Comparison of tree transpiration under wet and dry canopy conditions in a Costa Rican premontane tropical forest

Abstract

Spatial and temporal variation in wet canopy conditions following precipitation events can influence processes such as transpiration and photosynthesis, which can be further enhanced as upper canopy leaves dry more rapidly than the understory following each event. As part of a larger study aimed at improving land surface modelling of evapotranspiration processes in wet tropical forests, we compared transpiration among trees with exposed and shaded crowns under both wet and dry canopy conditions in central Costa Rica, which has an average 4200 mm annual rainfall. Transpiration was estimated for 5 months using 43 sap flux sensors in eight dominant, ten midstory and eight suppressed trees in a mature forest stand surrounding a 40‐m tower equipped with micrometeorological sensors. Dominant trees were 13% of the plot's trees and contributed around 76% to total transpiration at this site, whereas midstory and suppressed trees contributed 18 and 5%, respectively. After accounting for vapour pressure deficit and solar radiation, leaf wetness was a significant driver of sap flux, reducing it by as much as 28%. Under dry conditions, sap flux rates (Js) of dominant trees were similar to midstory trees and were almost double that of suppressed trees. On wet days, all trees had similarly low Js. As expected, semi‐dry conditions (dry upper canopy) led to higher Js in dominant trees than midstory, which had wetter leaves, but semi‐dry conditions only reduced total stand transpiration slightly and did not change the relative proportion of transpiration from dominant and midstory. Therefore, models that better capture forest stand wet–dry canopy dynamics and individual tree water use strategies are needed to improve accuracy of predictions of water recycling over tropical forests. Copyright © 2016 John Wiley & Sons, Ltd.

Upscaling transpiration in diverse forests: Insights from a tropical premontane site

“…In our study, stand transpiration amounted to only 1.4 ± 0.7 mm day −1 due to persistent low radiation, evaporative demand, and frequent wet canopy conditions. These results matched transpiration estimates of Aparecido et al () nearby site, which were 1.4 ± 0.5 mm day −1 over a 12‐month period in 2014.…”

“…Diameters (diameter at breast height) of all 92 trees within the plots were measured (Figure a). To determine active‐xylem sapwood depth, we visually inspected 5‐mm cores with the aid of safranin dye as described in Aparecido et al (). Xylem depth was assessed on each tree equipped with sap flow sensors and an additional 26 trees spanning the full range in diameters.…”

“…It is possible that vertical variation in leaf wetness alters the proportioning of transpiration in the canopy. Despite clear patterns of reduced wetness duration at the top of the canopy, Aparecido et al () did not find large differences in the transpiration response to leaf wetness between trees in the overstory and less exposed midstory (i.e., both proportionally decreased), which suggests plants may have developed adaptations to the frequently wet environment. By contrast, understory trees had a constant rate of water use independently of the canopy wetness condition.…”

“…The influence that leaf wetness exerts on transpiration in TMF canopies is largely unknown (Eller, Burgess, & Oliveira, ; Horna, Schuldt, Brix, & Leuschner, ; O'Brien, Oberbauer, & Clark, ). However, in a recent study, leaf wetness was shown to reduce transpiration by an additional 28%, after accounting for the effects of incident solar radiation ( R s ) and vapour pressure deficit ( VPD ; Aparecido, Miller, Cahill, & Moore, ). Energy in the form of solar radiation is a limiting variable, and evapotranspiration is closely coupled with it in lowland tropical forests across a wide range of tree species (Meinzer, ; Meinzer, James, Goldstein, & Woodruff, ).…”

Foliar water uptake in Amazonian trees: Evidence and consequences

“…The leaf wetness sensors give a continuous millivolt output in response to surface wetness, and typically a clearly defined threshold is selected U c = k fu Ψ surface −Ψ canopy P p Lt y in which the sensor is either wet or dry (Aparecido, Miller, Cahill, & Moore, 2016). The leaf wetness sensors give a continuous millivolt output in response to surface wetness, and typically a clearly defined threshold is selected U c = k fu Ψ surface −Ψ canopy P p Lt y in which the sensor is either wet or dry (Aparecido, Miller, Cahill, & Moore, 2016).…”