Riparian management and restoration are adaptive processes that should include feedback from monitoring. On top of budget considerations, monitoring must be done quickly to allow agile responses to changes in field conditions such as invasion by new species, changing regulatory requirements, and climate change. The use of high-resolution aerial imagery – from satellites or drones-- may solve some problems of monitoring cost, speed, and large-scale coverage. To be effective, the interpretation of aerial images depends on ground-level observations for learning and verification. In many areas, particularly along small streams that have relatively narrow riparian extents, vegetation communities have not been defined well enough to meet the needs for monitoring with aerial imagery. In 2019, we began classifying previously undefined riparian vegetation associations at the University of Idaho’s Rinker Rock Creek Ranch as part of a larger project to guide riparian restoration and livestock management in riparian areas. Riparian plant communities were first roughly identified from publicly available aerial imagery, then searched by small teams of botanists and ecologists. Square relevé plots (8m x 8m) were subjectively placed and divided into nested subplots to begin assessing species-area relationships.
Results/Conclusions
We described riparian vegetation types in this area that will be used in conjunction with rapid field methods informing the use of drone imagery for large-scale riparian assessment and monitoring. We identified six riparian plant community types: meadow, low wet meadow, streambank, willow shrubland, aspen woodland, and riparian transition. Communities grouped well in NMDS, based on percent cover or presence/absence. Meadow and low-meadow types overlapped and may be considered as a single community for evaluation from aerial images. Relevé plots of 64 square meters generally captured most of the species in the areas of the plots, but plots in different drainages were needed to sample most of the total species. Rapid methods we incorporated hold promise for supporting monitoring with drone imagery. Future analyses will explore the functional attributes of these riparian community types and the ability of drone imagery to successfully discriminate among community types.