Camo-maps: An efficient method to assess and project riparian vegetation colonization after a major river flood

Abstract

Major river floods provide powerful physical disturbances that create and expand gravel bars and islands. Their barren surfaces are suitable for seedling colonization by plants including poplars (cottonwoods) and willows that grow to contribute rich wildlife habitats and other ecosystem services. Conversely, in locations such as through bridges, riparian woodlands would impede future flows, elevating flood stages and increasing damage. It is consequently useful to understand where opportunity versus hazard locations occur and this encourages monitoring and predictive modeling of vegetation colonization. Colonization is related to the river inundation patterns but hydrodynamic modeling is difficult for complex channels and irregular features. Here we present an alternative empirical approach based on mapping of actual inundation patterns from a sequence of aerial photographs at different river flows. Based on field inventory of seedling distributions of different plants, we estimated inundation thresholds for six progressively lower riparian cover types: woodland, shrubland, perennial herbaceous, ruderal annuals, transition and barren. These were coordinated with the inundation patterns, enabling predictive mapping of the different vegetation classes to produce ‘camo-maps’. These resemble camouflage with the irregular shapes and classification colors matching the vegetation types. The derived camo-maps were then compared with a lidar-based mapping of topography and vegetation structure. As an illustrative case study, we present the method for a large and complex gravel bar upstream from a bridge sequence and an expanded island below another bridge in Calgary Canada, where a hundred-year flood in June 2013 caused severe damage. The camo-maps reasonably represented the actual colonization patterns, although low flow years enabled some vegetation survival at lower positions than predicted. We present this as an effective and efficient empirical method to provide an alternative or complement to riparian vegetation projections based on hydrodynamic modeling.