Ecosystem-level responses of carbon and energy from a tropical wet forest in Costa Rica

Whether tropical forests are sources, sinks, or neutral with respect to their carbon balance with the atmosphere remains unclear. To address this issue, estimates of net ecosystem exchange of carbon and energy (NEE) were made for 3 years (1998–2000) using the eddy-covariance technique in a tropical wet forest in Costa Rica. Mean daytime NEE was ca. −18 μmol CO₂ m−2 s−1(uptake) and mean nighttime NEE 4.6 μmol CO₂ m−2 s −1 (efflux). However, because ∼80% of the nighttime data in this forest were collected during laminar flow conditions (<0.2 m s −1), nighttime NEE was likely underestimated. Using an alternative analysis, mean nighttime NEE increased to 6.9 μmol CO₂ m −2 s−1. Incident radiation accounted for ∼51% of the variation in the daytime fluxes, with temperature and vapor pressure deficit together accounting for another ∼20%. This forest was a slight negative carbon sink in 1998 (−0.08 to −1.42 t C ha−1 y−1), a moderate sink in 1999 (−1.65 to −3.21 t C ha−1 y−1), and a strong sink in 2000 (−6.1 to −8.1 t C ha−1 y−1 ). This trend is interpreted as relating to the dissipation of warm-phase El Niño effects over the course of this study.; The effects of net radiation (R_n), vapor pressure deficit (VPD), and surface conductances on energy balance and evapotranspiration (ET) were also determined for this forest. Sensible (H) and latent heat (λE) fluxes were estimated as the sum of above canopy eddy-covariance fluxes and changes in below-canopy heat profiles. Albedo was ∼12% of incident radiation and did not differ seasonally. R_n was significantly different among years, explaining ∼79% of the variation in each of the H and λE fluxes. The effects of VPD did not explain any additional variation in heat fluxes. λE was always greater than H (when R_n exceeded 40 W m−2). The dimensionless decoupling coefficient, Ω was always >0.5 and peaked at 0.7, suggesting that ET for this the forest was generally decoupled from physiological controls. There was better precision in estimating λE flux using the Priestly-Taylor model rather than using the more physiologically-based Penman-Monteith model. Annual ET was 54–66% of bulk precipitation and utilized ∼88–97% of available energy (R _n).