Partitioning of evapotranspiration and its controls in four grassland ecosystems: Application of a two-source model
Introduction
Evapotranspiration (ET) is an important process for ecosystem water budgets and energy balance, and is closely linked to ecosystem productivity (Law et al., 2002, Scott et al., 2006). Vegetation transpiration (T) and soil water evaporation (E), controlled by different biotic and physical processes, are the two major components of ET. To accurately predict the climatic response of ecosystem functions and processes, quantifying the partitioning of ET and its controls are critical (Williams et al., 2004, Lauenroth and Bradford, 2006). The common approaches used to partition ET currently are measurements using lysimeter, sap flow, infrared thermometers and isotopes (Evett et al., 1994, Williams et al., 2004, Scott et al., 2006, Moran et al., 2009) and modeling (Shuttleworth and Wallace, 1985, Kustas, 1990, Brenner and Incoll, 1997, Kemp et al., 1997, Reynolds et al., 2000). Among these methods, modeling is becoming more and more popular because of its exclusive advantage in addressing ecosystem processes over a spectrum of timescales (Shugart, 2000). The Shuttleworth–Wallace model (S–W model) has been widely used for its simple and accurate consideration of hydrological processes, and good performance (Sene, 1994, Tourula and Heikinheimo, 1998, Brisson et al., 1998, Iritz et al., 1999, Anadranistakis et al., 2000, Kato et al., 2004).
Nevertheless, there are still some insufficiencies in the application of the S–W model. First, most of its applications were undertaken on croplands, with far fewer reports on its use on grasslands (Lafleur and Rouse, 1990, Nichols, 1992, Stannard, 1993). Grasslands occupy nearly 40% of global land surface and play a very important role in global energy balance and carbon budgets (White et al., 2000). In China, also about 40% of the country is covered by grasslands (Fan et al., 2008). Studies have indicated that the grassland regions in China are highly sensitive to global climate change (Ding et al., 2006). Until now very few reports have addressed the partitioning of ET in Chinese grasslands using the S–W model or other methods. The second problem is that the S–W model has mostly been used with short time periods (less than one growing season). Its long-term performance, when more processes are involved, has not been fully tested. Furthermore, due to the lack of investigating the model long-term performance, our knowledge of the partitioning of ET and its controls is critically limited. This limitation complicates our ability to predict the climatic responses of ecosystem carbon and water processes (Williams et al., 2004, Lauenroth and Bradford, 2006, Scott et al., 2006).
Many studies have indicated that E/ET, the indicator of ET partitioning, was controlled by canopy conductance at the diurnal timescale and by leaf area index (LAI) at the seasonal timescale (Sakuratani, 1987, Liu et al., 2002, Kato et al., 2004, Scott et al., 2006, Sauer et al., 2007). However, a question persists: what factor is dominantly responsible for the year-to-year and site-to-site variations? Further, due to the dearth of research on inter-site comparisons, how the effects of canopy stomatal conductance and LAI on E/ET would change across an environmental gradient remains unclear.
Using multi-year measurements of carbon and water vapor fluxes with the eddy covariance technique for four grassland ecosystems in China, we sought to address the following questions: (1) is the S–W model applicable to Chinese grasslands over long-term periods of time? (2) What is the importance of the soil water evaporation for whole ecosystem water vapor fluxes in Chinese grasslands? (3) How do the canopy stomatal conductance and LAI affect E/ET at different spatiotemporal scales? (4) How do the responses of E/ET to changes in canopy stomatal conductance and LAI vary across an environmental gradient? The ecosystems in this study belong to the main Chinese grassland types, and they illustrate an obvious water availability gradient (Fan et al., 2008), which enables us to investigate the effects of the environment over diverse spatial scales. There are obvious seasonal and inter-annual variations in ET in each ecosystem. This provides us a good opportunity to test the applicability of the S–W model.
Section snippets
Study sites
Four ecosystems with distinct water availability conditions and vegetation types were selected from ChinaFLUX eddy covariance (EC) tower stations (Yu et al., 2006a, Yu et al., 2006b). Shidi alpine swamp meadow (SD) is located at the Haibei alpine grassland station on the Qinghai-Tibet Plateau (37°37′N, 101°20′E; 3160 m a.s.l.). The climate at this site is characterized by strong solar radiation, with long cold winters and short cool summers. Annual mean air temperature is −1.7 °C. Annual mean
Performance of S–W model at the half-hour time scale
We ran the S–W model with the best parameter set which was the mean of the 20 most successful parameter sets (Table 2). In general, the model successfully simulated ET at the four ecosystems with the R2 between measured and modeled ET above 0.8 (Fig. 2). According to the slope k, the model generally overestimated ET by 8–15% at all the sites except DX. We also examined diurnal simulation in three distinct phases: pre-growing season (April), peak growing season (August) and late growing season
Model performance and uncertainties
The accurate estimate of ET at the diurnal and seasonal time scales in this study confirmed the ability of the S–W model to make accurate predictions for Chinese grasslands. The good performance of the S–W model was also confirmed in many other ecosystems (Tourula and Heikinheimo, 1998, Sene, 1994, Iritz et al., 1999, Kato et al., 2004). Compared with previous studies, the performance of the S–W in this study was generally better. The following three reasons may be responsible for the
Conclusion
With multi-year measurements of ecosystem evapotranspiration with eddy covariance systems, this study confirmed the good long-term performance of the S–W model at the four grassland ecosystems in China. Our study indicates that taking into account canopy rainfall interception and using varied parameter sets in different years may improve the model performance. The results of this study highlight the importance of soil water evaporation in water vapor fluxes for Chinese grasslands, and future
Acknowledgements
This research was jointly funded by National Natural Science Foundation of China (Grant Nos. 30590381, 30800151 and 30700110), the Knowledge Innovation Program of the Chinese Academy of Sciences (Grant No. KZCX2-YW-432), the National Key Research and Development Program (Grant No. G2002CB412501), and the “Hundred Talents” Program of the Chinese Academy of Sciences. We thank two anonymous reviewers for their valuable comments to improve this manuscript. Special thanks to Mr. David C. Brill at
References (42)
- et al.
Crop water requirements model tested for crops grown in Greece
Agricultural Water Management
(2000) - et al.
The effect of clumping and stomatal response on evaporation from sparsely vegetated shrublands
Agricultural and Forest Meteorology
(1997) - et al.
Parameterization of the Shuttleworth–Wallace model to estimate daily maximum transpiration for use in crop models
Ecological Modelling
(1998) - et al.
Gap filling strategies for defensible annual sums of net ecosystem exchange
Agricultural and Forest Meteorology
(2001) - et al.
Test of a modified Shuttleworth–Wallace estimate of boreal forest evaporation
Agricultural and Forest Meteorology
(1999) - et al.
Estimation of evapotranspiration, transpiration ratio and water-use efficiency from a sparse canopy using a compartment model
Agricultural Water Management
(2004) - et al.
Application of an energy combination model for evaporation from sparse canopies
Agricultural and Forest Meteorology
(1990) - et al.
Environmental controls over carbon dioxide and water vapor exchange of terrestrial vegetation
Agricultural and Forest Meteorology
(2002) - et al.
Determination of daily evaporation and evapotranspiration of winter wheat and maize by large-scale weighing lysimeter and micro-lysimeter
Agricultural and Forest Meteorology
(2002) - et al.
Partitioning evapotranspiration in semiarid grassland and shrubland ecosystems using time series of soil surface temperature
Agricultural and Forest Meteorology
(2009)