Arid
基于GIS和RBFN模型的新疆土壤风蚀度研究
其他题名Study on Wind Erodibility of Soils in Xinjiang based on GIS and RBFN Model
杨光华
出版年2009
学位类型硕士
导师包安明
学位授予单位中国科学院新疆生态与地理研究所
中文摘要新疆是我国土壤风蚀危害极为严重的区域之一。频繁发生的土壤风蚀给新疆自然和生态环境造成了极大的破坏,直接导致土地荒漠化,致使生态环境恶化,耕地面积萎缩、耕地质量下降、经济损失惨重,同时也给人民的生命财产带来巨大的损失。研究新疆土壤风蚀度的区域分布差异,对水土保持及土地资源的合理开发和利用无疑具有十分重要的理论和实践意义。\n 本研究依据土壤风蚀的研究理论和前人的研究成果,结合新疆自然地理条件的特点,以1000m×1000m为基本研究单元,选取风场强度、土壤可蚀性、植被覆盖度、干燥度指数和地形起伏度5个指标作为土壤风蚀度的影响因子,并运用GIS技术提取各指标的空间数据;参照土壤风蚀强度分级指标,将新疆土壤风蚀度划分为轻风蚀度、中风蚀度、强风蚀度和极强风蚀度4个等级。在此基础上,建立了径向基函数网络模型(Radial Basis Function Network,RBFN),并根据已有研究成果,在不同土壤风蚀度等级的代表性分布区域内选取训练样本,通过网络训练确定了RBFN模型参数,对新疆进行了土壤风蚀度研究的初步尝试。研究结果表明: \n(1)新疆极强土壤风蚀度和强土壤风蚀度区域面积占全疆面积的50.85%,表明全疆超过半数的土地土壤风蚀危害已经相当严重;中土壤风蚀度的区域面积居第二,预示着全疆超过1/4的土地正面临着土壤风蚀危害加剧的威胁。\n(2)新疆土壤风蚀度的空间分布格局呈现“东南高,西北低;盆地高,高山低”的特点,这与该区气候和地理分布相吻合;土壤可蚀性越大、干燥度指数越大、离风口、风区越近的区域,土壤风蚀度越高;植被覆盖度越低(水域除外)、地形起伏度越小的区域,土壤风蚀度越高。\n(3)极强土壤风蚀度地区在东疆主要分布于吐鲁番—哈密盆地戈壁区、哈顺沙漠、南湖戈壁、淖毛湖风区,在南疆主要分布于塔里木盆地沙漠区、库姆塔格沙漠、昆仑山中部高山戈壁区,在北疆主要分布于阿拉山口大风口、达坂城—托克逊沿线风区和吉木萨尔—奇台—木垒沿线风区;强土壤风蚀度地区主要分布于准噶尔盆地半固定沙丘、天山以南、昆仑山和阿尔金山山区的荒漠戈壁区以及东疆风场强度稍弱的荒漠戈壁区;中土壤风蚀度地区主要分布于准噶尔盆地北缘至阿尔泰山山区、塔城—额敏盆地中低盖度草地、伊吾盆地中低盖度草地、绿洲盐碱地、罗布泊盐壳、天山和昆仑山等中低盖度草地区域;轻土壤风蚀度地区主要分布于阿尔泰山森林、塔城盆地高盖度草地、绿洲农业区、伊吾盆地高盖度草地、天山山脉、伊犁谷地以及昆仑山极高山区。\n(4)通过与前人研究和长期观测成果(新疆土壤风蚀强度分区、沙尘源空间分布和大气降尘量空间分布等)的对比验证表明,尽管本研究与前人的研究角度和方法不同,但关于新疆土壤风力侵蚀的总体分布格局却是一致的,从而证明了使用GIS-RBFN集成技术、在1000m×1000m尺度下研究得出的新疆土壤风蚀度具有较高的准确性。
英文摘要Xinjiang is one of the regions where wind erosion of soils is very serious in China. The frequence of wind erosion of soils destroy to natural and ecological environment of Xinjiang, which directly causes land desertification, and leads to ecological environment deterioration, decrease of cultivated land area, decline of farmland quality and seriously economic damaged, meanwhile, wind erosion of soils brings to great loss of people′s life and property. So, studying the regional distribution regularity of wind erodibility of soils in Xinjiang, which has a very important theoretical and practical significance to soil and water conservation and reasonable development and utilization of land resources.\nThis study based on the theory of wind erosion of soils and predecessor’s research data, combined with physiographic condition of Xinjiang, using 1000m×1000m as basic unit, selected wind intensity, soil erodibility, vegetation coverage, index of soil dryness and relief degree of land surface as the impact factors to study wind erodibility of soils, and extracted spatial data of each index by GIS technology. Referenced to the classifying index of wind erosion intensity, wind erodibility of soils in Xinjiang was divided into four levels(slight, moderate, severe and very severe). On this basis, RBFN model was established. According to the existing research results, the author selected the training samples in typical distribution areas of the different level of wind erodibility of soils, and then determineed the parameters of RBFN model through network training. The results showed as follows: \n(1) The region area of very severe and severe wind erodibility of soils accounted for 50.85% of the whole area of Xinjiang, which showed that wind erosion of more than a half land area of Xinjiang has been quite severely. The region area of moderate wind erodibility of soils occupied second place, which foreshowed there were more than a quarter land area of Xinjiang was confronted with the grave threats of wind erosion. \n(2) The spatial distribution pattern of wind erodibility of soils in Xinjiang indicated that wind erodibility of soils was high in the southeastern and basin region, northwestern and alpine was low. It was consistent with the region’s climate and geographical distribution. The higher soil erodibility, index of soil dryness and wind intensity was, the higher wind erodibility of soils was. The lower vegetation coverage except for waters and relief degree of land surface was, the higher wind erodibility of soils was. \n(3)The region of very severe wind erodibility of soils distributed mainly in the Eastern Xinjiang (gobi of Turpan-Hami Basin, Hashun desert, South Lake gobi, Naomaohu fetch), in the Southern Xinjiang(desert area of Tarim Basin, Kumtag desert, alpine gobi in central Kunlun mountain), and in the Northern Xinjiang(Arla mountain gap, along the Dabancheng-Tuokexun fetch and the Jimusar-Qitai-Mori fetch). The region of severe wind erodibility of soils distributed mainly in semi-fixed sand dune of Dzungaria Basin, Gobi desert terrain of mountain area in south of Tianshan, Kunlun mountain and Arkin mountain. The region of moderate wind erodibility of soils distributed mainly from north of Dzungaria Basin to Altai mountain area, middle and low coverage grassland of Tacheng-Emin Basin,Yiwu Basin,Tianshan and Kunlun mountain, saline land of oasis and saline shell of Lop Nur. The region of slight wind erodibility of soils distributed mainly in Altai mountain forest, high coverage grassland of Tacheng Basin, oasis agricultural area, Yiwu Basin, Tianshan mountains, Ili valley and extreme high mountain area of Kunlun.\n (4)Through contrasting with previous research results about partition of wind erosion intensity, spatial distribution of dust resources and atmo-spheric deposition, we can find that although this paper used different study methods from the previous research, the result was consistent with the entire distribution pattern of wind erosion in Xinjiang. Thereby wind erodibility of soils in Xinjiang under the scale of 1000m×1000m had a good accuracy by using GIS-RBFN integration technology.
中文关键词GIS ; 径向基函数网络模型 ; 新疆 ; 土壤风蚀度
英文关键词GIS RBFN Xinjiang Wind erodibility of soils
语种中文
国家中国
来源学科分类地图学与地理信息系统
来源机构中国科学院新疆生态与地理研究所
资源类型学位论文
条目标识符http://119.78.100.177/qdio/handle/2XILL650/286792
推荐引用方式
GB/T 7714
杨光华. 基于GIS和RBFN模型的新疆土壤风蚀度研究[D]. 中国科学院新疆生态与地理研究所,2009.
条目包含的文件
条目无相关文件。
个性服务
推荐该条目
保存到收藏夹
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[杨光华]的文章
百度学术
百度学术中相似的文章
[杨光华]的文章
必应学术
必应学术中相似的文章
[杨光华]的文章
相关权益政策
暂无数据
收藏/分享

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。