Chapter 24 Automated Predictive Mapping of Ecological Entities

Publisher Summary Ecological land classification is a widely practiced activity that partitions space into successively smaller and more homogeneous spatial entities. This chapter uses the U.S. National Hierarchical Framework of Ecological Units (NHFEU) as an example to illustrate the size and scale at which different levels in a hierarchical classification operate and the principal criteria used to differentiate ecological units at various scales. This example was selected to emphasize the importance placed on landform, topography, slope gradient, aspect, and slope position as key criteria for differentiating the lower level units in the system. The chapter describes a case study based on procedures used to produce predictive ecosystem maps (PEM) at a scale of 1:20,000. These maps cover over 3.5 million ha of the 8.5 million ha of forested land in the former Cariboo Forest Region of British Columbia, Canada. For consistency and comparison, the methods and results illustrated in the chapter use the small Baranja Hill data-set. The landscape-based approach to ecological land classification is a form of pattern recognition, based on ecological theory. The unit areas delineated represents hypotheses that arise from expert heuristic knowledge of what has been deemed ecologically important for land management. The methods described in the chapter comprise a hybrid of automated, semi-automated, and manual procedures that develop and apply heuristic, rule-based conceptual models of ecological—landform and soil—landform relationships in a manner similar to the CLORPT or SCORPAN approaches. The procedures attempt to directly parallel, or mimic, the logic and decision-making process followed by local ecological experts.