黄土高原北洛河上游流域退耕还林草的水文响应模拟

发布时间：2018-03-23 19:51

本文选题：径流泥沙　切入点：模型模拟　出处：《西北农林科技大学》2017年博士论文

【摘要】：黄土高原地区水土流失严重,水资源短缺,生态系统脆弱。为了遏制严重的土壤侵蚀,1999年后,我国在黄土高原地区实施了大范围的退耕还林还草等水土保持措施。土地利用及土地覆盖发生的显著变化,深刻影响着黄土高原的生态系统-水文响应过程,成为多领域专家学者关注的热点地区。本研究以黄土高原丘陵沟壑区北洛河上游吴起水文站控制范围为研究区域,以实测数据为基础,用Mann-kendall非参数检验法等水文统计法对1963-2012年流域的降雨、径流和输沙量进行了趋势和阶段分析;解译Landsat TM影像,分析了流域1995、2007、2010和2014年等不同时期植被覆盖度和土地利用的时空变化;对SWAT模型进行参数率定和模型验证,定量评价了土地利用和气候变化对径流量变化的贡献程度,模拟流域内径流、蒸发和土壤含水量等地表生态水文过程并进行时空变化分析;利用修正通用流失方程RUSLE模拟流域内土壤侵蚀时空变化趋势;最后通过设定不同土地利用和气候的变化情景,模拟分析了不同情景下的生态水文响应。本研究取得的主要成果如下:(1)采用1979年和2002年为阶段划分时间节点,分析了流域气象水文要素的年变化趋势和阶段变化特征。北洛河上游流域1963-2012年间平均降水量为440.3mm。50年间降雨量的变化表现出不显著的下降趋势(P0.05)。1963-1978年间降雨量均值为416.9mm,但年降雨量呈不显著下降趋势,趋势线性率为-2.3 mm·a-1;1979-2001年期间降雨量均值为441.9mm,呈不显著上升趋势,趋势线性率为0.2 mm·a-1;2002-2012年间降雨量均值为449.1mm,呈不显著下降趋势,下降斜率为-0.4 mm·a-1。(2)流域1963-2012年的径流量和输沙量均呈现出显著减少趋势(P0.01),年均变率分别为-0.32 mm·a-1、-170.18 t·km-2·a-1。1963-1978年间径流量均值为31.6mm,呈不显著下降趋势,趋势线性率为-0.9mm·a-1;1979-2001年期间径流量均值为28.5mm,呈不显著上升趋势,趋势线性率为0.3 mm·a-1;2002-2012年间径流量均值为16.5mm,呈显著下降趋势(P0.05),下降斜率为-0.7 mm·a-1。1963-1978年间年输沙模数均值为13828.1 t·km-2·a-1,呈不显著下降趋势,趋势线性率为-851.1 t·km-2·a-1;1979-2001年期间年输沙模数均值为10375.1 t·km-2·a-1,呈不显著上升趋势,趋势线性率为148.9 t·km-2·a-1;2002-2012年间年输沙模数均值为2023.3 t·km-2·a-1,呈显著下降趋势(P0.05),下降斜率为-376.8 t·km-2·a-1。(3)1995年北洛河上游流域内以耕地和草地为主,其面积分别占总面积的46.79%和49.94%,呈现出以农、牧地为主的土地利用结构特征。2000年,耕地和草地为主要用地类型的格局没有变化,耕地面积稍有减少。1995-2000年,耕地面积减少了8.8%,林、草地面积分别增加了5.1%、8.5%。而2010年耕地面积锐减68.8%,林、草地面积扩大了6.5倍和27.0%。研究区农、林、草用地面积的数量结构,从1990年的1:0.1:1.1,2000年的1:0.1:1.3,改变为2010年的1:1.4:4.3。2000-2010年间耕地、草地和林地的变化幅度显著大于1995-2000期间。从植被覆盖度分析来看,北洛河上游流域(刘家河站控制范围)在1995-2014年平均覆盖度从20.21%增加到51.22%。通过对植被覆盖度进行分级,并对不同级别间的转化进行分析发现,1995年到2007年,≤40%的低、中低盖度级别植被的面积数量迅速和持续减少,先转化为40-60%的中盖度植被,到2014年,又迅速向盖度60-80%和80%的植被转化的特征。海拔越低,坡度越陡,越向阴坡,低盖度植被向高盖度植被转化程度越大,面积越广,三时期植被覆盖差异越大。(4)SWAT模拟流域近30年中气候和土地利用变化下的水文响应。以1986-1995年为基准期,1996-2005年和2006-2012年分别为气候和人类活动综合影响下的两个变化期。SWAT模拟结果表明,人类活动变化对径流减少的贡献程度从第二阶段的56.5%增加至第三阶段的77.8%。随着流域内植被覆盖度逐渐增加,流域时段平均地表径流量从1986-1995年间的26.8mm,减少至1996-2005年间20.9mm,2006-2012年间的11.9mm。相应地,流域时段平均土壤含水量41.8mm,减少到33.7mm和26.9mm。而蒸发量从362.7mm,增加到370.1mm和406.7mm。空间上地表径流量和土壤含水量均呈现出由东北部向西南部、南部和东部呈逐渐递减的趋势。而蒸发量在空间上的变化趋势则与地表径流量和土壤含水量相反,表现出由北部、东北部地区逐渐向南部和西南部地区递增趋势。(5)采用RUSLE模型模拟了1990、2000和2010年三时期土壤侵蚀时空变化特征。结果表明,三时期间流域平均土壤侵蚀模数从15220.09、6471.08,减小到2591.94t·km-2·a-1。3时期的土壤侵蚀量在地形上的分布表现出趋同性,即高程上均在1475-1575m内平均侵蚀模数和侵蚀量表现出峰值。随着坡度增加,平均侵蚀模数增加,流域内75%以上的侵蚀量均来自于15°区域。3时期平均侵蚀模数均遵循阳坡半阳坡半阴坡阴坡的规律。(6)分析流域土地利用的地形地貌分布,进行优化设置,模拟产流和侵蚀过程。结果表明,空间优化配置情景下,与原2010年土地利用相比,耕地由原来的14.59%降到9.99%;草地由原来的63.43%下降到49.79%;林地则呈上升趋势,由20.95%增加到了39.20%。相对于2010年土地利用和土地覆盖,流域径流深降低了16.98%;平均侵蚀模数降低了18.65%。采用2010年土地利用,分析未来气候情景下的水文响应发现,在年尺度下,RCP4.5和RCP8.5两个情景下的2021-2050模拟径流深与历史期模拟值相比均有不同程度的增加。在月尺度下,未来2021-2050年内,RCP4.5和RCP8.5两个情景下的气候数据预测径流深表明,径流深在1-3月份和10-11月份将会呈现出不同程度的减少趋势,而6-8月份的径流深在RCP4.5情景下可能呈现增加的趋势,在RCP8.5情景下可能呈现出减少的趋势。
[Abstract]:Serious soil erosion in the Loess Plateau area, the shortage of water resources, fragile ecosystem. In order to curb soil erosion serious, after 1999, China in the Loess Plateau region, the implementation of a large range of returning farmland to forest and grass for soil and water conservation measures. The land use and land cover change was happened, a profound impact on ecosystem of Loess Plateau the hydrological response process, has become a hot area of attention of scholars. This study in Hilly and gully area of Loess Plateau Wuqi Beiluohe river hydrological station control as the study area, based on the measured data, using Mann-kendall nonparametric test method of watershed hydrology statistics 1963-2012 years rainfall, analyzes the trend and stage the runoff and sediment yield; the interpretation of Landsat TM images, analyzed the basin in 199520072010 and 2014 in different periods including vegetation coverage and land use spatial and temporal variation of SWAT; Model parameter calibration and model validation, quantitative evaluation of land use and climate change on the contribution of runoff, runoff simulation, analysis of temporal and spatial variation of evaporation and soil moisture and surface hydrological process; revised universal soil loss equation RUSLE simulation of watershed soil erosion variation with time and space utilization; finally through the change different scenarios of land use and climate, simulation analysis of the eco hydrological response under different scenarios. The main results are as follows: (1) by 1979 and 2002 for the stage division between nodes, analyzed the basin meteorological and hydrological factors change trend and stage change characteristics. The upstream watershed of Beiluohe River 1963-2012 years average precipitation changes during 440.3mm.50 precipitation showed a decrease trend is not significant (P0.05) between.1963-1978 rainfall mean 416.9mm, but The annual rainfall was not significantly decreased, linear trend rate of -2.3 mm and A-1; 1979-2001 years of rainfall mean 441.9mm, showed no significant upward trend, linear trend rate was 0.2 mm - A-1; 2002-2012 years of rainfall mean 449.1mm, was not significantly decreased, decreased mm a-1. slope of -0.4 (2) basin 1963-2012 years of runoff and sediment discharge showed a significant decreasing trend (P0.01), the annual change rate was -0.32 mm A-1, -170.18 t KM-2 a-1.1963-1978 from runoff mean 31.6mm, was not significantly decreased, linear trend rate of -0.9mm - A-1; during the 1979-2001 year runoff mean 28.5mm that was not a significant upward trend, linear trend rate was 0.3 mm - A-1; 2002-2012 years of runoff mean 16.5mm, decreased significantly (P0.05), -0.7 mm a-1.1963-1978 is down slope during the annual sediment transport modulus value was 13828.1 T - KM-2 - A-1 That was not a significant downward trend, linear trend rate was -851.1 T - KM-2 - A-1; during the 1979-2001 years mean annual sediment transport modulus is 10375.1 T - KM-2 - A-1, was not a significant upward trend, linear trend rate of 148.9 T - KM-2 - A-1; 2002-2012 years annual sediment transport modulus value was 2023.3 t KM-2. A-1 was significantly decreased (P0.05), -376.8 t down slope as KM-2 a-1. (3) 1995 the Beiluohe river basin is dominated by farmland and grassland, the area accounted for 46.79% of the total area and 49.94%, showing to the agricultural, rangeland based land use structure for.2000 years, cultivated land and grassland the main land use pattern has not changed, cultivated land area decreased slightly in.1995-2000, arable land decreased by 8.8%, forest, grassland area increased by 5.1% 8.5%., while the cultivated land area in 2010 dropped 68.8%, forest, grassland area increased by 6.5 times and 27. Research area of agricultural, forest, grass land The number of product structure, 1:0.1:1.3 from the 1990 1:0.1:1.12000 year, changed to cultivated land during 1:1.4:4.3.2000-2010 2010, variation of grassland and woodland was higher than 1995-2000. During the period from vegetation coverage analysis, the upstream watershed of Beiluohe River (Liu Jia He Railway Station control range) in 1995-2014 years the average coverage increased from 20.21% to 51.22%. were graded by the vegetation coverage, and the transformation between different levels of analysis, from 1995 to 2007, low than 40%, the number of low level coverage area of vegetation rapid and sustained reduction, first converted into 40-60% in 2014, the characteristics of vegetation coverage, and to the rapid transformation of 60-80% and 80% of the vegetation coverage. The higher the altitude low, the slope is steep, the closer to the north, low coverage vegetation to the high vegetation conversion degree is bigger, more wide area, the three period of vegetation cover the greater the difference. (4) SWAT simulation basin nearly 30 In the year of climate and land use change. The hydrological response to 1986-1995 years for the reference period of 1996-2005 years and 2006-2012 years respectively, the comprehensive effects of climate and human activities under the two changes to the.SWAT simulation results show that the contribution of human activities change on runoff reduction from the second phase of the 56.5% stage with 77.8%. increased to third watershed vegetation coverage increased gradually, the average surface runoff from the basin during the 1986-1995 years of 26.8mm, reduced to 1996-2005 20.9mm, 2006-2012 11.9mm. corresponding to the time river basin average soil water content of 41.8mm, 33.7mm and 26.9mm. to reduce the amount of evaporation from 362.7mm increased to 370.1mm and 406.7mm. space on the surface runoff and soil the water content showed from the northeast to the southwest, South and East is a gradually decreasing trend. While the change of evaporation in space and trend Surface runoff and soil moisture on the contrary, exhibited by the north, northeast to the South and southwest area gradually increasing trend. (5) using RUSLE model to simulate the spatial and temporal variation characteristics of soil erosion in 19902000 and 2010 three period. The results showed that during the three average soil erosion modulus decreased from 15220.096471.08 to distribution 2591.94t - KM-2 - a-1.3 during the period of soil erosion in the terrain shows convergence, including elevation are within the 1475-1575m average erosion modulus and erosion showed a peak. With the increase of slope, the average erosion modulus increased, erosion amount more than 75% within the basin are from 15 ~.3 period were regional average erosion modulus follow on sunny and semi sunny slopes semi shady shady rules. (6) analysis of topography distribution of land use, optimization settings, simulation of runoff and erosion processes. The results show that the optimal allocation of space Under the situation, compared with the original land use in 2010, arable land was reduced from 14.59% to 9.99%; grassland decreased from 63.43% to 49.79%; forest land increased by 20.95%, compared to 2010 increased to 39.20%. of land use and land cover, the runoff decreased by 16.98%; the average erosion modulus decreased by 18.65%. land in 2010 the analysis of future climate scenarios, hydrological response found in the annual scale, 2021-2050 RCP4.5 and two RCP8.5 simulated runoff scenarios and deep historical period simulation value compared to increase in varying degrees. In the month scale, the next 2021-2050 years, the climate data of runoff prediction and RCP8.5 RCP4.5 two under the situation of deep show that the depth of runoff in 1-3 month and 10-11 months will show different degrees of decrease, and the runoff 6-8 month deep under the RCP4.5 scenario may show an increasing trend, in RCP8.5. There may be a decreasing trend in the scene.

【学位授予单位】：西北农林科技大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：S157

【相似文献】