Knowledge Resource Center for Ecological Environment in Arid Area
| 基于遥感-能量平衡理论的地表空气温度与蒸散发估算研究 | |
| 其他题名 | Studies on Remote Sensed Surface Air Temperature Retrieval and Evapotranspiration Estimation Based on Remote Sensing-Energy Balance Theory |
| 刘素华 | |
| 出版年 | 2017 |
| 学位类型 | 博士 |
| 导师 | 苏红波 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 蒸散发（Evapotranspiration, ET），包括土壤蒸发和植被蒸腾，是土壤-植被-大气系统中能量和水分传输以及转换的主要过程，也是能量平衡和水量平衡的重要组成部分。准确和及时的获取蒸散发不但对全球气候变化和陆面过程机理研究有着重要的科学意义，而且对区域水资源管理和农业可持续发展均有重要的指导意义和价值。作为蒸散发估算的必要和重要输入变量，近地表2米处的空气温度是下垫面辐射交换的综合反映指标。因此，空气温度时空分布信息的获取对地表蒸散发的研究具有不可替代的作用。遥感技术可以非接触的获取大尺度地气界面的能量状况、下垫面特征参数以及水分供应等信息，是估算区域尺度空气温度和地表蒸散发的有效途径，因而日益受到研究者的关注。本文以地表能量平衡为基础并加入了对水平平流能量输入的考虑为主要理论依据，以区域尺度空气温度的反演和地表蒸散发的估算为研究内容，基于遥感数据、地表微气象观测数据等，改进了遥感反演空气温度的方法、提出了估算蒸散发的能量平衡法（加入水平平流能量输入的影响）和进行了蒸散发时间尺度扩展的探索性研究。本论文开展的主要研究工作如下：（1）针对反演空气温度的ADEBAT（the advection-energy balance for air temperature）模型，对模型的关键变量——平流因子的空间扩展进行了改进。ADEBAT模型是基于空气温度形成机理的模型，它既考虑了本地热驱动对空气温度形成的决定性作用，也考虑了水平平流能量输入对空气温度的影响，具有物理意义明确，普适性好的特点。然而，原始的ADEBAT模型对于平流因子的处理并不完善，因为它计算出的平流因子呈现出块状分布且每一个块状区域内数值是一样的。本文提出利用反距离权重法（IDW）对平流因子进行空间扩展以改进ADEBAT模型，也即是IADEBAT（the improved advection-energy balance for air temperature）模型，并利用黑河中游绿洲区2012年的遥感数据和气象观测数据对研究区内的空气温度进行反演。结果表明，IADEBAT模型的估算精度要优于ADEBAT模型，决定系数(R2)，均方根误差(RMSE)和平均绝对误差(MAE)分别从0.67，0.43K和0.34K提高到0.77，0.31K和0.24K，且估算值在空间分布上更符合实际。因此，IADEBAT模型对于平流因子的改进是合理的、可行的和有意义的。 （2）利用遥感反演得到的空气温度空间分布结果，对气象站的观测空气温度值进行空间代表性分析。由于地表空间异质性的存在，基于站点的空气温度观测值往往只能代表气象站周围很局限区域的温度状况。出于对气候变化等研究进行更加精确描述的考虑，对气象站观测空气温度的空间代表性进行评价是有必要的。本文在IADEBAT模型反演的黑河中游绿洲区空气温度结果基础上，采用基于站点与区域分布图统计评价法对研究区内选定的目标气象站空气温度观测值进行空间代表性分析。结果表明，站点的空间代表性范围与下垫面的复杂程度（相同距离半径内下垫面的种类数）密切相关。下垫面越均质，站点的空间代表性越好，空间代表性范围越大；随着距离观测站变远，观测值与周围区域内的估算值的标准偏差也变大，说明站点空间代表性随着距离增大而变差。（3）对遥感反演空气温度的方法进行对比。目前已经发展了多种遥感模型用于空气温度的反演，常见的有温度-植被指数（the temperature-vegetation index，TVX）法、单因子统计法、多因子统计法和IADEBAT模型。由于模型对比可以帮助鉴别模型优点、缺点以及帮助模型进行改进，本文利用以上四种方法在黑河中游地区进行空气温度的反演，并对不同方法的估算结果进行对比和分析。结果表明：TVX法会在低密度植被覆盖区产生较大的误差；统计法建模得到的经验方程更适合具有相似土地利用类型的区域；相较于单因子统计法，多因子统计法的精度由于受到多个变量的控制而有所提高；对于IADEBAT模型，虽然计算结果良好，但是由于需要输入大量的遥感反演变量和地面观测数据，而且模型推导过程比较复杂，故而会影响模型的应用和推广。 （4）基于能量平衡理论并考虑水平平流效应的影响，提出了估算蒸散发的Surface Energy Balance-Advection（SEB-A）模型。遥感估算蒸散发的模型多是基于地表能量平衡方程的，然而广泛存在的水平平流能量输入会对地表能量平衡的状态造成影响。本文提出了一种创新性的SEB-A方法用于估算蒸散发：该法以能量平衡方程为基础，但是加入了水平平流能量输入对蒸散发影响的效应分析。SEB-A方法认为最终消耗于蒸散发的能量可以分解为两部分，一部分来自于地表热驱动，可以用能量平衡方程来求解；另一部分来自于水平平流能量输入，可以通过站点观测资料进行确定。利用SEB-A模型原理，本文在黑河中游地区进行了蒸散发的估算，模型估算结果与涡度相关（Eddy Covariance，EC）观测结果经过验证具有良好的一致性，R2、MAE和RMSE分别为0.713，39.3 W/m2，54.6 W/m2；估算结果的百分比误差在0%到35%之间变化；与土地利用类型图的关系表明SEB-A模型可以准确的表达由于地表异质性差异所造成的蒸散发差异。（5）根据观测资料并结合以往研究成果，提出了对瞬时ET进行日扩展的新思路。由于遥感获取的蒸散发是基于卫星过境时刻的瞬时值，需要扩展成日尺度才具有更大的应用价值。本文基于地面观测资料，发现蒸散发日变化过程可以采用高斯拟合曲线进行刻画和描述，并据此提出了一种新的日尺度蒸散发获取思路：基于高斯拟合曲线法对瞬时蒸散发进行日尺度扩展，并将扩展结果与实测值进行对比验证以评价该方法。结果表明，高斯拟合曲线法可以很好的描述和刻画蒸散发日变化过程。黑河中游地区的案例应用则说明了高斯拟合曲线的扩展结果具有很好的精度: 四天的估算结果与EC实测值的相关性很好，R2、MAE和RMSE分别为0.82、0.41 mm和0.46 mm; 百分比误差在0到18%之间变化；与土地利用类型图的关系表明高斯拟合法获取的日蒸散发在空间分布可以很好的描述由于土地利用类型不同所导致的蒸散发空间差异。因此，高斯拟合曲线法可以作为一种有效的方法进行蒸散发日过程模拟和日尺度蒸散发的估算。 |
| 英文摘要 | Evapotranspiration (ET), which is a combined process of evaporation and transpiration, links energy and water transportation and exchange in the soil-plant-atmosphere continuous system and it is also a crucial component of the energy balance and water balance. Therefore, the accurate quantification of ET not only palys an important role in studying global climate change and land-surface process, but also is beneficial to the water resource management and the sustainable development of agriculture. Surface air temperature is a basic variable that is normally observed at the height of 2 meters by meteorological stations. It is a synthesized indicator for describing the energy exchange between the earth surface and the atmosphere. More importantly, surface air temperature is an essential and necessary variable in the physical process of evapotranspiration. Hence, it is necessary to acquire the spatial and temporal distribution of air temperature when make researches on the evapotranspiration. Remote sensing techniques provide a straightforward and consistent way to obtain the spatially distributed information for characterizing such as land surface interactions, unerlysing surface feature and moisture supply, and hence it is a fast developing and effective approach to retrieve surface air temperature and the regional evapotranspiration. Combinbing the remote sensing data with the ground-based data, this study took the surface energy balance model with considering the horizontal advection as the basic theory to make improvements on the ADEBAT (the advection-energy balance for air temperature) model, put forward the SEB-A (Surface Energy Balance-Advection) method and conduct exploratory research on derieving daily ET or monthly ETs. The main contents of this paper are as follows.(1) Based on the theory of the ADEBAT model, the paper made improvement on the advection factor which is the key variable of the model.The ADEBAT model, which is based on the formation mechanism of air temperature, takes into account the leading and conclusive effect of the local driving force–solar radiation and the exotic driving force –horizontal advection on the air temperature. The model has good portability and general applicability. However, the spatial expansion of the advection factor in the ADEBAT model is inappropriate because the advection factor would be shown as blocks with constant values, which is incorrect for that the advection also has spatial heterogeneity like other geographical elements. This paper made improvement on the ADEBAT model (i.e. IADEBAT) by using the inverse distance weighted (IDW) method to expand the advection factor. Applications in the middle reaches of Heihe River in China, 2012, showed that air temperature estimates calculated by the IADEBAT model with an R2（coefficient of determination）, an RMSE (root mean square error) and a MAE (mean average error) of 0.77，0.31 K and 0.24 K, were more accurate than that obtained by the original ADEBAT model with an R2, an RMSE and a MAE of 0.67，0.43 K and 0.34 K. In addition, the spatial distribution of estimates by the IADEBAT model was more accordant with the reality. As a result, the improvement on the advection factor proposed in the study is feasible, reasonable and valuable.(2) According to the spatial distribution of the air temperature retrieved by remote sensing, the paper conducted spatial representativeness analysis on air temperature monitoring stations. To provide more accurate descriptions of climate changes and hydrologic studies, it is of great significance to evaluate the spatial representativeness of air temperature measurements at meteorological stations. Statistical method is adopted in the study to analyze spatial representativeness based on the difference between air temperature of the target station and that of the circled neighborhood obtained in the Section (1). Analyses indicated that the size of the spatial representative range is intimately connected with the complexity of the underlying surface; the more uniform the underlying surface, the wider range the target station can represent. With increasing the radius of neighborhood, the representativeness of the same target station would be worse.(3) This paper made evaluatation on remote sensing methods for estimating surface air temperature. Four remote sensing methods—the temperature-vegetation index (TVX), the simple statistical approach, the advanced statistical approach and the improved advection-energy balance for surface air temperature (IADEBAT)—have been developed to acquire surface air temperature on a regional scale. Model comparison can help identify model benefits and inadequacies and can give suggestions on model development. This study applied the four methods to estimate the surface air temperature in northwestern China, and compared and analysed estimates obtained by them together. Results can be summarized as: the TVX would produce high errors at regions with sparse vegetation cover. The accuracy of the statistical approach greatly depends on the location of data acquisition; the obtained relationship would be more appropriate for similar land covers; the accuracy of the advanced statistical approach can be improved by multiple regressions because it is dominated by multiple variables related to surface air temperature. As for the ADEBAT, the needs of numerous variables would transfer errors and have error accumulation, and hence, the estimation precision of it, sometimes, is not the best when compared to the advanced statistical approach.(4) The proposal of the SEB-A method which is based on the energy balance theory and also considers the advection effects. Surface energy balance (SEB) models are widely used to simulate regional evapotranspiration with remote sensing. The presence of horizontal advection, however, perturbs the surface energy balance system and contributes to the uncertainty of energy influxes. Thus, it is vital to consider horizontal advection when applying SEB models to estimate evapotranspiration. This study proposes an innovative and simplified approach, the SEB-A method, which is based on the energy balance theory and also takes into account the horizontal advection to determine evapotranspiration by remote sensing. The SEB-A method considers that the actual evapotranspiration consists of two parts: the local ET that is regulated by the energy balance system and the exotic ET that arises from horizontal advection. To evaluate the SEB-A method, it was applied to the middle regions of the Heihe River in China. Instantaneous ET for three days were acquired and assessed with ET measurements from eddy covariance (EC) systems. The results demonstrated that the ET estimates had a high accuracy, with an R2 of 0.713, a MAE of 39.3 W/m2 and an RMSE of 54.6 W/m2 between the estimates and corresponding measurements. Percent error was calculated to more rigorously assess the accuracy of these estimates, and it ranged from 0% to 35%. The relationship between the ET estimates and land use types indicated that the ET estimates had spatial distributions that correlated with vegetation patterns and could well demonstrate the ET differences caused by different land use types.(5) According to the anlyses of measurements and existing researches, this paper proposed the approach for deriving daily ET from remote sensed instaneous ET. To have more practical application value, regional ETs through remote sensing that are instantaneous values need to be converted into daily totals. In this study, the Gaussian fitting method of deriving daily ET from remotely sensed instantaneous ET on the basis of diurnal course of daytime ET measurements is proposed. Validity verification shows that the Gaussian fitting curve closely follows the ET measurements during the daytime. Four days’ daily ETs in the Heihe River basin were derived and evaluated with ET measurements from EC system. High accuracy between daily ET estimates and measurements was obtained, with an R2 of 0.82, a MAE of 0.41mm and an RMSE of 0.46 mm. Percent errors were calculated to make more scientific assessments on the estimation accuracy, which were ranging from 0% to 18%. Analyses on the relationship between daily ET estimates and land use status showed that the spatial distribution of daily ET estimates acquired by the Gaussian fitting method was intimately linked with the vegetation patterns and could well demonstrate ET differences caused by land use types. |
| 中文关键词 | 能量平衡模型 ; 空气温度 ; 空间代表性分析 ; 蒸散发 ; 时间尺度扩展 |
| 英文关键词 | surface energy balance air temperature spatial representativeness evpotranpiration time scale extension |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 地图学与地理信息系统 |
| 来源机构 | 中国科学院地理科学与资源研究所 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/287871 |
| 推荐引用方式 GB/T 7714 | 刘素华. 基于遥感-能量平衡理论的地表空气温度与蒸散发估算研究[D]. 中国科学院大学,2017. |
| 条目包含的文件 | 条目无相关文件。 | |||||
| 个性服务 |
| 推荐该条目 |
| 保存到收藏夹 |
| 导出为Endnote文件 |
| 谷歌学术 |
| 谷歌学术中相似的文章 |
| [刘素华]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [刘素华]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [刘素华]的文章 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
