How we count counts: Examining influences on detection during shoreline surveys of marine debris.

Shoreline surveys are a common approach for documenting loads of marine macrodebris (≥ 2.5 cm). When surveys are conducted repeatedly over time and space, patterns in source, abundance, geographic distribution, and composition can be detected. Yet to realize their full potential, monitoring programs that rely on surveys must grapple with high variability in debris abundance, and appropriately manage uncertainty when reporting estimates of debris quantity. A potentially important source of bias in estimating debris loads from shoreline monitoring datasets is variability in debris detection rates. With this in mind, we conducted field experiments using common strip-transect marine debris survey protocols, designed to test detection of macrodebris. We quantified how protocol, shoreline, and debris characteristics influence the detectability of marine macrodebris. Detection rates varied according to debris distance from observer (0-5 m), number of observers, debris characteristics (size, color), and shoreline substrate. Our results highlight considerations for monitoring program design. Comparisons across datasets should be approached cautiously given differences in survey protocols and sources of bias that may affect debris density estimates should be quantified and addressed. We hope these results will inform marine debris monitoring efforts that are optimized for intended data use and impact.

Towards a North Pacific long-term monitoring program for ocean plastic pollution: A systematic review and recommendations for shorelines.

Increased organized monitoring is key to improving our understanding of marine debris on shorelines. Shorelines are demonstrated sinks for marine debris but efforts to quantify debris often fail to capture and report core variables and survey design techniques necessary to ensure study repeatability, comparability and to provide meaningful results. Here, we systematically review the available literature regarding marine debris distribution and abundance on shorelines of countries bordering the North Pacific Ocean (NPO), which are demonstrated to have unusually high marine debris abundance and diversity both at the ocean surface and stranded on shorelines. The majority of the 81 papers documenting shoreline debris in the NPO were studies that took place for less than one year (76.5%). Additionally, most sampling sites were visited only once (57.3%). Precise site locations (GPS coordinates) were provided in only 44.4% of the evaluated studies. Debris quantities were reported using nine different measurement units, with item counts per area and item counts per mass being most commonly reported for macro- and microplastics, respectively. Taken together, most of the reviewed studies could not be repeated by others given the information provided. We propose a series of guidelines with regard to marine debris shoreline sampling metrics, indicators, methods, and target goals in the NPO in order to improve comparability and repeatability. These follow the basic tenets of environmental survey design, which when not accounted for, can limit the applicability and value of large-scale shoreline monitoring efforts.

Walking the talk: The responsibility of the scientific community for mitigating conference-generated waste.

An estimated 19-23 million metric tons of global plastic waste reportedly entered aquatic environments in 2016 with mounting evidence that plastic marine debris causes ecological effects across all levels of biological organization in aquatic systems. Scientific conferences generate opportunities for waste through food and beverage services, giveaways, marketing and registration materials, poster and trade exhibits, attendee travel, lodging services, and local transportation. Zero waste measures instituted at the Sixth International Marine Debris Conference resulted in the avoidance of 76,300 single-use items. Zero waste is a process defined by a spectrum of actions ranging from no reduction whatsoever to generation of absolutely no waste. Achieving 100% zero waste is very difficult. Deciding where on the spectrum you wish to land and being comfortable with that target is paramount for event planning. Planning for reduced waste takes time, funding, and determination, but environmentally-themed organizations have a responsibility to lead by example.

Relative abundance of derelict fishing gear in the Hawaii-based pelagic longline fishery grounds as estimated from fishery observer data.

Derelict fishing gear (DFG) is abundant across the remote North Pacific Ocean, accumulating in convergence zones that coincide with the fishing grounds of the Hawai'i-based pelagic longline fishery. Longlines are prone to snagging DFG, providing an opportunistic, yet regular, reporting mechanism by longline fishery observers (fishery-dependent data). We apply a zero-inflated negative binomial model previously used to standardize catch per unit effort (CPUE) for bycatch and incidentally caught species in this fishery to estimate DFG relative abundance and qualitative trends within the longline fishing grounds. During 2008-2016, observers reported 1326 marine debris items intercepted by longlines, dominated by DFG at nearly 90%. Modeling results suggest that the relative abundance of DFG has declined ~66% from 2008-2016. DFG standardized CPUE was higher for longlines fished inside the North Pacific Subtropical Convergence Zone (versus outside) and increased moving eastward and northward toward the Great Pacific Garbage Patch. Despite substantially less effort in the shallow-set sector of the fishery (∼60 m depth), DFG standardized CPUE was four-fold greater than that of the deep-set sector (∼250 m) suggesting that marine debris observer reporting focused in this sector may be most effective. Some fishermen voluntarily stow snagged debris; incentivizing at-sea removal may elicit further cooperation.

Marine debris impacts to a tidal fringing-marsh in North Carolina.

We evaluated injuries to Spartina alterniflora by debris items common to North Carolina coastal waters as a function of debris type (wire blue crab pots, vehicle tires, and anthropogenic wood) and deployment duration, and monitored S. alterniflora recovery following debris removal. Injuries sustained by S. alterniflora and subsequent recovery, varied considerably between debris types. Differences were likely due to dissimilarities in the structure and composition of debris. Tires caused an immediate (within 3 weeks) and long-term impact to S. alterniflora; tire footprints remained devoid of vegetation 14 months post-removal. Conversely, crab pot impacts were not as abrupt and recovery was short-term (<10 months). We suggest that removal programs specifically target habitats that are susceptible to negative impacts (e.g., salt marsh) and prone to debris accumulation. Management would benefit from the inclusion of habitat information in removal databases.

Understanding uncertainty in seagrass injury recovery: an information-theoretic approach.

Vessel groundings cause severe, persistent gaps in seagrass beds. Varying degrees of natural recovery have been observed for grounding injuries, limiting recovery prediction capabilities, and therefore, management's ability to focus restoration efforts where natural recovery is unlikely. To improve our capacity for predicting seagrass injury recovery, we used an information-theoretic approach to evaluate the relative contribution of specific injury attributes to the natural recovery of 30 seagrass groundings in Florida Keys National Marine Sanctuary, Florida, USA. Injury recovery was defined by three response variables examined independently: (1) initiation of seagrass colonization, (2) areal contraction, and (3) sediment in-filling. We used a global model and all possible subsets for four predictor variables: (1) injury age, (2) original injury volume, (3) original injury perimeter-to-area ratio, and (4) wave energy. Successional processes were underway for many injuries with fast-growing, opportunistic seagrass species contributing most to colonization. The majority of groundings that exhibited natural seagrass colonization also exhibited areal contraction and sediment in-filling. Injuries demonstrating colonization, contraction, and in-filling were on average older and smaller, and they had larger initial perimeter-to-area ratios. Wave energy was highest for colonizing injuries. The information-theoretic approach was unable to select a single "best" model for any response variable. For colonization and contraction, injury age had the highest relative importance as a predictor variable; wave energy appeared to be associated with second-order effects, such as sediment in-filling, which in turn, facilitated seagrass colonization. For sediment in-filling, volume and perimeter-to-area ratio had similar relative importance as predictor variables with age playing a lesser role than seen for colonization and contraction. Our findings confirm that these injuries naturally initiate seagrass colonization with the potential to recover to pre-injury conditions, but likely on a decadal scale given the slow growth of the climax species (Thalassia testudinum), which is often the most severely injured. Our analysis supports current perceptions that sediment in-filling is critical to the recovery process and indicates that in order to stabilize injuries and facilitate seagrass recovery, managers should consider immediate restorative filling procedures for injuries having an original volume >14-16 m3.