Journal of Experimental Marine Biology and Ecology
Volume 561, April 2023, 151869
Bed-scale impact and recovery of a commercially important intertidal seaweed
Elliot M. Johnston a, Hannah N. Mittelstaedt a, Laura A. Braun a, Jessica F. Muhlin b, Brian J. Olsen a, Hannah M. Webber ac, Amanda J. Klemmer a
https://doi.org/10.1016/j.jembe.2023.151869
Abstract
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.
Introduction
The harvest and recovery of seaweeds (macroalgae) has been studied widely across littoral zones and latitudes (Keser et al., 1981; Mafra Jr. and Cunha, 2006; Ulaski et al., 2020; Westermeier et al., 2019). There is high interest in seaweeds’ response to harvest given their ecological role as foundation species (Dudgeon and Petraitis, 2005; Schmidt et al., 2011) and economic value for coastal communities (Rebours et al., 2014). Many seaweed species exhibit relatively fast growth rates (Mafra Jr. and Cunha, 2006; Reed et al., 2008), and the ability of some to recover biomass a few years after harvest is often highlighted in examples of sustainable fisheries (e.g., Marquez et al., 2014; Vea and Ask, 2011). However, there has been a paradigm shift toward ecosystem-based management (EBM) in recent decades—particularly in marine ecosystems—as resource managers recognize the importance of maintaining not only the population sizes of targeted resources, but also trophic interactions throughout associated food webs (Arkema et al., 2006; Pikitch et al., 2004).
Assessing the food-web impacts of commercial seaweed harvest requires that the scale of inquiry overlaps with the scales of both human harvest and the food web itself. The proportion of harvested individuals can change across spatial scales and result in variable impacts to habitat quality depending on the home range size and movement type of the organism under consideration (Grindal and Brigham, 1999; Leonard et al., 2008). Along many North Atlantic rocky intertidal zones, Ascophyllum nodosum (hereafter rockweed)—a canopy-forming fucoid seaweed—dominates the intertidal seaweed assemblage (Guiry and Morrison, 2013; Vadas et al., 2004). Rockweed has experienced a rise in global harvest over the past several decades and, as a result, there have been concerns about the ecological sustainability of harvesting practices (Lotze et al., 2019; Seeley and Schlesinger, 2012). Some jurisdictions report rockweed harvest by total landed weight or percent biomass removed across a landscape-scale area on the order of tens of square kilometers (i.e., sector; Ugarte and Sharp, 2012), yet these metrics lack context at the site-level scale over which a single harvest event occurs and some highly mobile organisms—such as birds and fish—utilize the ecosystem. Furthermore, impacts associated with small, intensely harvested patches (e.g., several m2; Walder, 2015) may be different than those associated with rockweed beds that experience spatially heterogeneous harvest typical of commercial practices (DFO, 1999). Quantifying harvest impacts at the harvest-site scale (i.e., hundreds to thousands of square meters) will allow EBM approaches to simultaneously assess population-level management of the harvested resource and the impacts of commercial harvest on mobile organisms that interact with these ecosystems at relatively large spatial scales.
In addition to assessing harvest impacts at different spatial scales, resource managers must consider the representativeness of research protocols to commercial harvest in their jurisdiction. For instance, some studies use harvest intensities that cut rockweed fronds below the legal harvest height in the area of study (Keser et al., 1981; Black and Miller, 1991; Fegley, 2001). Studies may also create uniformly harvested treatment plots with hand shears that are not typical of patchy commercial harvest (Keser et al., 1981; Fegley, 2001; Walder, 2015). Recovery thresholds may be crossed at these high harvest intensities, but species may show greater resilience at lower harvest intensities (Keser et al., 1981). As a result, harvest regulations informed by research that lacks representativeness to commercial practices can lead to uncertainty in the outcomes of management decisions.
Our objective was to design an experiment to assess the effects of commercial rockweed harvest by measuring impact and recovery at the same spatial scale as the harvest. We conducted a Before-After Control-Impact (BACI) experiment using commercial rockweed harvest practices at 38 sites across the coast of Maine (USA; Fig. 1). To assess impact and recovery, we measured mean bed height and biomass during three time periods: pre-harvest (2018), harvested (2019), and one year post-harvest (2020). Given the large site sizes in our study and the limitations of sampling their full spatial extent, we use a two-part approach in our analysis. First, we tested whether our sampling methodology detected harvest at impact sites relative to control sites. Second—contingent on detecting harvest—we quantified the impact and recovery of rockweed height and biomass at the rockweed bed scale. We hypothesized that harvest impacts to both rockweed characteristics will be spatially heterogeneous at a site, resulting in harvest impacts that are detectable at the site scale, but of smaller effect size than those reported in previous studies with smaller treatment plot sizes. Additionally, we hypothesized that sites with greater harvest intensity (i.e., biomass loss) will show less recovery one year later. We define recovery as a return to pre-harvest baseline values in height and biomass. While this approach defines impact and recovery strictly by these two morphological characteristics of the target resource, and does not consider any direct or indirect impacts on other species in the ecosystem, our study provides a harvest-scale perspective that can inform EBM recommendations in rockweed ecosystems.
Section snippets
Study design
The BACI study design (Stewart-Oaten et al., 1986) is commonly used to assess the impacts of disturbance, and has been frequently used to assess the impacts of rockweed harvest (e.g., Kay, 2015; Kelly et al., 2001; Trott and Larsen, 2012; Ugarte et al., 2006; Walder, 2015). BACI experiments allow for strong causal inference—spatial replication (control vs. impact) separates the effects of natural environmental variation and experimental treatment, while temporal replication (before vs. after)
Results
Mean holdfast count increased slightly at each consecutive time period, but the treatment x time period interaction was not statistically significant (F2,72 = 0.37, P = 0.69), indicating that changes in rockweed densities would not impact the inferences in our study. In addition, there was no statistically significant effect of harvest observation on the change in rockweed biomass between the pre-harvest and harvested time periods (F1,17 = 1.78, P = 0.20).
Discussion
We conducted a Before-After Control-Impact study with commercially harvested rockweed beds at 38 sites across five regions in Maine to quantify the bed-scale impact and short-term (i.e., one year) recovery from rockweed harvest. Our findings support our prediction that harvest impacts are spatially heterogeneous at the scale of the rockweed bed, resulting in declines in mean rockweed height and biomass that were relatively small in their effect sizes relative to the effects on the most impacted
Funding
This work was supported by funding from Maine Sea Grant and Pittman-Robertson funds awarded by the Maine Department of Inland Fisheries and Wildlife (grant number CT-09A-20170710*79). This project was also supported by the USDA National Institute of Food and Agriculture, Hatch Project Numbers ME0-21710, ME0-22207, and ME0-22322 through the Maine Agricultural and Forest Experiment Station.
CRediT authorship contribution statement
Elliot M. Johnston: Conceptualization, Methodology, Investigation, Formal analysis, Writing – original draft, Visualization. Hannah N. Mittelstaedt: Conceptualization, Methodology, Investigation, Writing – review & editing. Laura A. Braun: Investigation, Writing – review & editing. Jessica F. Muhlin: Conceptualization, Methodology, Writing – review & editing, Funding acquisition. Brian J. Olsen: Conceptualization, Methodology, Writing – review & editing, Project administration, Funding
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We acknowledge that our research takes place on the traditional homelands of the Penobscot and Passamaquoddy tribes, and we thank the Sipayik nation for access to several of our sites. We also thank the multitude of private landowners in Maine who allowed access to their property and permission to harvest rockweed for our study. Coordinating harvest at sites across the coast of Maine was a large undertaking and we are grateful for the help of S. Domizi, G. Tobey, B. Tobey, B. Morse, J. Grotton,