Evapotranspiration (ET) is an very important water balance component for the “Asia water tower”, Tibetan Plateau (TP). However, accurately monitoring and understanding the spatial and temporal variability of ET components (soil evaporation-Es, canopy transpiration-Ec, and intercepted water evaporation-Ew) over the TP remains very difficult, due to the scarcity of observational data for this remote area. Here, the 37 years (1982-2018) monthly ET components on the TP were produced using the MOD16-STM model, which uses the recently available datasets of soil properties, meteorological conditions, and remote sensing. The estimated ET correlates very well with measurements from 9 flux towers, with low root mean square errors (RMSE=13.48 mm/month) and mean bias (MB=2.85 mm/month), and the coefficient of determination (R2=0.83) and the index of agreement (IOA=0.92). The annual averaged ET for the whole TP (specified with an elevation higher than 2500m) is about 0.93±0.037 Gt/year. The main contribution of ET on the TP comes from the soil, with Es accounted for more than 84% of ET. The ET showed a significantly increasing trend with rates of about 1 to 4 mm/year (p<0.05) over most parts of the central and eastern TP, while a significantly decreasing trend with rates of -3 to -1 mm/year was shown in the northwestern TP. The increasing rate of ET in TP over the past 37 years is around 0.96 mm/year. The increase in ET over the entire TP from 1982 to 2018 can be explained by warming and wetting during the same period. MOD16-STM ET exhibited acceptable performance at TP, which was demonstrated by comparison with previous studies. It can adequately represent the actual ET and can be used in research on water resources management, drought monitoring, and ecological change, for example.

Frozen soil refers to a soil or rock mass with a temperature lower than or equal to 0 ° C and containing ice. It is particularly sensitive to temperature and its physical and mechanical properties change significantly with temperature. The frost heaving deformation and melt settlement deformation of frozen soil are the most common frozen soil disasters. Their occurrence is mainly caused by the change of the inherent temperature of frozen soil due to the frozen soil engineering activities. Therefore, the protection of frozen soil is mainly to protect the temperature of frozen soil. , to maintain it in the closest state before the engineering activities. The main method for obtaining the temperature of the frozen land is to embed the temperature measuring cable. Through the data acquisition function of the CR3000, the resistance value of the temperature measuring cable is obtained at different times, and the temperature value is calculated by the correspondence between the calibration coefficient and the resistance value. According to the sensitive characteristics of frozen soil to temperature, the change of ground temperature can reflect the change of climate, and can also analyze the influence mechanism and degree of human activities on the stability of frozen soil in combination with other factors, so as to guide the later engineering activities. Upgrading and upgrading of frozen soil protection measures.

The data set covers 599 meteorological stations in five Central Asian countries, including the following elements: * daily maximum temperature, * daily minimum temperature, * observed temperature, * Precipitation (i.e. rain, melting snow), covering the following dates: 1980-1986; 1996-2005; 2010; 2014; 2015 The data comes from ghcn-d, a data set containing global land area daily observation data, which integrates climate records. The data is a direct measurement of surface temperature, without interpolation or model assumptions, and contains many long-term site records. The disadvantage is uneven space coverage. Due to changes in observation time, site location, and the type of thermometer used, the records contain many heterogeneity. For more information about this dataset, see https://www.ncdc.noaa.gov/ghcnd-data-access

These datasets include mean annual ground temperature (MAGT) at the depth of zero annual amplitude (approximately 3 m to 25 m), active layer thickness (ALT), the probability of the permafrost occurrence, and the new permafrost zonation based on hydrothermal condition for the period of 2000-2016 in the Northern Hemisphere with an 1-km resolution by integrate unprecedentedly large amounts of field data (1,002 boreholes for MAGT and 452 sites for ALT) and multisource geospatial data, especially remote sensing data, using statistical learning modelling with an ensemble strategy, and thus more accurate than previous circumpolar maps.

This data set contains the results of the calculation of Net Primary Productivity (NPP) on the Tibetan Plateau based on ecological models and remote sensing data from 1982 to 2006. Ecosystem NPP of the Tibetan Plateau was generated based on the remote sensing Advanced Very High Resolution Radiometer (AVHRR) data and the Carnegie-Ames-Stanford Approach (CASA) model(1982-2006), the soil carbon content was generated based on the second soil census data, and the biomass carbon data were generated based on the High Resolution Biosphere Model (HRBM) model. Forest ecosystem NPP of the Tibetan Plateau (1982-2006): npp_forest82.e00，npp_forest83.e00，npp_forest84.e00，npp_forest85.e00，npp_forest86.e00， npp_forest87.e00，npp_forest88.e00，npp_forest89.e00，npp_forest90.e00，npp_forest91.e00， npp_forest92.e00，npp_forest93.e00，npp_forest94.e00，npp_forest95.e00，npp_forest96.e00， npp_forest97.e00，npp_forest98.e00，npp_forest99.e00，npp_forest00.e00，npp_forest01.e00， npp_forest02.e00，npp_forest03.e00，npp_forest04.e00，npp_forest05.e00，npp_forest06.e00 Grassland ecosystem NPP of the Tibetan Plateau(1982-2006)： npp_grass82.e00，npp_grass83.e00，npp_grass84.e00，npp_grass85.e00，npp_grass86.e00， npp_grass87.e00，npp_grass88.e00，npp_grass89.e00，npp_grass90.e00，npp_grass91.e00， npp_grass92.e00，npp_grass93.e00，npp_grass94.e00，npp_grass95.e00，npp_grass96.e00， npp_grass97.e00，npp_grass98.e00，npp_grass99.e00，npp_grass00.e00，npp_grass01.e00，npp_grass02.e00，npp_grass03.e00，npp_grass04.e00，npp_grass05.e00，npp_grass06.e00. Biomass carbon and soil carbon of the Tibetan Plateau： Biomass.e00，Socd.e00. The soil carbon content data （Socd） are generated based on data of the second soil census of China and Soil Map of China (1:1,000,000) by soil subclass interpolation. The NPP data are generated from the CASA model and AVHRR data simulation: Potter CS, Randerson JT, Field CB et al. Terrestrial ecosystem production: a process model based on global satellite and surface data. Global Biogeochemical Cycles, 1993, 7: 811–841. The biomass carbon data are generated via HRBM model simulation： McGuire AD, Sitch S, et al. Carbon balance of the terrestrial biosphere in the twentieth century: Analyses of CO2, climate and land use effects with four process-based ecosystem models. Global Biogeochem. Cycles, 2001, 15 (1), 183-206. The raw data are mainly remote sensing data and field observation data with high accuracy; the verification and adjustment of the measured data in the field during the production were undertaken to maintain the error of the simulation results and the field measured data within the acceptable range as much as possible; the verification results of the NPP data and the field measured data show that the error remains within 15%. The spatial resolution is 0.05°×0.05° (longitude×latitude).

The data set mainly includes typical rare earth deposits in China, such as Maoniuping and Lizhuang rare earth deposits in Mianning, Western Sichuan, and Gansha OBO rare earth deposits in Gansu Province. These rare earth deposits are genetically related to carbonate alkaline rock complex. In situ U-Pb dating, whole rock major and trace elements, Sr nd Pb radioisotopes, C-O-B-Ca stable isotopes and mineral in situ major and trace elements contents of rocks or ores in these complexes were analyzed. The major elements were measured by X-ray fluorescence spectrometer (XRF), the trace elements were measured by inductively coupled plasma mass spectrometry (ICP-MS), and the isotopes were mainly measured by mc-icp-ms. The main conclusions are as follows: (1) it is revealed that the magma source area of alkaline carbonate type REE deposit experienced the addition of strong subduction material, and its formation depth may be deeper than previously thought（2) It is revealed that the aegirization may be related to carbonatite and alkaline magmatism, and there may be differences in the aegirization between the two types of magma（3) The later reformation of the rare earth deposits with younger age may be relatively weak, while the rare earth deposits with older age are easy to be influenced by the later geological process and difficult to distinguish.

Paleoecological and paleolimnological studies can provide a long-term perspective on changes in environmental and ecosystem processes. The sediments documented both direct and indirect impacts of climate change and human activities on aquatic ecosystems. The fossils of zooplankton remain and pigments in lake sediments can reflect community structure changes of primary producers and primary consumers. The authors reconstructed the zooplankton and algal community changes during the past 600 years using carapaces of A. tibetiana and resting eggs of D. tibetana and pigments from the sediments of Dagze Co, in the central Tibet Plateau. Using total nitrogen and total phosphorus reconstructed the nutrient changes. These results suggest that algal community structure and changes in production can be attributed to alterations in the zooplankton community, with important implications for Tibetan aquatic ecosystems.

Global land cover data are key sources of information for understanding the complex interactions between human activities and global change. FROM-GLC (Finer Resolution Observation and Monitoring of Global Land Cover) from Tsinghua is the 30 m resolution global land cover maps produced. The Global land cover data of all 34 key nodes of pan-third pole region are produced through analyse by argis. The classfication system is crop(10), forest(20), grass(30), shrbu(40), wetland(50), water(60), tundra(70), impervious(80), Bareland(90), snow/ice(100), cloud(120). Finally, This data set serves as the research basis for all remote sensing data and provides baseline data for the project.

This data set is daily surface albedo product over Tibet plateau region from 2002 to 2020 with a spatial resolution of 0.00425°. The MODIS reflectance data product was used to retrieve the Extended Multi-Sensor Combined BRDF Inversion (EMCBI) Model which has coupled with topographic effects with assistance of a BRDF priori-knowledge. The daily BRDF was retrieved in a 5-day period to collect multi-angular information from MODIS observations. And then the daily albedo is estimated, where the black sky albedo was calculated at local noon. MODIS surface reflectance data (MOD09GA and MYD09GA) are downloaded from the official website. The albedo product is quality-controlled with better temporal and spatial continuity in Tibet plateau area. The validation results show that it meets the accuracy requirements of albedo application with higher precisions comparing to the other similar products. And thus, this product is useful for the long-term environmental monitoring and radiation energy budget research study.

The data defines LC classes using a set of classifiers. The system was designed as a hierarchical classification, which allows adjusting the thematic detail of the legend to the amount of information available to describe each LC class, whilst following a standardized classification approach. As the CCI-LC maps are designed to be globally consistent, their legend is determined by the level of information that is available and that makes sense at the scale of the entire world. The “level 1” legend – also called “global” legend – presented in Table 3-1 meets this requirement. This legend counts 22 classes and each class is associated with a ten values code (i.e. class codes of 10, 20, 30, etc.). The CCI-LC maps are also described by a more detailed legend, called “level 2” or “regional”. This level 2 legend makes use of more accurate and regional information – where available – to define more LCCS classifiers and so to reach a higher level of detail in the legend. This regional legend has therefore more classes which are listed in Appendix 1. The regional classes are associated with nonten values (i.e. class codes such as 11, 12, etc.). They are not present all over the world since they were not properly discriminated at the global scale.

In 1970, land use was visually interpreted from MSS images, with an overall interpretation accuracy of more than 90%. Land classification was carried out in accordance with the land use classification system of the Chinese Academy of Sciences. For detailed classification rules, please read the data description document. The 2005 and 2015 data sets were collected from the European Space Agency (ESA) Data acquisition of global land cover types includes five Central Asian countries (Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan) and Xinjiang, China. There are 22 land use types in the data set. The IPCC land use classification system is adopted. Please refer to the documentation for specific classification details.

This data set consists of tree ring carbon and oxygen data in East Asian monsoon region and Qilian Mountain region of China. Tree rings in Qilian mountain area include 4 tree cores, the tree species is Sabina przewalskii, and the measured isotopic data is 921. Cellulose was extracted from tree ring logs by chemical treatment, and the obtained cellulose samples were wrapped in a silver cup. The isotopic ratio was measured by Delta V advantage stable isotope mass spectrometer, and the analysis error was less than 0.21 ‰. The experimental analysis was completed in the laboratory of soil structure and minerals, Institute of Geology and Geophysics, Chinese Academy of Sciences. This data has certain significance for the study of paleoclimate in East Asian monsoon region.