KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| 内陆河高寒山区洪水发生过程研究 | |
| Alternative Title | Study on the occurrence process of inland river in the alpine mountainous area |
| 段永超 | |
| Subtype | 博士 |
| Thesis Advisor | 刘铁 ; Philippe De Maeyer |
| 2020-06-30 | |
| Degree Grantor | 中国科学院大学 |
| Place of Conferral | 北京 |
| Degree Discipline | 理学博士 |
| Keyword | 洪水 融雪 融冰 积温 SWAT Flood Snow-melt Ice-melt Accumulated temperature SWAT |
| Abstract | 水是一切生物存在和生长以及人类生产生活的基本条件。但全球水资源分布严重不均，特别是在干旱、半干旱地区，由于降水稀少，其水资源对维持生态系统环境和保障经济发展尤为重要。新疆位于干旱、半干旱地区，水资源 90%以上来自山区径流，产汇流的气象成因更加复杂；由于山区降雪量较大，有大面积的永久性积雪和冰川的存在，导致融雪融冰性洪水灾害频发。本研究针对高寒山区洪水发生过程和机理研究不足、 传统水文模型在高海拔山区精度较低等问题，在现有水文模型的基础上，结合温度与积雪冰川的关系，改进了降水形态判定模块和融雪算法，增加了独立的融冰计算模块，并对模型适用性进行了验证。研究结果对深入认识和把握气候变化对干旱区水循环的影响机理以及防灾减灾效益的提高和气候变化适应性对策与策略的制定提供科学支撑。高寒山区累积的冰川和积雪是融水型洪水的主要来源，目前国内外水文模型大多基于湿润区或者平原区水文过程而建立，对冰雪径流的模拟表现不足。 目前已有的水文模型在进行融雪和融冰计算时，并未考虑山区地形以及区域地理条件的影响，对山区融雪、融冰过程中的气象要素和关键参数的时空异质性考虑不足；且大多数模型将冰雪作为一体模拟其融水径流过程，较少耦合独立的冰川融水算法和模块，导致对融水型洪水水量、峰值与峰现事件等的模拟精度较低。因此，本研究充分考虑山区积温条件对降水形态和融雪、融冰过程的影响，在原有SWAT 模型的基础上，改进了降水形态判定模块和融雪计算方法，增加独立的融冰计算模块，并模拟不同纬度的高海拔山区流域融雪融冰径流过程，开展精度评估和适用性比较。主要研究结果如下：（1）在 SWAT 模型的基础上， 增加了积温条件对降水形态判定和融雪计算的影响， 充分考虑了高海拔山区，积温对水文过程的影响。通过与站点实测数据对比发现，增加积温条件后，降水形态判定的准确率达 79.49%以上，并确定出雨雪分离的温度阈值。 在融雪计算过程中， 通过增加积温阈值，确定出融雪发生的温度条件，充分考虑在植被稀疏的高海拔山区，积温对融雪过程的影响。 改进后模型计算的融雪量增加了 45%，融雪对河道径流的贡献率从 44.7%增加到54.07%。 在日尺度的基础上增加了小时尺度的模拟，小时尺度的模拟结果中，洪峰值的模拟精度提高了 77%， 洪峰出现的时间偏差从 11 小时减小到约 1 小时。（2） 基于改进的模型，模拟分别位于阿尔泰山、天山和昆仑山的 4 个典型流域径流过程，分析高山冰雪融水对下垫面植被生长的影响，评估模型改进效果及其在不同纬度山区流域的适用性。结果表明，改进后模型在三大山系的典型流域中表现良好， 模拟精度较改进前有明显提高。在冰川覆盖较少的大青河流域改进效果尤为明显。 由于该流域存在大量积雪覆盖，改进后融雪模块计算更加准确；在另外三条流域中有冰川覆盖，融雪模块计算冰川融水量有一定误差，这为开发独立融冰计算模块做下铺垫。 在四条典型流域中， 洪峰值的模拟平均增加了56.19%，融雪对河道的贡献率增加了 27.02%。（3） 为实现高海拔山区融冰与融雪的单独计算， 提高模型冰川融水的计算精度，在 SWAT 模型的基础上开发独立的融冰计算模块，并对高海拔山区的融雪和融冰型洪水过程进行模拟。与此同时， 结合流域植被 NDVI 变化， 利用冰雪融水与植被生长的相关性， 对改进后的模型的计算精度进行验证。通过增加融冰模块， 模型计算的洪峰值大小增加了 77.65%，洪峰过境时间精度提高了 82.93%，冰雪融水对河道的贡献率增加了 18.73%。（ 4）为揭示干旱半干旱区高山融雪融冰与植被生长的关系， 研究了植被NDVI 与融水量之间的相关性，利用生态学的知识验证模型改进的合理性。 根据改进后模型在新疆三大山系的四条典型流域模拟中的表现， 融雪量与 NDVI 的相关系数从 0.27 增加到 0.68。 通过开发独立的融冰计算模块， 模型计算的冰雪融水量与植被的 NDVI 相关性有了显著提高。本研究对新疆山区洪水过程进行了相关研究，主要的创新点可归纳为在SWAT 模型基础上，改进冰雪消融相关结构和算法，构建高寒山区洪水预报模型，并在多流域验证其有效性。 通过模型的改进与应用，本研究可以为高寒山区洪水过程的相关研究提供重要参考， 同时为洪水灾害的预报提供一定的技术支撑。 |
| Other Abstract | Water is the basic condition for the existence and growth of all living things, aswell as human production and life. However, the global water resources are severelyunevenly distributed, especially in arid and semi-arid areas. Due to the scarcity ofprecipitation, the water resources are particularly important for maintaining theecosystem environment and ensuring economic development. Xinjiang is located inarid and semi-arid areas, more than 90% of water resources come from mountain runoff.The meteorological causes of runoff and confluence are more complex; because of thelarge amount of snowfall in mountain areas, large areas of permanent snow and glaciers,resulting in frequent snow melting and ice melting flood disasters. In this study, aimingat the shortage of the research on the process and mechanism of flood occurrence in thealpine area, the traditional hydrological model has low accuracy in the high altitudearea. On the basis of the existing hydrological model, combined with the relationshipbetween temperature and snow glacier, the precipitation pattern judgment module andsnow melting algorithm has been modified, and the independent ice melting calculationmodule has been newly added, and the applicability of the model was verified. Theresearch results provide scientific support for understanding scientifically and graspingthe impact of climate change on the water cycle in arid regions, as well as providingsupport for improving disaster prevention and mitigation benefits and formulatingclimate change adaptive countermeasures and strategies.A large number of glaciers and snow accumulated in mountainous areas are themain sources of meltwater floods. At present, most of the hydrological models are basedon the hydrological process of humid or plain areas, and the simulation of ice and snowrunoff is a weak link. In the calculation of snow melting and ice melting, the mountainterrain and regional geographical conditions have not been considered in the existinghydrological models, and the temporal and spatial heterogeneity of meteorologicalelements and key parameters in the process of snow melting and ice melting in themountain area were not taken into account. Moreover, The ice and snow are used as awhole to simulate its meltwater runoff process, and independent glacial meltwateralgorithms and modules are less coupled, resulting in lower simulation accuracy ofmeltwater flood water volume, peak and peak occurrence events in most of the modelsTherefore, this study fully considers the influence of mountain accumulated Water is the basic condition for the existence and growth of all living things, aswell as human production and life. However, the global water resources are severelyunevenly distributed, especially in arid and semi-arid areas. Due to the scarcity ofprecipitation, the water resources are particularly important for maintaining theecosystem environment and ensuring economic development. Xinjiang is located inarid and semi-arid areas, more than 90% of water resources come from mountain runoff.The meteorological causes of runoff and confluence are more complex; because of thelarge amount of snowfall in mountain areas, large areas of permanent snow and glaciers,resulting in frequent snow melting and ice melting flood disasters. In this study, aimingat the shortage of the research on the process and mechanism of flood occurrence in thealpine area, the traditional hydrological model has low accuracy in the high altitudearea. On the basis of the existing hydrological model, combined with the relationshipbetween temperature and snow glacier, the precipitation pattern judgment module andsnow melting algorithm has been modified, and the independent ice melting calculationmodule has been newly added, and the applicability of the model was verified. Theresearch results provide scientific support for understanding scientifically and graspingthe impact of climate change on the water cycle in arid regions, as well as providingsupport for improving disaster prevention and mitigation benefits and formulatingclimate change adaptive countermeasures and strategies.A large number of glaciers and snow accumulated in mountainous areas are themain sources of meltwater floods. At present, most of the hydrological models are basedon the hydrological process of humid or plain areas, and the simulation of ice and snowrunoff is a weak link. In the calculation of snow melting and ice melting, the mountainterrain and regional geographical conditions have not been considered in the existinghydrological models, and the temporal and spatial heterogeneity of meteorologicalelements and key parameters in the process of snow melting and ice melting in themountain area were not taken into account. Moreover, The ice and snow are used as awhole to simulate its meltwater runoff process, and independent glacial meltwateralgorithms and modules are less coupled, resulting in lower simulation accuracy ofmeltwater flood water volume, peak and peak occurrence events in most of the modelsTherefore, this study fully considers the influence of mountain accumulated error in the calculation of glacier melting water by snow melting module, which alsopaves the way for the development of independent ice melting module. In the fourtypical basins, the flood peak value increased by 56.19% on average, and thecontribution rate of snowmelt to the river increased by 27.02%.(3) In order to separate ice and snow melt in high-altitude mountainous areas,accurately calculate and simulate glacial meltwater, the new ice melting module wasadded based on the original SWAT model, as well as the simulating snow and icemelting flood in high-altitude mountain areas was processed. Meanwhile, the accuracyof the modified model was verified by combining the NDVI of the basin and therelationship between the ice and snow melting and ecology. By adding the ice meltingmodule, the size of flood peak increased by 77.65%, the time precision of flood peakincreased by 82.93%, and the contribution rate of ice and snow melting to riverincreased by 18.73%.(4) In order to reveal the relationship between alpine snow melting and ice meltingand vegetation growth in arid and semi-arid areas, the correlation between NDVI andmelting water was studied. At the same time, the rationality of model modification wasverified by using this ecological knowledge. In the four typical basins of three mountainsystems in Xinjiang, the correlation coefficient between snowmelt and NDVI increasedfrom 0.27 to 0.68 through model modification. By comparing the NDVI correlationcoefficient between snow and ice melt and different vegetation, it was found that thecorrelation between snow and ice melt calculatedly by the improved model andvegetation NDVI have been significantly improved. After the modification of theindependent ice melting calculation module, the correlation between the ice and snowmelting water calculated by the model and the NDVI of vegetation has beensignificantly improved.The main innovation of this study is to modify the structure and algorithm of snowand ice melting based on SWAT model, to build the flood forecast model of alpine area,and to verify its effectiveness in multi basin. Through the modification and applicationof the model, it was found that this study could be an important reference for researchon flood processes in alpine regions and would also provide technical support for flooddisaster forecasting. |
| 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/365004/15440 |
| Collection | 中国科学院新疆生态与地理研究所 研究系统 |
| 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 | 段永超. 内陆河高寒山区洪水发生过程研究[D]. 北京. 中国科学院大学,2020. |
| 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