大的图谱变化模式为“林地--林地--林地--林地”；前期变化型次之,变化面积6758.28hm2,占研究区总面积的3.1005%,面积比例最大的变化模式是“林地--灌木--灌木--灌木”；后期变化型面积比例最大的变化模式是“林地--林地--林地--灌木”；中间过渡型面积最大的变化模式是“林地--林地--灌木--灌木”；反复变化型面积变化类型比例最大的变化模式是“林地--灌木--草地--林地”；持续变化型是面积最小的图谱单元,其面积最大变化模式是“未利用地--草地--灌木林地--林地”,形成草地-灌木林地-林地等交错的亚高山灌丛草甸带的格局现状。
(8)粒度效应的研究所采用的景观指标中：斑块数量、斑块边界密度、蔓延度、分维数和景观斑块聚集度有明显的粒度效应,它们随粒度的增加表现出有规律的变化趋势；斑块边界密度、蔓延度和景观斑块的聚集度没有出现粒度突变效应；景观多样性指数和分离度在这次研究中没有出现明显的粒度效应,对粒度变化的敏感程度较低。
(9)林地景观的聚集度最大,分离度最小,其次是草地,表明林地在所有景观类型中集中分布程度最高,构成了区域的控制性景观,草地在所有景观类型中集中分布程度较高,构成了区域的主要景观类型；水域景观的聚集度指数最小,分离度较大；居民地的分维数最小。
(10)聚集度指数呈逐年减少的趋势,蔓延度的指数呈逐年增加的趋势,且平均斑块面积减小,区域景观破碎化程度加剧,结果造成景观多样性、景观均匀度呈现递增,景观优势度减小,景观多样性和优势度两指数呈现出一定的负相关关系,景观多样性和均匀度两指数基本上呈现出一定的正相关关系,反映了各景观类型所占比例差异减小,而且景观类型组成成分分布较均匀,景观类型的分布由集中趋向分散交错,景观斑块分布的均匀程度较大,景观的完整性减弱,连通性降低,景观的异质性增强。
(11)草地、未利用地作为研究区内面积比重较大的景观组分类型,其保留率一直较高；林地和灌木林地具有较高的转入率和转出率,转入率要大于转出率,且双向变化最剧烈；水域保留率高的原因主要是受林区深切峡谷的地形庇护；居民地等更容易受到人为的影响,其保留率得以维持或快速变化。
(12)从预测的面积来看,居民地面积一直呈现增加的趋势,耕地和未利用地呈现减少的趋势,草地、林地和水域呈现先减少而后增加的趋势,灌木林地呈现先增加而后减少的趋势,表明经过18年的的恢复,人工林和次生天然林的逐步演替,其生物量将会增长,其生态屏障功能,如水源涵养、水土保持、径流调节、生物多样性保护等也将逐步恢复,区域生态环境的质量得到了好转,再次表明天然林的生态恢复是一个很缓慢的过程。
(13)计算土地利用变化的终极状态概率,并将其换算为各用地类型占地面积,本文首次将终极状态概率和CA-Markov相结合,创造性地对研究区域的土地利用进行模拟和预测,其最终结果表明：耕地、灌木林地和未利用地面积呈现减少的趋势,灌木林地和未利用地减少的幅度远大于耕地,草地和林地的面积呈现增加的趋势,林地增加的幅度较大,水域的面积基本上没有发生太大的变化。
(14)强度侵蚀的主导植被类型是耕地和草地区域,中度侵蚀区域集中分布在灌木林地地带,轻度侵蚀和微度侵蚀的主导植被类型是落叶阔叶林和针阔混交林,其中微度侵蚀主要集中分布于灌木林地,无侵蚀和微度侵蚀的主要植被类型是针叶林,总体而言,从控制土壤侵蚀强度的能力大小看：针叶林>落叶阔叶林>针阔混交林>灌丛>草地>农用地。
(15)杂谷脑河流域生态脆弱性评价,可以划分为5个等级,依次分别为：EVI低于25为微度脆弱,其所占的面积比例为2.7040%；EVI在26-40之间为低度脆弱,其面积比例为36.0134%；EVI在41-55之间为中度脆弱,面积比例为29.8752%；EVI在56-70之间为高度脆弱,面积比例为19.5512%；EVI高于70为极度脆弱,面积比例为11.8562%,其中微度和低度脆弱区所占比例之和仅为38.7174,说明区域生态系统总体运行状态一般,且有31.4074%左右的面积属于重度和极度脆弱区,表明生态环境问题依然严峻,气候干燥、土壤贫瘠、干旱缺水和地质灾害贫乏是其生态脆弱性形成的自然基础,人类活动加剧尤其是不合理放牧和大量垦荒是生态脆弱性进一步加剧的重要人为根源。
(16)现有的生态脆弱性评价,几乎都是单向性的,没有对评价结果进行独立的验证,在查阅文献的基础上,本文创造性地提出了生态脆弱性评价结果的验证方法,通过网格的中心点提取平均盖度和环境脆弱度评价值,对两者进行回归分析,研究两者之间的关系,该结果表明植被指数决定了大于3/4左右的生态脆弱性区域,从某种意义上讲区域植被覆盖密度很大程度上可以决定着一个地区的生态脆弱性程度,通过研究植被盖度与生态脆弱度之间的这种关系,为我们提供了根据植被指数来验证生态脆弱度评价结果的可能性,表明评价结果能够较好地反映研究区域生态脆弱性实际情况。
(17)对不同海拔、坡向生境脆弱度的三维趋势面模型的分析可知,生境脆弱度较低的区域集中在半山半高山区域,在海拔3400-3800米之间是其主要的分布区；其次是河谷区,在该区域随着海拔的升高,脆弱度呈现减少的趋势；在亚高山区和高山区,随着海拔的升高,脆弱度呈现增加的趋势。各海拔带生境脆弱度随坡向的变化不是特别明显,总体上是呈现缓慢增加的趋势。
(18)生境脆弱度在坡度方向呈现“U,,变化曲线,而在坡向方面呈现缓慢递增的趋势,生境脆弱度较低的区域主要集中分布于坡度在21-420和坡向在60-1600之间的交叉区域,坡度在42-900区间,随着坡度的增加,生境脆弱度呈现增加的趋势；坡度600-900范围内,随着坡向的增大,evi呈现“U,,变化曲线,生境脆弱度在阳坡和半阳坡(180-2100)较小,而在阴坡和半阴坡(28-800和320-3600)较大。
(19)生境脆弱度在高程方面呈现“U,,变化曲线,而在坡度方面呈现弧度较大的下降圆弧曲线,生境脆弱度较低的区域主要集中分布于坡度在21-420的半山半高山区。
Minjiang upstream in Sichuan province and western regions is an important ecological barrier. Also it is the significant experimental subject of Natural Forest Protection Programme and the vital ecological barrier of the headwater of Yangtse River. For a long time, that the unreasonable resources development and utilization ways of human cause the contradiction among population, resources and environment increasingly. The contradiction limits the regional economic social development and gives rise to the week ecological environment which has endanger the ecological security of the Yangtse River basin. Based on this background, in this paper I choose arid valley of Minjiang River-Zagunao River as the subject, use space information technology(RS, GIS and RS), treat remote sensing data from 1980 to 2006 as the main data source and combine with terrain analysis to deeply analyse the subject of land use and landscape pattern in time series and space scale. Moreover, I dissect the quantity change of land use type, spatio:temporal evolution and space transfer law by using land use mapping model. Then by assessment of ecological environmental vulnerability in study area and exploration of ecological restoration approach in habitat vulnerability biota, in ecology protection and sustainable development of subject I expect to give the rationale and scientific advice which will effect an optimistic strategic influence in restoration and reconstruction of the ecological environment and sustainable development of river basin. The conclusions will be given in the following passages.
(1) In the study area, proportion of landscape type, quantity of plaque and dominance distribute lopsidedly. Regional natural landscape (such as forest, grassland, shrub land and rocks exposed areas) account for the absolute dominant position, taking up over 90%. However, half natural landscape and anthropogenic landscape (such as cut over land, agricultural land and settlement place) occupy smaller proportion. It illustrates that in Zagunao River the dominated regional landscape pattern is natural scenery, and man-made landscape is obviously in ascendancy.
(2) From 1980 to 2006, the transformation of land use pattern in Zagunao River area displayed in two ways. Firstly, the proportion of forest and unused land had reduced. The area of forest declined maximally, and that of unused land decreased minimally. Secondly, the acreage of shrub land, grass land, agricultural land and water area remained escalation. The increasing proportion of shrub land made up the biggest part of all,in the meanwhile, that of water area took up the smallest one.
(3) Under the support of ArcGIS, I use 5* 5km grid as the statistical unit to get the spatial clustering of synthetic degree of land use by method of Natural Break. It could give us the space distribution pattern. From that, we can see the areas of high land use comprehensive level mainly locate in the place along Zagunao River. The main reasons why coast wise of Zagunao River display the high level of land use is the climate which is suit for living and the big degree of human interference. Also that the area which expresses the pattern status of reasonable use of various lands is very important.
(4) The change of woodland barycenter coordinate is relatively complex. And it often changes from circumpolar latitude to low latitude, then back to circumpolar latitude. Grassland barycenter coordinate transforms low latitude into circumpolar latitude. Distribution frequency of high altitude gradient increases gradually. Shrubland barycenter coordinate turns low latitude into circumpolar latitude and high elevation curvilinearly. Unused land barycenter coordinate usually converts the areas of circumpolar latitude and low elevation into the places of low latitude and high elevation.
(5) Valley area (from 1980 to 2006) had formed the regional landscape pattern which is forest-cultivated land-waters interlaced. In 2006, it had shaped the regional landscape pattern of stagger forest, shrub land and agricultural land. Half-high-mountain ares mainly formed sub-alpine acerola forest landscape pattern mixing with forest, shrub land and agricultural land. However, the main pattern in half-high-mountain and semi-mountain area was grassland-shrub land-forest interlaced sub-alpine bush meadow belt landscape pattern. This place was transitional zone from forest landscape to meadow landscape. High-mountain area primarily formed the high-mountain meadow landscape pattern with grassland, unused land and forest mixed.
(6) Settlement place scatters in every slope partition, but mainly distributed in the flat slopes. The distribution of waters mainly concentrated in flat slope area. And the acreages of settlement place and waters arrived the peak in the area without slope. Both the forest and the shrub land stayed in the positive status in abrupt slope. Most forest was always in the area of partial adret slope and partial shady slope. And the distribution of shrub land was usually in the adret slope, not the shady slope. However, that of the grassland was in the opposite side. Also at the slope, the difference of distribution in grassland and shrub land is obvious.
(7) The period from 1980 to 2006 witnessed a fact that the stable mapping unit occupied large proportion of units of every variation pattern map, making up 92.5245% in whole study area. And the variation pattern map, the area of which remained the top of the table, was'forest-forest-forest-forest'. The changing area of prophase type was 6758.28hm", ranked second with taking up 3.1005% of all. In this type, the variation pattern map which had the largest acreage was'forest-shrub land-shrub land-shrub land'. However, that in later period was'forest-forest-forest-shrub land'. The model with the biggest change of area in transitional type was'forest-forest-shrub land-shrub land'. In reiterative change model the variation pattern, the acreage of which is the biggest one, was'forest-shrub land-grassland-forest'. The sustainable change model was the mapping unit with the smallest area in whole. And in that type, the model with the largest change of acreage was' unused place-grassland-shrub land-forest'. Then the pattern status of sub-alpine shrub-meadow area with grassland-shrub land-forest interlaced had formed.
(8) The study of grain-size effect chose several landscape indexes.And in these indexes, the grain-size effect was obviously found in plaques number, plaques boundary density, spread degrees,fractal dimension and landscape plaques gathered degrees which expressed the regular trend of variation.However, there was no granularity mutation effect in plaques boundary density, plaques shape index and landscape plaques gathered degrees. Landscape diversity index did not result in any obvious granularity mutation effect. So it can be said that landscape diversity index had low sensitivity of that.
(9) Woodland landscape had the high gathered degree and the low separation degree, and the second was grassland. It expressed that in whole type of landscape, forest which constituted the dominant scenery had the highest degree of concentration distribution. And grassland had the higher degree of concentration distribution, constituting the main scenery. The gathered degree of waters landscape was the smallest of all, but the separation degree of that was the biggest. The settlement place had the smallest HFD in whole.
(10) There were some facts. The gathered degree index turned to decline year by year. Spread degree index turned to increase, and average plaques area decreased. Regional landscape fragmentation degree intensified. And these result in the accumulation of landscape diversity and evenness and the decline of landscape dominance index. There was a negative correlation between landscape diversity and landscape dominance index. Also there was a positive correlation between landscape diversity and landscape evenness. It reflected that different landscape types made up the similar proportion. And the compositions of landscape types were uniformly distributed. The distribution of landscape types turned concentrated into scattered and crisscross. Most landscape patched distributed evenly. There was a decrease in landscape integrity and connectivity. Nevertheless, landscape heterogeneity increased.
(11) The retention rate of grassland and unused places which were treated as the landscape types with large proportion of area in subject always stayed in the high level. Forest and shrub land had the higher switching-in rate and switching-out rate which varied most drastically. And the switching-in rate was more bigger the the switching-out rate. The result of high retention rate of waters was protected by the deep canyon in forest. The settlement place was effected by human beings easily. So its retention rate can remain stabilization or change sharply.
(12) From the expected acreage, the area of settlement place always remain increase. The acreage of plowland and unused places turned to decline. That of the grassland, forest and waters decreased at first and then rose. However the area of shrub land stood in the opposite side. That expressed the gradual succession of plantation and secondary forest after the restoration of 18 years:the biomass live weight would increase, the ecological barrier function (such as water conservation, soil and water conservation, runoff regulation and biodiversity conservation) would recover stepwise, the quality of ecotope began to improve. Also that expressed ecological restoration of natural forest should experience a slow process.
(13) To calculate the posibility of the ultimate state of the land use which is changed, and to converse the posibility to the area of the various land types, this paper will combine the posibility of the ultimate state with the CA-Markov firstly, and creatively simulate and predict the land use of studied area, its final result shows that:the area of plow land, shrub land and used places turned to decrease. That of shrub land and used places dropt sharply than that of plow land. The area of grass land and forest had the trend of increase. And the rise of the area of forest was more rapid. The area of waters remain stabilization nearly.
(14) The dominant vegetation types in deep erosion region ware plow land and grassland. However, that in moderate erosion area mainly distributed in shrub land. The dominant vegetation type in mild erosion and micro degree erosion area was the broad leaved deciduous forest and the leaved and coniferous mixed forest. And the micro degree erosion area mainly distributed in the shrub land. The dominant vegetation type in both no erosion region and micro degree erosion region was coniferous forest. All in all, from the level of ability to control soil erosion intensity:coniferous forest>broad leaved deciduous forest>leaved and coniferous mixed forest>shrub land>grassland>agricultural land.
(15) The assessment of ecological vulnerability of the Zagunao river can be divided into five levels, as follows:
EVI belows 25 as the slight vulnerability, the proportion of the area is 2.7040%;
EVI between 26-40 as the light vulnerability, the proportion of the area is 36.0134%;
EVI between 41-55 as the medium vulnerability, the proportion of the area is 29.8752%;
EVI between 56-70 as the high vulnerability, the proportion of the area is 19.5512%;
EVI higher than 70 as the extreme vulnerability, the proportion of the area is 11.8562%. The
sum of the proportion of the area of slightly vulnerable and light vulnerable is only 38.7174. It indicates that the overall operation state of the ecological system is general, and there are the high and extreme areas about 31.4074%. It shows that the environmental problems are still severe. The dry climate, the Poor soil, the drought and the water shortage and the fregeuntly geological disaster are the natrual basis of the formation of the ecological vulnerability. Human origin of the severe ecological vulnerability. More and more human activities,especially unreasonable herd and abundant assart is the essentially.
(16) There is no independent validation for the existing assessment of ecological vulnerability which is almost one-way. Based on the literature review, this paper creatively puts forward the method to verify the results of the ecological vulnerability assessment. Then analyze the relationship of the rate of the average coverage and the value of the environmental vulnerability evaluation which are extracted by the center point of the grid. The result of it shows that the vegetation index decides almost 3/4 of the ecologically vulnerable zone. In a sense, the density of ragional vegetation can largly decise the extent of the ecological vulnerability in a region. The research of the relation between the rate of vegetation covering and the level of ecological vulnerability enables us verify the result of the assessment of ecological vulnerability according to vegetation index. Moreover, the research dedicates that assessment result can well reflect the actual situation of the regionally ecological vulnerability which is studied.
(17) From the 3D trend surface model analysis of habitat fragile degree in different elevation and aspect, the area with low habitat fragile degree mainly distributed in the half-high-mountain and semi-mountain area, especially in the area with the elevation between 3400m to 3800m. The secondary is valley area. In this area, habitat fragile degree had the negative correlation with elevation. In sub-alpine area and alpine area, with the increase of elevation, the habitat fragile degree turned to become more higher. With the change of aspect, the habitat fragile degree in different elevation did not have any obvious transformation, whereas it expressed a trend of low increase in whole.
(18) In gradient, habitat fragile degree expressed'∪'curve. Nevertheless, it expressed the trend of slow increase in aspect. The area with low habitat fragile degree mainly distribute in the cross area between 21-420 gradient and 60-1600 aspect. If gradient between 42-900, habitat fragile degree was proportional to the aspect. If gradient between 600-900, when aspect rose, EVI expressed'∪'curve. In the sunny slope and the semi-sunny slope, habitat fragile degree was more lower. However, that in the other site conditions was more higher.
(19) In elevation, habitat fragile degree expressed'∪'curve. However, in gradient it expressed decline arc curve with large radian. The area with poor habitat fragile degree mainly distributed in the half-high-mountain and semi-mountain area with gradient between21-420.
‘(1)研究区域内各景观类型面积、斑块数量与优势度分布不均衡,差异明显。区域自然景观(森林景观、草地、灌木林地、岩石裸露地等)占绝对主导地位,面积总和均达到90%以上,而半自然景观、人为景观(采伐迹地、农地、居民地等)占据较小的比例,说明杂谷脑河流域以自然景观主导区域景观格局特征,人为景观则明显处于支配地位。
(2)1980-2006年杂谷脑河流域的土地利用类型变化表现为林地和未利用地面积减少；灌木林地、草地、耕地、居民地和水域面积增加的态势,林地面积减少的最大,未利用地面积减少的最小；灌木林地面积增加的最大,水域面积增加的最小
(3)采用5x5km的格网作为统计单元,并在ArcGIS的支持下,对土地利用综合程按自然断裂法(Natural Break)对土地利用综合度进行空间聚类,获取其空间分布格局,土地利用综合程度较高的区域主要在杂谷脑河流域沿岸的区域,归因于区域气候适宜,适宜人口生存,人类的干扰程度较大,呈现出不同土地利用兼备的格局现状。
(4)林地的重心坐标变化相对较为复杂,变化趋势为高纬度----低纬度----高纬度；草地重心坐标由低纬度向高纬度转变,高海拔梯度的分布频率逐渐增加；灌木林地的重心坐标由低纬度区域向高纬度、高海拔区域转变,变化较为曲折性；未利用地的重心坐标由高纬度低海拔区域向低纬度高海拔区域转变。
(5)河谷区(1980-2001年)形成森林-耕地-水域等交错的区域景观格局；2006年形成森林-灌木林地-耕地等交错的区域景观格局；亚高山区主要形成森林-灌木林地-草地等交错的亚高山针叶林带景观格局；半高山半山区主要形成草地-灌木林地-林地等交错的亚高山灌丛草甸带景观格局,该地带是森林景观向草地景观的过渡地带；高山区主要形成草地-未利用地-林地等交错的高山草甸带景观格局。
(6)居民地在各个坡度分区上都有分布,主要集中分布在平坡上,水域主要集中分布在平坡区域,居民地和水域都是在无坡向区域的面积达到最大；林地和灌木林地都是在陡坡上处于优势地位,林地主要分布于半阳坡和半阴坡区域,灌木林地在阳坡上的分布比阴坡偏多,草地的分布刚好相反,且草地和灌木林地在坡向上的分布差异明显。
(7)1980-2006年稳定型图谱单元在各类变化模式图谱单元中占比例较大,占研究区总面积的92.5245%,面积最大的图谱变化模式为“林地--林地--林地--林地”；前期变化型次之,变化面积6758.28hm2,占研究区总面积的3.1005%,面积比例最大的变化模式是“林地--灌木--灌木--灌木”；后期变化型面积比例最大的变化模式是“林地--林地--林地--灌木”；中间过渡型面积最大的变化模式是“林地--林地--灌木--灌木”；反复变化型面积变化类型比例最大的变化模式是“林地--灌木--草地--林地”；持续变化型是面积最小的图谱单元,其面积最大变化模式是“未利用地--草地--灌木林地--林地”,形成草地-灌木林地-林地等交错的亚高山灌丛草甸带的格局现状。
(8)粒度效应的研究所采用的景观指标中：斑块数量、斑块边界密度、蔓延度、分维数和景观斑块聚集度有明显的粒度效应,它们随粒度的增加表现出有规律的变化趋势；斑块边界密度、蔓延度和景观斑块的聚集度没有出现粒度突变效应；景观多样性指数和分离度在这次研究中没有出现明显的粒度效应,对粒度变化的敏感程度较低。
(9)林地景观的聚集度最大,分离度最小,其次是草地,表明林地在所有景观类型中集中分布程度最高,构成了区域的控制性景观,草地在所有景观类型中集中分布程度较高,构成了区域的主要景观类型；水域景观的聚集度指数最小,分离度较大；居民地的分维数最小。
(10)聚集度指数呈逐年减少的趋势,蔓延度的指数呈逐年增加的趋势,且平均斑块面积减小,区域景观破碎化程度加剧,结果造成景观多样性、景观均匀度呈现递增,景观优势度减小,景观多样性和优势度两指数呈现出一定的负相关关系,景观多样性和均匀度两指数基本上呈现出一定的正相关关系,反映了各景观类型所占比例差异减小,而且景观类型组成成分分布较均匀,景观类型的分布由集中趋向分散交错,景观斑块分布的均匀程度较大,景观的完整性减弱,连通性降低,景观的异质性增强。
(11)草地、未利用地作为研究区内面积比重较大的景观组分类型,其保留率一直较高；林地和灌木林地具有较高的转入率和转出率,转入率要大于转出率,且双向变化最剧烈；水域保留率高的原因主要是受林区深切峡谷的地形庇护；居民地等更容易受到人为的影响,其保留率得以维持或快速变化。
(12)从预测的面积来看,居民地面积一直呈现增加的趋势,耕地和未利用地呈现减少的趋势,草地、林地和水域呈现先减少而后增加的趋势,灌木林地呈现先增加而后减少的趋势,表明经过18年的的恢复,人工林和次生天然林的逐步演替,其生物量将会增长,其生态屏障功能,如水源涵养、水土保持、径流调节、生物多样性保护等也将逐步恢复,区域生态环境的质量得到了好转,再次表明天然林的生态恢复是一个很缓慢的过程。
(13)计算土地利用变化的终极状态概率,并将其换算为各用地类型占地面积,本文首次将终极状态概率和CA-Markov相结合,创造性地对研究区域的土地利用进行模拟和预测,其最终结果表明：耕地、灌木林地和未利用地面积呈现减少的趋势,灌木林地和未利用地减少的幅度远大于耕地,草地和林地的面积呈现增加的趋势,林地增加的幅度较大,水域的面积基本上没有发生太大的变化。
(14)强度侵蚀的主导植被类型是耕地和草地区域,中度侵蚀区域集中分布在灌木林地地带,轻度侵蚀和微度侵蚀的主导植被类型是落叶阔叶林和针阔混交林,其中微度侵蚀主要集中分布于灌木林地,无侵蚀和微度侵蚀的主要植被类型是针叶林,总体而言,从控制土壤侵蚀强度的能力大小看：针叶林>落叶阔叶林>针阔混交林>灌丛>草地>农用地。
(15)杂谷脑河流域生态脆弱性评价,可以划分为5个等级,依次分别为：EVI低于25为微度脆弱,其所占的面积比例为2.7040%；EVI在26-40之间为低度脆弱,其面积比例为36.0134%；EVI在41-55之间为中度脆弱,面积比例为29.8752%；EVI在56-70之间为高度脆弱,面积比例为19.5512%；EVI高于70为极度脆弱,面积比例为11.8562%,其中微度和低度脆弱区所占比例之和仅为38.7174,说明区域生态系统总体运行状态一般,且有31.4074%左右的面积属于重度和极度脆弱区,表明生态环境问题依然严峻,气候干燥、土壤贫瘠、干旱缺水和地质灾害贫乏是其生态脆弱性形成的自然基础,人类活动加剧尤其是不合理放牧和大量垦荒是生态脆弱性进一步加剧的重要人为根源。
(16)现有的生态脆弱性评价,几乎都是单向性的,没有对评价结果进行独立的验证,在查阅文献的基础上,本文创造性地提出了生态脆弱性评价结果的验证方法,通过网格的中心点提取平均盖度和环境脆弱度评价值,对两者进行回归分析,研究两者之间的关系,该结果表明植被指数决定了大于3/4左右的生态脆弱性区域,从某种意义上讲区域植被覆盖密度很大程度上可以决定着一个地区的生态脆弱性程度,通过研究植被盖度与生态脆弱度之间的这种关系,为我们提供了根据植被指数来验证生态脆弱度评价结果的可能性,表明评价结果能够较好地反映研究区域生态脆弱性实际情况。
(17)对不同海拔、坡向生境脆弱度的三维趋势面模型的分析可知,生境脆弱度较低的区域集中在半山半高山区域,在海拔3400-3800米之间是其主要的分布区；其次是河谷区,在该区域随着海拔的升高,脆弱度呈现减少的趋势；在亚高山区和高山区,随着海拔的升高,脆弱度呈现增加的趋势。各海拔带生境脆弱度随坡向的变化不是特别明显,总体上是呈现缓慢增加的趋势。
(18)生境脆弱度在坡度方向呈现“U,,变化曲线,而在坡向方面呈现缓慢递增的趋势,生境脆弱度较低的区域主要集中分布于坡度在21-420和坡向在60-1600之间的交叉区域,坡度在42-900区间,随着坡度的增加,生境脆弱度呈现增加的趋势；坡度600-900范围内,随着坡向的增大,evi呈现“U,,变化曲线,生境脆弱度在阳坡和半阳坡(180-2100)较小,而在阴坡和半阴坡(28-800和320-3600)较大。
(19)生境脆弱度在高程方面呈现“U,,变化曲线,而在坡度方面呈现弧度较大的下降圆弧曲线,生境脆弱度较低的区域主要集中分布于坡度在21-420的半山半高山区。
Minjiang upstream in Sichuan province and western regions is an important ecological barrier. Also it is the significant experimental subject of Natural Forest Protection Programme and the vital ecological barrier of the headwater of Yangtse River. For a long time, that the unreasonable resources development and utilization ways of human cause the contradiction among population, resources and environment increasingly. The contradiction limits the regional economic social development and gives rise to the week ecological environment which has endanger the ecological security of the Yangtse River basin. Based on this background, in this paper I choose arid valley of Minjiang River-Zagunao River as the subject, use space information technology(RS, GIS and RS), treat remote sensing data from 1980 to 2006 as the main data source and combine with terrain analysis to deeply analyse the subject of land use and landscape pattern in time series and space scale. Moreover, I dissect the quantity change of land use type, spatio:temporal evolution and space transfer law by using land use mapping model. Then by assessment of ecological environmental vulnerability in study area and exploration of ecological restoration approach in habitat vulnerability biota, in ecology protection and sustainable development of subject I expect to give the rationale and scientific advice which will effect an optimistic strategic influence in restoration and reconstruction of the ecological environment and sustainable development of river basin. The conclusions will be given in the following passages.
(1) In the study area, proportion of landscape type, quantity of plaque and dominance distribute lopsidedly. Regional natural landscape (such as forest, grassland, shrub land and rocks exposed areas) account for the absolute dominant position, taking up over 90%. However, half natural landscape and anthropogenic landscape (such as cut over land, agricultural land and settlement place) occupy smaller proportion. It illustrates that in Zagunao River the dominated regional landscape pattern is natural scenery, and man-made landscape is obviously in ascendancy.
(2) From 1980 to 2006, the transformation of land use pattern in Zagunao River area displayed in two ways. Firstly, the proportion of forest and unused land had reduced. The area of forest declined maximally, and that of unused land decreased minimally. Secondly, the acreage of shrub land, grass land, agricultural land and water area remained escalation. The increasing proportion of shrub land made up the biggest part of all,in the meanwhile, that of water area took up the smallest one.
(3) Under the support of ArcGIS, I use 5* 5km grid as the statistical unit to get the spatial clustering of synthetic degree of land use by method of Natural Break. It could give us the space distribution pattern. From that, we can see the areas of high land use comprehensive level mainly locate in the place along Zagunao River. The main reasons why coast wise of Zagunao River display the high level of land use is the climate which is suit for living and the big degree of human interference. Also that the area which expresses the pattern status of reasonable use of various lands is very important.
(4) The change of woodland barycenter coordinate is relatively complex. And it often changes from circumpolar latitude to low latitude, then back to circumpolar latitude. Grassland barycenter coordinate transforms low latitude into circumpolar latitude. Distribution frequency of high altitude gradient increases gradually. Shrubland barycenter coordinate turns low latitude into circumpolar latitude and high elevation curvilinearly. Unused land barycenter coordinate usually converts the areas of circumpolar latitude and low elevation into the places of low latitude and high elevation.
(5) Valley area (from 1980 to 2006) had formed the regional landscape pattern which is forest-cultivated land-waters interlaced. In 2006, it had shaped the regional landscape pattern of stagger forest, shrub land and agricultural land. Half-high-mountain ares mainly formed sub-alpine acerola forest landscape pattern mixing with forest, shrub land and agricultural land. However, the main pattern in half-high-mountain and semi-mountain area was grassland-shrub land-forest interlaced sub-alpine bush meadow belt landscape pattern. This place was transitional zone from forest landscape to meadow landscape. High-mountain area primarily formed the high-mountain meadow landscape pattern with grassland, unused land and forest mixed.
(6) Settlement place scatters in every slope partition, but mainly distributed in the flat slopes. The distribution of waters mainly concentrated in flat slope area. And the acreages of settlement place and waters arrived the peak in the area without slope. Both the forest and the shrub land stayed in the positive status in abrupt slope. Most forest was always in the area of partial adret slope and partial shady slope. And the distribution of shrub land was usually in the adret slope, not the shady slope. However, that of the grassland was in the opposite side. Also at the slope, the difference of distribution in grassland and shrub land is obvious.
(7) The period from 1980 to 2006 witnessed a fact that the stable mapping unit occupied large proportion of units of every variation pattern map, making up 92.5245% in whole study area. And the variation pattern map, the area of which remained the top of the table, was'forest-forest-forest-forest'. The changing area of prophase type was 6758.28hm", ranked second with taking up 3.1005% of all. In this type, the variation pattern map which had the largest acreage was'forest-shrub land-shrub land-shrub land'. However, that in later period was'forest-forest-forest-shrub land'. The model with the biggest change of area in transitional type was'forest-forest-shrub land-shrub land'. In reiterative change model the variation pattern, the acreage of which is the biggest one, was'forest-shrub land-grassland-forest'. The sustainable change model was the mapping unit with the smallest area in whole. And in that type, the model with the largest change of acreage was' unused place-grassland-shrub land-forest'. Then the pattern status of sub-alpine shrub-meadow area with grassland-shrub land-forest interlaced had formed.
(8) The study of grain-size effect chose several landscape indexes.And in these indexes, the grain-size effect was obviously found in plaques number, plaques boundary density, spread degrees,fractal dimension and landscape plaques gathered degrees which expressed the regular trend of variation.However, there was no granularity mutation effect in plaques boundary density, plaques shape index and landscape plaques gathered degrees. Landscape diversity index did not result in any obvious granularity mutation effect. So it can be said that landscape diversity index had low sensitivity of that.
(9) Woodland landscape had the high gathered degree and the low separation degree, and the second was grassland. It expressed that in whole type of landscape, forest which constituted the dominant scenery had the highest degree of concentration distribution. And grassland had the higher degree of concentration distribution, constituting the main scenery. The gathered degree of waters landscape was the smallest of all, but the separation degree of that was the biggest. The settlement place had the smallest HFD in whole.
(10) There were some facts. The gathered degree index turned to decline year by year. Spread degree index turned to increase, and average plaques area decreased. Regional landscape fragmentation degree intensified. And these result in the accumulation of landscape diversity and evenness and the decline of landscape dominance index. There was a negative correlation between landscape diversity and landscape dominance index. Also there was a positive correlation between landscape diversity and landscape evenness. It reflected that different landscape types made up the similar proportion. And the compositions of landscape types were uniformly distributed. The distribution of landscape types turned concentrated into scattered and crisscross. Most landscape patched distributed evenly. There was a decrease in landscape integrity and connectivity. Nevertheless, landscape heterogeneity increased.
(11) The retention rate of grassland and unused places which were treated as the landscape types with large proportion of area in subject always stayed in the high level. Forest and shrub land had the higher switching-in rate and switching-out rate which varied most drastically. And the switching-in rate was more bigger the the switching-out rate. The result of high retention rate of waters was protected by the deep canyon in forest. The settlement place was effected by human beings easily. So its retention rate can remain stabilization or change sharply.
(12) From the expected acreage, the area of settlement place always remain increase. The acreage of plowland and unused places turned to decline. That of the grassland, forest and waters decreased at first and then rose. However the area of shrub land stood in the opposite side. That expressed the gradual succession of plantation and secondary forest after the restoration of 18 years:the biomass live weight would increase, the ecological barrier function (such as water conservation, soil and water conservation, runoff regulation and biodiversity conservation) would recover stepwise, the quality of ecotope began to improve. Also that expressed ecological restoration of natural forest should experience a slow process.
(13) To calculate the posibility of the ultimate state of the land use which is changed, and to converse the posibility to the area of the various land types, this paper will combine the posibility of the ultimate state with the CA-Markov firstly, and creatively simulate and predict the land use of studied area, its final result shows that:the area of plow land, shrub land and used places turned to decrease. That of shrub land and used places dropt sharply than that of plow land. The area of grass land and forest had the trend of increase. And the rise of the area of forest was more rapid. The area of waters remain stabilization nearly.
(14) The dominant vegetation types in deep erosion region ware plow land and grassland. However, that in moderate erosion area mainly distributed in shrub land. The dominant vegetation type in mild erosion and micro degree erosion area was the broad leaved deciduous forest and the leaved and coniferous mixed forest. And the micro degree erosion area mainly distributed in the shrub land. The dominant vegetation type in both no erosion region and micro degree erosion region was coniferous forest. All in all, from the level of ability to control soil erosion intensity:coniferous forest>broad leaved deciduous forest>leaved and coniferous mixed forest>shrub land>grassland>agricultural land.
(15) The assessment of ecological vulnerability of the Zagunao river can be divided into five levels, as follows:
EVI belows 25 as the slight vulnerability, the proportion of the area is 2.7040%;
EVI between 26-40 as the light vulnerability, the proportion of the area is 36.0134%;
EVI between 41-55 as the medium vulnerability, the proportion of the area is 29.8752%;
EVI between 56-70 as the high vulnerability, the proportion of the area is 19.5512%;
EVI higher than 70 as the extreme vulnerability, the proportion of the area is 11.8562%. The
sum of the proportion of the area of slightly vulnerable and light vulnerable is only 38.7174. It indicates that the overall operation state of the ecological system is general, and there are the high and extreme areas about 31.4074%. It shows that the environmental problems are still severe. The dry climate, the Poor soil, the drought and the water shortage and the fregeuntly geological disaster are the natrual basis of the formation of the ecological vulnerability. Human origin of the severe ecological vulnerability. More and more human activities,especially unreasonable herd and abundant assart is the essentially.
(16) There is no independent validation for the existing assessment of ecological vulnerability which is almost one-way. Based on the literature review, this paper creatively puts forward the method to verify the results of the ecological vulnerability assessment. Then analyze the relationship of the rate of the average coverage and the value of the environmental vulnerability evaluation which are extracted by the center point of the grid. The result of it shows that the vegetation index decides almost 3/4 of the ecologically vulnerable zone. In a sense, the density of ragional vegetation can largly decise the extent of the ecological vulnerability in a region. The research of the relation between the rate of vegetation covering and the level of ecological vulnerability enables us verify the result of the assessment of ecological vulnerability according to vegetation index. Moreover, the research dedicates that assessment result can well reflect the actual situation of the regionally ecological vulnerability which is studied.
(17) From the 3D trend surface model analysis of habitat fragile degree in different elevation and aspect, the area with low habitat fragile degree mainly distributed in the half-high-mountain and semi-mountain area, especially in the area with the elevation between 3400m to 3800m. The secondary is valley area. In this area, habitat fragile degree had the negative correlation with elevation. In sub-alpine area and alpine area, with the increase of elevation, the habitat fragile degree turned to become more higher. With the change of aspect, the habitat fragile degree in different elevation did not have any obvious transformation, whereas it expressed a trend of low increase in whole.
(18) In gradient, habitat fragile degree expressed'∪'curve. Nevertheless, it expressed the trend of slow increase in aspect. The area with low habitat fragile degree mainly distribute in the cross area between 21-420 gradient and 60-1600 aspect. If gradient between 42-900, habitat fragile degree was proportional to the aspect. If gradient between 600-900, when aspect rose, EVI expressed'∪'curve. In the sunny slope and the semi-sunny slope, habitat fragile degree was more lower. However, that in the other site conditions was more higher.
(19) In elevation, habitat fragile degree expressed'∪'curve. However, in gradient it expressed decline arc curve with large radian. The area with poor habitat fragile degree mainly distributed in the half-high-mountain and semi-mountain area with gradient between21-420.
引文
[1]乔青.川滇农牧交错带景观格局与生态脆弱性评价[D].北京：北京林业大学博士论文,2007.
[2]王成.局地土地利用变化及景观格局研究--以重庆市沙坪坝区为例[D].重庆：西南大学博士论文,2007.
[3]刘京涛.岷江上游植被蒸散时空格局及其模拟研究[D].北京：中国林业科学研究院博士论文,2006：1-2.
[4]严代碧.眠江上游干旱河谷区退化植被特征及其恢复重建的研究[D].北京：北京林业大学博士论文,2006：2-4.
[5]刘茹,张晶晶,雷蕾.岷江上游生态脆弱性的模糊评价[J].资源开发与市场,2006,22(6)：500-533.
[6]叶正伟.基于生态脆弱性的淮河流域水土保持策略研究[J].水土保持通报,2007,27(3)：141-145.
[7]周毅,李旋旗,赵景柱.中国典型生态脆弱带与贫困相关性分析[J].北京理工大学学报.2008,28(3)：260-262.
[8]邬建国.景观生态学--概念与理论[J].生态学杂志,2000,19(1)：42-43.
[9]李娟娟.上海城市景观格局演变及其生态安全影响研究[D].上海：复旦大学博士论文,2007.
[10]缪宁.川西亚高山红桦---岷江冷杉天然次生林的空间格局分析[D].北京：中国林业科学研究院博士论文,2009.
[11]高凯.多尺度的景观空间关系及景观格局生态效应的变化研究[D].武汉：华中农业大学博士论文,2010.
[12]徐广才,康慕谊,贺丽娜等.生态脆弱性及其研究进展[J].生态学报,2009,29(5)：2578-2580.
[13]邓向瑞.北京山区景观格局格局及其尺度效应研究[D].北京：北京林业大学博士论文,2007：2-4.
[14]王景伟,王海泽.景观指数在景观格局描述中的应用--以鞍山大麦科湿地自然保护区为例[J].水土保持研究,2006,13(2)：230-231.
[15]赵福强,代力民,于大炮等.长白山露水河林业局森林景观格局动态[J].应用生态学报,2010,21(5)：1180-1184.
[16]刘宇,吕一河,傅伯杰.景观格局--土壤侵蚀研究中景观指数的意义解释及局限性[J].生态学报,2011,31(1)0267-0275.
[17]万荣荣,杨桂山.太湖流域土地利用与景观格局演变研究[J].应用生态学报,2005,16(3)：475-476.
[18]张志明,罗亲普,王文礼.2D与3D景观指数测定山区植被景观格局变化对比分析[J].生态学报,2010,30(21)：58865893.
[19]马燕飞,沙占江,郭丽红等.基于NDVI及DEM的青海湖北岸景观格局空间自相关分析[J].遥感应用,2010,6(10)：95-119.
[20]王瑶,刘前进,于兴修.沂蒙山区土壤侵蚀垂直景观格局的分形特征[J].农业工程学报,2010,26(11)：304-309.
[21]李玉凤,刘红玉,郑囡等.基于功能分类的城市湿地公园景观格局--以西溪湿地公园为例[J].生态学报,2011,31(4)：1021-1028.
[22]张盼盼,胡远满,肖笃宁等.地形因子对喀斯特高原山区潜在石漠化景观格局变化的影响分析[J].土壤通报,2011,41(6)：1305-1310.
[23]赵冰,张杰,孙希华.基于GIS的淮河流域桐柏—大别山区生态脆弱性评价[J].水土保持研究,2009,16(3)：135-138.
[24]樊哲文,刘木生,沈文清.江西省生态脆弱性现状GIS模型评价[J].地球信息科学学报,2009,11(2)：202-208.
[25]宋国利,李玉宝,付春雷等.基于RS与GIS的乐清湾湿地景观格局变化分析[J].东北林业大学学报,2010,38(12)：80-83.
[26]周永娟,仇江啸,王姣等.三峡库区消落带生态环境脆弱性评价[J].生态学报,2010,30(24)：6727-6728.
[27]付博,姜琦刚,任春颖等.扎龙湿地生态脆弱性评价与分析[J].干旱区资源与环境,2011,25(1)：50-52.
[28]万星,周建中,丁晶等.岷江上游生态脆弱性综合评价的集对分析[J].中国农村水利水电,2006,12：33-39.
[29]孔博,陶和平,李爱农,等.汶川地震灾区生态脆弱性评价研究[J].水土保持通报,2010,30(6)：180-181.
[30]卢冬爱,谈树成,夏既胜.基于景观格局和水土流失敏感性的大理市生态脆弱性分析[J].云南地理环境研究,2009,21(2)：92-96.
[31]李滨勇,陈海滨,唐海萍等,基于AHP和模糊综合评判法的北疆各地州生态脆弱性评价[J.北京师范大学学报(自然科学版),2010,46(2)：197-201.
[32]戴其文,赵雪雁.生态补偿机制中若干关键科学问题--以甘南藏族自治州草地生态系统为例[J].地理学报,2010,65(4)：495-505.
[33]乔青,高吉喜,王维等.生态脆弱性综合评价方法与应用[J].环境科学研究,2008,21(5)：117-123.
[34]王丽婧,席春燕,付青等.基于景观格局的三峡库区生态脆弱性评价[J].环境科学研究,2010,23(10)：1269-1270.
[35]余坤勇,刘健,,黄维友等.基于GIS技术的闽江流域生态脆弱性分析[J].江西农业大学学报,2009,31(3)：568-573.
[36]蔡海生,刘木生,陈美球等.基于GIS的江西省生态环境脆弱性动态评价[J].水土保持通报,2009,29(5)：190-196.
[37]李红丽,智颖飙,张荷亮等.新疆生态足迹与环境压力的时空分异[J].生态学报,2010,30(17)：4676-4684.
[38]周彬,余新晓,陈丽华等.北京山区森林景观格局与环境关系的CCA研究[J].水土保持通报,2010,30(6)：148-152.
[39]刘宏文.山西省庞泉沟森林群落特征与景观格局特征分析的研究[D].北京：北京林业大学博士论文,2007：1-2.
[40]姚建,张军,朱莞.岷江上游生态脆弱性驱动力分析[J].资源开发与市场,2007,23(11)：977-982.
[41]郝璐,王静爱,张化.北方草地畜牧业生态系统健康综合评价与诊断[J].生态学报,2008,28(4)：1456-1465.
[42]邱彭华,徐颂军,谢跟踪等.基于景观格局和生态敏感性的海南西部地区生态脆弱性分析[J].生态学报,2007,27(4)：1257-1264.
[43]彭立,苏春江,徐云等.岷江上游生态环境现状、存在问题及治理对策[J].江西农业大学学报(社会科学版),2007,6(1)：80-84.
[44]余妹萍,刘国东,吴嫒等.岷江上游日径流过程分维分析及其生态脆弱性表征[J].西南民族大学学报(自然利学版),2005,31(1)：80-84.
[45]宫兆宁,张翼然,宫辉力.北京湿地景观格局演变特征与驱动机制分析[J].地理学报,2011,66(1)：77-88
[46]刘艳芬,张杰,马毅.1995--1999年黄河三角洲东部自然保护区湿地景观格局变化[J].应用生态学报,2010,21(11)：2904-2911.
[47]刘世荣,孙鹏森等.岷江上游生态水文图集[M].北京：中国地图出版社,2008.
[48]史培军,宫鹏,李晓兵,等.土地利用／覆盖变化研究的方法与实践[M].北京：科学出版社,2000.
[49]刘纪远.中国资源环境遥感宏观调查与动态研究[M].北京：中国科学技术出版社,1996.
[50]杜培军.遥感原理与应用[M].北京：中国矿业大学出版社,2006.
[51]梅安新,彭望禄,秦其明,等.遥感导论[M].北京：高等教育出版社,2001.
[52]章孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州：浙江大学出版社,1997.
[53]张永生.遥感图像信息系统[M].北京：科学出版社,2000.
[54]姜小光,干长耀,王成.成像光谱数据的光谱信息特点及最佳波段的选择-以北京顺义区为例[J].干旱地区研究,2000,23(3)：214-220.
[55]刘建平,赵英时.高光谱遥感数据解译的最佳波段选择方法研究[J].中国科学院研究生院学报,1999,6(2)：153-161.
[56]鲍文东.基于GIS的土地利用动态变化研究[D].济南：山东科技大学博士论文,2007.
[57]李虎杰.岷江上游生态环境建设与经济可持续发展[J].四川环境,2001.20(4)：51-56.
[58]刘洋.峨江源头区植被景观与流城土城怪蚀的动态相关性[D].成都：中国科学院成都生物研究所博士论文,2007.
[59]何国金,李克鲁,胡德永,等.多卫星遥感数据的信息融合：理论、方法与实践[J].中国图像图形学报,1999,7(9)：744-750.
[60]J.G..Liu.Smoothing Filter-based Intensity Modulation:a spectral preserve imagefusion Techninue for improving spatial details[J].International Journal of RemoteSensing,2000, 21(18):3461:3472.
[61]贾凌,都金康,赵萍,等.基于TM的海南省土地利用／覆盖动态变化的遥感监测与分析[J].国土资源遥感,2003,(1)：22-26.
[62]孙家柄,舒宁,关泽群.遥感原理,方法和应用[M].北京：测绘出版社,1997.
[63]朱述龙,张占睦.遥感图像获取与分析[M].北京：科学出版社,2000.
[64]许榕峰.基于遥感的龙海市土地利用变化动态监测与专题研究[D].福州：福州大学硕士论文,2003.
[65]Congalton R G.A review of assessing the accuracy of classification of remotely sensed data[J].Remote Sensing of Environment,1991,37:35-46.
[66]Lucas I F,Frans J M,Wel V D.Accuracy assessment of satellite derived land-cover data:a review[J].Photogrammetric Engineering&Remote Sensing,1994,60(4):410-432.
[67]王妍.基于DEM的地形信息提取与景观空间格局分析[D].重庆：西南大学硕士论文,2007.
[68]Miller,C.and Laflamme.The Digital Terrain Model-Theory and Application[J]. Photogrammetric Engineering.1958,(24):433-442.
[69]Makarovic, B. Progressive Sample for DTM[J]. ITC Jounal,1973,40:397-416.
[70]黎景良,后斌,危双峰,等.基于DEM的广东省山区土地利用变化分析[J].测绘通报,2007,(6)：53-57.
[71]Tang GA. A Research on the Accuracy of DEM Beijing[M.] New York:Science Press,1996.
[72]Ebish, K.Effect of Digital Elevation Resolution on the Properties of Contours[J]. Technical Paper, ASP-ACSM Fall Convention,1984,4():424-434.
[73]Hannah,M.Error detection and correction in digital terrain models[J]. Photogrammetric Engineering and Remote sensing,1981,47(l):36-69.
[74]张云红,王道杰,江晓波,等.岷江上游土地利用变化及其对生态环境的影响[J].西南大学学报(自然科学版),2010,32(3)：102-107.
[75]亓兴兰.基于3S技术闽江流域土地利用演化及其空间格局过程研究[D].福州:福建农林硕士大学,2006.
[76]杨丽,谢高地,甄霖等.径河流域土地利用格局的时空变化分析[J].资源科学,2005,27(4)：27-32.
[77]朱会义,李秀彬,何书金,等.环渤海地区土地利用的时空变化分析[J].地理学报,2001,56(3)：253-260.
[78]王秀兰.包玉海.土地利用动态变化研究方法探讨[J].地理科学进展,1999,18(1)：81-87.
[79]于兴修.中尺度流域土地利用灌被变化及其水环境效应研究—以太湖上游西菩溪流域为例[D].南京：南京中国科学院南京地理与湖泊研究所博士学位论文,2003.
[80]郑国强.基于RS的无锡市土地利用/覆被变化信息提取及其时空演变研究[D].南京：南京大学博士学位论文,2004.
[81]刘纪远,布和敖斯尔.中国土地利用变化现代过程时空特征的研究--基于卫星遥感数据[J].第四纪研究,2000,2(3)：229-239.
[82]刘庆.亚高山针叶林生态学研究[M].成都：四川大学出版社,2002.
[83]张明阳,王克林,刘会玉,等.喀斯特生态脆弱区桂西北土地变化特征[J].生态学报,2009,29(6)：3106-3116.
[84]刘耀林,焦利民.顾及尺度效应和景观格局的土地利用数据综合指标研究[J].测绘学报,2009,38(6)：549-554.
[85]王思远,刘纪远,张增祥,等.中国土地利用时空特征分析[J].地理学报,2001,56(6)：631-639.
[86]朱会义,李秀彬.关于区域土地利用变化指数模型方法的讨论[J].地理学报,2003,58(5)：643-650.
[87]樊玉山,刘纪元主编.西藏自治区土地利用[M].北京：科学出版社,1994.
[88]任志远,张艳芳.土地利用变化与生态安全评价[M].北京：科学出版社,2003.
[89]钟成.集成LiDAR和遥感影像生成高质量真正射影像研究[J].测绘学报,2011,40(1)
[90]向双.岷江江上游干旱河谷葡萄的土宜和生态布局研究[D].四川雅安：四川农业大学,2002.
[91]樊宏.岷江上游近50 a土地覆被的变化趋势[J].山地学报,2002,20(1)：64-69
[92]沈泽昊,岳超,赵俊,等.南水北调中线水源区土地利用/土地覆被的空间格局[J].地理学报,2006,61(6)：633-644.
[93]李增加,马友鑫,李红梅,等.西双版纳土地利用/覆盖变化与地形的关系[J].植物生态学报,2008,32(5)：1091-1103.
[94]赖长鸿.基于GIS的川西道孚林区森林景观格局研究[D.四川雅安：四川农业大学,2006.
[95]焦超卫,赵牡丹,汤国安,等.基于GIS的植被空间格局特征与地形因子的相关关系[J].水土保持通报,2005,25(6)：19-23.
[96]Buheaosier, Liu Jiyuan.Using Remote SensedStudy on Land Cover Change in China Based on Seasonal and Longitudinal CharacteristicsData. Asahikaw:IGU-LUCC Press,1997, 20-24
[97]谢高地.人口增长胁迫下的全球地利用变化研究[J].自然资源学报,1999,14(3)：233-241.
[98]王良健,包浩生.基于遥感和GIS的区域土地利用变化的动态监测研究[J].经济地 理,2000,20(2)：47-50.
[99]Lucas J F J, Accuracy assessment of satellite derived land cover data:a review. PE&RS, 1994,60(4):410-432.
[100]Taylor J C, Brewer T R, A C Bird. Monitoring landscape change in the national parks of England and Wales using aerial photo interpretation and GIS. Int. J. Remote Sensing, 2000,21(13):2737-2752.
[101]J F Mas. Monitoring land-cover changes./nt. J. Remote Sensing,1999,20(1):139-152.
[102]龚文峰.基于RS和GIS松潘地区天然林景观动态过程与分类研究[D]。哈尔滨,东北林业大学博士论文,2007.
[103]廖克.地球信息图谱与数字地球[J].地理研究,2001,20(1)：55-61.
[104]叶庆华,刘高焕,陆洲,等.基于GIS的时空复合体--土地利用图谱模型研究方法[J].地理科学进展,2002,21(4)：349-357
[105]张国坤.新开河流景观格局动态变化及图谱分析[D].长春：东北师范大学博士论文,2008.
[106]姚建.峨江上游生态脆弱性分析及评价[D].成都：四川大学博士论文,2004.
[107]包维楷,张镱锂,王乾,等.大渡河上游林区森林资源退化及其恢复与重建[J].山地学报,2002,20(2)：194-198.
[108]阎建忠,张镜铿等.基于植被演替的土地覆被变化研究—大渡河上游的森林采伐、更新和退化[J].中国科学D辑,地球科学,2005,35(11)：1060-1073.
[109]Forman R T T.. Some general principles of landscape and regional ecology. Landscape. 1995,10(3):133-142.
[110]Soberg S,Jain A K,Taxt T.Multisource Classification of remotely sensed data:Fusion of LandsatTM and SAR Images[J].IEEETransactionon Geoscience and Remote Sensing,1994,32(4):768-778
[111]PAN Daiyuan, G Domon, D Marceau, et al. Spatial pattern of coniferous and deciduous forest patchesin an Eastern North America agricultural landscape:the influence of land use and physical amibutes[J].Landscape Ecoloav.2001.16:99-110.
[112]TumerMQRuseherCL.ChangesinlandseaPePatternsinGeorgia,USA[J],LandseaPeEeol, 1988,1(4):241-251.
[113]TumerMG. Land sea Peeh an gesinni erural counties of Georgia[JlPhotograEng RemoteSensing,1990,56:379-386.
[114]郭晋平,王俊田,李世光.关帝山林区景观要素沿环境梯度分布趋势的研究[J]植物生态学报,2000,24(2)：135-140.
[115]郭晋平,薛俊杰,李志强,等.森林景观恢复过程中景观要素斑块规模的变化[J].生态学报,2000,20(2)：218-223.
[116]钱乐祥,陶黎新.福建森林植被景观空间格局分异研究[J].热带地理,199717(4)： 397-404
[117]Ehrlich D, Lambin E F. Broad scale land-cover classification and Internal climatic change.International journal of Remote Sensing.1996,17:845-862.
[118]王让会,张慧芝,游先祥,等.塔里木河流域生态景观格局的遥感信息提取与分析[J].北京林业大学学报,2003,25(2)：43-47.
[119]毕晓丽,周睿,刘丽娟,等.径河沿岸景观格局梯度变化及驱动力分析[J].生态学报,2005,25(5)1041-1048.
[120]贾宝全,王国柱,慈龙骏.景观格局分布特征的区域差异分析—以鄂尔多斯高原伊金霍洛旗为例[J].干旱区研究,2003,20(3)：168-174.
[121]马媛,师庆东,潘晓玲.西部干旱区生态景观格局动态分[J].干旱区地理,2004,27(4)：516-519.
[122]熊金国,王世新,周艺等.利用景观格局指数的空间分辨率对水域面积提取影响的分析[J].武汉大学学报·信息科学版,2011,36(1)：98-103.
[123]杨德伟,张云红,王道杰.等岷江上游土地利用的时空分异[J].山地学报,2009,27(5)：579-584.
[124]高小红,王一谋,杨国靖.基于RS与GIS的榆林地区景观格局动态变化研究[J].水土保持学报,2004,18(1)：168-171.
[125]角媛梅,马明国,肖笃宁.黑河流域中游张掖绿洲景观格局研究[J].冰川冻土,2003,25(1)：94-99.
[126]徐达.中山陵景区SPOT5融合及分类技术研究[D].南京：南京林业大学硕士论文,2005.
[127]史培军,潘耀忠,陈晋,等.深圳市土地利用／覆盖变化与生态环境安全分析[J].自然资源学报,1999,14(4)：293-299.
[128]陈利顶,傅伯杰,王军.黄土丘陵区典型小流域土地利用变化研究[J].地理科学,2001,21(1)：46-51.
[129]TurnerB L, D S, Sanderson S,etal. Land-Use and Land-CoverChange:Science/Research Plan[M]. IGBP ReportNo.35, HDP ReportNo.7, IGBP andHDP, Stockholm and Geneva,1995.
[130]倪绍祥.土地利用／覆被研究的几个问题[J].自然资源学报,2005,20(6)：932-937.
[131]邬建国.景观生态学----格局、过程、尺度与等级[M].北京：高等教育出版社,2000.
[132]TurnerM G, GardnerR H, O'NeillR V. Landscape Ecology:Theory and Practice[M]. New York:Springer-Verlag,2001.
[133]赵晶.上海城市土地利用与景观格局的空间演变研究[D].上海：华东师范大学,博士论文,2005.
[134]赵文武,傅伯杰,陈利顶.景观指数的粒度变化效应[J].第四纪研究,2003,23(3)：326-333.
[135]曾辉,高凌云,夏洁.基于修正的转移概率方法进行城市景观动态研究—以南昌市区为例[J].生态学报,2003,23(11)：2201-2209.
[136]刘殿伟.过去50年三江平原土地利用/覆盖变化的时空特征与环境效应[D].长春：吉林大学博士论文,2006.
[137]孙武,南忠仁.生态脆弱带内部空间分异结构与脆弱度划分[J].生态学报,2002,22(4)：445-451.
[138]Naveh Z.& Liebervan A.S., Landscape ecology theory and application[M].New York:Springer-Verlag,1984.
[139]Iverson,L.lt,Land-Use Changes in Dlinois,USA:the Influence of Landscape Attributes on Current and Historic Land Use[J].Landscape Ecology,1988,2:45-62.
[140]肖寒,欧阳志云,赵景柱等.海南岛景观空间结构分析[J].生态学报,2001,21(1)：20-26.
[141]刘学录.盐化草地景观中的斑块形状指数及其生态学意义[J].草业科学,2000,17(2)：50-56.
[142]Clarke K C. loose-coping growth prediction for San Information Science,1998a cellular automation model and G1S[J]:Francisco and Washington/Baltimore,1998,(53):293-308.
[143]李吉英.基于GIS的土地利用变化及驱动力研究[D].哈尔滨：东北农业大学硕士论文,2006.
[144]Voinov A,Costanza R,Wainger L,et al.Patuxent Landscape ecological and economic modeling of awater-Shed[J],Environ.Mod.Soft,1999,14:473-491.
[145]De.Koning.GH.J,Verburg.PH,Veldkamp.A,et al.Multi-scale modeling of land use change dynamics in Ecuador[J],Agric.Syst,1999,61:77-93
[146]杨莉.基于系统动力学和元胞自动机的土地利用结构优化研究—以贵州省黔西县为例.贵阳,贵州大学硕士论文,2009.
[147]李景刚,何春阳,史培军,等.中国北方未来土地利用变化情景模拟[J].地理学报,2004,59(2)：274-282.
[148]刘继生,陈彦光.基于GIS的细胞自动机模型与人地关系的复杂性探讨[J].地理研究,2002,21(2)：155-162.
[148]DENG Xiangzheng SU Hongbo,ZHAN Jinyan. Integration ofmultiple data sources to simulate the dynamics of land systems [J]. Sensors,2008,(8):620-634.
[149]邓祥征,林英志,黄河清土地系统动态模拟方法研究进展[J].生态学杂志,2009,28(10)：2123-2129.
[150]魏云洁,甄霖,邓祥征,等.中国生物能源政策情景对土地利用格局的影响[J].资源科学,2009,31(4)：552-558.
[151]LuijtenJC.ASystematieMethodforGeneratingLandUsePatternsUsingStoehastieRulesand BasieLandseapeCharacterlsties:Result5foraColombianHillsideWatershod[J].Agrieulture EcosystemS & Environment,2003(95):427-441
[152]刘耀林,刘艳芳,张玉梅.基于灰色—马尔柯夫链预测模型的耕地需求量预测研究[J].武汉大学学报(信息科学版),2004,29(7)：575-579
[153]熊利亚等.基于地理元胞自动机的土地利用变化研究[J].资源科学,2005,27(4)：38--39
[154]CostantitiM,FarinaA,ZirilliF.The Fusion of different resolutionSARImages[J].In: Proceedings of the IEEE,1997,85(1):139-145.
[155]侯西勇,常斌,于信芳.基于CA--Markov的河西走廊土地利用变化研究[J].农业工程学报,2004,20(5),28-19
[156]杨国清,刘耀林,吴志峰.基于CA-Markov模型的土地利用格局变化研究[J].武汉大学学报,2007,32(5)：414-418
[157]高志强,刘纪远.基于遥感和GIS的中国土地潜力资源的研究[J].遥感学报,2000,4(2)：136-140.
[158]汪雪格.吉林西部生态景观格局变化与空间优化研究[D].长春：吉林大学博士论文,2008.
[159]刘琼,欧名豪,彭晓英.基于马尔柯夫过程的区域土地利用结构预测研究—以江苏省昆山市为例[J].南京农业大学学报,2005,25(3)：107-112.
[160]贾科利.基于遥感、GIS的陕北农牧交错带土地利用与生态环境效应研究.西北农林科技大学博士学位论文,2007
[161]朱德举.土地评价[M].北京：中国大地出版社,2002.
[162]黄跃进,唐锦春,孙柄楠.基于GIS的农用地土地适宜性评价模型的建立[J].浙江林学报,1999,16(4)：406-408.
[163]陈健飞,刘卫民.FuzZy综合评判在土地适宜性评价中的应用[J].资源科学,1999,21(4)：71-74
[164]宋如华,齐实,孙保平,等.区域土地资源的适宜性评级和空间布局[J].土壤侵蚀与水土保持学报,1997,3(3)：23-30.
[165]廖和平.未利用适宜性评价方法研究—以攀枝花市仁和区为例[J].西南大学学报(自然科学版),1997,22(2)：199-205.
[166]Clarke K C,Gazulis N,Dietzel C K,et al.A Decade of SLEUTHing:Lessons Learned from Applicarions of a Cellular Automation Land use Change Model.In Fisher,P.F.(Eds.)Classics from IJGIS:Twenty Years of the Internation journal of Geographica Information Science and Systems.Boca Raton,FL:CRC P ress,2006.
[167]史忠植.高级人工智能[M].北京：科学出版社,1998.
[168]胡广录,赵文智,刘鹄.内陆河小流域综合治理对景观格局的影响--以童子坝河流域为例[J].中国沙漠,2010,30(6)：1398-1404.
[169]包维楷,吴宁.滇西北德钦县高山、亚高山草甸的人为干扰状况及其后果[J].中国草地,2003,25(2)：1-8.
[170]谢尚春,刘慧,王海涛,等.岷江上游生态安全综合评价[J].安徽农业科学,2010,38(11)：5807-5809.
[171]赵兵.岷江上游城镇脆弱性生态环境的敏感性分析与灾后重建[J].西南民族大学学 报·自然科学版,2009,35(1)：165-169.
[172]王瑞燕.县域尺度环境脆弱性演变及其土地利用／覆盖效应[D].泰安：山东农业大学博士论文,2009.
[173]王万忠,焦菊英.中国的土壤侵蚀因子定量评价研究[[J].水土保持通报,1996,16(5)：1-20.
[174]王万忠,焦菊英,赫小品.中国降水侵蚀力R值的计算与分布[[J].水土保持学报,1995,9(4)：5-18.
[175]王听,岷江上游泥石流活动的分布特征研究,重庆师范学院学报,2000,17(4)：87-93
[176]唐晓春.川西北山地的泥石流活动与水土流失[J].水土保持学报,1991,5(4)：18--23
[177]谢强,岷江上游地区生态环境现状遥感解译分析[J].四川环境,2002,21(1)：51-54
[178]冶民生,关文彬,谭辉等.岷江干旱河谷灌丛a多样性分析[J].生态学报,2004,24(6)：1123-1130
[179]叶延琼,樊宏,陈国阶.岷江上游土地退化及其防治对策[]J.水土保持通报,2002,22(6)：56-70
[180]岳书平,张树文,闫业超,等.公主岭市土地利用变化的生态效应研究[J].资源科学,2006,28(6)：161-166.
[181]魏琦.北方农牧交错带生态脆弱性评价与生态治理研究—以内蒙古林西县为例[D].北京：中国农科院研究生院博士论文,2010.
[182]王兮之,陆龙,梁钊雄.青海涅水流域植被覆盖度时空变化分析[J].水土保持研究,2010,17(6)：172-175.
[183]吴宁,刘照光.青藏高原东部亚高山森林草甸植被地理格局的成因探讨[J].应用与环境生物学报,1998,4(3)：290-297.
[184]包维楷,陈庆恒,陈克明.岷江上游干旱河谷植被恢复环境优化调控技术研究川.应用生态学报,1999,10(5)：542-544.
[185]包维楷,吴宁.慎西北德钦县高山、亚高山草甸的人为干扰状况及其后果[J].中国草地,2003,25(2)：1-8.
[186]岳书平,张树文,闫业超,等.东北典型农区土地利用变化的生态效应研究-以公主岭市为例[J].干旱区资源与环境,2007,21(7)：64-68.
[2]王成.局地土地利用变化及景观格局研究--以重庆市沙坪坝区为例[D].重庆：西南大学博士论文,2007.
[3]刘京涛.岷江上游植被蒸散时空格局及其模拟研究[D].北京：中国林业科学研究院博士论文,2006：1-2.
[4]严代碧.眠江上游干旱河谷区退化植被特征及其恢复重建的研究[D].北京：北京林业大学博士论文,2006：2-4.
[5]刘茹,张晶晶,雷蕾.岷江上游生态脆弱性的模糊评价[J].资源开发与市场,2006,22(6)：500-533.
[6]叶正伟.基于生态脆弱性的淮河流域水土保持策略研究[J].水土保持通报,2007,27(3)：141-145.
[7]周毅,李旋旗,赵景柱.中国典型生态脆弱带与贫困相关性分析[J].北京理工大学学报.2008,28(3)：260-262.
[8]邬建国.景观生态学--概念与理论[J].生态学杂志,2000,19(1)：42-43.
[9]李娟娟.上海城市景观格局演变及其生态安全影响研究[D].上海：复旦大学博士论文,2007.
[10]缪宁.川西亚高山红桦---岷江冷杉天然次生林的空间格局分析[D].北京：中国林业科学研究院博士论文,2009.
[11]高凯.多尺度的景观空间关系及景观格局生态效应的变化研究[D].武汉：华中农业大学博士论文,2010.
[12]徐广才,康慕谊,贺丽娜等.生态脆弱性及其研究进展[J].生态学报,2009,29(5)：2578-2580.
[13]邓向瑞.北京山区景观格局格局及其尺度效应研究[D].北京：北京林业大学博士论文,2007：2-4.
[14]王景伟,王海泽.景观指数在景观格局描述中的应用--以鞍山大麦科湿地自然保护区为例[J].水土保持研究,2006,13(2)：230-231.
[15]赵福强,代力民,于大炮等.长白山露水河林业局森林景观格局动态[J].应用生态学报,2010,21(5)：1180-1184.
[16]刘宇,吕一河,傅伯杰.景观格局--土壤侵蚀研究中景观指数的意义解释及局限性[J].生态学报,2011,31(1)0267-0275.
[17]万荣荣,杨桂山.太湖流域土地利用与景观格局演变研究[J].应用生态学报,2005,16(3)：475-476.
[18]张志明,罗亲普,王文礼.2D与3D景观指数测定山区植被景观格局变化对比分析[J].生态学报,2010,30(21)：58865893.
[19]马燕飞,沙占江,郭丽红等.基于NDVI及DEM的青海湖北岸景观格局空间自相关分析[J].遥感应用,2010,6(10)：95-119.
[20]王瑶,刘前进,于兴修.沂蒙山区土壤侵蚀垂直景观格局的分形特征[J].农业工程学报,2010,26(11)：304-309.
[21]李玉凤,刘红玉,郑囡等.基于功能分类的城市湿地公园景观格局--以西溪湿地公园为例[J].生态学报,2011,31(4)：1021-1028.
[22]张盼盼,胡远满,肖笃宁等.地形因子对喀斯特高原山区潜在石漠化景观格局变化的影响分析[J].土壤通报,2011,41(6)：1305-1310.
[23]赵冰,张杰,孙希华.基于GIS的淮河流域桐柏—大别山区生态脆弱性评价[J].水土保持研究,2009,16(3)：135-138.
[24]樊哲文,刘木生,沈文清.江西省生态脆弱性现状GIS模型评价[J].地球信息科学学报,2009,11(2)：202-208.
[25]宋国利,李玉宝,付春雷等.基于RS与GIS的乐清湾湿地景观格局变化分析[J].东北林业大学学报,2010,38(12)：80-83.
[26]周永娟,仇江啸,王姣等.三峡库区消落带生态环境脆弱性评价[J].生态学报,2010,30(24)：6727-6728.
[27]付博,姜琦刚,任春颖等.扎龙湿地生态脆弱性评价与分析[J].干旱区资源与环境,2011,25(1)：50-52.
[28]万星,周建中,丁晶等.岷江上游生态脆弱性综合评价的集对分析[J].中国农村水利水电,2006,12：33-39.
[29]孔博,陶和平,李爱农,等.汶川地震灾区生态脆弱性评价研究[J].水土保持通报,2010,30(6)：180-181.
[30]卢冬爱,谈树成,夏既胜.基于景观格局和水土流失敏感性的大理市生态脆弱性分析[J].云南地理环境研究,2009,21(2)：92-96.
[31]李滨勇,陈海滨,唐海萍等,基于AHP和模糊综合评判法的北疆各地州生态脆弱性评价[J.北京师范大学学报(自然科学版),2010,46(2)：197-201.
[32]戴其文,赵雪雁.生态补偿机制中若干关键科学问题--以甘南藏族自治州草地生态系统为例[J].地理学报,2010,65(4)：495-505.
[33]乔青,高吉喜,王维等.生态脆弱性综合评价方法与应用[J].环境科学研究,2008,21(5)：117-123.
[34]王丽婧,席春燕,付青等.基于景观格局的三峡库区生态脆弱性评价[J].环境科学研究,2010,23(10)：1269-1270.
[35]余坤勇,刘健,,黄维友等.基于GIS技术的闽江流域生态脆弱性分析[J].江西农业大学学报,2009,31(3)：568-573.
[36]蔡海生,刘木生,陈美球等.基于GIS的江西省生态环境脆弱性动态评价[J].水土保持通报,2009,29(5)：190-196.
[37]李红丽,智颖飙,张荷亮等.新疆生态足迹与环境压力的时空分异[J].生态学报,2010,30(17)：4676-4684.
[38]周彬,余新晓,陈丽华等.北京山区森林景观格局与环境关系的CCA研究[J].水土保持通报,2010,30(6)：148-152.
[39]刘宏文.山西省庞泉沟森林群落特征与景观格局特征分析的研究[D].北京：北京林业大学博士论文,2007：1-2.
[40]姚建,张军,朱莞.岷江上游生态脆弱性驱动力分析[J].资源开发与市场,2007,23(11)：977-982.
[41]郝璐,王静爱,张化.北方草地畜牧业生态系统健康综合评价与诊断[J].生态学报,2008,28(4)：1456-1465.
[42]邱彭华,徐颂军,谢跟踪等.基于景观格局和生态敏感性的海南西部地区生态脆弱性分析[J].生态学报,2007,27(4)：1257-1264.
[43]彭立,苏春江,徐云等.岷江上游生态环境现状、存在问题及治理对策[J].江西农业大学学报(社会科学版),2007,6(1)：80-84.
[44]余妹萍,刘国东,吴嫒等.岷江上游日径流过程分维分析及其生态脆弱性表征[J].西南民族大学学报(自然利学版),2005,31(1)：80-84.
[45]宫兆宁,张翼然,宫辉力.北京湿地景观格局演变特征与驱动机制分析[J].地理学报,2011,66(1)：77-88
[46]刘艳芬,张杰,马毅.1995--1999年黄河三角洲东部自然保护区湿地景观格局变化[J].应用生态学报,2010,21(11)：2904-2911.
[47]刘世荣,孙鹏森等.岷江上游生态水文图集[M].北京：中国地图出版社,2008.
[48]史培军,宫鹏,李晓兵,等.土地利用／覆盖变化研究的方法与实践[M].北京：科学出版社,2000.
[49]刘纪远.中国资源环境遥感宏观调查与动态研究[M].北京：中国科学技术出版社,1996.
[50]杜培军.遥感原理与应用[M].北京：中国矿业大学出版社,2006.
[51]梅安新,彭望禄,秦其明,等.遥感导论[M].北京：高等教育出版社,2001.
[52]章孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州：浙江大学出版社,1997.
[53]张永生.遥感图像信息系统[M].北京：科学出版社,2000.
[54]姜小光,干长耀,王成.成像光谱数据的光谱信息特点及最佳波段的选择-以北京顺义区为例[J].干旱地区研究,2000,23(3)：214-220.
[55]刘建平,赵英时.高光谱遥感数据解译的最佳波段选择方法研究[J].中国科学院研究生院学报,1999,6(2)：153-161.
[56]鲍文东.基于GIS的土地利用动态变化研究[D].济南：山东科技大学博士论文,2007.
[57]李虎杰.岷江上游生态环境建设与经济可持续发展[J].四川环境,2001.20(4)：51-56.
[58]刘洋.峨江源头区植被景观与流城土城怪蚀的动态相关性[D].成都：中国科学院成都生物研究所博士论文,2007.
[59]何国金,李克鲁,胡德永,等.多卫星遥感数据的信息融合：理论、方法与实践[J].中国图像图形学报,1999,7(9)：744-750.
[60]J.G..Liu.Smoothing Filter-based Intensity Modulation:a spectral preserve imagefusion Techninue for improving spatial details[J].International Journal of RemoteSensing,2000, 21(18):3461:3472.
[61]贾凌,都金康,赵萍,等.基于TM的海南省土地利用／覆盖动态变化的遥感监测与分析[J].国土资源遥感,2003,(1)：22-26.
[62]孙家柄,舒宁,关泽群.遥感原理,方法和应用[M].北京：测绘出版社,1997.
[63]朱述龙,张占睦.遥感图像获取与分析[M].北京：科学出版社,2000.
[64]许榕峰.基于遥感的龙海市土地利用变化动态监测与专题研究[D].福州：福州大学硕士论文,2003.
[65]Congalton R G.A review of assessing the accuracy of classification of remotely sensed data[J].Remote Sensing of Environment,1991,37:35-46.
[66]Lucas I F,Frans J M,Wel V D.Accuracy assessment of satellite derived land-cover data:a review[J].Photogrammetric Engineering&Remote Sensing,1994,60(4):410-432.
[67]王妍.基于DEM的地形信息提取与景观空间格局分析[D].重庆：西南大学硕士论文,2007.
[68]Miller,C.and Laflamme.The Digital Terrain Model-Theory and Application[J]. Photogrammetric Engineering.1958,(24):433-442.
[69]Makarovic, B. Progressive Sample for DTM[J]. ITC Jounal,1973,40:397-416.
[70]黎景良,后斌,危双峰,等.基于DEM的广东省山区土地利用变化分析[J].测绘通报,2007,(6)：53-57.
[71]Tang GA. A Research on the Accuracy of DEM Beijing[M.] New York:Science Press,1996.
[72]Ebish, K.Effect of Digital Elevation Resolution on the Properties of Contours[J]. Technical Paper, ASP-ACSM Fall Convention,1984,4():424-434.
[73]Hannah,M.Error detection and correction in digital terrain models[J]. Photogrammetric Engineering and Remote sensing,1981,47(l):36-69.
[74]张云红,王道杰,江晓波,等.岷江上游土地利用变化及其对生态环境的影响[J].西南大学学报(自然科学版),2010,32(3)：102-107.
[75]亓兴兰.基于3S技术闽江流域土地利用演化及其空间格局过程研究[D].福州:福建农林硕士大学,2006.
[76]杨丽,谢高地,甄霖等.径河流域土地利用格局的时空变化分析[J].资源科学,2005,27(4)：27-32.
[77]朱会义,李秀彬,何书金,等.环渤海地区土地利用的时空变化分析[J].地理学报,2001,56(3)：253-260.
[78]王秀兰.包玉海.土地利用动态变化研究方法探讨[J].地理科学进展,1999,18(1)：81-87.
[79]于兴修.中尺度流域土地利用灌被变化及其水环境效应研究—以太湖上游西菩溪流域为例[D].南京：南京中国科学院南京地理与湖泊研究所博士学位论文,2003.
[80]郑国强.基于RS的无锡市土地利用/覆被变化信息提取及其时空演变研究[D].南京：南京大学博士学位论文,2004.
[81]刘纪远,布和敖斯尔.中国土地利用变化现代过程时空特征的研究--基于卫星遥感数据[J].第四纪研究,2000,2(3)：229-239.
[82]刘庆.亚高山针叶林生态学研究[M].成都：四川大学出版社,2002.
[83]张明阳,王克林,刘会玉,等.喀斯特生态脆弱区桂西北土地变化特征[J].生态学报,2009,29(6)：3106-3116.
[84]刘耀林,焦利民.顾及尺度效应和景观格局的土地利用数据综合指标研究[J].测绘学报,2009,38(6)：549-554.
[85]王思远,刘纪远,张增祥,等.中国土地利用时空特征分析[J].地理学报,2001,56(6)：631-639.
[86]朱会义,李秀彬.关于区域土地利用变化指数模型方法的讨论[J].地理学报,2003,58(5)：643-650.
[87]樊玉山,刘纪元主编.西藏自治区土地利用[M].北京：科学出版社,1994.
[88]任志远,张艳芳.土地利用变化与生态安全评价[M].北京：科学出版社,2003.
[89]钟成.集成LiDAR和遥感影像生成高质量真正射影像研究[J].测绘学报,2011,40(1)
[90]向双.岷江江上游干旱河谷葡萄的土宜和生态布局研究[D].四川雅安：四川农业大学,2002.
[91]樊宏.岷江上游近50 a土地覆被的变化趋势[J].山地学报,2002,20(1)：64-69
[92]沈泽昊,岳超,赵俊,等.南水北调中线水源区土地利用/土地覆被的空间格局[J].地理学报,2006,61(6)：633-644.
[93]李增加,马友鑫,李红梅,等.西双版纳土地利用/覆盖变化与地形的关系[J].植物生态学报,2008,32(5)：1091-1103.
[94]赖长鸿.基于GIS的川西道孚林区森林景观格局研究[D.四川雅安：四川农业大学,2006.
[95]焦超卫,赵牡丹,汤国安,等.基于GIS的植被空间格局特征与地形因子的相关关系[J].水土保持通报,2005,25(6)：19-23.
[96]Buheaosier, Liu Jiyuan.Using Remote SensedStudy on Land Cover Change in China Based on Seasonal and Longitudinal CharacteristicsData. Asahikaw:IGU-LUCC Press,1997, 20-24
[97]谢高地.人口增长胁迫下的全球地利用变化研究[J].自然资源学报,1999,14(3)：233-241.
[98]王良健,包浩生.基于遥感和GIS的区域土地利用变化的动态监测研究[J].经济地 理,2000,20(2)：47-50.
[99]Lucas J F J, Accuracy assessment of satellite derived land cover data:a review. PE&RS, 1994,60(4):410-432.
[100]Taylor J C, Brewer T R, A C Bird. Monitoring landscape change in the national parks of England and Wales using aerial photo interpretation and GIS. Int. J. Remote Sensing, 2000,21(13):2737-2752.
[101]J F Mas. Monitoring land-cover changes./nt. J. Remote Sensing,1999,20(1):139-152.
[102]龚文峰.基于RS和GIS松潘地区天然林景观动态过程与分类研究[D]。哈尔滨,东北林业大学博士论文,2007.
[103]廖克.地球信息图谱与数字地球[J].地理研究,2001,20(1)：55-61.
[104]叶庆华,刘高焕,陆洲,等.基于GIS的时空复合体--土地利用图谱模型研究方法[J].地理科学进展,2002,21(4)：349-357
[105]张国坤.新开河流景观格局动态变化及图谱分析[D].长春：东北师范大学博士论文,2008.
[106]姚建.峨江上游生态脆弱性分析及评价[D].成都：四川大学博士论文,2004.
[107]包维楷,张镱锂,王乾,等.大渡河上游林区森林资源退化及其恢复与重建[J].山地学报,2002,20(2)：194-198.
[108]阎建忠,张镜铿等.基于植被演替的土地覆被变化研究—大渡河上游的森林采伐、更新和退化[J].中国科学D辑,地球科学,2005,35(11)：1060-1073.
[109]Forman R T T.. Some general principles of landscape and regional ecology. Landscape. 1995,10(3):133-142.
[110]Soberg S,Jain A K,Taxt T.Multisource Classification of remotely sensed data:Fusion of LandsatTM and SAR Images[J].IEEETransactionon Geoscience and Remote Sensing,1994,32(4):768-778
[111]PAN Daiyuan, G Domon, D Marceau, et al. Spatial pattern of coniferous and deciduous forest patchesin an Eastern North America agricultural landscape:the influence of land use and physical amibutes[J].Landscape Ecoloav.2001.16:99-110.
[112]TumerMQRuseherCL.ChangesinlandseaPePatternsinGeorgia,USA[J],LandseaPeEeol, 1988,1(4):241-251.
[113]TumerMG. Land sea Peeh an gesinni erural counties of Georgia[JlPhotograEng RemoteSensing,1990,56:379-386.
[114]郭晋平,王俊田,李世光.关帝山林区景观要素沿环境梯度分布趋势的研究[J]植物生态学报,2000,24(2)：135-140.
[115]郭晋平,薛俊杰,李志强,等.森林景观恢复过程中景观要素斑块规模的变化[J].生态学报,2000,20(2)：218-223.
[116]钱乐祥,陶黎新.福建森林植被景观空间格局分异研究[J].热带地理,199717(4)： 397-404
[117]Ehrlich D, Lambin E F. Broad scale land-cover classification and Internal climatic change.International journal of Remote Sensing.1996,17:845-862.
[118]王让会,张慧芝,游先祥,等.塔里木河流域生态景观格局的遥感信息提取与分析[J].北京林业大学学报,2003,25(2)：43-47.
[119]毕晓丽,周睿,刘丽娟,等.径河沿岸景观格局梯度变化及驱动力分析[J].生态学报,2005,25(5)1041-1048.
[120]贾宝全,王国柱,慈龙骏.景观格局分布特征的区域差异分析—以鄂尔多斯高原伊金霍洛旗为例[J].干旱区研究,2003,20(3)：168-174.
[121]马媛,师庆东,潘晓玲.西部干旱区生态景观格局动态分[J].干旱区地理,2004,27(4)：516-519.
[122]熊金国,王世新,周艺等.利用景观格局指数的空间分辨率对水域面积提取影响的分析[J].武汉大学学报·信息科学版,2011,36(1)：98-103.
[123]杨德伟,张云红,王道杰.等岷江上游土地利用的时空分异[J].山地学报,2009,27(5)：579-584.
[124]高小红,王一谋,杨国靖.基于RS与GIS的榆林地区景观格局动态变化研究[J].水土保持学报,2004,18(1)：168-171.
[125]角媛梅,马明国,肖笃宁.黑河流域中游张掖绿洲景观格局研究[J].冰川冻土,2003,25(1)：94-99.
[126]徐达.中山陵景区SPOT5融合及分类技术研究[D].南京：南京林业大学硕士论文,2005.
[127]史培军,潘耀忠,陈晋,等.深圳市土地利用／覆盖变化与生态环境安全分析[J].自然资源学报,1999,14(4)：293-299.
[128]陈利顶,傅伯杰,王军.黄土丘陵区典型小流域土地利用变化研究[J].地理科学,2001,21(1)：46-51.
[129]TurnerB L, D S, Sanderson S,etal. Land-Use and Land-CoverChange:Science/Research Plan[M]. IGBP ReportNo.35, HDP ReportNo.7, IGBP andHDP, Stockholm and Geneva,1995.
[130]倪绍祥.土地利用／覆被研究的几个问题[J].自然资源学报,2005,20(6)：932-937.
[131]邬建国.景观生态学----格局、过程、尺度与等级[M].北京：高等教育出版社,2000.
[132]TurnerM G, GardnerR H, O'NeillR V. Landscape Ecology:Theory and Practice[M]. New York:Springer-Verlag,2001.
[133]赵晶.上海城市土地利用与景观格局的空间演变研究[D].上海：华东师范大学,博士论文,2005.
[134]赵文武,傅伯杰,陈利顶.景观指数的粒度变化效应[J].第四纪研究,2003,23(3)：326-333.
[135]曾辉,高凌云,夏洁.基于修正的转移概率方法进行城市景观动态研究—以南昌市区为例[J].生态学报,2003,23(11)：2201-2209.
[136]刘殿伟.过去50年三江平原土地利用/覆盖变化的时空特征与环境效应[D].长春：吉林大学博士论文,2006.
[137]孙武,南忠仁.生态脆弱带内部空间分异结构与脆弱度划分[J].生态学报,2002,22(4)：445-451.
[138]Naveh Z.& Liebervan A.S., Landscape ecology theory and application[M].New York:Springer-Verlag,1984.
[139]Iverson,L.lt,Land-Use Changes in Dlinois,USA:the Influence of Landscape Attributes on Current and Historic Land Use[J].Landscape Ecology,1988,2:45-62.
[140]肖寒,欧阳志云,赵景柱等.海南岛景观空间结构分析[J].生态学报,2001,21(1)：20-26.
[141]刘学录.盐化草地景观中的斑块形状指数及其生态学意义[J].草业科学,2000,17(2)：50-56.
[142]Clarke K C. loose-coping growth prediction for San Information Science,1998a cellular automation model and G1S[J]:Francisco and Washington/Baltimore,1998,(53):293-308.
[143]李吉英.基于GIS的土地利用变化及驱动力研究[D].哈尔滨：东北农业大学硕士论文,2006.
[144]Voinov A,Costanza R,Wainger L,et al.Patuxent Landscape ecological and economic modeling of awater-Shed[J],Environ.Mod.Soft,1999,14:473-491.
[145]De.Koning.GH.J,Verburg.PH,Veldkamp.A,et al.Multi-scale modeling of land use change dynamics in Ecuador[J],Agric.Syst,1999,61:77-93
[146]杨莉.基于系统动力学和元胞自动机的土地利用结构优化研究—以贵州省黔西县为例.贵阳,贵州大学硕士论文,2009.
[147]李景刚,何春阳,史培军,等.中国北方未来土地利用变化情景模拟[J].地理学报,2004,59(2)：274-282.
[148]刘继生,陈彦光.基于GIS的细胞自动机模型与人地关系的复杂性探讨[J].地理研究,2002,21(2)：155-162.
[148]DENG Xiangzheng SU Hongbo,ZHAN Jinyan. Integration ofmultiple data sources to simulate the dynamics of land systems [J]. Sensors,2008,(8):620-634.
[149]邓祥征,林英志,黄河清土地系统动态模拟方法研究进展[J].生态学杂志,2009,28(10)：2123-2129.
[150]魏云洁,甄霖,邓祥征,等.中国生物能源政策情景对土地利用格局的影响[J].资源科学,2009,31(4)：552-558.
[151]LuijtenJC.ASystematieMethodforGeneratingLandUsePatternsUsingStoehastieRulesand BasieLandseapeCharacterlsties:Result5foraColombianHillsideWatershod[J].Agrieulture EcosystemS & Environment,2003(95):427-441
[152]刘耀林,刘艳芳,张玉梅.基于灰色—马尔柯夫链预测模型的耕地需求量预测研究[J].武汉大学学报(信息科学版),2004,29(7)：575-579
[153]熊利亚等.基于地理元胞自动机的土地利用变化研究[J].资源科学,2005,27(4)：38--39
[154]CostantitiM,FarinaA,ZirilliF.The Fusion of different resolutionSARImages[J].In: Proceedings of the IEEE,1997,85(1):139-145.
[155]侯西勇,常斌,于信芳.基于CA--Markov的河西走廊土地利用变化研究[J].农业工程学报,2004,20(5),28-19
[156]杨国清,刘耀林,吴志峰.基于CA-Markov模型的土地利用格局变化研究[J].武汉大学学报,2007,32(5)：414-418
[157]高志强,刘纪远.基于遥感和GIS的中国土地潜力资源的研究[J].遥感学报,2000,4(2)：136-140.
[158]汪雪格.吉林西部生态景观格局变化与空间优化研究[D].长春：吉林大学博士论文,2008.
[159]刘琼,欧名豪,彭晓英.基于马尔柯夫过程的区域土地利用结构预测研究—以江苏省昆山市为例[J].南京农业大学学报,2005,25(3)：107-112.
[160]贾科利.基于遥感、GIS的陕北农牧交错带土地利用与生态环境效应研究.西北农林科技大学博士学位论文,2007
[161]朱德举.土地评价[M].北京：中国大地出版社,2002.
[162]黄跃进,唐锦春,孙柄楠.基于GIS的农用地土地适宜性评价模型的建立[J].浙江林学报,1999,16(4)：406-408.
[163]陈健飞,刘卫民.FuzZy综合评判在土地适宜性评价中的应用[J].资源科学,1999,21(4)：71-74
[164]宋如华,齐实,孙保平,等.区域土地资源的适宜性评级和空间布局[J].土壤侵蚀与水土保持学报,1997,3(3)：23-30.
[165]廖和平.未利用适宜性评价方法研究—以攀枝花市仁和区为例[J].西南大学学报(自然科学版),1997,22(2)：199-205.
[166]Clarke K C,Gazulis N,Dietzel C K,et al.A Decade of SLEUTHing:Lessons Learned from Applicarions of a Cellular Automation Land use Change Model.In Fisher,P.F.(Eds.)Classics from IJGIS:Twenty Years of the Internation journal of Geographica Information Science and Systems.Boca Raton,FL:CRC P ress,2006.
[167]史忠植.高级人工智能[M].北京：科学出版社,1998.
[168]胡广录,赵文智,刘鹄.内陆河小流域综合治理对景观格局的影响--以童子坝河流域为例[J].中国沙漠,2010,30(6)：1398-1404.
[169]包维楷,吴宁.滇西北德钦县高山、亚高山草甸的人为干扰状况及其后果[J].中国草地,2003,25(2)：1-8.
[170]谢尚春,刘慧,王海涛,等.岷江上游生态安全综合评价[J].安徽农业科学,2010,38(11)：5807-5809.
[171]赵兵.岷江上游城镇脆弱性生态环境的敏感性分析与灾后重建[J].西南民族大学学 报·自然科学版,2009,35(1)：165-169.
[172]王瑞燕.县域尺度环境脆弱性演变及其土地利用／覆盖效应[D].泰安：山东农业大学博士论文,2009.
[173]王万忠,焦菊英.中国的土壤侵蚀因子定量评价研究[[J].水土保持通报,1996,16(5)：1-20.
[174]王万忠,焦菊英,赫小品.中国降水侵蚀力R值的计算与分布[[J].水土保持学报,1995,9(4)：5-18.
[175]王听,岷江上游泥石流活动的分布特征研究,重庆师范学院学报,2000,17(4)：87-93
[176]唐晓春.川西北山地的泥石流活动与水土流失[J].水土保持学报,1991,5(4)：18--23
[177]谢强,岷江上游地区生态环境现状遥感解译分析[J].四川环境,2002,21(1)：51-54
[178]冶民生,关文彬,谭辉等.岷江干旱河谷灌丛a多样性分析[J].生态学报,2004,24(6)：1123-1130
[179]叶延琼,樊宏,陈国阶.岷江上游土地退化及其防治对策[]J.水土保持通报,2002,22(6)：56-70
[180]岳书平,张树文,闫业超,等.公主岭市土地利用变化的生态效应研究[J].资源科学,2006,28(6)：161-166.
[181]魏琦.北方农牧交错带生态脆弱性评价与生态治理研究—以内蒙古林西县为例[D].北京：中国农科院研究生院博士论文,2010.
[182]王兮之,陆龙,梁钊雄.青海涅水流域植被覆盖度时空变化分析[J].水土保持研究,2010,17(6)：172-175.
[183]吴宁,刘照光.青藏高原东部亚高山森林草甸植被地理格局的成因探讨[J].应用与环境生物学报,1998,4(3)：290-297.
[184]包维楷,陈庆恒,陈克明.岷江上游干旱河谷植被恢复环境优化调控技术研究川.应用生态学报,1999,10(5)：542-544.
[185]包维楷,吴宁.慎西北德钦县高山、亚高山草甸的人为干扰状况及其后果[J].中国草地,2003,25(2)：1-8.
[186]岳书平,张树文,闫业超,等.东北典型农区土地利用变化的生态效应研究-以公主岭市为例[J].干旱区资源与环境,2007,21(7)：64-68.