Water availability impacts on evapotranspiration partitioning

Authors: Scott, Russell - Russ; Knowles, John; NELSON, J.A. - Max Planck Institute For Biophysical Chemistry; GENTINE, P. - Columbia University; LI, X. - Peking University; Bryant, Ross; Biederman, Joel

Submitted to: Agricultural and Forest Meteorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/16/2020
Publication Date: 2/15/2021
Publication URL: https://handle.nal.usda.gov/10113/7193961
Citation: Scott, R.L., Knowles, J.F., Nelson, J., Gentine, P., Li, X., Bryant, R.B., Biederman, J.A. 2021. Water availability impacts on evapotranspiration partitioning. Agricultural and Forest Meteorology. 297, Article 108251. https://doi.org/10.1016/j.agrformet.2020.108251.
DOI: https://doi.org/10.1016/j.agrformet.2020.108251

Interpretive Summary: Knowing how evapotranspiration (ET) is partitioned into abiotic and biotic pathways is essential to understanding how water affects ecosystem plant growth. The growing availability of ET data as well as methods to partition allows us to better understand how water availability affects ET partitioning. ARS researchers in Tucson, Arizona working with researchers at various universities used four partitioning techniques and long-term (14 years) ET records from regionally co-located grassland and woodland sites that share the same semiarid climate but differ in their water availability. They found that the amount of total plant transpiration (T) to total ET (T/ET) was much smaller and more variable at the grassland where the vegetation does not have access to groundwater. In contrast, T/ET was unimodally distributed at the groundwater-dependent woodland where high T/ET levels were maintained throughout both the dry and wet summer periods. Annual transpiration at both sites closely tracked water availability as quantified by ET, whereas abiotic evaporation was nearly constant. All methods yielded similar patterns of seasonal and interannual partitioned ET fluxes although they differed in their magnitudes by around 5 – 10% in T/ET differences. These results represent novel insights into the effects of water availability on ET partitioning variability that will help to untangle the mechanisms by which water is transported through various ecosystems and their differing effects on terrestrial carbon and water exchange.

Technical Abstract: Knowing how evapotranspiration (ET) is mediated by abiotic and biotic pathways is essential to understanding how water affects ecosystem productivity. Recent synthesis and partitioning-model studies have been used to better understand how the average transpiration fraction (T/ET) changes across sites and biomes, but the temporal variability of the partitioning and its controls are less understood with respect to changing water availability. Here, we examine how the seasonal to interannual variability of ET partitioning is affected by water availability. To do this, we leverage four partitioning techniques and long-term (14 years) eddy-flux records from regionally co-located grassland and woodland sites that share the same semiarid climate but differ in their water availability. T/ET was much smaller and more variable at the grassland where the vegetation does not have access to groundwater. Seasonally, a pronounced dry foresummer T/ET depression at the rainfall-dependent grassland resulted in a bimodal T/ET distribution. In contrast, T/ET was unimodally distributed at the groundwater-dependent woodland where high T/ET levels were maintained throughout both the dry and wet summer periods. Annual transpiration at both sites closely tracked water availability as quantified by ET, whereas abiotic evaporation was nearly constant. All methods yielded similar patterns of seasonal and interannual partitioned ET fluxes although they differed in their magnitudes by around 5 – 10% in T/ET differences. At both sites, transpiration was more closely correlated to plant metrics (photosynthesis, leaf area index) and ET, and evaporation was more correlated to surface moisture variables (soil moisture, precipitation). However, substantial differences between monthly and annual correlations demonstrate variable controls on monthly partitioning throughout the year. These results represent novel insights into the effects of water availability on ET partitioning variability that will help to untangle the mechanisms by which water is transported through various ecosystems and their differing effects on terrestrial carbon and water exchange.