KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| 基于水文响应单元下流域土地利用/覆被变化对水文过程影响研究——以阿克苏河流域为例 | |
| 孟凡浩 | |
| Subtype | 博士 |
| Thesis Advisor | 刘铁 |
| 2018-06-05 | |
| Degree Grantor | 中国科学院大学 |
| Place of Conferral | 新疆乌鲁木齐 |
| Degree Discipline | 理学博士 |
| Keyword | SWAT 模型 水文响应单元 土地利用/覆被变化 水文响应 阿克苏河流域 SWAT Model Hydrological Response Unit Land Use/Cover Change Hydrological Process Aksu River Basin |
| Abstract | 受到全球气候变化以及人类活动的双重影响，干旱、半干旱区面临着一系列的生态问题，比如水资源短缺、植被退化、水土流失严重等，阿克苏河流域作为典型内陆河—塔里木河的主要水源，对阿克苏河流域下游和塔里木河干流的生态环境、社会经济等方面起到非常重要的作用，但是由于阿克苏河流域草地退化和耕地扩张增加了水文循环过程的不确定性。为了进一步了解干旱区土地利用/覆被变化对水文过程的影响，本论文结合“青年千人计划”《塔里木河流域缺资料区域的水文建模方法》，以阿克苏河流域作为研究区域，通过分析同一时间段内阿克苏河流域的地表覆被类型的变化、降水变化以及径流量的变化趋势，厘清土地利用/覆被变化和降水变化与径流变化的相关关系，阐明其影响，并利用基于水文响应单元的 SWAT 模型模拟不同土地利用/覆被变化的水文过程响应情况，以期为阿克苏河流域土地利用管理和水资源调控提供理论依据。主要研究成果如下：（1）研究时段内阿克苏河流域稀疏草地、落叶阔叶林、灌木林、湿地、冰/雪覆被类型呈减少趋势，草甸、草原、水体、建设用地、裸地、水田和旱地呈增加趋势；稀疏草地、草原、草甸覆被类型有 80%以上分布在海拔 2000 m 至 4000 m 区域，而耕地则主要分别在 1000 m 至 2000 m 的区域，裸地有 60%分布在 3500 m至 4000 m 的区域；耕地 90%以上分布在平原、微斜坡和缓斜坡，即坡度小于 5°以下，而草地主要分布在斜坡和陡坡上，即坡度在 5°-35°之间；所有土地利用/覆被类型在四个坡向上均有分布，虽然比例相差不多，但阴坡所占比例相对较大；而从空间转换来看，稀疏草地主要转出为裸地、耕地和草原等。（2）研究时段内阿克苏河流域降水量呈显著上升趋势（p_value<0.05），但上升趋势并不一致；通过分析沙里桂兰克、协合拉、源流之和、西大桥、上下游径流之差五个径流指标的线性趋势，发现前四者均呈现显著增加趋势（p_value<0.05），而上下游径流之差虽有增加趋势，但不显著（p_value=0.1035），说明下垫面环境可能已发生变化；利用双累积曲线法和 Mann-Kendall 检验法将阿克苏河流域研究时段划分为参考时段（1961-1990 年）和显著变化时段（1991-2011 年）两个阶段，40.89%的径流变化是由土地利用/覆被变化引起的，而其余的 59.11%来自于降水变化的影响，不过统计方法并未考虑人工建筑（堤坝等）对水文过程影响，因此导致高估气候变化和低估土地利用/覆被变化对径流变化的影响。（3）基于水文响应单元的 SWAT 模型模拟阿克苏河流域水文过程，通过 LH-OAT算法和 SUPI-2 算法对流域模型参数进行敏感性分析，发现降水梯度、温度梯度、相关融雪因子等参数对流域非常敏感；通过多尺度土地利用/覆被数据的模型结果对比发现阿克苏河流域 90 m 土地利用/覆被数据是流域模型适宜尺度；利用阿克苏河流域适宜尺度土地利用/覆被数据输入模型，模拟流域的径流量在日尺度、月尺度以及年尺度上与实测径流拟合效果均比较满意。总体上，相关性系数 R2以及纳什效率系数 ENS 均在 0.6 以上，相对误差小于 20%。（4）通过分析基于水文响应单元的 SWAT 模型对流域土地利用/覆被变化的水文响应的模拟缺陷，提出了一套子流域分割法来优化 LUCC 水文响应分析的准确度；通过设置不同情景（荒漠化、耕地扩张、草地恢复）和对比阿克苏河流域变化期和基准期土地利用/覆被变化对水文过程影响，发现春季洪峰量有明显增加，地表径流组分中坡面径流贡献比例增大，说明稀疏草地的退化和裸地与耕地扩张有增加地表径流的影响，并且改进后方法削弱了增加的量；此外在空间上可以看出改进后方法对坡面径流变化区域有较大影响。 |
| Other Abstract | Because of the dual impacts of climate change and human activities, the arid regionsuffers considerable ecological problems, such as water shortages and serious soilerosion. Tarim River Basin is the largest typical inland river in China. As one of themain headwater of the Tarim River Basin, the Aksu River Basin plays an important rolein the ecological environment and social economy to the downstream as well as themainstream of the Tarim River. However, due to the degradation of grassland andexpansion of farmland in the Aksu River Basin, the uncertainties of the hydrologicalcycle processes are enlarged. In order to further understand the impact of land use/coverchange on hydrological processes in arid regions, the Aksu River Basin was selected asthe study area. By using the trend analysis methods, the effects of land use/cover changeand precipitation variations on the runoff process are described. The SWAT model,which is based on the hydrological response unit, was used to simulate the response ofhydrological process to different land use/cover changes scenarios. The results of ourstudy is expect to provide a theoretical basis for land use management and waterresources regulation in the Aksu River Basin. The main results of this study weresummarized as follows.(1) The sparse grassland, deciduous broad-leaved forest, shrubbery, wetland, andice/snow cover in the Aksu River Basin presented a decreasing trend while meadows,grasslands, water bodies, construction land, bare land, paddy fields, and dry landspresented an increasing trend during the study period. More than eighty percentage ofthe sparse grassland, grassland and meadow cover types are distributed in the area withelevation range from 2000 m to 4000 m. However, the cultivated land is mainlydistributed in the area with elevation between 1000 m and 2000 m and sixty percentageof the bare land spread over the area with elevation range from 3500 m to 4000 m.Meanwhile, ninety percentage of the cultivated land is distributed in the plains area andin the areas with micro and gentle slopes (less than five degree). The grassland is mainly distributed on ramp and scarp with slope located between 5 degree and 35 degree. Allof the land use/cover types could be found on the four aspects. Although proportions ofthe land use/cover distributions are similar, relative larger percentage was presented inthe cloudy slopes rather than the sunny slopes. The type of sparse grassland is mainlytransferred to bare land, farmland and grassland when the spatial transformation wasconcerned.(2) The precipitation in the Aksu River Basin showed a significant increasing trend(p_value<0.05) with spatial difference. By the analysis of the long-term linear trends ofthe runoff in SLGLK, XHL, XDQ, the sum of the sources flow and the flow differencebetween upstream and downstream during the study period, a significant increasingtrend (p_value<0.05) was discovered with exception of the flow difference. Theincreasing trend of the flow difference didn’t pass the significance test at 95%confidence interval (p_value=0.1035). It indicated that the environment of theunderlying surface was likely to be changed. By using double-accumulation curvemethod and Mann-Kendall test method, the study period of Aksu River Basin wasdivided into two phases: the reference period (1961-1990) and the changing period(1991-2011). The contribution of land use/cover change and the increased precipitationto runoff changes were 40.89% and 59.11%, respectively. However, the statisticalmethod does not take into account the influence of artificial buildings (dikes, etc.) onhydrological processes, and which have led to an underestimation of climate changeand an overestimation of the impact of land use/cover change on runoff changes.(3) The SWAT model was used to simulate the hydrological process of the AksuRiver Basin and the sensitivity of the model parameters was analyzed by the LH-OATand the SUPI-2 algorithm. It was found that the gradients of precipitation andtemperature and the parameters related to snow melting factors were very sensitive tothe watershed. By comparing the model results driven by multi-scale land use/coverdata, the land use/cover data with 90 m spatial resolution in the Aksu River Basin isexpected to the suitable scale for the SWAT model. The simulated runoff driven by theland use/cover data with 90 m grids showed satisfactory performance at the daily,monthly, and yearly scales with low relative error (less than 20%) and high R2 (over0.6) as well as NS (over 0.6).(4) By analyzing the shortcoming of the SWAT model in simulating thehydrological response to land use/cover change, a sub-basin segmentation method wasproposed for the purpose of accuracy improvement. By setting different scenarios(desertification, expansion of cultivated land and restoration of grassland), the effectsof land use/cover change on hydrological processes during the changing period werecompared with the reference period. The results showed that the amount of the springflood peak increased significantly and the contribution of the slope runoff to surfacerunoff also increased. It indicated that the degeneration of sparse grassland and theexpansion of bare land and farmland are able to increase the surface runoff, while theimproved method weakened the amount of increase; In addition, it can be also foundthat the improved method has a great impact on the area with slope runoff change. |
| Subject Area | 地图学与地理信息系统 |
| Language | 中文 |
| Document Type | 学位论文 |
| Identifier | http://ir.xjlas.org/handle/365004/14933 |
| Collection | 研究系统_荒漠环境研究室 |
| Affiliation | 中国科学院新疆生态与地理研究所 |
| First Author Affilication | 中国科学院新疆生态与地理研究所 |
| Recommended Citation GB/T 7714 | 孟凡浩. 基于水文响应单元下流域土地利用/覆被变化对水文过程影响研究——以阿克苏河流域为例[D]. 新疆乌鲁木齐. 中国科学院大学,2018. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment