Postprint of: Skovsgaard, JP & Vanclay, JK 2013, 'Forest site productivity: a review of spatial and temporal variability in natural site conditions', Forestry, vol. 86, no. 3, pp. 305-315.
Indicators of forest site productivity may exhibit considerable spatial and temporal variability that should be considered in sustainable forest management. It is generally assumed that natural site conditions and, in turn, site productivity changes gradually and predictably. Our review illustrates many exceptions to this paradigm. Consequently, uni-dimensional productivity indicators such as the commonly used site index (estimated based on stand height) is not always sufficient to characterize site productivity for apparently homogeneous forest stands. To alleviate this problem, we suggest a hierarchical procedure for the estimation of forest site productivity including site mapping, unthinned reference stands (against which to measure growth performance) and adaptive modelling. The level and detail at which site mapping should be conducted (region, forest, management unit or subunit), depends on the objective (research vs. operational forestry), forest type and expected deviations in site productivity estimates compared with the cost of site mapping. Unthinned reference plots should preferably be maintained in the long term and the number of plots should increase with increasing site or stand heterogeneity (for homogeneous land we recommend one plot in 10 ha, two in 100 ha, etc.). With adaptive modelling site specific parameters can be updated at any time when new information has become available. Finally, the review indicates a need to re-define traditional measurement procedures to achieve a contemporary and rational statistical basis for the estimation of site productivity.