A novel approach to partitioning evapotranspiration into evaporation and transpiration in flooded ecosystems. (13th November 2021)

• Record Type: Journal Article
• Authors: Eichelmann, Elke; Mantoani, Mauricio C.; Chamberlain, Samuel D.; Hemes, Kyle S.; Oikawa, Patricia Y.; Szutu, Daphne; Valach, Alex; Verfaillie, Joseph; Baldocchi, Dennis D.
• Is Part Of: Global change biology. Volume 28:Number 3(2022)
• Page Start: 990
• Page End: 1007
• Publication Date: 2021-11-13
• Subjects: artificial neural networks -- eddy covariance -- latent energy -- machine learning -- terrestrial water cycle -- vapor pressure deficit -- wetlands; Climatic changes -- Environmental aspects -- Periodicals; Troposphere -- Environmental aspects -- Periodicals; Biodiversity conservation -- Periodicals; Eutrophication -- Periodicals; 551.5
• DOI: 10.1111/gcb.15974
• Languages: English
• ISSNs: 1354-1013

Abstract

Abstract: Reliable partitioning of micrometeorologically measured evapotranspiration (ET) into evaporation ( E ) and transpiration ( T ) would greatly enhance our understanding of the water cycle and its response to climate change related shifts in local‐to‐regional climate conditions and rising global levels of vapor pressure deficit (VPD). While some methods on ET partitioning have been developed, their underlying assumptions make them difficult to apply more generally, especially in sites with large contributions of E . Here, we report a novel ET partitioning method using artificial neural networks (ANNs) in combination with a range of environmental input variables to predict daytime E from nighttime ET measurements. The study uses eddy covariance data from four restored wetlands in the Sacramento‐San Joaquin Delta, California, USA, as well as leaf‐level T data for validation. The four wetlands vary in their vegetation make‐up and structure, representing a range of ET conditions. The ANNs were built with increasing complexity by adding the input variable that resulted in the next highest average value of model testing R 2 across all sites. The order of variable inclusion (and importance) was: VPD > gap‐filled sensible heat flux ( H _gf) > air temperature ( T air ) > friction velocity ( u ∗ ) > other variables. The model using VPD, H _gf, T air, and u ∗ showed the best performance during validation with independent data and had a mean testing R 2 value of 0.853 (averaged across all sites, range from 0.728 to 0.910). In comparison to other methods, our ANN method generated T /ET partitioning results which were more consistent with CO2 exchange data especially for more heterogeneous sites with large E contributions. Our method improves the understanding of T /ET partitioning. While it may be particularly suited to flooded ecosystems, it can also improve T /ET partitioning in other systems, increasing our knowledge of the global water cycle and ecosystem functioning. Abstract : We currently lack evapotranspiration partitioning methods that can be applied reliably in more complex evaporation dominated ecosystems and without the use of carbon exchange data. In this manuscript we partition evapotranspiration measured via eddy covariance into its components of evaporation and transpiration using machine learning methods, specifically artificial neural networks (ANNs), in flooded wetland sites in the Sacramento‐San Joaquin River Delta, California, USA. The ANN based partitioning method produces reliable and robust evaporation and transpiration estimates for the wetland sites, providing a new method to increase our understanding of ecosystem water exchange and the global water cycle.