KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| 新疆蒸散的估算及不确定性研究 | |
| 崔俊杰 | |
| Subtype | 硕士 |
| Thesis Advisor | 李龙辉 |
| 2018-06-01 | |
| Degree Grantor | 中国科学院大学 |
| Place of Conferral | 新疆乌鲁木齐 |
| Degree Discipline | 理学硕士 |
| Keyword | 不确定性分析 蒸散 TCH 方法 主成分回归分析 敏感性 Uncertainty analysis Evapotranspiration Three Cornered Hat method Principle component regression Sensitivity |
| Abstract | 蒸散是干旱半干旱区生态耗水的主要方式。 新疆深居亚欧大陆内部，垂直海拔高度落差较大，降水空间分布不均匀，植被覆盖空间异质性高， 绿洲农田区灌溉取水频繁。 此外，为适应干旱缺水的特殊环境，荒漠植被进化出特殊的形态特征（深根系，高根冠比等）和更加灵活复杂的植物用水策略等， 这些为蒸散的估算带来挑战和不确定性。国际上不同蒸散产品在新疆地区模拟结果存在差异。因此， 本研究选取中分辨率成像光谱仪（MOD16）蒸散产品、全球陆地蒸散阿姆斯特丹模型（Global LandEvapotranspiration Amsterdam Model， GLEAM）蒸散产品和通用陆面过程模型（Common Land Model， CoLM）的蒸散产品为代表，利用三角帽（Three CorneredHat， TCH）方法定量评价 3 种蒸散产品在新疆干旱区蒸散模拟结果的空间不确定性。 为进一步说明引起蒸散变化的原因，本研究通过分析不同蒸散产品的蒸散结果和气候因子的相关性， 以及利用主成分回归分析（Principle ComponentRegression, PCR）和多元非线性回归方法分析不同蒸散产品蒸散结果对主要气候因子（降水、 辐射和温度） 的敏感性大小， 同时说明影响新疆地区蒸散的主要气候因子，为提高新疆蒸散的模拟精度提供参考。 结果表明：（1） GLEAM 和 MOD16 蒸散产品的蒸散空间分布规律相似，山区蒸散值均显著高于平原区； CoLM 模型蒸散在阿尔泰山、天山蒸散值显著高于平原区。MOD16 在裸土区不进行蒸散值计算。 全疆多年平均蒸散值分别为： GLEAM（142.4±11.1 mm·a-1）、 CoLM（126.7±19.1 mm·a-1）、 MOD16（78.6±4.6mm·a-1）。（2）基于 TCH 方法的全疆蒸散模拟的不确定性结果大小依次为： CoLM 模型（24.2 mm·a-1）， GLEAM 蒸散产品（21.6 mm·a-1）， MOD16 蒸散产品（17.7mm·a-1）。 CoLM 模型和 GLEAM 在新疆不同植被功能类型、不同地形地貌及不同干湿状况条件下，不确定性规律表现一致， MOD16 蒸散不确定性与两者差异较大。在灌丛，天山和半湿润区， 3 种蒸散产品的不确定性显著不同，其中 CoLM不确定性最大（33.5 mm·a-1、 30.5 mm·a-1、 32.1 mm·a-1）， MOD16 不确定性最小（9.5 mm·a-1、 14.8 mm·a-1、 14.3 mm·a-1）。（3）全疆范围， 3 种蒸散产品模拟结果均显示蒸散与降水、比湿具有显著正相关， 与辐射、气温、气压和风速等气候因子相关关系并不突出。 水分因子对新疆蒸散的估算影响较大。（4） 蒸散对 3 种气候因子的总敏感性指数较大的区域主要分布在昆仑山东部，天山东部以及阿尔泰山部分地区。权重敏感性空间分布与总敏感性指数空间分布相似，可以用权重敏感性表示总敏感性指数。全疆大部分地区呈现蒸散对降水的敏感程度较大，如新疆南部平原区域和天山一带等，且蒸散对降水的敏感性系数为正值，蒸散随降水的增加而增加。蒸散对辐射变化的敏感程度较大的区域主要分布在阿尔泰山、天山东部和昆仑山东部等地区。蒸散对温度变化敏感程度较大的区域较少，主要分布在昆仑山部分地区等。 |
| Other Abstract | Evapotranspiration (ET) is an indispensable part of water resource consumptionin arid and semi-arid region. In the interior of the Eurasian continent, the verticalelevation is big different, and the spatial distribution of the precipitation is uneven.The vegetation here is extremely sparse and has relatively higher spatial heterogeneity,and the irrigation in the Oasis is frequent. In addition, in order to adapt to the specialenvironment of arid such as water scarcity, the desert vegetation has evolved specialmorphological characteristics (deep root, Kogan ratio, etc.), and more flexible andcomplex plant water strategy, which brings the challenge and uncertainty to theestimation of evapotranspiration.There were differences in the simulation results of different evapotranspirationproducts in Xinjiang. Therefore, the study selected Common Land Model (CoLM),Global Land Evapotranspiration Amsterdam Model (GLEAM) products and ModerateResolution Imaging Spectroradiometer (MOD16) products. The spatial uncertaintiesof simulation results of three kinds of evapotranspiration products in arid region ofXinjiang were quantitatively evaluated by using three Cornered Hat (TCH) method,which did not require a priori knowledge of the observed ET. In order to furtherexplain the causes of evapotranspiration, the correlation between evapotranspirationand climatic factors was analyzed, and the the dominant role of evapotranspirationresponse to main climatic factors (precipitation, radiation, tempreture) was analyzedby the Principal Component Regression (PCR) and multivariate nonlinear regressionmethod, which try to explain the main climatic factors affecting evapotranspiration,and provide reference for improving the simulation of evapotranspiration in Xinjiang.The results shows as follows:（1） The spatial patterns of GLEAM and MOD16 ET were similar, and the ETvalues in the mountains were significantly higher than that in the plain area. The ETvalues of CoLM in Altay and Tianshan Mountains were significantly higher than thatin the plain area. The MOD16 was not calculated by evapotranspiration in the baresoil area. In addition, the mean annual ET values in Xinjiang from 2000 to 2014differed markedly: GLEAM (142.4±11.1 mm/a), CoLM (126.7±19.1 mm/a), MOD16(78.6±4.6 mm/a), respectively.（2） The uncertainty in Xinjiang based on the TCH method of CoLM was the highest (24.2 mm/a), moderate of GLEAM (21.6 mm/a), and lowest of MOD16 (17.7mm/a). The uncertainty of CoLM and GLEAM was consistent in Xinjiang, and theuncertainty of MOD16 evapotranspiration was quite different. In shrubland, TianshanMountain and semi-humid regions, the uncertainties of CoLM (33.5 mm/a, 30.5 mm/a,32.1 mm/a) were significantly greater than that of GLEAM (28.3 mm/a, 24.7 mm/a,24.2 mm/a) and MOD16 (9.5 mm/a, 14.8 mm/a, 14.3 mm/a).（3） The simulation results of three evapotranspiration products showed that theevapotranspiration had significant positive correlation with precipitation and humidity,and it was not prominent in relation to climate factors such as radiation, airtemperature, air pressure and wind speed. The water factor had great influence on theestimation of evapotranspiration in Xinjiang.（4） The total sensitivity index of evapotranspiration to three climatic variableswas mainly distributed in the Kunlun Mountain, Eastern Tianshan and some parts ofAltai Mountain. The spatial distribution of weight sensitivity was similar to that oftotal sensitivity index, and the total sensitivity index can be expressed by weightsensitivity. In most areas of Xinjiang, the sensitivity of evapotranspiration toprecipitation was large, such as the southern Plain region and the Tianshan Mountains,and the sensitivity coefficient of evapotranspiration is positive, and evapotranspirationincreased with the increaseing of precipitation. The regions with large sensitivity toradiation change were mainly distributed in Altai Mountain, eastern Tianshan andKunlun Mountain. The areas with greater sensitivity to temperature change were less,mainly distributed in some parts of Kunlun Mountain. |
| 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/365004/14961 |
| 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 | 崔俊杰. 新疆蒸散的估算及不确定性研究[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