As the value of ecosystem-based management (EBM) approaches is increasingly recognized in marine ecosystems, it is critical that the impacts of resource harvest are assessed at various spatial scales. This is particularly true for habitat-forming resources, such as wild seaweeds, that act as foundation species by physically structuring ecosystems. The impacts of spatially heterogeneous harvest may change with scale and have different management implications based on the ecosystem process or organism under consideration. Ascophyllum nodosum (hereafter rockweed) is a canopy-forming fucoid seaweed endemic to rocky coastlines in the North Atlantic Ocean that has been harvested for centuries. We conducted a Before-After Control-Impact study of commercial rockweed harvest at 38 sites across the coast of Maine (USA) from 2018 to 2020 in an effort to understand impact and one-year recovery of two rockweed bed structural characteristics, height and biomass, at a scale similar to a single harvest event. Our results indicate that rockweed harvest is spatially heterogeneous at the scale of the rockweed bed, and as a result, the effect sizes of harvest at this scale are smaller than those reported in previous studies that assessed smaller spatial scales. Mean rockweed biomass recovered to pre-harvest values after one year of recovery, but mean rockweed height remained lower at impacted sites. While post-harvest recovery was generally high in our study, sites that experienced higher intensities of harvest were less likely to fully recover height or biomass one year post-harvest. Our findings provide resource managers with a bed-scale perspective that can inform EBM approaches, particularly for population-level management of harvested resources and impacts of harvest on highly mobile organisms—such as birds and fish—that interact with these ecosystems at larger spatial scales.

All data were collected in the field between 2018 and 2020 in Maine (USA).

All datasets were cleaned and organized in program R.

Datasets are .csv files and can be opened in any text editor. Code for analyses can be opened in program R.