KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| APEX 和 RWEQ 风蚀模型在塔里木盆地的适用性分析与改进 | |
| 邳华伟 | |
| Subtype | 博士 |
| Thesis Advisor | 雷加强 ; 冯广龙 |
| 2017-05-01 | |
| 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 | Apex Rweq 塔里木盆地 风洞 Tarim Basin Wind Tunnel |
| Abstract | APEX 和 RWEQ 作为目前通用的土壤风蚀预报模型，在许多国家和地区得到了广泛的应用。本文选取塔里木盆地阿克苏绿洲的枣园、棉田和沙漠（荒漠绿洲过渡带）三种典型的下垫面对比分析了 APEX 和 RWEQ 在该研究区的适用性。在多场大风事件中，通过野外试验监测，获得 APEX 和 RWEQ 模型参数、以及土壤流失量和风蚀沙通量的实测值。再通过参数率定，模型模拟，获得土壤流失量和风蚀沙通量的模拟值。对比实测值和模拟值，对模型进行模型适用性分析。结果显示：1. 两个模型在不同下垫面的预测精度各异。初始的 APEX 模型对棉田点（模型一致性指数 (d)=0.81, R2=0.50）和荒漠绿洲过渡带（d=0.81, R2=0.91）的土壤流失量模拟较好。然而，初始的 APEX 模型在枣园点对土壤流失量和沙通量模拟性能欠佳，其首要原因可以归结为该模型高估了植被在风沙运动中的保护作用（过程参数 VCF 偏小）。2. 在所选取的三个下垫面中，初始的 RWEQ 模型对土壤流失量的模拟均表现不佳。模型对土壤流失量和风蚀沙通量的预测值偏低。在棉田下垫面，预测值偏低可以归因于该模型对 S（试验地长度制约参数）的模拟性能不佳。在枣园点，该模型对风蚀沙通量的模拟性能预报不佳可以归因为对 Qmax的预报精度不够。3. 不区分下垫面，合并分析 23 个大风事件，我们发现初始的 APEX 模型模拟土壤侵蚀（包括土壤流失量和风蚀的沙通量）的性能要高于 RWEQ 模型。对比其它主要风蚀区域的模拟性能，例如美国华盛顿州的哥伦比亚高原的休耕轮作的小麦下垫面，初始的 APEX 模型对土壤流失量模拟效果也不理想，模型一致性指数为 0.18，低于该模型在塔里木盆地的模拟效果。初始的 RWEQ 模型模拟土壤流失量的性能良好，模型一致性指数 0.76，回归的拟合优度（R2）为0.40，高于该模型在塔里木盆地的模拟效果。模型参数校准是提高模型模拟性能的一个重要方法。本文选取的 APEX 和RWEQ 模型均包含大量的经验参数，这些非可测量的经验参数是模型本地应用精度不高的原因之一，也是模型率定的重点。本文运用模型参数暴力破解校准法 (brute force calibration technique)对模型参数进行校准，为更好的比较 APEX和 RWEQ 两个模型在不同下垫面的模拟性能，选取同样的大风事件来校准APEX 和 RWEQ 模型的过程参数。模型校准的结果显示： 通过模型过程参数的校准，除了在枣园下垫面的 RWEQ 模型，APEX 和 RWEQ 两个模型在所选取的其它下垫面，均有较高的模型性能。校准后的 APEX 和 RWEQ 模型，在棉田和荒漠绿洲过渡带点的模拟性能良好。APEX 和 RWEQ 模型在研究区适用性的提高，可以更好地为塔里木盆地的风蚀预报及沙尘治理工作服务。风洞试验可以定量控制风蚀影响因子，包括风速，沙源，边界条件等在短时间内取得较多的观测数据，本文利用可移动野外风蚀风洞在不同种农田耕作方式下测试了RWEQ 模型的性能。RWEQ 模型的一些积分方程和模型参数源自风洞试验，所以相对于野外实测的风蚀试验，RWEQ 模型在风洞试验中对土壤损失量的模型模拟性能更好（d>0.65 , 且 R2>0.90）。 |
| Other Abstract | The APEX (Agricultural policy Enviromental eX tender) and RWEQ (Revised Wind Erosion Equation) are the two most popular wind erosion model, which have been tested widely in many countries and regions of the world. The study tested the APEX and RWEQ under three contrasting land use types (cotton field, red date orchard, and desert-oasis ecotone) in Northwest region of the Tarim Basin. According to the conduct of the field experiment, input parameters for APEX and RWEQ, wind erosion soil loss and flux was measured in three landscapes during the spring of 2012 and 2013, then both models were assessed based on the measured erosion and simulated erosion by both models. The main conclusions were as follows:1. APEX and RWEQ appeared different performance in the three contrasting land use types. Adequate performance in simulating soil loss was achieved using the original APEX model in respectively a cotton field and desert-oasis ecotone in the Tarim Basin based on the index of agreement (d)=0.81, R2=0.50, and d=0.81,R 2=0.91. However, Inadequate performance in simulating soil loss was achieved using the original APEX model in the red date orchard, this can be due to the red date orchard had the smallest SEF and VCF of the three land use types in the Tarim Basin.The APEX wind erosion model appear overestimate the impact of vegetation in simulating soil discharge and loss from the red date orchard.2. The RWEQ inadequately simulated soil discharge and loss for the three land use types in the Tarim Basin as a result of d<0.6. The negative difference of mean between simulated and measured soil discharge and loss indicated the model underestimated erosion. RWEQ under-predicted maximum mass transport and thus soil discharge was underestimated in the red date orchard.3. The performance of both models were compared across multiple land use and soil types and time by consolidating these erosion events observed in this study. Based upon the consolidation of erosion events, the original APEX model better simulated erosion. Compared with other wind erosion study, such as the winter wheat - summer fallow (WW-SF) field in the Columbia Plateau, the original APEX inadequately simulated soil loss based on the low index of agreement (d<0.18), lower than in the Tarim Basin. While, the original RWEQ adequately simulated soil loss based on the low index of agreement (d=0.18, and R2>0.4), higher than in the Tarim Basin.Models are often calibrated to overcome constraints associated with the spatial or temporal resolution of application and thereby improve performance. The APEX model and RWEQ also contain a large number of empirical parameters. Therefore, we calibrated both models using the brute force calibration technique which has been applied to environmental models. In order to compare the model performance, we choose the identical erosion events (two events in April 2012 and the 17-19April2013 event) for each of the three land use types in the Tarim Basin. The calibrated APEX model and RWEQ adequately simulated erosion from the cotton field and desert–oasis ecotone in the Tarim Basin, the calibrated APEX model also adequately simulated erosion from the red date orchard in the Tarim Basin. The improved simulation of the calibrated APEX model in the red date orchard is likely due to a twofold decrease in the VCF. The Wind tunnel experiments provide opportunities to assess wind erosion under controlled conditions, we test the performance of the RWEQ in simulating soil loss from a portable wind tunnel. The original RWEQ adequately simulated soil loss based on the high index of agreement d>0.65 , and R2>0.90, these can due to some algorithms or parameters of RWEQ were derived from wind tunnel experiments. |
| Subject Area | 自然地理学 |
| Language | 中文 |
| Document Type | 学位论文 |
| Identifier | http://ir.xjlas.org/handle/365004/14788 |
| 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 |
| Recommended Citation GB/T 7714 | 邳华伟. APEX 和 RWEQ 风蚀模型在塔里木盆地的适用性分析与改进[D]. 新疆乌鲁木齐. 中国科学院大学,2017. |
| 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