Modelling tree stem‐water dynamics over an Amazonian rainforest. Issue 1 (10th December 2019)

• Record Type: Journal Article
• Authors: Yan, Binyan; Mao, Jiafu; Dickinson, Robert E.; Thornton, Peter E.; Shi, Xiaoying; Ricciuto, Daniel M.; Warren, Jeffrey M.; Hoffman, Forrest M.
• Is Part Of: Ecohydrology. Volume 13:Issue 1(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-12-10
• Subjects: Amazonian rainforest -- Community Land Model -- drought -- tree stem‐water model; Ecohydrology -- Periodicals; Hydrology -- Periodicals; Water -- Environmental aspects -- Periodicals; 577.6
• DOI: 10.1002/eco.2180
• Languages: English
• ISSNs: 1936-0584

Abstract

Abstract: A novel tree stem‐water model was developed to capture the dynamics of stem‐water storage and its contribution to daily transpiration. The module was incorporated into the Community Land Model (CLM), where it was used to test model sensitivity to stem‐water content for an evergreen rainforest site in Amazonia, that is, the BR‐Sa3 eddy covariance site. With the inclusion of the stem‐water storage, CLM produced greater dry‐season latent heat flux that was closer to observations, facilitated by easier canopy access to a nearby stem‐water source, rather than solely dependent on soil water. The simulated stem‐water content also showed seasonal variations in magnitude, along with the seasonal variations in sap flow rate. Stored stem‐water of a single mature tree was estimated to contribute 20–80 kg/day of water to transpiration during the wet season and 90–110 kg/day during the dry season, thereby partially replacing soil water and maintaining plant transpiration during the dry season. Diurnally, stem‐water content declined as water was extracted for transpiration in the morning and then was refilled from soil water beginning in the afternoon and through the night. The dynamic discharge and recharge of stem storage was also shown to be regulated by multiple environmental drivers. Our study indicates that the inclusion of stem capacitance in CLM significantly improves model simulations of dry‐season water and heat fluxes, in terms of both magnitude and timing.