岷江上游泥石流胁迫下山区聚落人口易损性评价

发布时间：2018-05-21 13:17

本文选题：岷江上游 + 七盘沟村　；参考：《西南科技大学》2017年硕士论文

【摘要】：在不同尺度下,人口易损性的影响因素不同,所用评价方法也会有所区别。因此本文选取了岷江上游和七盘沟村为研究对象,其中岷江上游为中尺度范围,七盘沟村为小尺度范围,即典型区域,并根据实际情况筛选评价指标,建立岷江上游人口易损性和七盘沟村人口易损性的评价指标,分别分析了两种尺度下人口易损性的不同。岷江上游的中尺度评价基于信息量模型,得出整个岷江上游人口易损性区划图。小尺度七盘沟村为典型区,基于MATLAB的SOM模型和ArcGIS支持下,获取了研究区各居户在泥石流威胁下聚落易损性高低(易损性大小)的分布图。本文的主要研究内容和研究成果如下:(1)本文通过筛选,选取了修正人口密度、人均GDP、万人医生数、万人病床数、监测点、劳动人口比例6个评价因子作为研究区的人口易损性评价指标,以5050?的网格为评价单元,将岷江上游划分为9192246个网格单元,并通过信息量模型的计算得到区域易损图。(2)通过分析发现,岷江上游整体人口易损性偏低,低易损区所占比例为91.09%;高人口易损区所占比例为3.78%。高易损区主要集中在汶川县、茂县、理县境内。(3)通过对各评价因子图和岷江上游人口区划图的分析对比可知:人口易损性基本与修正人口密度成正相关,与万人病床数、万人医生数、人均GDP、监测点、劳动人口比例呈负相关。一个地区修正人口密度越大,人口密度会增加系统人口易损性,而在其他因素综合作用下会降低系统人口易损性。人口密度、万人病床数、万人医生数、人均GDP是人口易损性的关键因素,对人口易损性的影响较大,劳动人口比例和监测点的影响相对较低。(4)在实地考察的基础上,结合影像的解译分析,选择了家庭规模、年龄结构、健康状况、民族特征、文化程度,与泥石流沟的距离6个指标作为典型区-七盘沟村人口易损性的评价因子,应用自组织神经网络模型(Self organized maps,简称SOM)进行研究区人口易损性评价,SOM模型的优点在于能够回避人为主观因素对各评价指标权重的影响。(5)通过研究发现:七盘沟村易损性高的区域主要集中在居户密集处,易损性以为沟口最大,向四周递减;极高、高易损性区域所占比例分别为3.52%、4.90%,比例最小;低、极低易损性区域所占比例分别为22.98%、60.06%,比例最大。根据SOM模型得出U-矩型阵与18个变量的平面信息图分析可以发现,人口易损性较高的多为中等家庭或大家庭、少数民族,教育程度较低,距离泥石流沟0-300m的住户;年龄结构与身体状况对人口易损性有一定影响,但是影响程度较低,与研究区实际情况基本符合。
[Abstract]:At different scales, the factors affecting the vulnerability of the population are different, and the evaluation methods will be different. So this paper selects the upper reaches of Minjiang River and Qipangou Village as the research object, in which the upper reaches of Minjiang River is mesoscale range, Qipangou Village is small scale range, that is, typical area, and the evaluation index is screened according to the actual situation. The evaluation indexes of population vulnerability in upper reaches of Minjiang River and Qipangou Village were established and the differences of population vulnerability in two scales were analyzed respectively. The mesoscale evaluation of the upper reaches of Minjiang River is based on the information quantity model, and the map of vulnerability of the whole population in the upper reaches of Minjiang River is obtained. The small scale Qipangou village is a typical area. Based on the SOM model of MATLAB and the support of ArcGIS, the distribution map of vulnerability (vulnerability) of families in the study area under the threat of debris flow is obtained. The main research contents and results of this paper are as follows: (1) through screening, we select the revised population density, per capita GDP, the number of doctors per 10,000, the number of beds per 10,000 people, and the monitoring points. The ratio of labor force to six evaluation factors is used as the evaluation index of population vulnerability in the study area, taking 5050,050,5050,500? The upper reaches of Minjiang River are divided into 919,246 grid units, and the regional vulnerability map is obtained by the calculation of the information content model. The analysis shows that the overall population vulnerability of the upper reaches of Minjiang River is relatively low. The proportion of low vulnerable area is 91 09, and that of high population vulnerable area is 3. 78%. The high vulnerability areas are mainly concentrated in Wenchuan County, Maoxian County and Lixian County. (3) through the analysis and comparison of each evaluation factor map and the population zoning map of the upper reaches of Minjiang River, it can be seen that the population vulnerability is positively related to the revision of population density and is related to the number of hospital beds per 10,000 people. The number of doctors per capita, the monitoring points, and the proportion of the working population were negatively correlated. The more the population density is corrected in a region, the more vulnerable the system population will be, and the lower the vulnerability of the system population will be under the combined action of other factors. The population density, the number of hospital beds, the number of doctors and the GDP per capita are the key factors of the vulnerability of the population. The impact on the vulnerability of the population is greater, and the impact of the ratio of the working population and the monitoring points is relatively low. Combined with the interpretation and analysis of the image, 6 indexes, such as family size, age structure, health status, ethnic characteristics, education level and distance from debris flow gully, were selected as the evaluation factors of population vulnerability in Qipangou village, a typical area. Using self organized mapsto evaluate population vulnerability in study area the advantage of SOM model is that it can avoid the influence of human subjective factors on the weight of evaluation indexes. Areas with high vulnerability are concentrated in densely populated areas, The vulnerability is the largest in the gully, decreasing in the surrounding area, and the proportion in the extremely high and high vulnerable area is 3.52 and 4.90, respectively, the proportion is the smallest, and the proportion in the low, extremely low vulnerability area is 22.98 and 60.06, respectively, with the largest proportion. According to the SOM model, the U- matrix and 18 variables of the plane information map analysis can be found that the most vulnerable population is middle or large families, ethnic minorities, low education, 0-300m distance from debris flow gully households; The age structure and physical condition have certain influence on the vulnerability of population, but the influence degree is low, which is basically consistent with the actual situation in the study area.
【学位授予单位】：西南科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P642.23

【相似文献】