Flux variance similarity-based partitioning of evapotranspiration over a rainfed alfalfa field using high frequency eddy covariance data
Introduction
Quantifying evapotranspiration (ET) is fundamental for a better understanding of agro-ecosystems and allocation of scare water resources since ET is a key component of the hydrological cycle that accounts for up to 95% of the water budget in dry agriculture (Wilcox et al., 2003). In recent years, the eddy covariance (EC) method has been widely used to measure high frequency (i.e., 10 Hz or higher) observations of the exchange of carbon dioxide (c) and water vapor (q) fluxes simultaneously at the landscape level (Baldocchi, 2014). Although EC can provide direct and continuous measurements of ET, it cannot provide separate measurements of the two components of ET: evaporation (E, nonproductive water use) and transpiration (T, productive water use enhancing plant productivity). It is difficult to determine individual components E and T at the landscape level through measurement (Burt et al., 2005; Wang et al., 2016; Williams et al., 2004). The partitioning of ET has important implications not only for the water budget but also for a mechanistic understanding of biological and climatic controls of E and T (Ferretti et al., 2003; Wang et al., 2015) since these two components are controlled by different processes and respond differently to climatic factors (Kool et al., 2014). Thus, the partitioning of ET into E and T is crucial to minimize the nonproductive loss of water and to improve water management practices and productivity of agroecosystems. Partitioning can also offer greater insights into the function of agroecosystems by reducing uncertainties in the interpretation of the coupling between water and carbon/nutrient cycles (Austin et al., 2004). In addition, partitioning is useful for improving the performance of land surface models as they are poorly constrained due to a lack of observations of the diurnal and seasonal variations of ET partitioning (Lawrence et al., 2007). However, the partitioning of ET into E and T is still theoretically and technically challenging.
The exchanges of q and c are tightly coupled ecosystem processes (Morales et al., 2005). Direct measurements of carbon gain and water loss by EC allow us to quantify water use efficiency (WUE) at the ecosystem level (Law et al., 2002; Wagle and Kakani, 2014a), which reflects the trade-off between water loss and carbon uptake in carbon assimilation process. However, direct measurement of WUE solely based on T (i.e., productive water use) is not possible by EC due to the lack of separate measurements of E and T.
There are various methods to partition ET into E and T (Kool et al., 2014; Sutanto et al., 2014). The ET partitioning methods range from the conventional technique of integrating hydrometric T measurements (i.e., sap flow) with E measurements (i.e., weighing lysimeter) (Herbst et al., 1996; Kelliher et al., 1992) to more recent techniques based on analyzing the isotopic composition of liquid water and water vapor (Sutanto et al., 2012; Yepez et al., 2005), and to modeling approaches such as global land surface models (Miralles et al., 2011), the HYDRUS-1D model (Simunek et al., 2005), and two source surface energy balance (SEB) models (Norman et al., 1995). However, all these methods have limitations due to experimental difficulties or uncertainties or issues of spatial and temporal coverages in the measurement of E and T.
The Flux Variance Similarity (FVS) ET partitioning method had been proposed nearly a decade ago to partition E and T using high frequency EC data (Scanlon and Kustas, 2010, 2012; Scanlon and Sahu, 2008). A few recent studies have employed the FVS partitioning method to compare with other partitioning approaches (Klosterhalfen et al., 2019a; Palatella et al., 2014; Peddinti and Kambhammettu, 2019; Perez‐Priego et al., 2018; Sulman et al., 2016). Some studies evaluated the performance of the FVS partitioning method for different land cover types: a suburban grass field (Wang et al., 2016), citrus orchards (Peddinti and Kambhammettu, 2019), a Mediterranean cropping system (Rana et al., 2018), and across gradients of woody plant cover (Wang et al., 2010). A few studies have performed sensitivity analysis of partitioning results using various estimates of leaf-level WUE (Klosterhalfen et al., 2019b; Sulman et al., 2016). Although EC measurements are available across the world through several EC networks (e.g., FLUXNET, AmeriFlux, EUROFLUX, AsiaFlux, ChinaFlux), wider testing and validation of the FVS partitioning method is still scarce, most likely due to the computational complexity of analyzing high frequency EC data. To permit wider practical applicability, the open source software Fluxpart has recently been developed to implement the FVS partitioning method (Skaggs et al., 2018). Details on the FVS flux partitioning method are available in the above-mentioned studies.
This study employed Fluxpart to partition and quantify the dynamics of E and T by examining q-c correlations (Rqc), and to characterize the biological and physical processes controlling the temporal dynamics of E, T, and Rqc over an alfalfa field (Medicago sativa L.). The alfalfa field was harvested periodically (4–5 times per year) and the study period consisted of two contrasting years: dry year 2016 (~32% less rainfall compared to the 30-year, 1981–2010, mean of 925 mm) and wet year 2017 (~20% more rainfall compared to the 30-year mean). Thus, this study can serve as a suitable case study for testing FVS ET partitioning method using the q-c correlation from high frequency (10 Hz) EC data.
Section snippets
EC data description and processing
Using an EC system, high frequency (10 Hz) observations of exchange of c and q between a 48 ha Alfalfa field (cv. Cimarron 400 planted in Fall 2012) and the atmosphere were recorded from Apr 2016 to May 2018 at the United States Department of Agriculture-Agricultural Research Service, Grazinglands Research Laboratory, El Reno, Oklahoma, USA. The EC system, mounted at a height of 2.5 m above the ground surface, comprised of an open path infrared gas analyzer (LI-7500 RS, LI-COR Inc., NE, USA)
Patterns of ET, E, and T
The 2017 growing season was wetter than the 2016 growing season (Fig. 1). The site received total rainfall of 501 mm and 930 mm during the 2016 and 2017 growing seasons (Apr-Oct), respectively. Thus, cumulative forage yield of alfalfa was also higher in 2017 (~10 dry t ha−1) than in 2016 (~7.5 dry t ha−1) (Wagle et al., 2019b). Total rain for Jan-Apr 2018 was only 101 mm (~0.6 times less than the 30-year mean of ~250 mm). As a result, ET values were higher during the 2017 growing season than
Conclusions
The correlation between water vapor (q) and carbon dioxide (c) fluxes were examined with the FVS ET partitioning method using 10 Hz time series eddy covariance (EC) measurements over a non-irrigated alfalfa field. This ET partitioning method successfully reproduced the seasonal and inter-annual variations of partitioned E and T fluxes. Temporal variability of Rqc was consistent with the expected shifts in the dominance of T or E. The Rqc was strongly regulated by vegetation status, hay
Disclaimer
“Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.”
“The U.S. Department of Agriculture (USDA) prohibits discrimination in all its programs and activities on the basis of race, color, national origin, age, disability, and where applicable, sex, marital status, familial status, parental status, religion, sexual orientation, genetic
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This study was partly supported by a research grant (Project No. 2019-68012-29888) through the USDA-NIFA. We would like to thank a research technician, Shelby Robertson, for her assistance in screening the input files and running the Fluxpart partitioning codes.
References (52)
Evaporation, xylem sap flow, and tree transpiration in a New Zealand broad-leaved forest
Agric. Meteorol.
(1992)Sensitivity analysis of a source partitioning method for H2O and CO2 fluxes based on high frequency eddy covariance data: findings from field data and large eddy simulations
Agric. Meteorol.
(2019)A review of approaches for evapotranspiration partitioning
Agric. Meteorol.
(2014)Environmental controls over carbon dioxide and water vapor exchange of terrestrial vegetation
Agric. Meteorol.
(2002)A review of the molecular diffusivities of H2O, CO2, CH4, CO, O3, SO2, NH3, N2O, NO, and NO2 in air, O2 and N2 near STP
Atmos. Environ.
(1998)- et al.
Source approach for estimating soil and vegetation energy fluxes in observations of directional radiometric surface temperature
Agric. Meteorol.
(1995) - et al.
Dynamics of crop coefficients for citrus orchards of central India using water balance and eddy covariance flux partition techniques
Agric. Water Manag.
(2019) - et al.
CO2 and H2O flux partitioning in a Mediterranean cropping system
Agric. Meteorol.
(2018) - et al.
Partitioning carbon dioxide and water vapor fluxes using correlation analysis
Agric. Meteorol.
(2010) - et al.
Correlation-based flux partitioning of water vapor and carbon dioxide fluxes: method simplification and estimation of canopy water use efficiency
Agric.Meteorol.
(2019)