Uncertainty in carbon exchange modelling in a forest canopy due to kB−1 parametrizations


  • J. Hong,

    Corresponding author
    1. National Institute for Mathematical Sciences, Daejeon, South Korea
    • National Institute for Mathematical Sciences, Daejeon, Korea 305-811.
    Search for more papers by this author
  • J. Kim,

    1. Department of Landscape Architecture and Rural System Engineering, Seoul National University, South Korea
    Search for more papers by this author
  • Y. H. Byun

    1. National Institute of Meteorological Research, Seoul, South Korea
    2. Met Office Hadley Center, Exeter, UK
    Search for more papers by this author
    • The contribution of this author was written in the course of his employment at the Met Office, UK, and is published with the permission of the Controller of HMSO and the Queen's Printer for Scotland.


The difference in turbulent transfer efficiency between momentum and scalars, represented by kB−1, has been the subject of considerable interest in micrometeorology, and various parametrizations have been proposed to address this issue. The simple kB−1 parametrizations that are based on either empirical formulations or K-theory are still popularly used in the land surface models despite their theoretical deficiencies. Moreover, the impact of the uncertainty in this parameter on modelling surface carbon exchange has not been previously estimated. In this study, we examined the uncertainties of the simulated gross primary productivity (GPP), ecosystem respiration (RE), and net ecosystem exchange (NEE ≡ RE–GPP) for a forest canopy due to kB−1 parametrizations. The tested parametrizations included not only the schemes that set kB−1 as a constant or as a function of only the Reynolds number (popularly used in the land surface model), but also the schemes that express kB−1 as a function of plant phenology derived from the Lagrangian theory. Except for parametrizations that produced aerodynamic resistance as large as canopy resistance over a tall forest canopy, the variabilities of GPP and RE induced by kB−1 parametrizations were less than 2% of the annual GPP and RE, respectively. Nevertheless, the model produced approximately 10% variability of NEE values with changes in the kB−1 parametrizations (203±24 g C m−2year−1) because the simulated RE was less sensitive to the kB−1 parametrizations than the simulated GPP due to the negative feedback among kB−1, temperature, and RE. Our findings reveal that kB−1 parametrizations should be suitably applied in land surface modelling for better simulation of the global carbon cycle. Copyright © 2011 Royal Meteorological Society and British Crown Copyright, the Met Office