黄河中游多沙粗沙区侵蚀产沙变化特征及影响因子分析

发布时间：2018-01-02 18:24

  本文关键词：黄河中游多沙粗沙区侵蚀产沙变化特征及影响因子分析　出处：《中国科学院研究生院(教育部水土保持与生态环境研究中心)》2016年博士论文　论文类型：学位论文

  更多相关文章： 土壤侵蚀 植被分区 高程分区 侵蚀类型分区 归一化植被指数 偏最小二乘回归

【摘要】：研究黄土高原地区侵蚀产沙的时空变化特征及规律及其驱动因素,对黄土高原水土流失的防治和生态环境的改善具有重要的指导意义。因此,本研究以黄河中游多沙粗沙区为研究对象,利用归一化植被指数(NDVI)数据,分析对研究区侵蚀产沙强度影响较大的植被的变化趋势;在此基础上,利用水文站实测资料、收集到的已出版或发表的有关黄土高原地区流域侵蚀产沙的资料,运用Mann-Kendall检验分析方法,研究黄河中游多沙粗沙区侵蚀产沙的演变特征;根据研究区植被、地形地貌以及土壤侵蚀类型特征,将研究区划分为不同的植被分区、高程分区以及土壤侵蚀类型分区,分析不同分区的侵蚀产沙情况;选取合适的地形地貌、气候、土壤、土地覆盖等指标,利用偏最小二乘回归,排除变量间的共线性问题,定量识别影响黄河中游多沙粗沙区不同侵蚀类型分区产沙的主控因子。本研究的主要结论如下:(1)通过分析1982-2013年黄河中游多沙粗沙区NDVI的时空分布发现,四季NDVI值分别是,0.186(春季),0.306(夏季),0.275(秋季),0.123(冬季)。研究区生长季和年平均NDVI均为0.313,另外研究区南部和东部NDVI值要高于研究区西部NDVI值。从研究区NDVI值变化趋势来看,四季NDVI年增加速度由快到慢依次为夏季(0.0055)秋季(0.0053)春季(0.0040)冬季(0.0004),生长季NDVI和年NDVI年平均增长速度分别为0.0046和0.0052。通过Mann-Kendall方法检验分析发现,春季和秋季NDVI突变年份为2002年,冬季的突变年份为2008年;夏季、生长季和全年NDVI的突变年份均有两个,分别是1998至1999年之间和2002年。1982-2013年研究区气候年变化呈现降雨和温度增加的“暖湿化”趋势。利用偏相关分析发现,研究区NDVI均与降雨量呈正相关关系;冬季研究区NDVI的变化与温度呈负相关关系,而其它各研究时段内NDVI的变化与温度呈正相关关系;并且,植被生长对温度的响应存在一个月的滞后期,而降雨无滞后效应。(2)通过分析黄河中游多沙粗沙区21条主要支流的平均侵蚀产沙强度发现,未治理期(1955-1969年)的研究区平均侵蚀产沙强度为9949.7t/km2?a,治理期(1970-2007年)的平均侵蚀产沙强度明显小于未治理期,平均侵蚀产沙强度仅为4710.5t/km2?a,较未治理期减少53.8%。其中,先治理期(1970-1989年)和后治理期(1990-2007年)研究区平均侵蚀产沙强度分别为6316.7t/km2?a和2925.8t/km2?a,较未治理期分别减少38.1%和71.2%。未治理期研究区侵蚀产沙强度≥5000t/km2?a的总面积是51752.5km2,先治理期研究区侵蚀产沙强度≥5000t/km2?a的总面积是36222.2km2,较未治理期减幅为30%。利用研究区8个水文站(偏关站、皇甫站、韩家峁站、横山站、青阳岔站、子长站、甘谷驿站和吉县站)1960-1989年的侵蚀产沙强度,通过mann-kendall方法检验分析发现,8个水文站年侵蚀产沙强度均呈下降趋势,突变年份大多在70年代和80年代后期。(3)利用黄河中游多沙粗沙区44个水文站1955-1989年和2006-2011年的水文资料,并且将黄河中游多沙粗沙区划分为不同的植被分区、高程分区和侵蚀类型分区,通过统计不同分区的侵蚀产沙情况发现,各植被分区、高程分区和侵蚀类型分区的侵蚀产沙强度随着时间的变化均呈现出降低的趋势。从不同植被分区来看,在各时间阶段(1970年前、1970-1979年、1980-1989年和2006-2011年)侵蚀产沙强度较高的植被分区是温带森林草原亚地带;从不同流域高程分区来看,在各时间阶段(1970年前、1970-1979年、1980-1989年和2006-2011年)侵蚀产沙强度均最小的高程分区为人类活动较少的高程1400m的流域;而侵蚀产沙强度较大的高程分区是人类活动较强烈的高程1200m的流域;从不同侵蚀类型分区来看,在各时间阶段(1970年前、1970-1979年、1980-1989年和2006-2011年)侵蚀产沙强度均最小的侵蚀类型区是森林黄土丘陵沟壑区;而侵蚀产沙强度在各阶段均较大的侵蚀类型区是黄土平岗丘陵沟壑区、黄土峁状丘陵沟壑区和黄土梁状丘陵沟壑区。(4)选取降雨、径流、土壤、地形地貌、土地利用和景观格局指数共28个指标,利用偏最小二乘回归分析方法,对研究区七个侵蚀类型分区侵蚀产沙强度的主控因子进行分析和辨别。研究发现,不同侵蚀类型区侵蚀产沙强度的主控因子有所不同。但整体而言,各侵蚀类型区侵蚀产沙强度的偏最小二乘回归模型第一和第二成分主要是由降雨、径流、平面曲率、土壤黏粒含量和砂粒含量、草地和林地百分比、最大斑块指数、蔓延度、香农多样性指数和辛普森多样性指数主导;通过自变量的变量投影重要性指标(VIP值)可以看出,降雨、径流、林地百分比、周长面积分维数和聚集度指数对不同侵蚀类型区侵蚀产沙强度具有较大影响。从自变量的回归系数来看,不同侵蚀类型区侵蚀产沙强度随降雨和径流的增加而增加;风沙黄土丘陵沟壑区、黄土平岗丘陵沟壑区、黄土峁状丘陵沟壑区和森林黄土丘陵沟壑区侵蚀产沙强度随林地百分比的增加而减少;风沙黄土丘陵沟壑区、黄土梁状丘陵沟壑区、黄土峁状丘陵沟壑区和森林黄土丘陵沟壑区侵蚀产沙强度与周长面积分维数呈负相关关系;风沙黄土丘陵沟壑区、黄土梁状丘陵沟壑区和黄土峁状丘陵沟壑区侵蚀产沙强度与聚集度指数呈正相关关系。
[Abstract]:Study on temporal and spatial variation of erosion on loess plateau characteristics and regularity of sediment and its driving factors, has an important guiding significance for the prevention and control of soil erosion in the Loess Plateau and the improvement of the ecological environment. Therefore, the study on the coarse sand area in the middle reaches of the Yellow River as the research object, using the normalized vegetation index (NDVI) data analysis, trend of research soil erosion intensity of vegetation; on this basis, the use of hydrological data, data of erosion and sediment collected has been published in the relevant area of the Loess Plateau Watershed, use the analysis method of Mann-Kendall test, the evolution characteristics of erosion and sediment yield of the middle reaches of the Yellow River sandy area; according to the vegetation of the study area. The topography and soil erosion types, the study area is divided into different zones of vegetation, elevation and soil erosion type partition partition partition analysis Soil erosion; suitable topography, climate, soil, land cover index, using partial least squares regression, eliminate collinearity among the variables, the main control factors affecting quantitative identification of coarse sand area in the middle reaches of the Yellow River in different types of erosion and sediment yield of the partition. The main conclusions of this study are as follows: (1) through time and space NDVI analysis of the distribution of coarse sand area in the middle reaches of the Yellow River 1982-2013 found that four NDVI values are 0.186 (spring), 0.306 (summer), 0.275 (Autumn), 0.123 (winter). The study area was 0.313 NDVI the average growth of Ji Henian, also of the southern and Eastern NDVI was higher than that of the western part of the study area NDVI from the research value. The NDVI value change trend, four NDVI increase speed from fast to slow in summer and autumn (0.0055) (0.0053) (spring 0.0040) and winter (0.0004), NDVI and NDVI in growth season, the average annual growth rate of 0.0046 and 0.005 respectively. 2. by the method of Mann-Kendall test analysis showed that the NDVI mutation in spring and autumn for the year 2002, the 2008 winter change year; summer growing season and the annual NDVI mutation year has two, respectively, between 1998 and 1999 and 2002.1982-2013 on climate change presents the annual rainfall and temperature increase the "warm" trend using partial correlation analysis showed that the NDVI of the study area were positively correlated with rainfall and temperature changes in winter; the study area of NDVI is negatively correlated, while the other during the period of study. The changes of NDVI positive relationship with temperature; and the vegetation growth responses to temperature are one month lag, but no rain lag effect. (2) through the analysis of the coarse sand area in the middle reaches of the Yellow River River and 21 major tributaries of the average erosion intensity is not found, governance period (1955-1969 years) of Qu Ping erosion intensity 9949.7t/km2? A, treatment period (1970-2007 years) the average erosion intensity is significantly lower than that of non treatment period, the average sediment yield intensity is only 4710.5t/km2? A, is not governance reduce 53.8%. the first treatment period (1970-1989 years) and after the treatment period (1990-2007 years) of the average sediment yield intensity are 6316.7t/km2? A and 2925.8t/km2? A, compared with non governance period were reduced by 38.1% and 71.2%. governance study area of soil erosion and sediment yield strength is not less than 5000t/km2? A total area is 51752.5km2, the first sand erosion strength more than 5000t/km2 period in the study area? A total area of a is 36222.2km2, with a decrease of 30%. during the period of governance the study area of 8 hydrological stations (Panguan station, Huangfu station, Mao Han station, Hengshan station, Qingyang Station Branch, sub station and Gangu station long, Jixian station) 1960-1989 years of erosion intensity, by the method of Mann-Kendall test analysis showed that 8 hydrological stations in erosion The strength decreased, the mutation year mostly in 70s and late 80s. (3) the use of coarse sand area in the middle reaches of the Yellow River 1955-1989 years and 2006-2011 years hydrological data of 44 hydrological stations, and divides the middle reaches of the Yellow River sandy area for different vegetation zoning, zoning zoning and Gao Cheng erosion, through statistics in different areas of erosion the sediment is found, the vegetation regionalization, erosion intensity and erosion of the Gao Cheng type partition partition with the change of time showed decreasing trend. From different vegetation regionalization, at each time stage (before 1970, 1970-1979, 1980-1989 and 2006-2011) erosion intensity high temperate vegetation regionalization the forest steppe zone; Gao Cheng basin from different district, in different phases (before 1970, 1970-1979, 1980-1989 and 2006-2011) Gao Cheng erosion sediment yield intensity were the smallest The elevation of 1400m basin partition for human activities less; and the erosion of sediment elevation intensity of the larger watershed partition is the elevation of 1200m human activity is more strong; from the partition different erosion types, in different phases (before 1970, 1970-1979, 1980-1989 and 2006-2011) erosion intensity are the smallest type of erosion the forest area in Loess Hilly Gully Region; and the erosion intensity in different stages of large erosion type region is in Loess Hilly and gully region of the Loess Pinggang, hilly loess region and Liang hilly and gully region. (4) rainfall, runoff, soil, topography, land use and landscape pattern index 28 indicators, using partial least squares regression analysis method, the seven types of erosion area erosion on the main control factors of sediment yield strength were analyzed and identified. The study found that different types of erosion soil erosion intensity The main control factors are different. But on the whole, the erosion area erosion and sediment yield intensity of partial least squares regression model of the first and the second component is mainly composed of rainfall, runoff, surface curvature, soil clay content and sand content, grassland and woodland percentage, largest patch index, contagion, diversity index and Shannon Simpson the diversity index of leading indicators; through the importance of the independent variable projection (VIP) can be seen, rainfall, runoff, land percentage, perimeter area fractal dimension and the aggregation index has a great influence on the different types of erosion soil erosion intensity. The independent variables from the regression coefficient, increases with the increasing of rainfall and runoff and sediment yield the intensity of erosion in different soil erosion area; sandy loess hilly and gully region, the Loess Hilly and gully area of Ping Gang, the intensity of sediment erosion in Loess Hilly Gully Region of loess hilly gully region and forest With the increase of the percentage of woodland decreased; sandy loess hilly and gully region, the Loess Liang hilly and gully region, there was a negative correlation between the fractal dimension of sediment strength and perimeter area erosion in Loess Hilly Gully Region and hilly forest in Loess Hilly Gully Region; sandy loess hilly and gully region, the intensity of sediment erosion in Loess Hilly Gully Region and shaped beam the Loess Hilly Gully Region and the aggregation index was positively correlated.

【学位授予单位】：中国科学院研究生院(教育部水土保持与生态环境研究中心)
【学位级别】：博士
【学位授予年份】：2016
【分类号】：S157
，

本文编号：1370399