Land surface hydrothermal conditions (LSHCs) reflect land surface moisture and heat conditions, and play an important role in energy and water cycles in soil-plant-atmosphere continuum. Based on comparison of four evaluation methods (namely, the classic statistical method, geostatistical method, information theory method, and fractal method), this study proposed a new scheme for evaluating the spatial heterogeneity of LSHCs. This scheme incorporates diverse remotely sensed surface parameters, e.g., leaf area index-LAI, the normalized difference vegetation index-NDVI, net radiation-Rn, and land surface temperature-LST. The LSHCs can be classified into three categories, namely homogeneous, moderately heterogeneous and highly heterogeneous based on the remotely sensed LAI data with a 30 m spatial resolution and the combination of normalized information entropy (S′) and coefficient of variation (CV). Based on the evaluation scheme, the spatial heterogeneity of land surface hydrothermal conditions at six typical flux observation stations in the Heihe River Basin during the vegetation growing season were evaluated. The evaluation results were consistent with the land surface type characteristics exhibited by Google Earth imagery and spatial heterogeneity assessed by high resolution remote sensing evapotranspiration data. Impact factors such as precipitation and irrigation events, spatial resolutions of remote sensing data, heterogeneity in the vertical direction, topography and sparse vegetation could also affect the evaluation results. For instance, short-term changes (precipitation and irrigation events) in the spatial heterogeneity of LSHCs can be diagnosed by energy factors, while long-term changes can be indicated by vegetation factors. The spatial heterogeneity of LSHCs decreases when decreasing the spatial resolution of remote sensing data. The proposed evaluation scheme would be useful for the quantification of spatial heterogeneity of LSHCs over flux observation stations toward the global scale, and also contribute to the improvement of the accuracy of estimation and validation for remotely sensed (or model simulated) evapotranspiration.

地表热液条件（LSHCs）反映了地表水分和热条件，并在土壤-植物-大气连续体的能量和水循环中起着重要作用。在比较四种评估方法（即经典统计方法，地统计方法，信息论方法和分形方法）的基础上，本研究提出了一种评估LSHCs空间异质性的新方案。该方案结合了多种遥感表面参数，例如叶面积指数-LAI，归一化差异植被指数-NDVI，净辐射-Rn和地表温度-LST。LSHC可以分为三类，即同类，S'和变异系数（CV）。基于评价方案，评价了黑河流域六个典型通量观测站植被生长期陆地表面热液条件的空间异质性。评估结果与Google Earth影像所显示的地表类型特征以及高分辨率遥感蒸散数据所评估的空间异质性相一致。降雨和灌溉事件，遥感数据的空间分辨率，垂直方向的异质性，地形和稀疏植被等影响因素也可能影响评价结果。例如，可以通过能量因子来诊断LSHCs空间异质性的短期变化（降水和灌溉事件），而可以通过植被因子来指示长期变化。当降低遥感数据的空间分辨率时，LSHCs的空间异质性降低。所提出的评估方案将有助于量化通量观测站向全球范围内的低烟高碳排放物的空间异质性，并有助于提高遥感（或模拟模型）蒸散量估计和验证的准确性。