%0 Dataset %T Heihe River eco hydrological remote sensing experiment: multi-scale observation experiment of surface evapotranspiration on non-uniform underlying surface - flux observation matrix data set (thermal diffusion liquid flow meter) (June September 2012) %J National Cryosphere Desert Data Center %I National Cryosphere Desert Data Center(www.ncdc.ac.cn) %U http://www.ncdc.ac.cn/portal/metadata/d5ee524c-c3ff-48cb-8818-d460dd43cdab %W NCDC %R 10.12072/ncdc.NIEER.db0044.2020 %A Sun Rui %A Liu shaomin %A Li xin %K Thermal diffusive flow meter (TDP);Liquid flux;transpiration;Flow rate;hydrometeor %X The data set contains the observed data of trunk sap flow in the flux observation matrix from June 14 to September 21, 2012. The study area is located in Yingke irrigation area, Zhangye City, Gansu Province. According to the different heights and DBH of the shelter forest, sample trees are selected to install TDP (thermal dispersion sap flow velocity probe), and domestic TDP pin type thermal diffusion plant stem flow meter is adopted, with the model of tdp30. The sample plots are TDP-1, tdp-2 and tdp-3, which are located near las4-s, point 6 and point 8 respectively. The height of the sample tree is tdp-2, TDP-1 and tdp-3 from high to low, and the DBH is tdp-2, tdp-3 and TDP-1 from large to small, which represents the whole area for trunk sap flow measurement. The installation height of the probe is 1.3m, and the installation directions are southeast, southwest and due north of the sample tree.The original observation data of TDP is the temperature difference between probes. The acquisition frequency is 30 seconds and the average time is 10 minutes. The published data are calculated and processed trunk sap flow data, including sap flow rate (cm / h), sap flow flux (cm3 / h) and daily transpiration (mm / D) every 10 minutes. Firstly, the liquid flow rate and liquid flow volume are calculated according to the temperature difference between the probes, and then the transpiration Q per unit area of the shelterbelt is calculated according to the shelterbelt area and tree spacing of the observation point. At the same time, the post-processing of the calculated rate and flux values: (1) eliminate the data obviously beyond the physical meaning or beyond the instrument range( 2) Missing data are marked with - 6999( 3) Suspicious data caused by probe failure and other reasons shall be identified in red font, and the data confirmed to be problematic shall be eliminated.For multi-scale observation test or site information, please refer to Liu et al. (2016), and for observation