Investigating the resilience of kelp forests to harvest amid a rapidly changing ocean

While kelp-related activities have become a major component of the global blue economy, the resilience of kelp forests to forest-scale harvest remains understudied. To determine how kelp harvest affects ecological and biophysical characteristics of an entire kelp forest, we conducted a seascape-scale harvest experiment of giant kelp (Macrocystis pyrifera). In collaboration with Indigenous kelp harvesters, knowledge holders, and leaders from the Kwakiutl First Nation, in addition to government and university scientists, we established harvested and unharvested control sections in three large kelp forests which ranged from 6.91 to 25.62 hectares. Harvesters then decided which kelp forest to harvest based on a 20% harvest quota, sea condition, kelp condition, and travel time. While harvesters consistently removed an average of 7100 +/- 325 kg of kelp from the top 5.81 +/- 2.17m of the forest canopy per harvest event, overall removing <10% of surface canopy over the season, the occurrence, timing, and magnitude of kelp harvest varied across each experimental site. Post-harvest, we detected a significant reduction in surface frond density at one of the three kelp forests that experienced early season and sequential press harvest effort. Moreover, we detected significant yet variable and ephemeral responses in benthic light intensity, seawater temperature, and flow to sequential harvest at two experimental kelp forests. Counter to our predictions, benthic light intensity within one kelp forest dropped following harvest likely due to the accumulation of kelp detritus. While we did not detect an effect of a single pulse harvest event on benthic light intensity or seawater flow, we did detect a significant cooling in seawater temperature at one of two harvested sections of kelp forest following sequential harvest. Bryozoan coverage was found to be significantly impacted by kelp harvest, however the seasonality and directionality of those impacts varied between kelp forests. Lastly, we did not detect an effect of harvest on individual kelp reproduction or new frond growth. Our results underscore the resilience, yet context-dependence, of kelp forests to harvest. More broadly, the co-design, co-production, and co-implementation of this experiment, conducted at ecologically and socially relevant scales, models an equitable way to inform a more resilient and just blue economy.