Calibration and validation of a simplified process-based model for the prediction of the carbon balance of Scottish Sitka spruce (Picea sitchensis) plantations

There is increasing recognition that forestry provides a low cost and robust means of climate change abatement through carbon sequestration and substitution. However, current understanding of forest ecosystem carbon exchange and forest-atmosphere interactions are often inadequately characterized by existing empirical growth models with resulting poor representation for regional extrapolations. In this paper, we describe the parameterisation and independent validation, against both eddy covariance and forest growth experimental data, of a process-oriented model 3PGN to provide assessments of carbon sequestration of Sitka spruce (Picea sitchensis (Bong.) Carriere) plantations across Scotland. In comparison with eddy covariance measurements, the model predicted all of the major annual carbon fluxes, i.e., gross primary production (P-G), net ecosystem production (P-E), and ecosystem respiration (R-E), with biases lower than 10%. At a monthly time step, only P-G and P-E were accurately estimated, whereas R-E was not. At longer time scales (i.e., several decades), the model reliably represented the major patterns of the carbon balance. Soil type was identified as the important factor influencing site productivity; fertilization practices did not alter long-term site nutritional status. The analyses also highlighted the potential impact of carbon loss from carbon-rich soils, which can result in differences between optimal rotation length for carbon sequestration and for timber production.