The No. 2 hydrological section is located at 312 Heihe River Bridge (38°59′51.71″ N, 100° 24′38.76″ E, 1485 m a.s.l.) in the middle reaches of the Heihe River Basin, Zhangye, Gansu Province. The dataset contains observations from the No.2 hydrological section from 19 June, 2012, to 24 November, 2012. This section consists of two river sections, i.e., the east section is marked as No. 1 and the west section is marked as No. 2. The width of this section is 90 meters. This section consists of a gravel bed; the cross-sectional area is unstable because of human factors. The water level was measured using SR50 ultrasonic range and the discharge was measured using cross-section reconnaissance by the StreamPro ADCP. The dataset includes the following sections: Water level (recorded every 30 minutes) and Discharge. The data processing and quality control steps were as follows: 1) The water level data which collected from the hydrological station were averaged over intervals of 10 min for a total of 144 records per day. The missing data were denoted by -6999. 2) Data out the normal range records were rejected. 3) Unphysical data were rejected. For more information, please refer to Liu et al. (2016) (for multi-scale observation experiment or sites information), He et al. (2016) (for data processing) in the Citation section.

This dataset contains the flux measurements from the barren-land station eddy covariance system (EC) in the lower reaches of the Heihe hydrometeorological observation network from 10 July to 31 December, 2013. The site (101.133° E, 41.999° N) was located in the barren-land surface, Ejin Banner in Inner Mongolia. The elevation is 878 m. The EC was installed at a height of 3.5 m, and the sampling rate was 10 Hz. The sonic anemometer faced north, and the separation distance between the sonic anemometer and the CO2/H2O gas analyzer (CSAT3&Li7500) was 0.15 m. The raw data acquired at 10 Hz were processed using the Edire post-processing software (University of Edinburgh, http://www.geos.ed.ac.uk/abs/research/micromet/EdiRe/), including the spike detection, lag correction of H2O/CO2 relative to the vertical wind component, sonic virtual temperature correction, coordinate rotation (2-D rotation), corrections for density fluctuation (Webb-Pearman-Leuning correction), and frequency response correction. The EC data were subsequently averaged over 30 min periods. The observation data quality was divided into three classes according to the quality assessment method of stationarity (Δst) and the integral turbulent characteristics test (ITC), as proposed by Foken and Wichura [1996]: class 1 (level 0: Δst<30 and ITC<30), class 2 (level 1: Δst<100 and ITC<100), and class 3 (level 2: Δst>100 and ITC>100), which represent high-, medium-, and low-quality data, respectively. In addition to the above processing steps, the half-hourly flux data were screened using a four-step procedure: (1) data from periods of sensor malfunction were rejected; (2) data collected before or after 1 h of precipitation were rejected; (3) incomplete 30 min data were rejected when the missing data constituted more than 3% of the 30 min raw record; and (4) data were rejected at night when the friction velocity (u*) was less than 0.2 m/s. There were 48 records per day, and the missing data were replaced with -6999. Suspicious data were marked in red. Due to the malfunction of CO2/H2O gas analyzer and CF card storage problem, data during 17 July to 13 September and 6 December to 11 December were missing. The released data contained the following variables: data/time, wind direction (Wdir, °), wind speed (Wnd, m/s), the standard deviation of the lateral wind (Std_Uy, m/s), virtual temperature (Tv, ℃), H2O mass density (H2O, g/m^3), CO2 mass density (CO2, mg/m^3), friction velocity (ustar, m/s), stability (z/L), sensible heat flux (Hs, W/m^2), latent heat flux (LE, W/m^2), carbon dioxide flux (Fc, mg/ (m^2s)), quality assessment of the sensible heat flux (QA_Hs), quality assessment of the latent heat flux (QA_LE), and quality assessment of the carbon flux (QA_Fc). In this dataset, the time of 0:30 corresponds to the average data for the period between 0:00 and 0:30; the data were stored in *.xls format. For more information, please refer to Li et al. (2013) (for hydrometeorological observation network or sites information), Liu et al. (2011) (for data processing) in the Citation section.

The dataset of ground-based RPG-8CH-DP microwave radiometers (6.925H/V, 18.7H/V and 36.5H/V) and ground truth observations for snow was obtained in the Binggou watershed foci experimental area on Mar. 24 (time-continuous from 11:42 to 17:28 BJT) and Mar. 25, 2008 (short-time multi-angle observations). A gentle slope of 10° was chosen as the observation site, where there was firn snow and the snow layer and the ice layer appeared alternately. The radiometer beam was set from -20° to -55°, with the steplength 5°. Observation items included: (1) The brightness temperature by the microwave radiometer in .BRT and .txt (the ASCII format). Each row in .txt was listed by year, month, date, hour, minute, second, 6.925GHz (h), 6.925GHz (v), 10.65GHz (h), 10.65GHz (v) , 18.7GHz (h), 18.7GHz (v), 36.5GHz (h), 36.5GHz (v), the elevation angle, and the azimuth angle. Values for 6.925GHz and 10.65GHz were zero due to the absence of these two radiometers. (2) Snow parameters including the snow profile temperature by the probe thermometer and the handheld infrared thermometer, the snow grain size by the handheld microscope, snow moisture, snow density, and snow permittivity by the snow fork. Five subfolders are archived, including the brightness temperature and the profiles of liquid water content, the snow grain size, snow density and the snow temperature.

The dataset of ground truth measurements for snow synchronizing with the airborne microwave radiometers (K&Ka bands) mission was obtained in the Binggou watershed foci experimental area on Mar. 30, 2008. Those provide reliable data for retrieval of snow parameters and properties, especially for dry and wet snow identification. Observation items included: (1) Snow density, snow complex permittivity, snow volumetric moisture and snow gravimetric moisture by the snowfork in BG-A; (2) Snow parameters including snow depth, the snow surface temperature synchronizing with the airborne microwave radiometers (K&Ka bands), the snow layer temperature, the snow grain size and snow density in BG-A (10 points), BG-B (6 points), BG-F (12 points), BG-H (21 points) and BG-I (20 points); For each snow pit, the snowpack was divided into several layers with 10-cm intervals of snow depth. The layer depth (by the ruler), the snow grain size (by the handheld microscope), snow density (by the cutting ring) and the snow temperature (by the probe thermometer) were obtained at each snow pit. Two files including raw data and the preprocessed data were archived.

This dataset contains the flux measurements from site No.10 eddy covariance system (EC) in the flux observation matrix from 4 June to 17 September, 2012. The site (100.39572° E, 38.87567° N) was located in a cropland (maize surface) in Yingke irrigation district, which is near Zhangye, Gansu Province. The elevation is 1534.73 m. The EC was installed at a height of 4.8 m; the sampling rate was 10 Hz. The sonic anemometer faced north, and the separation distance between the sonic anemometer and the CO2/H2O gas analyzer (CSAT3&Li7500) was 0.17 m. Raw data acquired at 10 Hz were processed using the Edire post-processing software (University of Edinburgh, http://www.geos.ed.ac.uk/abs/research/micromet/EdiRe/), including spike detection, lag correction of H2O/CO2 relative to the vertical wind component, sonic virtual temperature correction, coordinate rotation (2-D rotation), corrections for density fluctuation (Webb-Pearman-Leuning correction), and frequency response correction. The EC data were subsequently averaged over 30 min periods. Moreover, the observation data quality was divided into three classes according to the quality assessment method of stationarity (Δst) and the integral turbulent characteristics test (ITC), which was proposed by Foken and Wichura [1996]: class 1 (level 0: Δst<30 and ITC<30), class 2 (level 1: Δst<100 and ITC<100), and class 3 (level 2: Δst>100 and ITC>100), representing high-, medium-, and low-quality data, respectively. In addition to the above processing steps, the half-hourly flux data were screened in a four-step procedure: (1) data from periods of sensor malfunction were rejected; (2) data before or after 1 h of precipitation were rejected; (3) incomplete 30 min data were rejected when the missing data constituted more than 3% of the 30 min raw record; and (4) data were rejected at night when the friction velocity (u*) was less than 0.1 m/s. There were 48 records per day; the missing data were replaced with -6999. Moreover, suspicious data were marked in red. The released data contained the following variables: data/time, wind direction (Wdir, °), wind speed (Wnd, m/s), the standard deviation of the lateral wind (Std_Uy, m/s), virtual temperature (Tv, ℃), H2O mass density (H2O, g/m^3), CO2 mass density (CO2, mg/m^3), friction velocity (ustar, m/s), stability (z/L), sensible heat flux (Hs, W/m^2), latent heat flux (LE, W/m^2), carbon dioxide flux (Fc, mg/ (m^2s)), quality assessment of the sensible heat flux (QA_Hs), quality assessment of the latent heat flux (QA_LE), and quality assessment of the carbon flux (QA_Fc). In this dataset, the time of 0:30 corresponds to the average data for the period between 0:00 and 0:30; the data were stored in *.xlsx format. For more information, please refer to Liu et al. (2016) (for multi-scale observation experiment or sites information), Xu et al. (2013) (for data processing) in the Citation section.

Soil bulk density, porosity, water content, water characteristic curve, saturated hydraulic conductivity, particle analysis, infiltration rate, and sampling point location information in the upper reaches of the Heihe River Basin. 1. The data is for 2014 supplementary sampling for 2012, using the ring knife to take the original soil; 2. The soil bulk density is the dry bulk density of the soil and is measured by the drying method. The original ring-shaped soil sample collected in the field was thermostated at 105 ° C for 24 hours in an oven, and the soil dry weight was divided by the soil volume (100 cubic centimeters) , unit: g/cm 3 . 3. Soil porosity is obtained according to the relationship between soil bulk density and soil porosity; 4. Soil infiltration analysis data set, the data is the field experimental measurement data from 2013 to 2014. 5. The infiltration data is measured by “MINI DISK PORTABLE TENSION INFILTROMETER”, and the approximate saturated hydraulic conductivity under a certain negative pressure is obtained. 6. Soil particle size data was measured at the Grain Granulation Laboratory of the Key Laboratory of the Ministry of Education of Lanzhou University. The measuring instrument is a Malvern laser particle size analyzer MS2000. 7. The saturated hydraulic conductivity is measured according to the enamel hair self-made instrument of Yi Yanli (2009). The Marioot bottle was used to maintain the head during the experiment; at the same time, the Ks measured at the time was converted to the Ks value at 10 °C for analysis and calculation. 8. Soil water content data is measured using ECH2O, including 5 layers of soil water content and soil temperature. 9. The water characteristic curve is measured by the centrifuge method: the undisturbed soil of the ring cutter collected in the field is placed in a centrifuge, and each of the speeds is measured at 0, 310, 980, 1700, 2190, 2770, 3100, 5370, 6930, 8200, 11600. The secondary rotor weight is obtained.

30m month compositing Fraction Vegetation Cover (FVC) data set of Heihe River Basin provides the results of monthly FVC synthesis in 2011-2014. The data constructs multi-angle observation data sets by using China's domestic satellite HJ/CCD data with high temporal resolution (2 days after networking) and spatial resolution (30m) , and divides the country into different vegetation divisions and land types. The conversion coefficients of NDVI and FVC are calculated respectively, and use the calculated conversion coefficient lookup table and monthly compositing NDVI to produce the regional monthly compositing FVC products. The 30m month compositing FVC product in the Heihe River Basin can directly obtain the vegetation coverage ratio through high-resolution data, and mitigate the influence of low-resolution data heterogeneity; in addition, selecting the typical period of vegetation growth change, by fitting the vegetation index of each pixel time series to obtain the growth curve parameters that correspond to each pixel; then the land use map and the vegetation classification map are combined to find the representative uniform pixels of the region for training the conversion coefficients of the vegetation index. Compared with the ASTER reference FVC results, the 30m/month compositing FVC product in the Heihe River Basin is slightly higher than the ASTER reference result, but the overall deviation is not large, and the maximum value of the root mean square error (RMSE) of the product and the reference value is less than 0.175. In addition, compared with the ground survey data of Huailai experimental site in Hebei Province, the 30 m/month compositing FVC products generally reflect the seasonal variation of vegetation growth, and the deviation from the ground survey data is less than 0.1. At the same time, compared with the ground measurements of vegetation coverage in many watersheds in Northeast, North China and Southeast China, the overall error between the compositing FVC products and the ground measurements is less than 0.2. In all, the 30m/month compositing FVC data set of Heihe River Basin comprehensively utilizes multi-temporal and multi-angle remote sensing data to improve the estimation accuracy and time resolution of FVC parameter products, so as to better serve the application of remote sensing data products.

According to the sample survey data, in August 2013, 30 forest plots were set up in the Tianlaochi watershed, with a plot size of 10 m×20 m. The long side of the plot was parallel to the slope of the hillside, including 26 blocks of Picea crassifolia forest. 2 blocks of Sabina Przewalsskii forest and 2 mixed forests of Picea and Sabina. In the plot, the diameter of the breast of each tree (the diameter of the trunk at a height of 1.3 m) is measured by a diameter tape, and the height of each tree and the height under the branches (the height of the first live branch at the lower end of the canopy) is measured by a hand-held ultrasonic altimeter. The north-south direction and the east-west crown width are measured with a tape measure, and the sample site is positioned by differential GPS. The parallel version of HASM-AD algorithm is used to simulate the classified LIDAR point cloud data. DEM is generated from ground points, DSM is generated from all points, and the height of surface features is obtained by differential operation between DSM and DEM. In forest area, it is called Canopy Height Model (CHM). A circular window with a given search radius is used to find the local maximum value on CHM. If the central pixel value is the maximum value, it is determined as the crown vertex. The pixel attribute value of the tree vertex is the tree height, and the spatial resolution is 1m.

This dataset was acquired on May 25, 2008 by the L&K-band airborne microwave radiometer at the Linze-Biandukou flight area.The L-band frequency is 1.4 GHz, the rear view is 35 degrees, and the dual-polarization (H and V) information is obtained; the K-band frequency is 18.7 GHz, with zenith angle observation, and there is no polarization information. The plane took off from Zhangye Airport at 9:51 (Beijing time, the same below) and landed at 15:01. The observation from 10:10 to 12:30 was in the Linze area, the flight altitude is about 1800m, and the flight speed is about 250km/hr. The plane flew low over Linze Reservoir from 12:31 to 12:38. The plane works in the Bianduko aerophotography region from13:13 to 14:35, the flight altitude is about 3000m, and the flight speed is about 250km/hr. The original data is divided into two parts: microwave radiometer data and GPS data. The L and K bands of microwave radiometer are all from non-imaging observation, the digital values obtained from instantaneous observation are recorded by text files, the longitude and latitude of flight and the attitude parameters of aircraft are recorded by GPS data. At the same time, through the respective clock records of the microwave radiometer and GPS, the microwave observation can be linked with the GPS record, and the microwave observation can be matched with the geographical coordinate information. Due to the relatively low resolution of the microwave radiometer, the leeway, welter and pitching of the aircraft are generally neglected in data processing. According to the target of use and relative flight altitude (H), after calibration and coordinate matching, the observation information can be rasterized. The resolution (x) of the L and K bands can be considered consistent with the observation footprint. The reference resolution is: L band, x = 0.3H; K band, x = 0.24H. After the above steps, products that can be directly used by users can be obtained.

This data set includes observation data of meteorological elements in the downstream desert station of Heihe Hydrometeorological Observation Network from January 1, 2016 to December 31, 2016. The site is located in the desert beach of Ejina Banner, Inner Mongolia, and the underlying surface is desert. The latitude and longitude of the observation point is 100.9872E, 42.1135N, and the altitude is 1054m. The air temperature and relative humidity sensors are installed at 5m and 10m, facing the north; the barometer is installed at 2m; the tipping bucket rain gauge is installed at 10m; the wind speed sensor is set at 5m, 10m, and the wind direction sensor is set at 10m, facing the north; the four-component radiometer is installed at 6m, facing south; two infrared thermometers are installed at 6m, facing south, the probe orientation is vertically downward; the soil temperature probe is buried in the ground surface 0cm and underground 2cm, 4cm, 10cm, 20cm 40cm, 60cm and 100cm, in the south of the 2m from the meteorological tower; soil moisture sensors are buried in the underground 2cm, 4cm, 10cm, 20cm, 40cm, 60cm and 100cm, in the south of the 2m from the meteorological tower; soil heat flux plates (3 pieces) are buried in the ground 6 cm in order. Observation items include: air temperature and humidity (Ta_5m, RH_5m, Ta_10m, RH_10m) (unit: centigrade, percentage), air pressure (Press) (unit: hectopascal), precipitation (Rain) (unit: mm), wind speed (WS_5m, WS_10m) (unit: m / s), wind direction (WD_10m) (unit: degree), four-component radiation (DR, UR, DLR_Cor, ULR_Cor, Rn) (unit: watts / square meter), surface radiation temperature (IRT_1, IRT_2 ) (unit: centigrade), soil heat flux (Gs_1, Gs_2, Gs_3) (unit: watts/square meter), soil moisture (Ms_2cm, Ms_4cm, Ms_10cm, Ms_20cm, Ms_40cm, Ms_60cm, Ms_100cm) (unit: volumetric water content, percentage) and soil temperature (Ts_0cm, Ts_2cm, Ts_4cm, Ts_10cm, Ts_20cm, Ts_40cm, Ts_60cm, Ts_100cm) (unit: centigrade). Processing and quality control of the observation data: (1) ensure 144 data per day (every 10 minutes), when there is missing data, it is marked by -6999; (2) eliminate the moment with duplicate records; (3) delete the data that is obviously beyond the physical meaning or the range of the instrument; (5) the format of date and time is uniform, and the date and time are in the same column. For example, the time is: 2016-6-10 10:30; (6) the naming rules are: AWS+ site name. For hydrometeorological network or site information, please refer to Li et al. (2013). For observation data processing, please refer to Liu et al. (2011).

This data set is the acquisition of the super-site forest 3D structure of the scanning point cloud data and other ancillary data based on the ground-based lidar (LiDAR) . Data acquisition time is from June 4, 2008 to June 12, 2008. Riegl LMS-Z360i ground-based LiDAR was used. The super site is divided into 16 sub-samples of 25m×25m, LiDAR base station points are set in each sub-sample, and LiDAR acquisition 3D full coverage LiDAR point metadata is set at each base station point. The content of the data set: total station measurement coordinates (x, y, z) for each LiDAR data acquisition base station point, the instrument attitude measured by a digital slope meter and an angle meter when each station collects data, and the laser radar scanning point cloud data at each station. This data set can provide realistic 3D forest scenes, provide detailed ground observation data for the development and correction of various 3D forest remote sensing models, and provide ground verification data for airborne and spaceborne remote sensing data.

The data set contains the observation data of meteorological elements from the Barren-land Station,which is located along the lower reaches of the Heihe Hydro-meteorological Observation Network, and the data set covers data from January 1, 2014 to December 31, 2014. The station is located in Sidaoqiao,Dalaihubu Town, Ejina Banner, Inner Mongolia. The underlying surface is barren land. The latitude and longitude of the observation point is 101.1326E, 41.9993N, and the altitude is 878m. The four-component radiometer is installed 6 meters above the ground, facing South; two infrared thermometers are installed 6 meters above the ground, facing South, and the probe orientation is vertical downward; the soil temperature probes are buried respectively at 0cm on the ground surface, 2cm and 4cm under the ground, they are located 2 meters from the meteorological tower in the South; the soil moisture sensors (installed on March 15,2014) are buried 2cm and 4cm under the ground, 2 meters from the meteorological tower in the South; the soil heat flow boards (3 pieces) are buried 6cm under the ground, 2 meters from the meteorological tower in the South. Observed items include: four-component radiation (DR, UR, DLR_Cor, ULR_Cor, Rn) (unit: watt / square meter), surface radiation temperature (IRT_1, IRT_2) (unit: Celsius) , soil heat flux (Gs_1, Gs_2, Gs_3) (unit: watt / square meter), soil moisture (Ms_2cm , Ms_4cm) (unit: volumetric water content, percentage), soil temperature (Ts_0cm, Ts_2cm, Ts_4cm) (unit: Celsius). Processing and quality control of observation data: (1) Ensure 144 data per day (every 10 minutes), if there is missing data, it is marked as -6999. The surface radiation temperature IRT2 data during October 12,2014 to November 8,2014 is missing because of sensor problem; Some 2cm soil moisture data during March21 to March 29 and October 12 to November 8 is missing due to probe problem. (2) Eliminate moments with duplicate records; (3) Remove data that is significantly beyond physical meaning or beyond the measuring range of the instrument; (4) Data marked by red is debatable; (5) The formats of the date and time are uniform, and the date and time are in the same column. For example, the time is: 2014-9-10 10:30; (6) The naming rule is: AWS + site name. For hydro-meteorological network or site information, please refer to Li et al. (2013). For observation data processing, please refer to Liu et al. (2011).