The upper Yellow River water conservation area is an important section for ecological protection and high-quality social-economic development in the Yellow River Basin.The impact of climate change on the spatio-temporal distribution of its water resources will exacerbate the occurrence of natural disasters， which in turn will affect China's regional ecology and even the living environment of human beings.Based on the CN05.1 gridded air temperature， precipitation， and wind speed data from 1961 to 2016 and the actual evaporation data of GLEAM V3.5a from 1981 to 2020， this paper analyzed the changes of key climatic elements in the water conservation area of the upper Yellow River Basin in the past 60 years.The results are as follows： （1） The air temperature in the research area increased significantly， the variation trend of the air temperature can reach 0.34 °C·（10a）^-1， and the largest temperature rise occurred in winter.Maqu county and surrounding areas in Heihe region have seen the most dramatic warming.（2） The regional precipitation increased as a whole， the variation trend of the precipitation is about 9.3 mm·（10a）^-1.The growth rate was relatively rapid in spring and summer， and the slowest occurred in winter.The growth trend gradually decreased from the northwest to the southeast， with a significant increase in the arid regions of the west， and a relative decreasing trend in the humid regions of the southeast.（3） After experienced an anomalous rise in the early 1970s， the near-surface wind speed continued to decline， the variation trend of it is about -0.11 m·s^-1·（10a）^-1， and the wind speed decreased in most regions of the study area， especially in the northern Datong River area.（4） The actual evaporation showed an increasing trend， the variation trend of it is 11.89 mm·（10a）^-1， of which the relatively large increases occurred in spring and winter， and the growth rate showed a trend of gradual increasing from east to west.Among them， the increase of the actual evaporation in Zhaling Lake， Eling Lake area and the western part of Datong River area was the most significant， and only a few places in the middle and southeast decreased.（5） From the perspective of precipitation and evaporation， the growth rate of precipitation is slightly higher than that of the actual evaporation during the same period （from 1981 to 2016）， therefore the water conservation area in the upper reaches of the Yellow River Basin also has the trends of warming and wetting as the climate warms.

Greater attention has been received about the variation and trends in extreme climate events. Based on daily precipitation data recorded at 45 meteorological stations in the upper Yellow River Basin from 1970 to 2017, the upper Yellow River Basin was distributed into three regions according to topography, elevation, climate and other factors. For each region, wavelet analysis and Mann-Kendall abrupt change analysis were employed to analyze the spatial distribution and temporal trends of annual precipitation and extreme precipitation events. The results show that the distribution of average annual precipitation in the upper Yellow River Basin has obvious regional differences from the southeast to the northwest. The average annual precipitation and extreme precipitation in the upper Yellow River Basin had obvious periodic oscillation characteristics with 22a mostly, followed by 18a and 8a. For the whole region, the extreme precipitation and average annual precipitation were more consistent, but the precipitation in the upper of the Yellow River Basin has been increasing in recent years, while the frequency of extreme precipitation events has been decreasing. It is implicated that the spatial pattern of climate in China has been changed during the past 48 years.

By analyzing Soil Moisture Anomaly Percentage Index （SMAPI） at different soil layers， dry-wet evolution of the source region of the Yellow River （SRYR） during 2008 -2017 are investigated using observations from the Maqu-Ruoergai soil temperature and moisture monitoring network.To diagnose the water vapor transportation path and potential water vapor sources in different processes， the Lagrange Flexible Particle Dispersion Model （FLEXPART）， which is driven by reanalysis data （National Centers for Environmental Prediction Final， NECP FNL）， are used to simulate the backward trajectories of target particles.The results show that the water vapor transportation path can be divided into three categories： （1） South Branch transportation.The water vapor origins from the Indian Ocean and the Arabian Sea， and finally arrives at the SRYR by way of the Indian Peninsula and Bay of Bengal； （2） East Branch transportation.The water vapor is from the Pacific Ocean and the South China Sea， then passes through the Yangtze River Basin， and finally arrives at the SRYR from eastern and southern flank of the Tibetan Plateau； （3） North Branch transportation.The water vapor is from the Atlantic Ocean， the northern African continent， and the European continent， then arrives at the SRYR from the western or northern side of the Tibetan Plateau by way of the mid-latitude Eurasian continent.Moreover， the North Branch is dominant in dry period， whereas the South and East branches are prominent in wet period.The water vapor sources also show discrepancies for dry and wet periods.The water vapor sources of the Tibetan Plateau are mainly distributed around the Kunlun Mountains during wet period， and are scattered distributed from north to south during transitional period， and are located around the Tianshan during dry period.The intensity of the water vapor sources of the Iranian Plateau， Pamir Plateau， and the Bay of Bengal gradually strengthen from wet to dry period， the intensity of the water vapor sources of the Sichuan Basin-Qinling Mountains and south China enhanced first and then weakened， while the source of Qilian Mountain-Loess Plateau weakened after enhanced.The intensity of water vapor sources over the middle and lower reaches of the Yangtze River and around East China has been weakening from the wet period to the dry period.

Based on the precipitation data of 143 stations, temporal and spatial variation characteristics of precipitation and rain days in the Yellow River basin during 19612010 are analyzed. The results show that: (1)Multi-year average spatial distribution characteristics of annual precipitation and annual rainy days in the Yellow River Basin is less in the North and more in the South. (2)Annual precipitation and annual rainy days change trend have effective consistency in the Yellow River basin during the recent 50 years, both decreasingly. And the reduction trend of the annual rain days is more significant. The percentage of stations of annual precipitation with negative trend is 81.8%, and the percentage of stations of annual rainy days with negative trend is 88.8%. (3)About the seasonal variation in the Yellow River basin, the precipitation and rainy days in winter appear positive trend while they present negative trends in other three seasons, especially in autumn. The negative trend stations of precipitation in four seasons passing the significance test are autumn>spring>summer>winter, and the negative trend stations of rain days in four seasons passing the significance test are autumn>summer>spring>winter. (4)The unanimous mutation of both annual precipitation and annual rainfall in the basin is 19851986, and the decrease of precipitation and rain days mainly results from the changed atmospheric circulation. For example, the above normal precipitation and rain days in 1986 are due to the stronger monsoon which enables the effectively conveying of vapor and the wind convergence in the northwest. On the other hand, after the mutation, the weaker monsoon, the lack of effective moisture transport and the anticyclonic circulation in the area from the Mongolia to Hetao may result in less precipitation and rain days.

The Source Area of the Yellow River is located in the northeastern part of the Qinghai-Xizang Plateau， and the meteorological stations are sparsely distributed in this basin， the study of the applicability of various precipitation data products has an important values in promoting the hydrological modeling in the basin.Based on the China Meteorological Assimilation Datasets for SWAT model Version1.1 （CMADS V1.1）， the Tropical Rainfall Measurement Mission （TRMM） precipitation datasets （3B42 Version7） and the Soil and Water Assessment Tool （SWAT） driven by these precipitation data， respectively， and the SWAT-CUP （SWAT Calibration and Uncertainty Program） and SUFI-2 （Sequential Uncertainty Fitting2） algorithm 27 sensitivity parameters were rate in simulating the variation of multi-year monthly average runoff， the simulated results were compared with the observations to evaluate the accuracy of CMADS and TRMM 3B42 precipitation data products and the applicability of SWAT model were evaluated in the Source Area of the Yellow River source area.The results show that： （1） The distribution of all three precipitation datasets showed an increasing trend from the west to the east， and TRMM 3B42 was in better agreement with the measured precipitation than CMADS data set in terms of annual and monthly variation.（2） The sensitivity analysis of the parameters showed that the sensitivity degree of SCS （Soil Conservation Service） runoff curve number， groundwater lagging coefficient， and soil evaporation compensation coefficient were stronger than that of the others.（3） The simulated runoff by using the CMADS and TRMM 3B42 precipitation datasets had better results than that by using the measured precipitation data， with the correlation coefficients R ² of 0.93， 0.92 and 0.88 for the rate period at the three hydrological stations， respectively， while the results of the TRMM 3B42 simulation were the next best， with the coefficients of correlation （R） of the rate-period and validation-period of above 0.80， and the Nash-Sutcliffe efficiency coefficient （NSE） of the simulations is above 0.50.This research demonstrates the applicability of CMADS datasets and SWAT model for runoff simulation in high-altitude areas with complex landscape types and sensitive to climate change， and provides a replacement solution for improving the hydrological models in areas where there are sparely meteorological stations.

The ground-based observational dataset is applied to evaluate the performance of 5 GCMs from the Coupled Model Intercomparison Project Phase 6 （CMIP6） in the Upper Yellow River Basin during 1995 -2014.And then， precipitation trends in the near term （2021 -2040）， mid-term （2041 -2060）， and long term （2081 -2100） under 7 SSP-RCP scenarios are projected， respectively.The results show that： （1） Multi-model ensemble mean can capture the inner-annual distribution of precipitation in the Upper Yellow River Basin， and the characteristic of more precipitation in the South and less precipitation in the north can also be captured.The spatial correlation coefficient between the simulated data and the observed data is above 0.9.That is to say， the spatial-temporal characteristics of precipitation in the Upper Yellow River Basin can be simulated satisfactorily by an ensemble mean of 5 GCMs.（2） In the 21^st century， annual precipitation in the Upper Yellow River Basin will demonstrate a significant increase tendency with obvious inter-decadal fluctuations.Relative to the baseline period 1995 -2004， the ascended annual precipitation will be faster in the near-term and then slow down in the 21^st century under SSP1-1.9 and SSP1-2.6 scenarios.Annual precipitation will rise continuously from near term to the end of 21^st under SSP2-4.5， SSP3-7.0， and SSP5-8.5 scenarios.Under SSP4-3.4 and SSP4-6.0 scenarios， annual precipitation will decrease slightly in the near term， but a turning point is detected in the mid-term， and precipitation will increase afterward.Spatially， the largest increase will be in the areas where precipitation is relatively small， including the headstream region above Huangheyan station and the region between Lanzhou and Toudaoguai.（3） Seasonal precipitation in the 21^st century in the Upper Yellow River Basin will show an overall rising trend with fluctuation， and the growth rate varies with scenarios and seasons.Except for the SSP4-6.0 scenario， precipitation trend under high radiation forcing scenarios will be greater than that of low radiation forcing scenarios.The growth rate is the largest for winter precipitation， while the growth rate is the smallest for summer precipitation， both passing the 0.1 significance level.Spatially， the highest growth of spring and autumn precipitation is projected in the headstream region above Huangheyan station and the region between Lanzhou and Toudaoguai.While the smallest growth of spring and autumn precipitation will be in the region between Huangheyan and Lanzhou.The area with the highest increase of winter precipitation is projected in the region between Lanzhou and Toudaoguai， and the smallest increase in the region between Huangheyan and Lanzhou.Summer precipitation will decrease in most of the Upper Yellow River Basin under SSP4-3.4 and SSP4-6.0 scenarios， but it will increase in most of the Basin with the highest growth in the headstream region above Huangheyan under all other scenarios.

The seasonal characteristics of water and heat exchange in the alpine grasslands are significant， and the freezing‐thawing process has an important impact on the land‐atmospheric water and heat exchange.Based on the observation data of the land surface process in the Tangchama small watershed in the source area of the Yellow River from May 2014 to May 2015， this research divides the soil freezing‐thawing process into thawed stage （TT）， frozen stage （FF）， thawing to freezing （T-F） and freezing to thawing （F-T）， and the changes in the different states and period of the net radiation， sensible heat flux， latent heat flux and surface heat flux of the underlying surface of the alpine grassland are analyzed to explore the characteristics of water and heat exchange between the land‐atmosphere in the soil freezing‐thawing process.The results are as follows： （1） The average value of the net radiation flux in the thawed stage is generally greater than that of the other three stages， and the maximum value reaches 203.7 W·m^-2.The frozen soil melts in the freezing‐thawing stage， and the soil moisture content gradually increases.The radiation ratio increased significantly during the frozen stage， the net radiation diurnal variation was the largest in the thawed stage， reaching 717.6 W·m^-2， and the frozen stage was the smallest， followed by the freezing‐thawing stage.（2） The proportion of sensible heat flux and latent heat flux is different in the thawed and frozen stages.When completely thawed， due to precipitation and soil moisture content， the net radiation is mainly converted into latent heat flux.The maximum diurnal variation of latent heat flux is 193.7 W·m^-2， while the sensible heat flux is only about 80.0 W·m^-2.The diurnal average of sensible heat and latent heat in the thawing‐freezing phase， the freezing‐thawing period and the frozen period is not much different.The mean latent heat in the three period is 21.9 W·m^-2， and the sensible heat is 20.3 W·m^-2； The diurnal variation is greater than the latent heat in the three period， the soil suffers a freezing‐thawing cycle， the soil temperature difference is small， and the water content changes， and the net radiation is mainly converted into sensible heat during this period； the diurnal variation of sensible heat was greater than that of latent heat in the three stages.The freezing-thawing cycle occurred in the soil， the difference between ground and air temperature was small， and the moisture content changed.During this period， the net radiation was mainly converted to sensible heat.（3） The soil heat flux is positive （negative） in thawed （frozen） state， indicating that the surface soil absorbs （releases） heat from the atmosphere， and its daily variation range is large （small）.The above results show that the state and process of soil freezing and thawing have different characteristics for the water and heat exchange process between the land and atmosphere.

As the main mechanism of extratropical cyclogenesis， moist baroclinic instability plays a central role in the study of cyclone thermodynamics， which can be further divided into four categories： dry baroclinic instability， moist instability， diabatic Rossby wave and Type C cyclogenesis （tropopause intrusion）.The '7·20' heavy rainstorm was caused by the eastward movement of a Yellow River cyclone into North China after its rapid formation on July 18， 2016.Compared with the mature stage of the cyclone， the mechanism of the initial stage is still unclear.This article uses ERA5 reanalysis data and WRF model to study the moist baroclinic instability of the cyclogenesis event numerically.The results show that mid-lower troposphere presented diabatic Rossby wave pattern， that is， the eastward movement of the system was mainly driven by the cycle of vertical motion and diabatic effect.The vertical motion on which the wave relied was more provided by vorticity advection.The PV sink and the ageostrophic wind in the upper layer delayed the eastward movement of the tropopause intrusion PV， maintaining the phase difference between upper and lower layers.Finally， a PV column formed throughout the troposphere in front of dry intrusion.Using piecewise PV inversion， several sensitivity runs are designed to remove unbalanced circulation， tropopause dry intrusion PV and lower-level diabatic-produced PV from the initial field， respectively.Combined with the analysis of generalized omega equation， it shows that the baroclinic wave in this process must be coupled with the diabatic process with the help of sufficient water vapor to develop strongly.The cyclogenesis was suppressed when the latent heat was turned off.Dry baroclinic instability cannot explain this process.The removal of initial unbalanced field did not affect the baroclinic instability but will delay development of the system.Limited by humidity and mesoscale circulation structure， the active area of lower-level unbalanced flow was controlled by dry baroclinic dynamics.In this case， the gradient of {\theta }_{bottom} was too small to organize eastward diabatic Rossby wave by relying only on the initial lower-level PV.Nor can strong lower-level diabatic heating generate as in Type C cyclogenesis by tropopause intrusion.For this Yellow River cyclogenesis case， it is required the initial lower-level PV anomaly to be strong enough to counteract the suppression of the cooling subsidence in front of dry intrusion； on the other hand， it is also required that the dry intrusion， in an appropriate initial phase difference with the lower system， strengthened the ascending motion east of the low-level PV in form of vorticity advection through vertical penetration， so as to promote the eastward momentum of diabatic Rossby wave to enter north China with more saturated environment.None of dry baroclinic instability， diabatic Rossby wave and Type C cyclogenesis could independently explain this cyclogenesis event， which was an initial optimal perturbation growth under the combined effect of diabatic Rossby wave and tropopause dry intrusion.

Based on the observed atmospheric forcing data from the grassland station of Erling Lake in the source region of the Yellow River from May 1， 2014 to August 31， 2015， the model CLM4.5 was driven， and the influence of plateau snow cover on the soil hydrothermal process was carried out.By comparing and analyzing the numerical simulation results， the main conclusions are as follows： （1） With the increase of snow on the plateau， the time of soil thawing lags behind， the more snow accumulated， the later the soil begins to thaw， the faster the thawing rate was， and the shorter the duration of soil thawing process was.（2） Snow cover has a certain heat preservation effect during the frozen period.When snow cover is reduced， the heat preservation effect is weakened， the heat transfer from the soil to the atmosphere is increased， and the heat is mainly transferred to the atmosphere in the form of sensible heat.Snow cover has a certain cooling effect when soil is in the thawing period.The cooling effect lasts until June.（3） During the soil thawing period， snow cover has a certain moisturizing effect.The higher moisture soil brought by snow melt can increase the radiation energy absorbed by the surface， and the higher moisture soil has a large latent heat transfer， which makes the sensible heat flux lower.The higher moisture soil caused by melting snow will last until after June.

Based on the field observation data in the source region of the Yellow River from October 2011 to December 2017, the characteristics of soil freezing-thawing duration, soil temperature and moisture, surface energy components in high snowfall year and low snowfall year were analyzed in this paper. The results show that the snowy year has higher albedo, lower net radiation, and lower surface heat transfer than the less snowfall year. The time of the snowy year from the heat "source" to the heat "sink" is later than the less snowfall year. Snow accumulation can reduce the radiant energy absorbed by the soil, reduce the sensible heat transfer on the surface, increase the surface latent heat transport during the frozen period and the thawing period of the soil, reduce the heat transfer from the soil to the atmosphere during the frozen period, and reduce the heat transfer from the atmosphere to the soil during the thawing period. The snow cover has a cooling effect in the freezing period, which makes the soil freeze earlier in the snowy years and the soil temperature is lower in the same period. It has the function of heat preservation in the frozen period, which makes the soil temperature higher in the snowy years. In the thawing period, it has the effect of heat preservation ("cooling"), which makes the ablation later and the soil temperature is low in the same period. During the whole freezing-thawing period, the shallow soil moisture in the snowy year is higher than that in the less snowy year, indicating that the snow has a moisturizing effect on the shallow soil. Snow accumulation makes the soil to start freeze earlier and the start of ablation later, which can extend the number of days that the soil is completely frozen.

Water and heat & CO₂ exchanges between alpine wetland and atmosphere are main causes of the regional climate change.In this paper, the energy and mass exchanges between the alpine wetland and atmosphere have been explored by using the datasets collected from the land surface process field experiment conducted from 16 July to 16 October, 2013 in the source region of the Yellow River.The diurnal characteristics of the sensible heat and latent heat & CO₂ fluxes had been analyzed by statistical methods, the energy closure had also been analyzed.The results have shown:The diurnal variations of the sensible and latent heat fluxes in alpine wetland revealed diurnal characteristics every months, the sensible and latent heat flux increased after sun rise, the peak value of sensible heat flux appeared around 12:00-16:00 and reached 150.0 W·m^-2, the latent heat flux appeared around 16:00 in July, and reached 300.0 W·m^-2.During the whole summer period, the latent heat was the main consumption of net radiation in the wetland system.The energy closure reached 48.8%.The CO₂ exchange flux displayed a "U" pattern during the whole growing period, the CO₂ flux was downward after sun rise, and reached a peak in the middle daytime, the minimum value was -0.55 mg·m^-2·s^-1 at 15:30 on 21 July.Further analysis has shown that CO₂ flux was controlled by air temperature, solar radiation and canopy.

Based on observational data, encryption data of automatic station, NCEP reanalysis and satellite data, the structure characteristics of MCC occurred on 1 April and 2 August 2015 over the Yellow River midstream was comparatively analyzed. The results are as follows:(1) The formation stage of spring MCC was fast and mature was slow. It has forward propagation feature. Precipitation was stable, rain clusters moved slowly, and the rainstorm was mainly caused by long duration of precipitation. While the formation stage of summer MCC was slow and mature was rapid. It has backward propagation feature. It was mainly convective precipitation, rain clusters were mobility, and heavy rainfall mainly caused by short-period strong precipitation. Maximum hourly rainfall occurred in different areas at different stages of their life time. (2) The spring MCC was formed under the background of the southwest airflow at the whole layers and strong baroclinic property. The spatial structure characteristics of the divergence field was vertical. But the summer MCC was formed under the background of flat westerly circulation on 500 hPa, front of anticyclone on 200 hPa, and the weak baroclinic property. The spatial structure characteristics of the divergence field was declining. In their later development, there were significant differences in water vapor, thermal and dynamic structure. (3) Two MCC were formed and developed in the coexistence region of conditional, convective, and symmetric instability. The formation of MCC was related to the instability and the unstable energy, but their development were more closely related to the continuous increase of the unstable energy, and continuous enhancement of symmetric instability, especially in midsummer. (4) Intrusion of dry cold air at middle and upper layer, continuous strengthening of convergent ascending motion caused by the curvature vorticity at the whole layers, and symmetric instability were important triggering mechanisms of MCC in spring. Mesoscale convergent ascending motion at lower layer caused by shear vorticity, symmetric instability, and gravity wave propagation were the main triggering mechanisms of MCC in midsummer.

Based on the precipitation data of 92 stations of the Yellow River Basin and NCEP /NCAR daily reanalysis data, the spatial and temporal changes and period characteristic of summer rain in the Yellow River Basin and its relation to the east Asia subtropical westerly jet in last 50 years has been analyzed. The spatial anomaly features of summer precipitation in the Yellow River Basin are mainly three kinds: wet or dry in all area pattern, wet (or dry) in southeast and dry (or wet) in northwest pattern, wet (or dry) in southwest and dry (or wet) in northeast pattern. When anomalies north of the east Asian subtropical westerly jet center, more summer rainfall in the upper reaches of the Yellow River Basin and less in the lower reaches of the Yellow River Basin, when the anomalies south of the east Asian subtropical westerly jet center, the summer precipitation in the upstream region of the Yellow River Basin less than normal and the downstream region is above normal. Anomalies east of the east Asian subtropical westerly jet center, the summer rainfall of the Yellow River Basin in upstream region above normal and less in the middle and lower reaches of the Yellow River, when anomalies west of the east Asian subtropical westerly jet center, the middle and lower reaches of the Yellow River Basin precipitation above normal, the upper reaches of the Yellow River Basin rainfall less than normal.

A heavy rain in the middle of Yellow River (Shanxi) on 9 July 2013 was analyzed through conventional observation data, FY-2E TBB, and ground encryption automatic weather station data. Evolution characteristics of shear line at low and near Mesoscale turbulence, dynamic and thermal structure characteristics, water characteristics, and nature and function of partial northeast air flow were analyzed by using output data of the high-resolution Weather Research and Forecasting (WRF) model. The results show that the heavy rain was directly caused by four Mesoscale convective cloud clusters aroused near shear line at 700 hPa. The occurrence and development of the Mesoscale convective cloud clusters were together triggered by the turbulent vortex formed near shear line at low, the ground Mesoscale dew front, and the ground Mesoscale convective line. The warm air overhead heavy rain area were forced uplift along northwest to southeast, so that form an inclined upward flow area along west cold air climbing affected by pinch of west and northeast cold air coming from low level. Two secondary circulations in the opposite direction were formed in two sides of inclined upward flow area. The convergence center of water was located near boundary layer. But it was not main reason causing heavy rain. The important reason of causing this heavy rain were that convergence ascending motion at low level was continued strength, water was collected to the heavy rain area by inflow of southerly winds, water was transported to high lever by ascending flow so causing the whole layer water of continued increasing over heavy rain area, and instability of vortex layer was strengthened due to symmetrical and convective instability were coexist. And under the background of strong instability, water was lifted along inclined upward flow, condensed, and falling by high efficiency way. The northeast airflow at low level was an interface of dry and cold air and warm. Various meteorological elements were uniform distribution on the interface. But in the center of the interface, it was a zero area of advection, vertical velocity, vorticity, and divergence at low level. There formed an anti-cyclonic secondary circulation around the northeast flow. There had a great vertical wind shear in two sides of the northeast flow. With it goes deep into Hetao area and Shanxi province, convergence of ascending flow at low level was continually strengthened. At the same time, subsidence became active. It was one of important triggering mechanism.

As the alpine wetland underlying surface is moist and sensitive to climate change, quantitative assessments of environmental factors on the influence of latent heat flux over there have important scientific significance. Environmental factors on the influence of latent heat flux are potential to be expressed as sums of the atmosphere and surface factors. By using the datasets collected from the land surface process field experiment which was conducted in the source region of the Yellow River from June 1 to August 31, 2014, the characteristics of solar radiation and water vapour pressure deficit in controlling latent heat flux density were analyzed, and further assessment was evaluated by a defined controlling factor. The main results are as follows:(1) The average values of relative atmospheric controlling factors on latent heat flux density exercised by solar radiation and water vapour pressure deficit are 0.98 and 0.02. Relative atmospheric controlling factor on latent heat flux density is mainly exercised by solar radiation, and the role of water vapour pressure deficit is only marginal. (2) The average values of relative surface controlling factor over latent heat flux density exercised by solar radiation and water vapour pressure deficit are 0.12 and -0.31. The former is larger in the morning and evening, and smaller in the noon. The absolute value of the latter is smaller in the morning and evening, and larger in the noon. (3) The average values of absolute total controlling over the latent heat flux density exercised by solar radiation and water vapour pressure deficit are 0.22 and -0.06 W·m^-2·Pa^-1. The average values of relative total control are 1.10 and -0.29. (4) Solar radiation controls the latent heat flux mainly through a direct effect (atmospheric factor). While the water vapour pressure deficit controls latent heat flux mainly through the indirect effect of altering surface resistance (surface factor). (5) The average value of the decoupling factor over the alpine wetland is 0.38. It implied that the coupling between the alpine wetland and atmospheric is low. It is of practical, the solar radiation is the main factor to control the latent heat flux over alpine wetland. This study provides a new research approach for the study of the parameterization of latent heat flux and evaporation under the context of global climate change.

In this paper， two kinds of daily satellites precipitation products（TMPA 3B42 and IMERG-Final） are used to drive SWAT hydrological model.The results are compared and analyzed， and the applicability and simulation potential of the new satellite precipitation in the source region of Yellow River （SRYR） are evaluated.The results indicate that： （1） for large-scale watersheds， the results of sensitivity analysis and calibration of parameters for multiple sub-watersheds at the same time are not applicable to each station.Therefore， we choose to make sensitivity analysis and calibration of parameters in each station.Finally， the Nash-Sutcliffe efficiency coefficient （NSE） of the runoff simulation results of the three stations in the validation period are all above 0.50， and the determination coefficients （R²） are all above 0.60.（2） IMERG-Final products are better than TMPA 3B42 products in simulation results.Both of them can simulate the main trends of monthly runoff change in SRYR， but they have overestimated the peak values of runoff.The precision of the new satellite product （GPM） is higher than that of the predecessor satellite product （TRMM）.However， the accuracy of the GPM products in high altitude areas needs to be improved and further revised.

To drive Community Land Model version4.0 (CLM4.0) simulate soil moisture over the Source Region of the Yellow River, the atmosphere forcing data (3h/1°×1° resolutions) from 1961 to 2010 of the Princeton University were used, the default soil organic matter content in CLM4.0 was replaced by using a variable converted from the MODIS/NDVI.The simulation results were compared to AMSR-E soil moisture product provided by Vrije Universiteit Amsterdam, the Netherlands.The in-situ measurements in the Maqu Soil moisture Network were used to validate the simulation results.The results show that:the simulation result could reveal the characteristic of soil moisture in spatial distribution and temporal trend, the distribution of simulated soil moisture was more reasonable with modified surface organic matter content and soil texture, and the simulation results have relatively lower values than that of the ground measurements, and the VUA AMSR-E soil moisture product.

Using daily mean runoff data at Tangnaihai hydrological station from 2002 to 2011 and earthquake records in East Asia, the series of beat vortex intensity index in the source region of Yellow River were calculated, and the relationship between the beat vortex index and the runoff at Tangnaihai hydrological station has been analyzed, aimed at searching the changes of underground thermal activity that control the flood discharge in the flood season, at affording warranty to the short-term forecasts of runoff, and at programming the use of water resources and preventing floods and drought in the upper reaches of Yellow River. The results show that the beat vortex has good correspondence to the abundant runoff with daily mean runoff anomaly percentage ≥50%, the time differences between the starting date of the abundant runoff period and the data of the beat vortex are mostly within a week, there is a positive correlation coefficient between the mean runoff during the abundant runoff periods and the beat vortex intensity indexes(α=0.01). It is of great significance to understand the coupling between earth and atmosphere.

The minimal Number of Required Sites (NRS) for regional soil moisture estimating under the certain accuracy requirements, and the representativeness of sites are studied by using the soil moisture data derive from "Maqu soil moisture observation network in Yellow River source region of the Northwest Institute of Eco-Environment and Resources, Chinese Academy Sciences"from July 2011 to August 2012. In addition, the best upscaling method which suitable for network has been selected by evaluating four commonly used methods. The results show that: (1) Under the accuracies R≥0.99, RMSD≤0.02 m³·m^-3, NRS at both 5 and 10 cm depth is 10. (2) Representativeness of the sites have been validated by time stability analyses, time sliding correlation analyses and optimal combination of sites. NST01 is the most representative site at 5 cm depth for the first two methods, and CST05 and NST02 are the most representative sites at 10 cm depth. The optimum combination sites at 5 cm depth are NST01, NST02, and NST07. NST05, NST08, and NST13 are the best group at 10 cm depth. (3) Linear fitting is the best up-scaling method."Single site"obtained by OCS has the greatest up-scaling effect, and time sliding correlation analyses takes the second place. (4) The linear fitting equations are used to estimate the variation of regional soil moisture during 3 July 2013 to 3 July 2014, when a large number of observed soil moisture data are lost.

The land surface radiation budget and the impacts of vegetation on water and heat exchanges are the keys of land surface process.By using the field observed data at Maduo from 2013 to 2014, the characteristics of surface radiation budget were analyzed under different types of synoptic conditions.The results show that: (1) The different value of upward shortwave radiation is about 7.76 W·m^-2, between 100% vegetation covered and the control test.However, the different value of upward longwave radiation is only-0.62 W·m^-2.(2) The difference value of simulated longwave radiation between bare land and controlled test is about 5.34 W·m^-2.Moreover, the difference value between 100% vegetation covered and the control test is just-0.62 W·m^-2.(3) The decrease of LAI value has an influence on the increasing land surface albedo.Furthermore, the radiation flux and the surface turbulence flux change little.The upward shortwave radiation increases just about 1.35 W·m^-2.Meanwhile, the latent heat decreases about 8.43 W·m^-2.(4) On the other hand, the increasing LAI have an impact on the decrease of upward shortwave radiation and upward longwave radiation.What's more, the variation of radiation flux is larger when LAI increasing than LAI decreasing.(5) The radiation flux reveals obvious diurnal variation.The net radiation varies from 200 W·m^-2 to 461 W·m^-2 under different types of synoptic conditions.The soil heat flux at different depths reaches the peak from June to July in 2013.

As two integral parts of the cryosphere,snow cover and frozen soil both have important impacts on climate system,and the interaction between them makes it complicated to analyze the their influences on climate change in areas where snow cover and frozen soil coexist.In order to investigate the influence of snow cover on the thermal regime and hydrological process of frozen soil,different soil freezing and thawing stages was divided into soil freezing stage,frozen stage and thawing stage by soil temperature data observed from October1, 2013 to May 31,2014 in source region of the Yellow River,then the effects of snowfall on soil temperature and moisture were analyzed in each stage.During soil freezing stage,the ground surface net heat flux decreased during the snowfall day and the following sunny days(with snow cover),which resulted in an obvious increase of the soil daily minimum temperature at depths of 5 cm and 10 cm and led to a rise of the soil daily minimum temperature at depth of 20 cm to above 0 ℃.This could delay the soil freezing process at the depth of 20 cm.During soil thawing stage,the ground surface net heat flux decreased during the snowfall day.As a result,the soil daily minimum temperature at depths of 5 cm and 10 cm dropped to below 0 ℃ and thus prolonged the process of soil thawing at these layers.In the snowfall process of the soil freezing and thawing stages,snow cover could not only change the soil moisture by thawing,but also change soil moisture by influencing soil temperature.During soil frozen stage,snow cover had an effect on soil temperature,but it had little effect on soil moisture.During the soil freezing-thawing,compared to the change of soil moisture caused by soil freezing and thawing,the change of soil moisture caused by snow thawing is smaller.

Using observational data from Zoige Station in the source region of the Yellow River, the model CLM3.5 was employed to do a single point simulation. Comparison of the observed and simulated values of soil temperature and moisture verifies the applicability of the model in the seasonal frozen soil area of the source region of the Yellow River. The simulated soil temperature is better during the non-freeze period, though the values are much higher at the deep layer than the observation. But a bias occurs during the freezing period. The simulated temperature is lower and the simulated freezing depth is deeper than the observation. The simulation of soil moisture indicates the start time of freeze and thaw is ahead of time. The thermal conductivity of soil matrix in the model is too large may be the main reason for the deviation. So the sensitive experiment which replaces the original parameterization scheme with the Johansen scheme was done. The new simulated result shows the significant improvement to the original. It matches the observation more closely in the freezing depth and the start time of freeze and thaw. The phenomenon which the simulated temperature is lower than the observation during the freezing period is also improved to some extent.

A three-dimensional mesoscale atmospheric model WRF is used to examine the characteristic of lake effect over Lake Ngoring in the upper reach of the Yellow River and the influence of  terrestrial environment change on the lake effect. Three groups of experiments have been simulated which include environmental improvement, status quo, degradation. The main results show: (1) From noon to nightfall in sunny summer, there is a clear lake-breeze circulation in Lake Ngoring; both sensible heat flux and latent heat flux are small over the lake in daytime, moreover, there is a strong cold (warm) lake effect over Lake Ngoring region in daytime (nighttime); the specific humidity over lake is larger than that over land in surface layer, but there is converse above it; the water vapor wall and high value areas of sensible flux can be found on land along the lake under the influence of lake breezes. (2) Environment degradation results in an enhanced lake-breeze circulation and higher water vapor wall. In addition, the difference of the boundary layer height increases between land and lake; for the different surface characteristics, the changes of sensible heat  and latent heat fluxes over land are much more than that over the lake. (3) Environment changes impact on the distributions of air temperature and specific humidity in the bottom and top of the boundary layer through the underlying surface and the lake breeze, respectively, as a result, there is a reverse trend in different heights.

Gansu-Yellow River basin is a natural-social-economic complex ecosystem intertwined with ecological and development issues.Soil conservation service （SCS） is an important safeguard for preventing soil erosion and promoting high-quality development in this region.Combining with the Revised Universal Soil Loss Equation（RULSE）， we obtained the annual SCS dataset for Gansu-Yellow River basin from the Standardized Vegetation Difference Vegetation Index， Land Cover Type product MCD12Q1， precipitation data， Digital Elevation Model data and the Harmonized World Soil Database version 1.1.The times span for this SCS dataset ranges from 2001 to 2015， with spatial coverage between 33°6′29″N -40°0′6″N and 97°23′38″E -108°42′38″E.This study offers a scientific basis for developing measures to improve SCS， and provides important data support for assessing ecological security and constructing ecological security pattern.The URL for obtaining the entity dataset is http： //www.ncdc.ac.cn/portal/metadata/a918f7ed-5988-44ea-80ad-ee14acab89aa.