Measures of structural complexity in digital images for monitoring the ecological signature of an old-growth forest ecosystem

Abstract

Conducting field samples for monitoring ecological dynamics across multiple spatiotemporal scales is a difficult task using standard protocols. One alternative is to measure a restricted set of variables which can serve as an ecological orientor (EO) for quantifying habitat change. The objective of this article is to derive from digital images a measure of structural complexity that may serve as a proximate EO for monitoring forest dynamics in space and time. The mean information gain (MIG) index was used as a measure of structural complexity in photographs taken directly in the field over the entire growing season. At a small scene extent, the complexity of light intensity variations in digital images was positively related to species richness. At larger scene extents, forest understorey and overstorey layers showed predictable ecological signatures in structural complexity. In general, intensity and chroma were the two color space components which yielded the greatest sensitivity to habitat change through time. Within the framework of a standardized photographic protocol, it seems therefore reasonable to consider MIG as a suitable EO for monitoring forest dynamics in both space and time. Our results support the idea that it is possible on one hand to adopt a more holistic view of ecological processes to gain, on the other hand, spatial and temporal degrees of freedom for testing multiple scale hypotheses in the field.