A site selection decision framework for effective kelp restoration

Abstract Highlights Site selection is one of the most important factors for ecosystem restoration success A spatial prioritization framework for application to kelp restoration in California The framework merges kelp metrics derived from in-situ surveys and satellite imagery Site prioritization classification for every kelp forest site in California This framework can be applied to other species and regions with similar datasets We present a decision support framework in the form of a spatially explicit site classification scheme to prioritize locations for conducting kelp restoration. The framework was created for the entire coast of California, where kelp has been lost and restoration projects are increasingly proposed, but the framework is broadly applicable to other coastal habitats or species that are being considered for restoration. We first created spatial distribution models using almost two decades of in situ kelp forest monitoring data and a comprehensive suite of environmental and biological variables, and used the outputs to evaluate the historical stability of kelp forests prior to a marine heatwave (MHW). We then used kelp canopy abundance data derived from satellite imagery to measure the impact of the MHW (i.e. extent of forest loss) and the recent state of kelp forests, including the trend of increase or decrease following the MHW. Finally, we integrated these site-specific kelp metrics to construct a classification tree for prioritizing restoration sites. Outputs of site prioritization are mapped across the study region, readily usable for managers and restoration practitioners with site-specific recommendations for restoration approaches. The framework can be updated due to knowledge of the important predictors of kelp and with new satellite imagery. Further, the framework can be adapted to other species and regions with similar data sets. This regional site selection framework is intended to be used in addition to socio-ecological, socio-economic, and administrative considerations.