This paper discusses the concept of spatial stratification (SS) as applied to forest modelling in general, and spatial forest modelling in particular. SS is a way of providing a geographically explicit forest description in forest modelling, or a way of accommodating spatially explicit management objectives and interventions. In the former, called a priori SS, stands of a forest landscape are spatially aggregated into a set of stand clusters which become input to forest modelling. The latter, called dynamic SS, utilizes stands as the input forest description upon which various spatial aggregations occur throughout forest modelling. Distinctions between the two alternative approaches are highlighted and implementation considerations are examined within the forest landscape management design context. The paper concludes that: (i) modelling techniques are directly linked to forest stratification approaches; and (ii) a priori stratification is seriously limited in spatial modelling for landscape management design where multiple and often conflicting spatial objectives exist. In view of these findings, the paper outlines an alternative spatial forest modelling approach using a combination of dynamic SS and heuristic optimization. Key words: spatial stratification, spatial forest modelling, heuristic optimization, forest management
Tamm Review: Ecological principles to guide post-fire forest landscape management in the Inland Pacific and Northern Rocky Mountain regions
