Simple Metrics of Ecological Connectivity for Application to the Design of Marine Protected Areas

While there are plenty of advanced tools to measure ecological connectivity, outputs from these approaches rarely translate to conservation measures. This disconnect is likely due to geographic bias for connectivity research, and a lack of translation of connectivity outputs to useable metrics for practitioners. In this study, we use a conservation-driven approach to estimate ecological connectivity patterns along the eastern Atlantic coast of Nova Scotia, Canada, a proposed marine protected area. Kelp beds are an important coastal feature along this coastline, however, our knowledge of their spatial distribution is limited. The objectives of this research are to: 1. predict the spatial distribution of kelp beds along the eastern shore of Nova Scotia and 2. compare simple metrics of ecological connectivity across the kelp beds in the region. We collated all presence and absence data for kelp beds in the study region (492 observations) and identified subpopulations for further connectivity calculations using a random forest model with depth, slope, distance from shore (Euclidean distance), bottom type, curvature, and bathymetric position index as the environmental predictors. Preliminary outputs from the random forest model indicate that bottom type, Euclidean distance and slope are the 3 most important environmental predictors using conditional variable importance. The resulting model was able to correctly classify presence/absence with an out of bag error rate of 37.2%. We used progressive vector plots, from both empirical measurements using an acoustic doppler current profiler (ADCP), and data from the finite volume community ocean model (FVCOM) to calculate local dispersal patterns of dominant kelp species, Asterias spp., Cancer irroratus, and Littorina spp. among kelp habitat patches. Data on ocean currents will also be used to calculate size and spacing based on a decision tree developed for management application. The goal of this research is to advance methods for the inclusion of ecological connectivity in the design and management of marine protected areas.