On the variability of the ecosystem response to elevated atmospheric CO2 across spatial and temporal scales at the Duke Forest FACE experiment

• Published in: Agricultural and forest meteorology 2017-01, Vol.232 (C), p.367-383
• Main Authors: Paschalis, Athanasios, Katul, Gabriel G., Fatichi, Simone, Palmroth, Sari, Way, Danielle
• Format: Article
• Language: English
• Subjects: Ecohydrological modeling; Elevated CO2; FACE; Spatio-temporal variability; Stomatal conductance model
• ISSN: 0168-1923; 1873-2240
• DOI: 10.1016/j.agrformet.2016.09.003

•Novel analysis of 10 years of leaf level data at the Duke FACE site.•Leaf-level flux variability depends on meteorological and internal ecosystem variability.•Meteorological variability highly impacts fine temporal scales.•Spatial ecosystem variability impacts coarser scales.•Upscaling of leaf level fluxes to the ecosystem scale is non-trivial. While the significance of elevated atmospheric CO2 concentration on instantaneous leaf-level processes such as photosynthesis and transpiration is rarely disputed, its integrated effect at ecosystem level and at long-time scales remains a subject of debate. In part, the uncertainty stems from the inherent leaf-to-leaf variability in gas exchange rates. By combining 10 years of leaf gas exchange measurements collected during the Duke Forest Free Air CO2 Enrichment (FACE) experiment and three different leaf-scale stomatal conductance models, the leaf-to-leaf variability in photosynthetic and stomatal conductance properties is examined. How this variability is then reflected in ecosystem water vapor and carbon dioxide fluxes is explored by scaling up the leaf-level process to the canopy using model calculations. The main results are: (a) the space-time variability of the photosynthesis and stomatal conductance response is considerable as expected. (b) Variability of the calculated leaf level fluxes is dependent on both the meteorological drivers and differences in leaf age, position within the canopy, nitrogen and CO2 fertilization, which can be accommodated in model parameters. (c) Meteorological variability is playing the dominant role at short temporal scales while parameter variability is significant at longer temporal scales. (d) Leaf level results do not necessarily translate to similar ecosystem level responses due to indirect effects and other compensatory mechanisms related to long-term vegetation dynamics and ecosystem water balance.