Recipient Organization
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON,WI 53715-1218
Performing Department
Biological Systems Engineering
Non Technical Summary
Every year, nearly 70 thousand cubic kilometers of water leave terrestrial ecosystems and enter the atmosphere via evapotranspiration (ET), which is one of the most critical fluxes in the global water cycle. Despite its importance, we are unsure whether global ET is increasing over time following the notion that the water cycle is accelerating, or decreasing and causing more river discharge. Land surface models represent our understanding of the terrestrial water cycle yet struggle to simulate the seasonal course of ET and its magnitude. This suggests that we need to improve our fundamental understanding of the processes that control ET.ET is made up of two different terms: transpiration (T) through plant stomata and evaporation (E) from soil and other non-stomatal surfaces. T and E respond differently to ongoing changes in atmospheric composition and climate. Plants for example respond to water scarcity by closing the stomata on their leaves to save water. This results in a decrease in T, which decreases evaporative cooling, increases surface temperature, and decreases plant carbon uptake to which the water cycle is coupled. T is therefore the fundamental variable to measure for understanding plant water stress and is a function of the physical factors that drive evaporation into the atmosphere as well as biological controls over these fluxes. Understanding how forested ecosystems regulate their water supplies by controlling T gives us insight into the amount of water that is available for groundwater recharge and runoff. T and E are relatively difficult to measure at the ecosystem scale compared with total ET, which is readily measured using a technique called eddy covariance. Rapid advancement in our understanding of the theory underlying surface-atmosphere water flux make it possible for the first time to measure T and E independently at the ecosystem scale at hourly to interannual time scales using eddy covariance. The central objective of this proposal is quantify T and E from eddy covariance ET observations across characteristic Wisconsin forested ecosystems using advanced techniques and to compare these observations against estimates from satellite remote sensing to improve our understanding of water cycling across space and time. This project will complete the first ever large-scale validation of T and E from remote sensing datasets and improve our understanding of how Wisconsin forests use water resources.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
Land surface models represent our understanding of the terrestrial water cycle yet struggle to simulate the seasonal course of evapotranspiration (ET) and its magnitude. This suggests that we need to improve our fundamental understanding of the processes that control ET. The objectives of the proposed project are: 1) to quantify the amount of water that enters the atmosphere through transpiration and evaporation in Wisconsin forested ecosystems across the vegetative growing season to fall transition and 2) to compare these observations to satellite remote sensing products to further improve them and thereby improve our understanding of the terrestrial water cycle. The central questions that we ask are: how do different forest management practices impact T and E and do remote sensing algorithms accurately predict T and E from forested ecosystems? If not, how can they best be improved?
Project Methods
A number of recent methods have been proposed to partition transpiration (T) and evaporation (E) from eddy covariance-based evapotranspiration (ET) measurements (Stoy et al., 2019). These products often agree with independent measurements of T, but often rely on assumptions about ecosystem functioning and plant water use that cannot always be justified. Of these methods, Scanlon and Kustas (2010) developed the 'flux variance similarity' (FVS) partitioning approach for T and E partitioning based on the notion that atmospheric eddies transporting CO2 and water vapor from stomatal processes (T and photosynthesis) and non-stomatal processes (E and respiration) independently follow flux variance similarity as predicted by Monin Obukhov Similarity Theory. In brief, there are two end-member scenarios for a parcel of air transported from a surface: one without stomata, and one with stomata. An eddy transported away from a surface that is respiring CO2 and evaporating water through pathways other than stomata will have deviations from mean CO2 concentration (c') and water vapor concentration (q') that are positively correlated. An eddy of air that interacts with a surface with stomata will have a negative relationship between c' and q' that is determined by a unique canopy water use efficiency (WUE). WUE can thus be used to establish a functional relationship between the variance of CO2 due to stomatal uptake (σcr) and the correlation between stomatal and non-stomatal CO2 exchange processes (ρcp, cr). ET can subsequently be partitioned into its T and E components by matching the observed correlation of q' and c' (ρq,c) to the corresponding value of ρcp,cr (Scanlon and Sahu 2008). Subsequent work on the FVS method by Skaggs et al. (2018) noted that there is an algebraic solution to terms that had previously been solved using optimization (namely σcr and ρcp,cr, Palatella et al. 2014) and created a Python module, fluxpart, to calculate T and E using the FVS approach. The original FVS approach of Scanlon et al. (2010) used a simple WUE formulation following Campbell and Norman (1998). Fluxpart originally allowed WUE to take a constant value or vary as a function of atmospheric vapor deficit (Skaggs et al., 2018) and now calculates WUE directly using optimization (Scanlon et al. 2019). We will proceed with the Skaggs et al. (2018) FVS approach with Scanlon et al. (2019) WUE calculation algorithm for this project given that they represent a technical improvement over the original algorithm (Scanlon et al., 2010). The original FVS approach was shown to realistically reproduce T/ET relationships over the growing period of a corn (maize) crop (Scanlon and Kustas 2012). The original FVS approach and fluxpart also give realistic interpretations of T/ET over dryland wheat fields in Montana (Stoy et al., 2019). FVS currently lacks an algorithm to calculate uncertainty in T and E outputs. Uncertainty estimates of the T and E calculations are necessary for a conservative interpretation of their values and to compare them against independent estimates from remote sensing. We will do so using a Monte Carlo approach following Stoy et al. (2006) for eddy covariance ET calculations that incorporates the sensitivity of T/ET values to the measured uncertainty of the parameters that enter the FVS calculation routine, particularly the WUE calculation that T and E estimates are sensitive to (Scanlon et al., 2019). We will then calculate T and E over the growing season for all 19 eddy covariance towers in CHEESEHEAD19, as well as their associated uncertainty. Doing so will help us understand how forest management practices impact hydrology; notice for example that the CHEESEHEAD19 database includes recent cut, middle aged, and mature aspen stands as well as different hardwood and softwood stands including a red pine plantation. By covering representative forest management strategies of northern Wisconsin, CHEESEHEAD19 gives us the opportunity to understand how land use decisions impact water cycling and water use by vegetation. Risks related to the availability of eddy covariance data are minimized by the robust existing CHEESEHEAD19 dataset that we propose to use here. Eddy covariance measurements have been collected from 19 towers, including 2 administered by PI Stoy, from June until October 2019 with excellent data coverage. The extensive calibration/validation network for the ECOSTRESS ET product from PT-JPL and existing observations during the CHEESEHEAD19 measurement period make it reasonable to assume that a full landscape-scale estimate of T and E across multiple months with uncertainty can be derived, and that this approach can be used to understand water use by Wisconsin forests. We therefore anticipate even if data collection or provision were to stop during the project timeframe, science goals of this work would still be achievable. References:Campbell GS, Norman JM (1998) An Introduction to Environmental Biophysics, 2nd edn. Springer, New YorkPalatella L, Rana G, Vitale D (2014) Towards a flux-partitioning procedure based on the direct use of high-frequency eddy-covariance data. Agricultural and Forest Meteorology 1-11Scanlon TM, Kustas WP (2010) Partitioning carbon dioxide and water vapor fluxes using correlation analysis. Agricultural and Forest Meteorology 150:89-99Scanlon TM, Kustas WP (2012) Partitioning evapotranspiration using an eddy covariance-based technique: Improved assessment of soil moisture and land-atmosphere exchange dynamics. Vadose Zone Journal 11:vzj2012.0025. https://doi.org/10.2136/vzj2012.0025Scanlon TM, Sahu P (2008) On the correlation structure of water vapor and carbon dioxide in the atmospheric surface layer: A basis for flux partitioning. Water Resources Research 44:W10418Scanlon TM, Schmidt DF, Skaggs TH (2019) Correlation-based flux partitioning of water vapor and carbon dioxide fluxes: Method simplification and estimation of canopy water use efficiency. Agricultural and Forest Meteorology 279:107732Skaggs T, Anderson R, Alfieri J, et al (2018) Fluxpart: Open source software for partitioning carbon dioxide and water vapor fluxes. Agricultural and Forest Meteorology 253:218-224Stoy PC, El-Madany T, Fisher JB, et al (2019) Reviews and syntheses: Turning the challenges of partitioning ecosystem evaporation and transpiration into opportunities. Biogeosciences 16:3747-3775Stoy PC, Katul GG, Siqueira MBS, et al (2006) Separating the effects of climate and vegetation on evapotranspiration along a successional chronosequence in the southeastern U.S. Global Change Biology 12:2115-2135