Analysis of water vapour flux between alpine wetlands underlying surface and atmosphere in the source region of the Yellow River

Yan Xie , Jun Wen , Rong Liu , Xin Wang , DongYu Jia

1. Key Laboratory of Land Surface Process and Climate Change in Cold and Arid Regions, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, Gansu 730000, China

2. University of Chinese Academy of Sciences, Beijing 100049, China

3. College of Atmospheric Sciences, Plateau Atmosphere and Environment Key Laboratory of Sichuan Province, Chengdu University of Information Technology, Chengdu, Sichuan 610225, China

ABSTRACT An underlying wetland surface comprises soil, water and vegetation and is sensitive to local climate change. Analysis of the degree of coupling between wetlands and the atmosphere and a quantitative assessment of how environmental factors influence latent heat flux have considerable scientific significance. Using data from observational tests of the Maduo Observatory of Climate and Environment of the Northwest Institute of Eco-Environment and Resource, CAS, from June 1 to August 31, 2014, this study analysed the time-varying characteristics and causes of the degree of coupling (Ω factor)between alpine wetlands underlying surface and the atmosphere and quantitatively calculated the influences of different environmental factors (solar radiation and vapour pressure deficit) on latent heat flux. The results were as follows： (1) Due to diurnal variations of solar radiation and wind speed, a trend developed where diurnal variations of the Ω factor were small in the morning and large in the evening. Due to the vegetation growing cycle, seasonal variations of the Ω factor present a reverse "U" trend. These trends are similar to the diurnal and seasonal variations of the absolute control exercised by solar radiation over latent heat flux. This conforms to the Omega Theory. (2) The values for average absolute atmospheric factor(surface factor or total) control exercised by solar radiation and water vapour pressure are 0.20 (0.02 or 0.22) and 0.005(-0.07 or -0.06) W/(m2·Pa), respectively. Generally speaking, solar radiation and water vapour pressure deficit exert opposite forces on latent heat flux. (3) At the underlying alpine wetland surface, solar radiation primarily influences latent heat flux through its direct effects (atmospheric factor controls). Water vapour pressure deficit primarily influences latent heat flux through its indirect effects (surface factor controls) on changing the surface resistance. (4) The average Ω factor in the underlying alpine wetland surface is high during the vegetation growing season, with a value of 0.38, and the degree of coupling between alpine wetland surface and atmosphere system is low. The actual measurements agree with the Omega Theory. The latent heat flux is mainly influenced by solar radiation.

Keywords: Alpine wetland; the source region of the Yellow River; latent heat flux; solar radiation; water vapour pressure deficit

1 Introduction

Energy and water exchange between the land and atmosphere have important influences on local climate change and are the core process of land surface studies. The land-atmospheric interaction can change the structure of the boundary layer above it, as well as its associated physical processes, and by changing the energy and material exchange between land and atmosphere, the surface energy affects regional and global climates (Trenberth et al., 2001; Mahrt and Vickers,2005; Oncley et al., 2007; Zhu and Lettenmaier,2007). Given different climatic backgrounds and underlying surfaces, there are considerable differences in the energy exchange processes between land and atmosphere (Hu et al., 1994; Li et al., 2000; Zhang and Cao, 2003; Bao et al., 2004). Climate change is sensitive to changes in latent heat flux, and vice versa(Jacobs and DeBruin, 1992). Additionally, there is a direct relationship between surface radiation budget,surface warming or cooling, evaporation, surface energy and the interaction between land and atmosphere (Chen et al., 1997).

Current research shows that latent heat flux is very important to near surface heat and water balances and is mainly affected by the interactions between environmental and surface factors. Heat (radiation and temperature) and water (soil moisture and atmospheric water vapour pressure deficit) factors are keys to controlling water and heat exchange between land and atmosphere (Jarvis and McNaughton, 1986;Kellner, 2001; Wever et al., 2002; Wang et al., 2008;Yu et al., 2008; Li, 2015). The degree of coupling between surface vegetation and the surrounding atmosphere can be represented by material fluxes and energy exchange capacities, which are mainly affected by aerodynamic and surface resistances (Zhang and Cao, 2003). In addition, when the degree of coupling is high, latent heat flux is mainly controlled by water vapour pressure deficit; when the degree of coupling is low, latent heat flux is mainly controlled by solar radiation. To study the spatial and temporal changes in latent heat flux for the given background global climate change, researching the degree of coupling between land and atmosphere and analysing the influence of environmental factors (e.g., solar radiation,water vapour pressure deficit) on latent heat flux are very important. Wetlands are a combination of water,soil and vegetation and are known as the "kidneys of the earth" (Li et al., 2009; Tan et al., 2013). Alpine wetland in the source region of the Yellow River is the main ecological barrier of the Tibetan Plateau and an important headwater; they are also the most fragile ecological environment and are the most sensitive to climate change (Li et al., 2012). The latent heat flux of the underlying wetlands are large during the vegetation growing season, and are larger at noon than that in the morning and evening; the monthly average latent heat fluxes were higher than those in nearby alpine meadows (Yu et al., 2008; Zhang et al., 2015).Therefore, researching the degree of coupling between land and atmosphere, and exploring the influence of solar radiation and water vapour pressure deficit on the latent heat flux of alpine wetlands of the Yellow River source area, play an important role in understanding the hydrological processes of the source area, climate change and the regional ecological environment.

At present, environmental control mechanisms of latent heat flux is heavily researched. Wang and Kellom?ki (2005), using the Penman-Monteith equation, found that solar radiation is the main environmental factor that controls latent heat flux of forest underlying surfaces, and that water vapour pressure deficit can be ignored. Through regression analysis,Yu et al. (2008) found that the main environmental impact factors of latent heat flux on Phragmites communis Trin. are net radiation, soil moisture content and relative humidity during the vegetation growing season, and net radiation, surface temperature and wind speed in the non-growing season. Wang et al.(2008) found that environmental variables impacting latent heat flux over lawns in semi-arid areas rank as follows： net radiation, difference between ground and atmospheric temperatures, ground temperature, relative humidity and atmospheric temperature. Ding et al.(2014), analysing the characteristics of heat and water fluxes and the main control factors over a maize field in an arid inland region, found that latent heat flux is the main energy consumer, and daily and interannual variations of latent heat flux are determined by net radiation and available energy, respectively. Li(2015) noted that the degree of influence of net radiation on the exchange of water and heat recedes with increasing time scale and the degree of influence of vapour pressure deficit and soil water content increases.

These studies, which focused on forest, grassland and farmland ecosystems, lack research on the effects of environmental factors controlling latent heat flux of alpine wetlands. In addition, research on the effects of these environmental factors on the latent heat flux is still qualitative and lacks quantitative evaluations and calculations.

Therefore, for alpine wetlands in the Yellow River source region of the Tibetan Plateau and considering the Penman-Monteith equation as the principal calculation, the characteristics and causes of the degree of coupling between alpine wetland surface and atmosphere was investigated. This study used observational data from the Maduo Observatory of Climate and Environment of the Northwest Institute of Eco-Envir-onment and Resources, CAS, to understand the influence of different environmental factors (e.g., solar radiation and vapour pressure deficit) and the atmosphere and surface factors controlling the latent heat flux over the alpine wetland. This has the potential to provide reference information for exploring the influences of environmental factors on latent heat flux of alpine wetlands in the Yellow River source region.

2 Materials and methods

2.1 Background of the site

The source region of the Yellow River is located in the northeastern part of the Tibetan Plateau and is characterised by many landscape types, e.g., basins,meadows, valleys, glaciers, lakes and permafrost.Qumarleb County of Qinghai Province is located to the southwest of the Yellow River; Yushu State is to the north and is located in the Qinghai-Tibet Plateau Yellow River source region, where the average altitude above sea level is 4,500.0 m (Figure 1). In this paper, we use observations from the Chinese site called the Environment of the Marlborough of the Chinese Academy of Sciences (Lat.： 35.02°N; Lon.：96.23°E; altitude above the sea： 4,313.0 m). Here, the cold season occupies seven to eight months and the warm season occupies four to five months. In addition, this region has a dry climate and experiences small annual temperature contrast but large daily temperature contrast. The annual averaged temperature is-3.3 °C, while annual averaged precipitation is 380-470 mm. This region has a typical plateau high cold climate, with a uniform surface of alpine meadows or seasonal wetlands, both of which are flat and open terrain (Chen et al., 2017).

Figure 1 The geographic location of the Maduo Observatory of Climate and Environment of the Northwest Institute of Eco-Environment and Resource, CAS

2.2 Eddy flux and radiation data processing

The flux datasets used in this study were obtained from observations of the eddy covariance system.This system consists of a CR5000 data acquisition device, a CSAT3 ultrasonic anemometer (Campbell),and a Li-7500 CO2/H2O analyser (Li-Cor Company).The whole system is installed 2.2 m from the ground,the instrument sampling frequency is 10 Hz, and the data collector automatically stores the original data.The observation times have a range within Beijing time of 00：00-23：30, recording every 0.5 hour, and the record is the average of the value measured ten minutes before and after the time (e.g., the record at 02：00 is the average of the actual measurements from 01：50 to 02：10).The data used in this study are from June 1 to August 31, 2014 (sample size is 227 after quality control). The system calculates the online flux(sensible and latent heat flux) using the eddy covariance principle and stores the time series of the average CO2flux, latent heat flux and sensible heat flux for 30 min, automatically adjusting the in-line flux with changes in the revised air density. To measure the temperature of air and surface, a Pt100 temperature sensor was installed in the ground radiation intensity meter. The radiation observation system with four component/net radiation sensors (Netherlands Hukse flux Company) were deployed to measure the solar short-wave radiation, ground short-wave upward radiation, ground long-wave upward radiation,and atmospheric long-wave downward radiation.

Due to weather-related factors, terrain conditions and physical limitations of the instrument, quality control of the observed data is needed. Quality controls based on the reference flux data when the universal standard is removed (Guo et al., 2004; Li et al.,2007; He, 2014); the specific methods used in this study are as follows：

1) Due to the influence of precipitation and cloud in relation to stability of the radiation budget, only datasets from clear days are used.

2) When turbulence is weak, there is large uncertainty of flux data. The friction velocity (u*) is a measure of turbulence intensity. All flux datasets where u*＞ 0.1 m·s are selected.

3) As turbulence is weak at night, the sensor probe is easily covered by dew condensation or frost, and the night-time water heat flux is small. Therefore, the datasets used must have a downward short-wave radiation greater than zero.

3 Methodology

According to the Penman-Monteith equation(Monteith and Unsworth, 1990), the latent heat flux density (λE) depends on available energy (FA), surface resistance (rc), aerodynamic resistance (ra), water vapour pressure deficit (D), air density (ρ), thermodynamic psychrometric constant (γ) and the rate of change of saturated vapour pressure with temperature(Δ), the equation of which can be expressed as follows：

where available energy (FA) can be expressed as(Rn-G), Rnand G represent net radiation and surface heat flux, respectively. Previous studies have shown that available energy (FA) is proportional to solar radiation (Rs) (Wang and Kellom?ki, 2005). The surface resistance (rc) can be calculated from Monteith (1990), Mackay et al. (2003) and Stewart(1988). Following Jacobs and De Bruin (1992), the influence of an independent variable on the dependent variable of quantitative calculation can be referred to as the control. To calculate the absolute control exercised by solar radiation and water vapour pressure deficit over latent heat flux density, they can,respectively, be expressed as the partial differential equation.

The coupling factor (ω) represents the degree of coupling between solar radiation (or water vapour pressure deficit) and latent heat flux density, which can be written as (Wang and Kellom?ki, 2005)

The effects of solar radiation and water vapour pressure deficit on latent heat flux through atmospheric factors is called direct control, while the effects of solar radiation and water vapour pressure deficit on latent heat flux through surface resistance is called indirect control. Total control is expressed as the sum control of atmospheric and surface factors, which work separately but simultaneously to contribute to the total control of latent heat flux. Detailed descriptions on how to calculate these variables are found in Cienciala et al. (1997) and Wang and Kellom?ki (2005).

If defined as Ω=1-ω, then the values of Ω range between 0.0 and 1.0 (Steduto et al., 1998). When Ω=0.0, the wetland-atmospheric system is completely coupled, and the latent heat flux is mainly affected by water vapour pressure deficit and surface resistance.When Ω=1.0, the wetland-atmospheric system is completely decoupled, and the latent heat flux is mainly influenced by solar radiation (or available energy).

4 Results

4.1 Typical sunny diurnal variation of the coupling between alpine wetlands and atmosphere

4.1.1 Diurnal variations of Ω factor

Figure 2 shows that the diurnal variation of Ω factor presents a trend of being small in the morning and large in the evening. This differs from meadow and field underlying surfaces. Figure 2a shows that maximum and daily average Ω factors are 0.42 and 0.27 on June 6, respectively. Because the alpine wetland vegetation has just turned green, the surface resistance is higher and the Ω factor is generally small.At this time, the degree of coupling between alpine wetland surface and atmospheric system is relatively strong. The daily variations of the Ω factor are affected by daily variations of wind speed at the same time. The sudden decrease in the Ω factor is due to the sudden increase in wind speed after 13：00; the continuous increase in the Ω factor is due to the decrease in wind speed after 16：00. The maximum and daily averages of wind speeds are 4.1 m/s and 2.0 m/s, respectively. Figure 2b shows that maximum and daily averages of the Ω factor are 0.58 and 0.33 on June 30,respectively. Alpine wetland vegetation is lush during this time. The degree of coupling between alpine wetland surface and atmospheric system is low. At this time, the wind speed is high, with maximum and daily average wind speeds of 8.3 m/s and 3.5 m/s, and the relatively dense vegetation makes the surface impedance decrease rapidly, causing the Ω factor to increase and the degree of coupling to weaken. Figure 2c shows that maximum and daily averages of the Ω factors are 0.60 and 0.37 on July 16, respectively,which implies that the degree of coupling between alpine wetland surface and atmospheric system has been very low. Alpine wetland vegetation is the most exuberant; thus, the Ω factor is larger. A fluctuating wind speed of approximately 2.5 m/s after 10：00 increases Ω factor fluctuations. Figure 2d shows that maximum and daily averages of the Ω factor are 0.49 and 0.31 on August 23, respectively. The degree of coupling between alpine wetland surface and atmospheric system is relatively poor. Alpine wetland vegetation has started to wither at this time, and the Ω factor begins to decrease. Like the diurnal variation of Ω factor on June 6, because of the sudden decrease in wind speed after 16：00, the Ω factor increases rapidly.

Figure 2 Typical sunny diurnal variations of the Ω factor and wind speed over alpine wetland surfaces in the source region of the Yellow River from June to August 2014 (lines and circle represent the Ω and wind speed at 2 m, respectively)

Figure 2 illustrates that the Ω factor is general small over the alpine wetland surfaces in the morning.The degree of coupling between alpine wetland surface and atmospheric system is relatively high, which is caused by weaker solar radiation, smaller canopy conductance and larger surface resistance. Although the wind speed is larger at noon, because the canopy conductance reaches a higher level and the surface resistance is small, the Ω factor is higher and the degree of coupling between alpine wetland surface and atmospheric system is low. Because the surface resistance keeps decreasing due to increasing water vapour pressure deficit and aerodynamic resistance keeps increasing due to decreasing wind speed, the Ω factor continues to grow in the afternoon. The degree of coupling between alpine wetland surface and atmospheric system remains low.

4.1.2 Diurnal variations of influence exercised by environmental factors over latent heat flux

Figure 3 shows the diurnal variations of absolute control exercised by solar radiation and water vapour pressure deficit on the latent heat flux. The diurnal variations of absolute control exercised by solar radiation over latent heat flux present a trend in which the absolute control is small in the morning and large in the evening, similar to the diurnal variation of the Ω factor. This conforms to the Omega Theory. After sunrise, with an increase in the Ω factor, the degree of coupling between alpine wetland surface and atmo-spheric system gradually drops, and the influence of solar radiation on latent heat flux increases gradually.The diurnal variations of absolute control of the water vapour pressure deficit over latent heat flux presents a trend in which the absolute control is small during the morning and evening and large at noon.

Figure 3 The diurnal variations of the absolute control exercised by solar radiation (Rs) and water vapour pressure deficit (D) over latent heat flux (written as ARs and AD, respectively)

4.2 Seasonal variation of coupling between alpine wetland and atmosphere

4.2.1 Seasonal variation of Ω factor

Figures 4c, 4d show the seasonal variation of the Ω factor, surface resistance and aerodynamic resistance during the vegetation growing season of alpine wetlands. Because the relative errors of surface resistances are at a minimum during the day, the daily value is the average from 12：00 to 14：00 for one day(Held et al., 1990). The Ω factor is small in early June, gradually increasing afterwards. The factor increases by 0.30 from early to late June, and peaks at 0.39 in mid- to late July, before gradually decreasing,and gradually decreases to approximately 0.30 in late August. Because alpine wetland vegetation has just turned green in early June, leading to a relatively high surface resistance, the Ω factor is small. Alpine wetland vegetation growth creates a decreased surface resistance, and the Ω factor gradually increases.Alpine wetland vegetation is the most robust, and surface resistance is generally small in mid- to late July.At the same time, due to small wind speeds, aerodynamic resistance is higher. Due to these two reasons,the Ω factor reaches its maximum value. Then, alpine wetland vegetation begins to wither, and the Ω factor also begins to decrease.

Figures 4a, 4b show that the correlation between the Ω factor and surface resistance is clear and that surface resistance plays a main role in seasonal variations of the Ω factor (relative to aerodynamic resistance). When the surface resistance is large, any increase (or decrease) of its value causes the Ω factor to rapidly decrease (or increase). The variations of surface resistance due to the withering of alpine wetland vegetation are the main reason for seasonal variations of the Ω factor. When the leaf area index of the vegetation is small, this effect is more significant.

4.2.2 Seasonal variations of influence of environmental factors on latent heat flux

Data analysis shows that the absolute control of solar radiation on latent heat flux reaches a maximum value of 0.22 in mid- to late July and a minimum value of 0.16 in early June, mirroring the seasonal variations of the Ω factor. This trend conforms to the Omega Theory. In mid- to late July, the Ω factor reache a maximum value of 0.39. The degree of coupling between alpine wetland surface and atmospheric system is very low. Therefore, the absolute control of solar radiation over latent heat flux reaches a maximum. In early June, the Ω factor reaches a minimum value of 0.27. The degree of coupling between alpine wetland surface and atmospheric system is relatively high. Therefore, the absolute control of solar radiation over latent heat flux reaches a minimum.

Figure 4 Typical sunny correlations relationship between the Ω factor and surface resistance (a) and aerodynamic resistance (b)and seasonal variations of them (c and d) over alpine wetlands from June to August in 2014 (using the average from 12：00 to 14：00)

During the vegetation growing season of alpine wetlands, solar radiation always plays a role in increasing latent heat flux, and the water vapour pressure deficit always plays a role in decreasing latent heat flux. The absolute control exercised by solar radiation over latent heat flux is always greater than the absolute control exercised by the water vapour pressure deficit. Because the minimum Ω factor is as high as 0.27, the degree of coupling between alpine wetland surface and atmospheric system is low during the vegetation growing season. The Omega Theory shows that latent heat flux is mainly influenced by solar radiation. The maximum (-0.12 W/(m2·Pa)) absolute control exercised by the water vapour pressure deficit over latent heat flux of alpine wetlands also appears in mid- to late July and the minimum (-0.05 W/(m2·Pa))is observed in early June.

4.3 Influence exercised by environmental factors over latent heat during vegetation growing season

4.3.1 Absolute atmospheric factor and surface factor controls of environmental factors

Results of the aforementioned formula are presented in Figure 5. Regarding the role of solar radiation in controlling latent heat flux, values for the absolute control of atmospheric factors mostly range from 0.07 to 0.39, with a mean of 0.20, and are inversely correlated with the ratio of surface resistance to aerodynamic resistance. Values of the absolute atmospheric controls of water vapour pressure deficit mostly range from 0.001 to 0.013 W/(m2·Pa), with an average of 0.005 W/(m2·Pa), and are closely inversely correlated with rc. These results show that absolute atmospheric factor controls exercised by solar radiation are closely correlated with the ratio of surface resistance to aerodynamic resistance and that exercised by the water vapour pressure deficit is closely correlated with surface resistance alone. Because there is a strong relationship between solar radiation and stomatal opening of vegetation, wind speed and stomatal opening of vegetation determine surface resistance and aerodynamic resistance, respectively. However, the water vapour pressure deficit is only closely related to vegetation transpiration, which is closely related to surface resistance.

Figure 5 Variations of the absolute atmospheric factor control exercised by solar radiation (a) and water vapour pressure deficit (b)over latent heat flux with ratio rc/ra and rc and variations of the absolute surface factor control exercised by the solar radiation (c) and water vapour pressure deficit (d) over latent heat flux with solar radiation and water vapour pressure deficit from June to August 2014 for alpine wetlands

The absolute surface factor control of solar radiation on latent heat flux has values ranging from 0.01 to 0.06, with a mean of 0.02 and is inversely correlated with solar radiation and the ratio of surface resistance to aerodynamic resistance. Values for the absolute surface factor control of water vapour pressure deficit mostly range from -0.16 to -0.01 W/(m2·Pa),with an average of -0.07 W/(m2·Pa), and the absolute value are closely correlated with water vapour pressure deficit.In calculating the absolute control exercised by solar radiation and water vapour pressure deficit, we use an empirical equation for surface resistance to estimate the surface factor control effects of solar radiation and water vapour pressure deficit on latent heat flux. Values for the surface factor control over latent heat flux are determined not only by the Ω factor but also by other factors, as described in the second terms of Equations (4)-(5). Consequently,even with a large value of ω (or low value of ω), due to the large value of the rc/raratio, the absolute surface factor control maintained by either solar radiation or water vapour pressure deficit can be low or high for this alpine wetland.

4.3.2 Absolute total control exercised by environmental factors

Values of the absolute total control exercised by solar radiation range from 0.08 to 0.42, with an aver-age of 0.22. These values are clearly correlated with rc/ravalues. Figure 6a shows that small values of rc/ra(＜50) produce absolute total control values attributable to solar radiation that decrease rapidly with an increase of rc/ra. Figure 6b shows that values of the absolute total control exercised by the water vapour pressure deficit range from -0.15 to -0.01 W/(m2·Pa),with an average of -0.06 W/(m2·Pa), and the absolute value increases with increasing water vapour pressure deficit. Their values are always less than zero, and the water vapour pressure deficit always weakens the latent heat flux of alpine wetlands. With an increase in the water vapour pressure deficit, the effect is more significant.

Figure 6 Variations of the absolute total control exercised by solar radiation (a) and water vapour pressure deficit (b) over latent heat flux with rc/ra and water vapour pressure deficit and variations of the absolute atmosphere factors, surface factors and total control exercised by solar radiation (c) and water vapour pressure deficit (d) over latent heat flux withsolar radiation and water vapour pressure deficit from June to August 2014 in alpine wetlands (subscripts 1 and 2 mark atmospheric factors and surface factors, respectively)

Figure 6c shows that no matter what the variations in solar radiation, the absolute total control it exercises over latent heat flux is mainly determined by atmospheric factors. Figure 6d shows that no matter what the variations in the water vapour pressure deficit are, the absolute total control it exercises over latent heat flux is mainly determined by surface factors. For the alpine wetland underlying surface,solar radiation primarily influences latent heat flux through direct actions (e.g., changing available energy). The water vapour pressure deficit primarily influences latent heat flux through indirect changes of the surface resistance (e.g., changing stomatal conductance).

Solar radiation and water vapour pressure deficit always act in opposite directions in the control of latent heat flux. However, the reaction of absolute total controls exercised by solar radiation and water vapour pressure deficit over latent heat flux is not one-to-one. As presented in Figure 7, a low absolute value of the absolute total control of the water vapour pressure deficit does not mean a correspondingly small value of solar radiation or vice versa. The absolute total control of solar radiation on latent heat flux is always greater than the absolute value exercised by the water vapour pressure deficit. The average of the Ω factor is 0.38 for alpine wetlands. The degree of coupling between alpine wetland surface and atmospheric system is relatively poor. This analysis shows that the absolute total control over latent heat flux is mainly attributable to solar radiation and that the role of the water vapour pressure deficit is only marginal.This finding is similar to the conclusions reached using the Omega Theory and represents the results of actual data. For the alpine wetland underlying surface with plenty of water, latent heat flux is mainly influenced by solar radiation.

Figure 7 Relationships of the absolute total controls exercised by solar radiation and water vapour pressure deficit in alpine wetlands from June to August 2014

5 Conclusions

Through the above analysis, the diurnal and seasonal variation characteristics and causes of the degree of coupling between alpine wetland underlying surface and atmosphere are investigated, and the influence on the latent heat flux of solar radiation and water vapour pressure deficit are calculated from aspects of atmospheric factors, surface factors and total control. The conclusions are as follows.

1) The diurnal variations of the Ω factor have a trend in which the Ω factor is small in the morning and large in the evening during the alpine wetland growing season. Because the increases in solar radiation and the water vapour pressure deficit after sunrise and decreased wind speed in the evening, the degree of coupling between alpine wetland surface and atmospheric system continues to be low. Because of the alpine wetland vegetation growing cycle, the seasonal variations of the Ω factor present a reversed "U"during the vegetation growing season here.

2) The diurnal and seasonal variations of the absolute controls exercised by solar radiation over latent heat flux are similar to the diurnal and seasonal variations of the Ω factor. This conforms to the Omega Theory.

3) Values for average absolute atmospheric factor(surface factor or total) control exercised by solar radiation and water vapour pressure are 0.20 (0.02 or 0.22) and 0.005 (-0.07 or -0.06) W/(m2·Pa), respectively. Generally speaking, solar radiation and water vapour pressure deficit exert opposite forces on latent heat flux.

4) For the alpine wetland underlying surface, solar radiation primarily influences latent heat flux through direct actions (e.g., changing available energy). Water vapour pressure deficit primarily influences latent heat flux through indirect effects of changing the surface resistance (e.g., by changing stomatal conductance).

5) The average value of the Ω factor in the alpine wetland underlying surface is 0.38, and the degree of coupling between alpine wetland surface and atmospheric system is relatively small. The actual measurements agree with the Omega Theory. The latent heat flux is mainly influenced by solar radiation.

As an evaluation index of the water vapour flux coupling, the Ω factor can represent the degree of coupling of water vapour pressure between the surface of vegetation and the atmosphere outside the boundary layer of vegetation. At the same time, this factor represents the relative importance of solar radiation and vapour flux coupling in controlling the rate of transpiration. This study analysed the diurnal and seasonal variation characteristics of the degree of coupling between alpine wetland underlying surface and atmosphere during the vegetation growing season,and quantitatively calculated the influence of solar radiation and vapour pressure deficit on latent heat flux using the Penman-Monteith Theory. We found that the observed results correspond with the Omega Theory. The Ω factor can be used to quantify the influence of surface resistance on latent heat flux, but the Ω factor cannot be used to quantify the absolute control exercised by solar radiation and water vapour pressure deficit on latent heat flux. The equations of the absolute control of solar radiation and water vapour pressure deficit on latent heat flux are non-linear and depend not only on the Ω factor but also on other factors, e.g., solar radiation, water vapour pressure deficit and response functions. This study is conducted at an observation point but is to be extended further across the Yellow River source region in the future through modelling in order to better understand the spatial and temporal variations of latent heat flux under the background of global climate change.

Acknowledgments:

This study was supported by funding from the National Natural Science Foundation of China (Grant Nos. 41530529 and 91737103). The authors are grateful to the anonymous reviewers for their constructive comments.