Transpiration and evaporation in a Californian oak-grass savanna: Field measurements and partitioning model results
Introduction
Transpiration (from vegetation) and evaporation (mainly from soil or other surfaces) are two main process-based components of evapotranspiration (ET), a critical parameter for understanding the carbon and water cycles on Earth (Schlesinger and Jasechko, 2014). With the eddy-covariance (EC) technique, we can now measure CO2 fluxes, but also H2O vapor fluxes (i.e., ET), intensively at the ecosystem level all over the world. Because of intrinsic interactions between the photosynthesis and transpiration processes, partitioning ET provides a fundamental data source for investigating water relations of carbon sequestration and their potential variability in changing climates (Nelson et al., 2018; Stoy et al., 2019). Specifically, we need the partitioned results to understand mechanisms of stomatal opening, leaf growth, energy absorption, and water balance in controlling photosynthesis and transpiration processes. In return, these understandings will help us to better predict the temporal and spatial variability of CO2 and H2O vapor fluxes. Thus, partitioning ET is critical for better developing process-based models and upscaling plant-based measurements (e.g., sap flow data) (Saugier et al., 1997; Wilson et al., 2001). Also, partitioning ET is highly demanded by validations of remote-sensing and top-down models, which are essential for water management in the area with limited water resources, such as in semi-arid or arid areas, or even in mesic areas but threatened by increased drought events due to climate change (Fisher et al., 2017; Humphrey et al., 2018; Kool et al., 2014).
Savanna is a type of ecosystem where trees unevenly distribute in herbaceous communities, forming the mosaic of woodlands and open grasslands. Such a landscape is often found in semi-arid areas, covering approximately 20% of the total land area on Earth (Scholes and Archer, 1997). The coexistence of trees and herbaceous communities enhance competitions to water sources. In California, oak trees and annual grasses coexist on the foothill of the Sierra Nevada Mountains. Located in the typical Mediterranean Climate zone, the oak-grass savanna experience wet, mild winters and dry, hot summers. In general, oak trees and annual grasses grow fast in the spring. As the rainy season stops, soil surface and shallow layers become drier. Meanwhile, oak trees can still maintain their necessary metabolic activities by accessing soil moisture in deeper soil layers or tapping groundwater; annual grasses die out and spread seeds dormant until the following rainy season. Differences in phenological and water accessible niches suggest that oak trees and annual grasses all contribute to the total ET via transpiration, but their contribution fractions should vary over the growing season (Baldocchi et al., 2004, 1997). Because the vegetation structures in the savanna ecosystem are complex, horizontally and vertically, testing the hypothesis in savanna ecosystems is quite challenging (Baldocchi et al., 2004; Kool et al., 2014).
ET measured with the eddy-covariance technique is a rich data source for further quantifying T and E (Kool et al., 2014; Stoy et al., 2019). As ET is measured from a single EC tower, the ET signal is a combination of both T and E signals, which is a barrier to obtain accurate values of transpiration or evaporation directly. A field solution is to derive T from multiple EC towers, such as installing two towers above and below the tree canopy (Baldocchi and Vogel, 1996; Baldocchi et al., 1987; Paul-Limoges et al., 2020; Scott et al., 2003). The primary concern of this approach is the possibility of insufficient turbulent conditions around the understory tower. However, since tree canopy is relatively open in the savanna area, chances of sufficient turbulent conditions for valid field measurements are higher than closed-canopy forests (Baldocchi and Meyers, 1991; Baldocchi et al., 1997; Launiainen et al., 2005; Misson et al., 2007; Scott et al., 2003). Many studies have compared understory tower measurements with results of other direct approaches (e.g., sap flow, stable isotope) or biophysical models (Black et al., 1996; Paul-Limoges et al., 2020; Roupsard et al., 2006; Scott et al., 2003; Wilson et al., 2001). We, therefore, gain enough confidence in using the multi-tower approach in the savanna area.
In addition, it has been pointed out that the understory tower signals also include the contributions of the understory vegetation layer (e.g., herbaceous communities), although an understory tower is often considered as a measure of soil evaporation (Holwerda and Meesters, 2019). Thus, we are here interested in testing whether the understory ET could be further partitioned into the transpiration of annual grasses and evaporation from soil and other wet surfaces (e.g., trunks, branches, leaves, and litters). Certainly, partitioning the understory ET needs the help of ET-partitioning models.
ET-partitioning models have been developed based on tower measurements (Li et al., 2019; Scott and Biederman, 2017; Wei et al., 2017; Zhou et al., 2016). A key thought behind the algorithm is whether first to solve the T/ET or E/ET ratio. Since T/ET + E/ET ≈ 1, either approach is possible to partition ET into T and E. For example, Zhou et al. (2016) proposed an algorithm based on leaf-level marginal water use efficiency, which is an application of the theory of stomatal controls on both photosynthesis and transpiration (more explanations in the Methods section). In contrast, Scott et al. (2017) solved the E/ET ratio with the assumption that transpiration should equal zero if gross primary productivity (GPP) equals zero. The assumption leads to the result that a mean value of E equals the interception of a valid linear relationship between ET and GPP. Since these two approaches are based on very different underlying principles, it will be interesting to compare them and see whether their partitioning results are comparable with what we learn from our field measurements.
Thus, our primary objectives for this study are: (1) to quantify the magnitude of transpiration of oak trees and annual grasslands with CO2 and H2O vapor flux measurements from three eddy-covariance towers; (2) to apply the two ET-partitioning models and compare their estimates of the T/ET ratio against our tower measurements; (3) to compare model results between each other. Based on differences in the phenology of oak trees and annual grasses, we hypothesize that maximum water loss via transpiration occurs in the spring when vegetation is in the most active photosynthesis, and consequently, transpiration contributes a large portion of the total ET. Moreover, the T/ET ratio would display seasonal variations, probably higher than the E/ET ratio during the growing season (Wei et al., 2018). For evaluating the model performance, we hypothesize that, at a minimum, they capture the correct seasonal patterns of T/ET as suggested by field measurements. In addition, we will discuss the model performance by comparing the magnitudes of partitioned results between different calculation approaches.
Section snippets
Study sites
Our two study sites are in an oak-grass savanna, representing an oak-dominated woody area (Tonzi Ranch, 38.438 N, 120.968 W) and open grassland area (Vaira Ranch, 38.418 N, 120.958 W). In the woodland area, oak trees are denser, and herbaceous communities coexist under the tree canopy and open space between trees (Fig. 1). The open grassland area is dominated by annual grasses with fewer oak trees scattered around the edge. The two sites are ~2 km apart, and the average elevation is 177 m above
Tower-measured ETover, ETunder, Toak, and T/ET ratios
ETover, ETunder, and Toak all varied over the growing season with differences in the magnitude and the timing of their maxima (Fig. 4a). Their decreasing rates after the peaks were also different. Toak peaked in the spring, and ETover matched Toak gradually as the understory annual-grass communities died out at the onset of dry summer, shown as a sharper decrease in ETunder with the decrease in soil moistures (Fig. 4c). It suggests that the total transpiration of the ecosystem (T) is dominated
Discussion
This study provides a basic idea of the percentage of T and E over the savanna landscape in California. In combination with field and statistical modeling approaches, our analysis shows that the percentage of oak canopy's transpiration contribution to the total ET in the oak woodland area is similar to the percentage that annual grasses’ transpiration in the open area, ~67%. This result is within the global range of 50% ~ 76% (Schlesinger and Jasechko, 2014; Wei et al., 2017) and consistent
Conclusions
In combinations with field measurements and ET-partitioning models, we learn that over this Californian savanna landscape, annual ET (± standard deviation) from the oak woodland was 419±85 mm. Among the total ecosystem-level evapotranspiration, oak canopy transpiration contributed ~67%, understory grasses transpiration contributed ~16%, and surface evaporation was ~17. In comparison, the open grassland has 324±43 mm in total annual ET, including ~67% water lost via grass transpiration and ~32%
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
The study sites are members of the AmeriFlux and Fluxnet networks. The research was supported in part by the Office of Science (BReco), U.S. Department of Energy, Grant No. DE-FG02-03Reco63638 and through the Western Regional Center of the National Institute for Global Environmental Change under Cooperative Agreement No. DE-FC02-03Reco63613. Other sources of support included the Kearney Soil Science Foundation, the National Science Foundation, and the Californian Agricultural Experiment Station
References (51)
Is friction velocity the most appropriate scale for correcting nocturnal carbon dioxide fluxes?
Agric. For. Meteorol.
(2009)- et al.
How plant functional-type, weather, seasonal drought, and soil physical properties alter water and energy fluxes of an oak-grass savanna and annual grassland
Agric. For. Meteorol.
(2004) The effect of land cover type and structure on evapotranspiration from agricultural and wetland sites in the Sacramento–San Joaquin River Delta, California
Agric. For. Meteorol.
(2018)- et al.
An approximate analytical model for footprint estimation of scalar fluxes in thermally stratified atmospheric flows
Adv. Water Resour.
(2000) Upscaling fluxes from tower to landscape: overlaying flux footprints on high-resolution (IKONOS) images of vegetation cover
Agric. For. Meteorol.
(2006)A review of approaches for evapotranspiration partitioning
Agric. For. Meteorol.
(2014)A simple and objective method to partition evapotranspiration into transpiration and evaporation at eddy-covariance sites
Agric. For. Meteorol.
(2019)- et al.
Slow ecosystem responses conditionally regulate annual carbon balance over 15 years in Californian oak-grass savanna
Agric. For. Meteorol.
(2016) - et al.
Inter-annual variability in carbon dioxide exchange of an oak/grass savanna and open grassland in California
Agric. For. Meteorol.
(2007) Partitioning forest carbon fluxes with overstory and understory eddy-covariance measurements: a synthesis based on FLUXNET data
Agric. For. Meteorol.
(2007)
Partitioning evapotranspiration with concurrent eddy covariance measurements in a mixed forest
Agric. For. Meteorol.
Partitioning energy and evapo-transpiration above and below a tropical palm canopy
Agric. For. Meteorol.
Partitioning carbon dioxide and water vapor fluxes using correlation analysis
Agric. For. Meteorol.
Transpiration in the global water cycle
Agric. For. Meteorol.
The understory and overstory partitioning of energy and water fluxes in an open canopy, semiarid woodland
Agric. For. Meteorol.
Fluxpart: open source software for partitioning carbon dioxide and water vapor fluxes
Agric. For. Meteorol.
Assessing soil CO2 efflux using continuous measurements of CO2 profiles in soils with small solid-state sensors
Agric. For. Meteorol.
Evapotranspiration partitioning for three agro-ecosystems with contrasting moisture conditions: a comparison of an isotope method and a two-source model calculation
Agric. For. Meteorol.
A comparison of methods for determining forest evapotranspiration and its components: sap-flow, soil water budget, eddy covariance and catchment water balance
Agric. For. Meteorol.
Seasonal variation in carbon dioxide exchange over a Mediterranean annual grassland in California
Agric. For. Meteorol.
Water use efficiency and evapotranspiration partitioning for three typical ecosystems in the Heihe River Basin, northwestern China
Agric. For. Meteorol.
Energy and CO2 flux densities above and below a temperate broad-leaf forest and a boreal pine forest
Tree Physiol.
Trace gas exchange at the floor of a deciduous forest: I. Evaporation and CO2 efflux
J. Geophys. Res., Atmos.
Canopy photosynthesis and water use efficiency in a deciduous forest
J. Appl. Ecol.
Seasonal variation of energy and water vapor exchange rates above and below a boreal jackpine forest
J. Geophys. Res.
Cited by (22)
Dynamics of evapotranspiration from concurrent above- and below-canopy flux measurements in a montane Sierra Nevada forest
2024, Agricultural and Forest MeteorologyEvapotranspiration partitioning based on underlying conductance in a complex tree-grass orchard ecosystem in the humid area of southern China
2024, Agricultural and Forest MeteorologyEvapotranspiration partitioning and agricultural drought quantification with an optical trapezoidal framework
2023, Agricultural and Forest MeteorologyInfluence of disturbance on transpiration and evaporation in tropical peat swamp forests
2023, Journal of HydrologyModeling Weather Effects on Plant Production in the California Annual Grassland
2023, Rangeland Ecology and ManagementDirect partitioning of eddy-covariance water and carbon dioxide fluxes into ground and plant components
2022, Agricultural and Forest Meteorology