The North Atlantic Gyre is a circular system of ocean currents in the Atlantic Ocean that, through the rotating pattern of currents and effects of wind and weather, amasses Sargassum in an area known as the Sargasso Sea. The Sargasso Sea is the only sea bounded by currents rather than by land (NOAA, 2019). The Gulf Stream acts as a conduit that transports Sargassum through the Caribbean, into the Gulf of Mexico, and off the coast of the southeastern United States, and evidence suggests that much of the Sargassum transported through the Gulf of Mexico originates from the North Equatorial Recirculation Region, a complex system of currents around western Africa and South America, in addition to the Sargasso Sea (Franks et al., 2016).

Since 2011, giant floating Sargassum mats in the Atlantic have increased in density and range to form an 8,850 kilometer-long belt, called the Great Atlantic Sargassum Belt, that can extend from West Africa to the Gulf of Mexico (Wang et al., 2019). High quantities of Sargassum have washed up on beaches and inundated coastlines in the Southeast U.S., Gulf of Mexico, and the Caribbean in recent years, presenting a logistical challenge for coastal communities to manage the tons of seaweed piling up on their shores (Lamb, 2018; Conley and Oliver, 2019). Satellite measurements suggest that Sargassum blooms start to develop in the Central Western Atlantic in February-March and are transported by winds and currents to the Caribbean as massive blooms from May-August (Wang and Hu, 2017). A changing climate may have unexpected impacts on Sargassum productivity and sinking as well as its tendency to be transported, aggregated, and scattered due to climate influences on algal productivity, weather patterns, and ocean currents (Sanchez-Rubio et al., 2018).

There have been recent efforts to quantify the amount of Sargassum on the sea surface through methods such as satellite imagery analysis (Wang and Hu, 2016), numerical models (Schamberger et al., 2022), and field measurements (Ody et al., 2019). A 2019 study that analyzed satellite remote sensing data from 2000 to 2018 found a significant increase in estimated Sargassum biomass in the Great Atlantic Sargassum Belt beginning in 2011, with the highest biomass estimated at more than 20 million metric tons in June 2018, the last year of the study (Wang et al., 2019). Given the economic impacts of these Sargassum blooms, new monitoring systems (Duffy et al., 2019; Valentini and Balouin, 2020) have been developed to better plan for and respond to Sargassum accumulation on coasts.

Sargassum is abundant in the Atlantic Ocean and forms an essential surface habitat that supports a diversity of marine organisms, including fish, invertebrates, sea turtles, marine birds, and marine mammals. Floating Sargassum mats serve as a primary nursery area for many fish species, some of which are commercially important (dolphinfishes, jacks, and amberjacks), and they provide a source of energy in an otherwise nutrient-poor area of the Atlantic (Casazza and Ross, 2010). These Sargassum mats provide essential habitat for approximately 120 species of fish and more than 120 species of invertebrates (Doyle and Franks, 2015).

Carbon sequestration is the process of storing carbon dioxide and other forms of carbon out of the atmosphere for long periods of time. It has been suggested that marine primary producers such as phytoplankton, macroalgae, mangroves, and seagrasses are more efficient at sequestering carbon than their terrestrial counterparts due to their high productivity and efficiency in trapping sediments and associated organic carbon (Mcleod et al., 2011; Arenas and Vaz-Pinto, 2015). Through photosynthesis, Sargassum converts sunlight, carbon dioxide, and ocean nutrients into sugars and other carbon compounds. This organic material, and the carbon it contains, can then end on a number of different pathways, such as washing up on beaches, being eaten by herbivores, or sinking to the bottom of the ocean. Once it ends up in deep ocean currents or seafloor sediments hundreds of meters below the surface, the carbon is prevented from being exchanged with the atmosphere over several hundred to several thousand years (Volk and Hoffert, 2013). Traditionally, seagrasses and mangroves have been considered the dominant form of oceanic carbon sequestration (Duarte and Cebrián, 1996); however, in recent years researchers have been looking at seaweeds like Sargassum’s role as important carbon sinks (N’Yeurt et al., 2012; Raven, 2017; Kokubu et al., 2019).

The gravitational sinking of Sargassum from the surface to the seafloor, and degradation of Sargassum within the water column and at the seafloor, are poorly understood processes. The mechanisms that deliver drifting seaweed to marine sediments include wind-induced Langmuir circulation that can transport floating seaweed fragments to a depth where pressure collapses its air bladders, rendering the seaweed negatively buoyant and removing it from the surface (Krause-Jensen and Duarte, 2016). Additionally, Sargassum living in the shaded understory of floating Sargassum may be prevented from photosynthesizing and lose buoyancy and die and start to degrade. It will also lose buoyancy with age and from encrustation of epizoans (Stoner, 1983). Once it achieves negative buoyancy, the Sargassum sinks at an estimated rate of 3.5 cm per second (Johnson and Richardson, 1977). At this rate, Sargassum would reach the seafloor at 1000 meters in just 8 hours. This is relatively fast for sinking particles and does not allow much time for remineralization, consumption, or degradation (Giering et al., 2020; Omand et al., 2020).

Sargassum sinking has been studied before using Autonomous Underwater Vehicle (AUV)-collected seafloor photographs in the southern part of the North Atlantic ocean (three stations between ~10-12° N and ~36-50° W), within the North Atlantic Subtropical Gyre where the Sargasso Sea is located. The biomass density of Sargassum deposited on the seafloor was estimated at 0.07 to 3.75 g/m², which is greater than the surface biomass as estimated through literature review at 0.024-0.84 g/m² (Baker et al., 2017). Krause-Jensen and Duarte (2016) estimated that 11 percent of macroalgal particulate organic carbon (POC) export, or 35 TgC/yr, reaches the deep sea globally, serving as an important carbon sink. Start-up companies and financial investors have noted the potential for macroalgae aquaculture for commercial products like bioplastics and food or to sink and bury on the deep seafloor to sequester carbon from the atmosphere (López Miranda et al., 2021; Oxenford et al., 2021). Sargassum in particular has been identified as a target species for such efforts because of its naturally high abundance and reproductive rate.

The factors that contribute to organic carbon export in the open ocean and its subsequent carbon sequestration are driven by a complex combination of ecological, biogeochemical, and physical oceanographic processes. Developing a predictive understanding of carbon export pathways for seaweeds including Sargassum is critical for understanding present and future rates of ocean carbon sequestration and informing the emerging marine carbon dioxide removal industry. This study provides an initial characterization of Sargassum on the seafloor by describing the number of Sargassum observed, its relationship to benthic composition, any interactions with benthic organisms, and geological features of the dive sites explored. We revisited a series of archived deep ocean exploration visual surveys that took place on seabed areas underlying known and likely Sargassum surface blooms as one step toward understanding the fate of sinking Sargassum in the marine system.

The National Oceanic and Atmospheric Administration (NOAA)’s Office of Ocean Exploration and Research is a U.S. federal program dedicated to exploring the unknown ocean for public benefit, filling in knowledge gaps about the marine environment through scientific discovery, technological advancements, and access to data. It conducts interdisciplinary ocean exploration expeditions on NOAA Ship Okeanos Explorer and other research vessels. Okeanos Explorer is a 68-meter research ship outfitted with an array of mapping sonars and a dual body Remotely Operated Vehicle (ROV) system: ROV Deep Discoverer (D2) and ROV Seirios. A unique and advantageous aspect of Okeanos Explorer operations is its ability to facilitate real-time communication and collaboration with shore-based scientists (Peters et al., 2019). Scientists on the ship and on land collaborate through a live video feed, text chatroom, shared conference phone line, and video annotation system, allowing participants not only to follow along live but to provide feedback and scientific expertise in real-time during the ROV dives (Kennedy et al., 2016; Selig et al., 2019).

Kennedy et al. (2019) describe the technical details on the Okeanos Explorer’s seafloor mapping and ROV systems and operations. All ROV dives examined in this study were conducted with NOAA’s dual body ROV system Deep Discover (D2) and Seirios, although only video from D2 was reviewed. The main capability of D2 is the ability to capture high-definition video, with its primary camera able to zoom in and provide close-up video of relatively small organisms (Kennedy et al., 2019). Seirios is directly tethered to the Okeanos Explorer via a fiber-optic cable and is also tethered to D2, a configuration that allows Seirios to absorb the sway of the ship while keeping D2 stable. The cable provides the ROVs with power and allows for data transfer between the ROVs and the ship. Both Seirios and D2 are equipped with a suite of sensors to measure environmental parameters like temperature, dissolved oxygen, salinity, and depth. Surveys conducted with this two-body system are exploratory in nature, providing baseline information on the broader biological, geological, and physical context of poorly explored regions of the ocean (Selig et al., 2019; Cantwell et al., 2020). Because the ROV time is maximized for a variety of objectives, true quantitative analyses of the imagery are not always possible, yet important insights may still be gleaned.

From 2018-2022, Okeanos Explorer conducted a series of expeditions as part of the Atlantic Seafloor Partnership for Integrated Research and Exploration (ASPIRE), a multi-national collaborative ocean exploration campaign to raise the collective knowledge and understanding of the North Atlantic Ocean. Three of the four expeditions reviewed – Océano Profundo 2018: Exploring Deep-sea Habitats off Puerto Rico and the U.S. Virgin Islands (EX1811), Windows to the Deep 2019: Southeast and Mid-Atlantic U.S Continental Margin, Port Canaveral, FL to Norfolk, VA (EX1903L2), and 2019 Southeastern U.S. Deep-sea Exploration (EX1907) – were part of this ASPIRE campaign.

A subset of expeditions and ROV dive sites were selected for this study based on the following considerations:

Details on bathymetric and ecological features are available in Supplementary Information Table 1 .

Okeanos Explorer ROV video footage is streamed live through SeaTube V3, a platform developed by Ocean Networks Canada (ONC) for scientists to view and annotate ROV footage both in real time and afterwards (Selig et al., 2019). SeaTube V3 provides open access and archives of dive videos, dive imagery, dive logs, navigational data, and metadata for deep ocean exploration (Jenkyns et al., 2013). An informal science chatroom developed by the Global Foundation for Ocean Exploration is also used for sidebar conversations between scientists and captures some of the annotations as well; the messages of the chat are then archived for future access ^{1 , 2} .

Annotations of Sargassum were done manually, by reviewing the selected footage from the Deep Discoverer ROV. Each of the dives selected for review was viewed in SeaTube V3 from the start of the ROV’s descent from the surface to the end of its ascent back to the surface to account for the possibility of Sargassum being observed in the water column. An object was identified as Sargassum based on its shape, color, texture, and movement. Figure 1 shows screen grabs of the ROV video to show the variability of what the Sargassum looks like when it reaches the seafloor.

When there was an object that was potentially Sargassum but identification was questionable, higher-resolution footage of that part of the dive was accessed and viewed in NOAA’s Ocean Exploration Video Portal ³ . If after reviewing the high-resolution footage the object could not with absolute certainty be identified as Sargassum, the annotation included the note “likely Sargassum” to indicate a degree of uncertainty. Annotations with “likely Sargassum” designation were counted as Sargassum in this report. All other observations were identified to the Sargassum genus referencing the World Register of Marine Species (WoRMS) system of classification and annotated in SeaTube V3.

For dives with abundant Sargassum, a new Sargassum annotation was made for every new frame of view that contained Sargassum. For instance, even if there were five clumps of Sargassum in the frame, this was counted as one Sargassum annotation. As soon as the ROV had Sargassum in its frame of view that was not clearly visible in the previous annotation, this was entered as a new annotation. Each Sargassum annotation was classified as “low,” “medium,” or “high” biomass based on visual estimation of the approximate amount of Sargassum within the frame of view on a scale from 1-8 ( Supplementary Information Table 2 ). Additionally, when mobile animals were observed on or in the near vicinity of the Sargassum detritus, this was noted, as well as whether the animal appeared to be grazing on the Sargassum. Bathymetric features (rocks, sediment, etc.) and other observations (high marine snow, presence of anthropogenic debris) were also noted in Supplementary Information Table 2 .

In addition to reviewing the ROV footage and accompanying audio commentary, the archived chatroom logs that include expert perspectives, and post-dive summaries, accessed through NOAA’s Institutional Repository ⁴ , were reviewed to help characterize each dive site.

Distance transited along the seafloor was calculated by importing spatial coordinates from the ROV recorded while at depth and plotting them in ESRI’s ArcGIS Pro 2.9 software. A line was then generalized based on the points and the length in meters was calculated for each dive. Distance transited varied from approximately 250 meters to over 1,000 meters per dive.