【作者】 孙泽祥；

【导师】 孙希华；

【作者基本信息】 山东师范大学 ， 地图学与地理信息系统， 2012， 硕士

【摘要】 土壤侵蚀是人类发展过程中所面临的重大环境问题，已经危及到人类的生存，越来越受到世界各国政府及科技工作者的关注。土壤侵蚀的研究，与人类自身的生存、发展息息相关，已经成为21世纪国际土壤学、农业科学及环境科学界共同关注的热点问题，开展土壤侵蚀的研究具有极其重要的意义。沂河发源于山东省淄博市沂源县，是山东省第二大河。沂源县地处沂河上游地区，是水利部确定的全国水土流失重点治理县。本研究以刘宝元等建立的中国土壤流失方程(China Soil Loss Equation)为基础，利用2009年4月份SPOT 10m分辨率多波段影像和2.5m分辨率全色影像为基础遥感数据，ASTER GDEM 30m分辨率DEM数据为地形数据来源，山东省1:50万土壤图为土壤数据来源，另外结合其它调查统计资料，在GIS软件和遥感图像处理软件的支持下，将所有基础数据，经过几何校正和投影变换等操作后，转换为相同的参考基准，便于进行数据的叠加分析。本研究利用水利部推广的网格抽样调查法，对沂源县整个研究区域进行4级网格划分，其中第三级网格为5km×5km网格，4级网格为将3级网格进行5×5划分，位于三级网格中心的1km×1km的网格称为基本抽样单元。通过对基本抽样单元的野外调查，利用1m×1m样方，获取样方的坐标，植被覆盖度，植被类型、耕作措施、水土保持措施等数据，在内业时根据野外调查的数据，对照遥感影像图，建立解译标志。分别从降雨侵蚀力、土壤可蚀性、坡长、坡度、植被覆盖度、工程措施和耕作措施等几个方面对影响土壤侵蚀的因子进行定量分析，找出其中的规律，进而计算出沂源县土壤侵蚀模数，最后将网格抽样调查法所获得的结果与传统计算方法取得的结果进行对比分析，对研究区内整体土壤侵蚀分布规律作出分析，并提出相应的水土保持治理建议。经过计算分析，研究区平均土壤侵蚀模数为2361 t/km~2 a，年流失量总量为346.42万t。其中轻度及以下侵蚀面积为1015.45 km~2，占研究区总面积的69.20%；中度侵蚀面积230.58km~2，占研究区总面积的15.71%，分布面积和比重相对较大；强烈侵蚀面积89.81 km~2，占研究区总面积的6.12%；极强烈侵蚀面积61.57 km~2，占研究区总面积的4.20%；剧烈侵蚀面积31.15 km~2，占研究区总面积的2.12%，是研究区中分布面积和比重最小的。中度及以上土壤侵蚀主要分布在海拔200 800m的丘陵、低山地带，该地区海拔相对较高，地势起伏较大是土壤流失的重点地区，该地区植被类型主要为人工植被，多为农作物，受人为因素干扰较重，因此土壤流失严重。将网格抽样调查法获得的结果与传统的土壤侵蚀计算方法获得的结果进行比较发现，利用网格抽样调查法计算得出抽样单元平局土壤侵蚀模数为2330 t/km~2 a；而采用传统计算方法得出的研究区平均土壤侵蚀模数为2361 t/km~2 a，两者相差仅31 t/km~2 a，说明完全可以采用网格抽样调查法来取代传统的计算方法对研究区进行监测，从而大大节省人力、物力和财力，并且可以收到比较好的效果。通过对三种不同网格抽样法的比较发现，规则网格抽样法获得的平均土壤侵蚀模数是最接近研究区整体计算结果的，为2330 t/km~2 a，而考察点抽样单元平均土壤侵蚀模数为1822 t/km~2 a，综合抽样单元的平均土壤侵蚀模数为2018 t/km~2 a。因此采用规则抽样单元比采用其它的抽样单元计算结果更接近真实值。更多还原

【Abstract】 Soil erosion is the major environmental problem human face in the process of developments,which has threatened the survival of the human race and attracted more and more attentions fromthe government, scientists and technologists in global world. Study on the soil erosion is closelyrelated to human survival and development, which has become the international issue soilscience, agricultural science and environmental science concern in the 21th century. It is of greatsignificance to carry out research on the soil erosion.Yihe River in Shandong province Zibo city Yiyuan County is the second great river. YiyuanCounty is located in the Yi River upstream regions, which is the key soil erosion control countythe ministry of water resources identified. The study is based on Chinese soil loss equationestablished by Liu Baoyuan The paper takes SPOT 10m resolution multispectral image and 2.5mresolution panchromatic images in the April of 2009 as the remote sensing data, ASTER GDEM30m resolution DEM data as terrain data sources and 1:500000 soil map in the Shandongprovince as soil data sources, combining the other survey data. It deals with all basic datathrough geometric correction and projection transformation operation, and then converts the datato the same reference to facilitate data overlay analysis.This study uses grid sampling survey method the Ministry of water resources promoted toconvert the study area of Yiyuan County in 4 levels. The third level is 5km×5km grid and thefourth level divides in 5×5 on the basis of the third level. Basic sampling unit is located on thecenter grid in the third level. Based on the field investigation in the basic sampling unit, thepaper makes use of 1m×1m quadrat to obtain the plot coordinates, vegetation coverage,vegetation types, tillage, soil and water conservation measures and other data. In order toestablish the interpretation signs, it is necessary to compare the remote sensing image with thefield survey data in the indoor room. It analysts quantitatively the rainfall erosion force, soilerodibility, slope length, slope, vegetation coverage, engineering measures, cultivation measuresand many other aspects that affect soil erosion to find the law. And then calculate the modulus ofsoil erosion in Yiyuan County. Finally, through comparing the grid sampling method results withthe results obtained from the traditional calculation method, it makes an analysis the overalldistribution of soil erosion law and puts forward the corresponding suggestions of soil and waterconservation management.From the studies, we know that the average soil erosion modulus is 2361t/km~2·a and the lossamount of the year is up to 3.4642 million ton. The mild and even worse erosion area is about1015.45 km~2, accounting for 69.2% of the total area of the study area. Moderate erosion area isabout 230.58 km~2, accounting for 15.71% and it has large distribution area and proportion. Strong soil erosion area is 89.81 km~2, accounting for 6.12%. Extreme intense erosion area is61.57 km~2, accounting for 4.20%. Severe erosion area is about 31.15 km~2, accounting for 2.12%and it has the smallest area and proportion. Moderate and above soil erosion areas are mainly inthe hills of an altitude of 200-800m and many low mountain regions, which are relatively highaltitude. The area full of low and high is the main district of soil erosion. Vegetation in the regionis main types of artificial vegetation, which are mostly crops and affected by human factorheavily. So the soil erosion is serious.Comparing grid sampling results with the traditional calculation results, it is shown that thesoil erosion modulus obtained from grid sampling investigation method is 2330 t/km~2·a. Theaverage soil erosion modulus acquired through the traditional calculation method is 2361 t/km~2·a.The difference between the two methods is 31 t/ km~2·a. This can monitor the study area throughgrid sampling method instead of traditional calculation method, which can save manpower,material and financial resources and can receive good results.By comparing the three different grid sampling methods, we can see that average modulusof soil erosion obtained from the regular grid sampling method is the most close to the calculatedresults of the overall study area, which is about 2330 t/km~2·a. But the average soil erosionmodulus in the inspected sampling unit is about 1822 t/km~2·a and the average soil erosion in theintegrated sampling unit is 2018 t/km~2·a. Therefore the calculated results in the use of regularsampling unit than the other sampling unit are closer to the true value.更多还原