Data for Next Generation Earth System Models: Profiles of Micrometeorological Variables in a Tropical Montane Rainforest Canopy

Abstract:

The complexity of the terrestrial processes represented in Earth system models is rapidly increasing, particularly with regards to tropical ecosystems. New data is needed to inform these models, which are beginning to implement a multi-level canopy approach in order to represent vertical variation in biophysical and micrometeorological responses. In this study, we measure an array of such variables at different heights throughout a tropical montane rainforest in Costa Rica over a year-long dataset collected from August 2014 to July 2015. The site, at the Texas A&M Soltis Center for Research and Education, is located on the Caribbean slope of the Cordillera Tilarán Mountains, in the transitional zone between lowland rainforest and montane cloud forest. The site receives MAP of 4200 mm; most occurs from May through December, leaving a relatively dry season from January to April. Ten minute average values of ρ_H2O, ρ_CO2, temperature, and VPD were measured at eight heights within and above the canopy while PAR and leaf wetness were measured at five.

The forest’s light regime was strongly influenced by both the occurrence of emergent trees and by the hillslope’s east-northeast aspect which shaded the lower canopy all afternoon. Carbon dioxide displayed a strong diurnal pattern, with a distinct increase in concentrations at the lowest, 1-m measurement level. Water vapor concentrations had a more complex pattern, indicating a midday peak in plant transpiration followed by an evening peak in evaporation due to afternoon storms (see figure). Combined, monthly variations in PAR, CO₂, and VPD profiles suggested that stomatal control over water loss was occurring during the dry months. Overall, precipitation events, and the wet leaf surfaces and cloud cover associated with them, were found to affect nearly every aspect of the forest’s micrometeorology, suggesting that significant errors in precipitation modeling in tropical regions could change the magnitude and direction of trace gas fluxes.