Large-scale seaweed cultivation and its purposeful sequestration is a potential ocean-based strategy for reducing atmospheric carbon dioxide (CO2) levels. Large-scale farming of seaweed would incorporate dissolved CO2 from the upper ocean into tissue that then can be sequestered at depth either by pumping biomass to depth or by its sinking through the water column. As many seaweeds grow, they release large amounts of dissolved organic carbon (DOC) into the upper ocean. Some fraction of that is thought to be recalcitrant to microbial activity over long timescales, enabling an additional pathway for seaweed cultivation to sequester carbon. Macrophyte biomass farms could be created on large scales in environments where there is sufficient solar illumination and available nutrients. Much progress has been made in the past decade in developing both commercial seaweed farms for human consumption and animal feed as well as pilot studies for the development of large-scale farms for biofuel production. The goal here is to assess the present state of knowledge and address what research and investments are needed to make purposeful seaweed cultivation and sequestration CO2 removal (CDR) worthy.
According to the criteria described in Chapter 1, the committee’s assessment of the potential for seaweed cultivation as a CDR approach is discussed in Sections 5.2–5.4 and summarized in Section 5.5. The research needed to fill gaps in understanding of seaweed cultivation and sequestration, as an approach to durably removing atmospheric CO2, is discussed in Section 5.6.
Studies on natural macrophyte-dominated ecosystems and in aquaculture facilities have provided much of the information needed to assess the CDR potential of seaweed cultivation. These include determinations of biomass density, carbon content, rates of net primary production (NPP), nutrient ratios, seasonality of growth, farming techniques that lead to higher yields, and release of DOC during production among others (e.g., Wheeler and North, 1981; Reed et al., 2008, 2015; Stewart et al., 2009; Rassweiler et al., 2018; Azevedo et al., 2019; Bak et al., 2020; Forbord et al.,
Considerably less is known about the fates of macrophyte carbon in natural ecosystems and how this carbon contributes to long-term carbon sequestration. Several studies have attempted to quantify the role of natural macrophyte ecosystems in global carbon sequestration (e.g., Smith, 1981; Chung et al., 2010; Wilmers et al., 2012; Krause-Jensen and Duarte, 2016; Watanabe et al., 2020). In a particularly influential contribution, Krause-Jensen and Duarte (2016) concluded that natural macroalgal ecosystems could make substantive contributions to global ocean carbon sequestration primarily through the export of plant biomass to depth and the seafloor and the production of recalcitrant DOC. Recalcitrant DOC is that fraction of the DOC pool that is resistant to rapid microbial degradation (e.g., Hansell, 2013). Via a synthesis of a broad range of previously published results over a wide range of taxa (Figure 5.1), Krause-Jensen and Duarte (2016) suggest that macroalgal ecosystems could at most sequester ~0.17 petagrams (Pg) C/yr or ~0.6 Gt CO2/yr globally with a wide range of uncertainty based upon the assumptions applied (roughly 0.06 to 0.27 Pg C/yr).1 They find that the export of recalcitrant DOC to below the mixed layer is the dominant contribution to their potential sequestration budget (roughly 70 percent of total sequestration).
The Krause-Jensen and Duarte (2016) paper attempts to quantify the carbon sequestration by naturally occurring macrophyte ecosystems if these ecosystems were to inhabit all regions in the world ocean where there is sufficient light on the benthos (see notes in Table 2 of Krause-Jensen and Duarte, 2016). Therefore, their estimate is the potential that natural macrophyte afforestation could
1 1 Pg C = 1,015 g C = 3.7 Gt CO2.
contribute to global carbon sequestration and not the actual amount that these ecosystems presently sequester. Here, the focus is on the large-scale cultivation of seaweed biomass and its purposeful sequestration in durable ocean reservoirs. Hence, an independent assessment of its carbon budget is made in Section 5.3. Afforestation of seaweed ecosystems and their potential as a CDR strategy are discussed in Chapter 6.
Purposeful macrophyte cultivation has recently become a particularly popular CDR strategy due to its potential to scale to CDR-relevant amounts with a wide range of enticing co-benefits (e.g., GESAMP, 2019; Gattuso et al., 2021; oceanvisions.org and similar websites). There are also important barriers to its viability as an effective CDR strategy that will be discussed below. One useful conceptualization of macrophyte cultivation as an ocean CDR strategy would be considering large-scale farming of seaweeds to assimilate CO2 from the surface ocean and then purposefully convey this fixed carbon to deep oceanic reservoirs that will remain out of contact with the atmosphere over some relevant planning time horizon, say 100 years. The air–sea CO2 equilibrium timescales relative to surface water residence times also need to be considered, as is the case with all marine CDR approaches considered here (see Section 1.3).
The elements of purposeful macrophyte cultivation as an ocean CDR strategy are depicted in Figure 5.2. Seaweed farming converts dissolved CO2 to seaweed biomass that is slowly replaced by air–sea fluxes from the atmosphere (e.g., GESAMP, 2019; Gattuso et al., 2021; Sala et al., 2021). The large-scale cultivation will also produce DOC during growth (Reed et al., 2008; Paine et al., 2021), some of which, it has been suggested, will be recalcitrant on decadal timescales (Krause-Jensen and Duarte, 2016). Both pathways may contribute to long-term carbon sequestration. In many regions of the world’s oceans, nutrients are naturally depleted in the upper layers of the water column but can be found just beneath the euphotic zone. Thus, it is likely that artificial upwelling devices will be needed to supply required nutrient concentrations in some settings, which brings with its application other consequences, such as upwelling of enriched dissolved CO2 deep waters and other possible ecological impacts (see Chapter 4).
The harvested biomass needs to be conveyed by some means deep enough in the water column that most of the carbon biomass, when remineralized back to CO2, will remain out of contact with the atmosphere. The leakage of sequestered material back to the sea surface will be driven by ocean circulation and mixing processes and hence will be a function of depth, location, and planning timeline (Siegel et al., 2021a). Further, the sequestered biomass will likely need to be injected to depth rapidly to ensure that little carbon is lost near the ocean surface, potentially requiring a conveyance device. The injected biomass will decompose back to CO2 and potentially recalcitrant DOC due to food web, particularly heterotrophic microbial, processes (Figure 5.2).
The cultivation and purposeful injections of seaweed biomass will likely affect ocean ecosystems both in the euphotic zone where the biomass is grown and at depth. In the near-surface ocean, these effects include reducing ambient nutrient levels and available light. Subsequently, that will likely reduce phytoplankton primary production rates, decrease carbon export from the surface ocean, and may affect trophic exchanges of energy that support fisheries and marine mammal populations. At depth, these perturbations may alter the natural balances of organic matter decomposition and remineralization, reducing oxygen concentrations (deoxygenation) and increasing subsurface CO2 and nutrient levels, leading to increases in acidification and eutrophication of these mesopelagic ecosystems. However, the strength of these impacts will depend upon the amount of CDR performed and the efforts made to displace their influences. There may also be several direct societal impacts of large-scale farming, including hazards to navigation and co-benefits such as reducing excess nutrients from fish and shellfish aquaculture facilities. Furthermore, the farming of macrophytes for products (e.g., food, biofuels, animal feed, etc.) may contribute to long-term carbon sequestration, possibly due to the production and release of recalcitrant DOC. The sequestration timescales of the fixed carbon in macrophyte biomass products are likely to be short (≤10 years at best) compared with the goal of CDR (≥100 years).
Much of the knowledge base for large-scale seaweed cultivation comes from experiences in large-scale seaweed farms for animal and human food as well as a potential carbon source for biofuels (Camus and Buschmann, 2017; Bak et al., 2018; Camus et al., 2018; Azevedo et al., 2019; ARPA-e, 2021a; Navarrete et al., 2021). Macroalgal farming and the products these farms create have become a multibillion dollar industry, with Asia being the main supplier (FAO, 2016). On a global basis, seaweed aquaculture production exceeds 30 million (wet) Mt and $5.6B on a monetary basis, and these totals are rapidly growing (FAO, 2016). Currently, farmed macroalgal uses include human and animal food, fertilizers, other products, and a feedstock for biofuel production (e.g., Milledge et al., 2014; ARPA-e, 2021a). The majority of algal biomass comes from a relatively small number of species (Milledge et al., 2014). Much focus has been placed on improving yields and the quality of the farmed macrophyte crops while lowering costs (e.g., Bak et al., 2018, 2020; Azevedo et al., 2019; Forbord et al., 2020; Matsson et al., 2021). Macrophyte farms are typically conducted on suspended longline ropes with embedded sporophytes moored in shallow (≤100 meters) coastal or estuarine waters (Peteiro and Freire, 2013; Camus and Buschmann, 2017; Camus et al., 2018). Existing farms are a few up to a few thousand hectares in size. Recent research supported by the U.S. Department of Energy Advanced Research Project Agency-e’s (ARPA-e’s) MARINER program is aimed at creating prototypes for scalable farmed systems that can be deployed in deeper waters for the purpose of growing biomass stocks for biofuels (ARPA-e, 2021a). Among the specific goals of MARINER are to develop technologies for farming macrophyte biomass that can be scaled to ≥100,000 hectares (≥1,000 km2) at production costs of ≤$80/dry metric ton biomass (ARPA-e, 2021a). The MARINER production cost goal is roughly equivalent to ~$75/metric ton CO2 in macrophyte biomass (assumes C content is 30 percent of the dry weight). Innovative farm designs even include systems that cycle vertically on a daily time course to optimize the acquisition of light energy during the day and subsurface nutrients at night (Navarrete et al., 2021). Important elements for successful macrophyte cultivation include selecting the cultivars to be farmed based
upon environmental conditions and local macrophyte strains; culturing spores to embryonic sporophytes and efficiently attaching these sporophytes on longline ropes; installing longline ropes in the field; permitting, installing, and maintaining farm facilities; monitoring the crop status as well as the environmental conditions; and developing an understanding of the effects of the environment on the crops and the crops on the environment, etc. (Camus et al., 2018; Bak et al., 2020; Bell et al., 2020; Visch et al., 2020; Matsson et al., 2021).
The knowledge base for the long-term fates of farmed macrophyte organic matter is far less known. It is important that the farmed macrophyte carbon be removed from the surface ocean so that the assimilated carbon will remain out of contact with the atmosphere on sequestration timescales (decades to centuries). The long-term incorporation of vast amounts of macrophyte carbon in ocean sediments either by purposeful placement on the seafloor or by sinking seaweed biomass from the surface seems highly unlikely. For example, Bernardino et al. (2010) studied the degradation of 100-kg bales of giant kelp placed on the seafloor at 1,670-meter depth in the Santa Cruz Basin in the North Pacific Ocean using remotely operated vehicles. They found significant changes after 6 months in macrofaunal community abundances and diversity near (within 1 meter) to the kelp bales, but very little discernible changes in sediment total organic carbon concentrations. Similar rapid decomposition rates of kelp biomass on the benthos (roughly 5 percent biomass per day) were found in shallow continental shelf depths (80–350 meters) off Carmel Canyon, California (Harrold et al., 1998) and at the bottom (1,300 meters) of the Santa Catalina Basin (Smith, 1983). Hence the water column seems the most likely reservoir for injected macrophyte carbon. Over centennial timescales appropriate for ocean CDR, this will be in the form of respired CO2.
A recent analysis of purposeful CO2 sequestration timescales using an ocean circulation inverse model showed that deeper discharge locations sequester injected CO2 for much longer than shallower ones and median sequestration times are typically decades to centuries, and approach 1,000 years in the deep North Pacific (Siegel et al., 2021a). Further, large differences in sequestration times occur both within and between major ocean basins. The Pacific and Indian basins generally have longer sequestration times than the Atlantic Ocean and Southern Ocean. Assessments made of the injected CO2 retained over a 100-year time horizon illustrate that most of the injected carbon will still be in the ocean at injection depths greater than 1,000 meters, with several geographic exceptions such as the western North Atlantic (see further descriptions of this work below). Ocean circulation and mixing thus place important constraints on the timescales over which injected CO2 remains in the water column.
Methodologies for the conveyance of macrophyte biomass to depth are just in their infancy. If the pneumatocysts are forced to burst (either by pressure due to being injected to depth or the tissue masticated), it is thought that the macrophyte biomass will sink until it reaches the seafloor (Krause-Jensen and Duarte, 2016; see also https://www.runningtide.com). Engineered solutions for pumping macrophyte biomass to depth via vertical tubes have also been developed, and a system for removing nuisance mats of Sargassum from the surface ocean has been prototyped recently (Gray et al., 2021). Technologies need to be developed to verify the delivery of organic carbon to depth with minimal losses.
Last, very little is known about the fate of DOC produced by macrophyte populations as they grow. Field work on natural giant kelp forests suggests that ~14 percent of macroalgal NPP is released as DOC of all labilities (Reed et al., 2015), the proportion of that DOC that is recalcitrant to degradation is largely unknown. Krause-Jensen and Duarte (2016) simply assume that the partitioning of macrophyte DOC into labile and recalcitrant fractions follows the ratio of global DOC export from the upper ocean to the global NPP, which results in an estimate for the production of recalcitrant DOC that is roughly 8 percent of macrophyte NPP. The validity of that assumption remains a major knowledge gap. A recent study of temperate Sargassum beds showed that 56 percent to 78 percent of the released DOC collected in these ecosystems was resistant to decomposition after
150 days (Watanabe et al., 2020). However, the timescales relevant for assessing CDR strategies are decadal to centennial, and more research in this area is clearly needed.
We can investigate the impact potential and scalability of purposeful macrophyte cultivation using a simple scaling analysis. A reasonably successful CDR goal would be to grow and sequester enough macrophyte biomass to remove 0.1 Gt CO2/yr (0.027 Pg C/yr = Seqgoal) from the upper ocean over a time horizon of more than 100 years. Clearly at these levels of CDR, seaweed cultivation is envisioned as contributing to a portfolio of terrestrial- and ocean-based CDR approaches. To assess the potential of purposeful macrophyte cultivation as a scalable CDR tool, we will estimate the areal size of a farm required to grow that amount over that time horizon and use this formulation to discuss the scalability of purposeful macrophyte cultivation to climate control–relevant scales.
Following the synthesis of Krause-Jensen and Duarte (2016), we consider the two primary pathways linking farmed macrophytes and carbon sequestration—the purposeful injection of particulate macrophyte carbon to depth and the release and eventual sequestration of recalcitrant DOC from the growing macrophyte farms, or
|Seqgoal = 0.1 Gt CO2/yr = 0.027 Pg C/yr = SeqBio + SeqDOC||(5.1)|
The C budget for natural macrophyte ecosystems created by Krause-Jensen and Duarte (2016) suggests that the SeqDOC should be 8 percent of global macrophyte NPP (fDOC; see Figure 5.1). Assuming this holds for a farmed system, SeqDOC should be
|SeqDOC = fDOC * NPPFarm||(5.2)|
where NPPFarm is the NPP of the farm. Some fraction (floss) of the farmed biomass will be lost due to herbivory, frond senescence, storms, and inefficiencies in the conveyance of biomass to depth, etc. We will assume that the value of floss is 20 percent, but it is recognized that this value is, at present, poorly constrained. Thus, the accumulation of macrophyte C biomass of the farm per year (SeqBio) will be equal to
|SeqBio = (1 − floss) * NPPFarm||(5.3)|
Thus, the amount of macrophyte NPP required to grow up each year to reach the 0.1 Gt CO2/yr goal would be
|NPPFarm = Seqgoal/(1 + fDOC − floss) = 0.033 Pg C/yr||(5.4)|
Contrasting the natural example above (Figure 5.1), the sequestration of macrophyte carbon from farm systems will be dominated by the direct injection of C biomass to depth as the entire crop (minus losses in the upper ocean) will be sequestered. Thus, the release of recalcitrant DOC on growth should be a small part of the total carbon sequestration budget (~9 percent of Seqgoal). This is much smaller than the Krause-Jensen and Duarte (2016) synthesis, which suggests that recalcitrant DOC release dominates (~70 percent) the total sequestration from natural macrophyte ecosystems. This difference arises because the biomass losses and remineralization in the surface ocean for CDR using seaweed cultivation and purposeful sequestration will be much smaller than in Krause-Jensen and Duarte’s (2016) synthesis of a natural seaweed ecosystem. Hence, the contri-
butions of recalcitrant DOC for CDR via seaweed cultivation will be much smaller than has been suggested for natural systems.
The areal extent of a seaweed farm capable of sequestering 0.1 Gt CO2/yr can be determined knowing NPPFarm, the number of crops per year that can get grown and sequestered (NCrop), the expected biomass density in the farm (Yield) and the C density of crop (CContent), or
|AFarm = NPPFarm/(Yield * CContent * NCrop)||(5.5)|
Assuming that the farmed system will maximize seaweed density and carbon quality, we use giant kelp observations of the maximum biomass observed from the Santa Barbara Coastal Long-Term Ecological Research data record (Rassweiler et al., 2018) or a maximum yield (Yield) equal to 1 kg DW/m2 and 30 percent of that dry weight (DW) will be carbon by DW mass (CContent). We also assume that one can grow and sequester on average 1.5 crops each year, which may be an optimistic assumption based on present macrophyte farms, particularly for higher-latitude sites. Together, the required area needed to sequester 0.1 Gt CO2/yr by seaweed cultivation comes to
|AFarm = 0.033 e(15 g C/yr)/([1e(3 g DW/m2)] * [0.3 g C/g DW] * [1.5 crops/yr]) = 7.3e(10 m2) = 73,000 km2 = 7.3 million hectares||(5.6)|
Thus, the size of a farm required to grow enough biomass to sequester 0.1 Gt CO2/yr would be a single square farm, 270 km on a side. These farms would logically be multiple farms spread out across the globe. This area is approximately equivalent to half the size of the State of Iowa or if one considers a 100-m-wide continuous belt of seaweed farm along all continents and islands, it would require 730,000 km of coastline. That is 63 percent of the global coastline. If placed along the coastline of the United States only, it would comprise a nearly 0.5-km-wide continuous belt of seaweed farm. The amount of ocean surface area required to sequester 0.1 Gt CO2/yr demonstrates the size of the engineering and logistical tasks at hand associated with scaling seaweed-cultivation CDR solutions to climate-relevant scales.
The formulation above enables the assessment of controls on the required size of macrophyte farms needed to grow enough biomass to sequester climate-relevant amounts of CO2 from the upper ocean. Increasing the sequestration goal (Seqgoal) will linearly increase the farmed area required, while increasing the yield, carbon content, and number of crops per year, or reducing the biomass losses before sequestration will similarly reduce the required farmed area. Work from ARPA-e’s MARINER project2 suggests that innovation could lead to increases in yields by up to fivefold from the 1-kg DW/m2 base value used here. However, only slight changes in the carbon content per unit biomass would be expected. Innovations could also help increase the number of crops per year for a given farm installation by dramatically increasing growth rates via cultivar selection and careful breeding of macrophyte strains and by engineering solutions that reduce biomass losses before sequestration.
There are many requirements to farm vast amounts of macrophyte carbon biomass. Optimal growth of macrophyte biomass requires adequate nutrient concentrations and light levels (e.g., Jackson, 1977, 1987; Gerard, 1982; Zimmerman and Kremer, 1986). Achieving both will be difficult because throughout most of the world’s oceans, vertical regions where there is enough solar radiation to drive photosynthetic processes (the euphotic zone) are often depleted in macronutrients, while vertical strata where adequate nutrients are available are often too deep to support growth, which is why macrophyte populations naturally inhabit nearshore habitats where both nutrients
and light are often adequate to support their growth. The nutrient requirements for concentrated seaweed farming will be particularly intense. A recent paper estimates that current seaweed production in China will by 2026 utilize all of the excess anthropogenic phosphorus discharged into Chinese waters (Xiao et al., 2017). To alleviate these issues, the MARINER program has made efforts to select cultivars that grow efficiently under low-nutrient conditions or employ artificial upwelling devices to supply nutrients from deeper depths (see details in Chapter 4) or implement novel mechanical means to bring the crop to depth at night (ARPA-e 2021a; Navarrete et al., 2021). Another important siting requirement is to create farming facilities that are robust to storms and protect macrophyte biomass and the farm infrastructure from storm losses. Access to nearby ports would also be essential to help reduce costs for farm operations and maintenance.
There remains the question of how the cultivated biomass should be sequestered to optimize the amount of respired CO2 that will be retained in the ocean before outgassing to the atmosphere. Macrophyte biomass will be recycled through oceanic food webs and eventually be respired to CO2 on decadal timescales. Thus, efforts must be made to deposit this biomass at depth where it will be assimilated by deep-sea ecosystems and respired back to CO2, but not in immediate contact with the atmosphere. Very little will likely end up in seafloor sediments; however, recent environmental DNA analysis has shown evidence of macrophyte DNA in surface sediment (Geraldi et al., 2019). Very little is known about the timescales of degradation of macroalgal carbon or DNA in seafloor sediments.
The timescales over which CO2 injected within the ocean interior remains sequestered from the atmosphere has recently been assessed using a model of steady-state global ocean circulation and mixing (Siegel et al., 2021a). This model shows challenges ahead for any purposeful water-column injections of CO2 aimed at sequestering CO2 from the atmosphere. First, there will be a wide range of sequestration times linking a discharge location with the sea surface due to the infinite number of pathways connecting them. The resulting probability distribution is highly skewed, with a large fraction of relatively young transit times and a long tail of very long transit times. Second, deeper discharge locations will sequester CO2 longer than shallower ones, and median sequestration times are typically decades to centuries. Third, large differences in sequestration times occur both within and between the major ocean basins, with the Pacific and Indian basins generally having longer sequestration times than the Atlantic Ocean and Southern Ocean. Last, assessments made over a 100-year time horizon illustrate that most of the injected carbon will be retained for injection depths greater than 1,000 meters, with several geographic exceptions such as the western North Atlantic (Figure 5.3). Retention is nearly ensured by depositing macrophyte biomass on the seafloor at depths greater than 2,000 meters (Figure 5.3). Conveyance apparatuses are likely needed to reduce the fragments of particulate carbon and DOC that could potentially be unintentionally released into the upper ocean.
Barriers to Implementation
The viability of seaweed cultivation as a CDR strategy has a range of barriers, incentives, and issues related to its implementation that need to be addressed. First, one must consider its potential environmental impacts. The sequestering of large amounts of organic matter at depth will surely have detrimental effects on the ecology of the deep sea, which is likely both the largest (by volume) and least understood biome on Earth (e.g., Martin et al., 2020). This is because the purposeful inputs of macrophyte organic matter at depth will eventually respire back to CO2, leading to local to regional increases in deoxygenation, acidification, and eutrophication. Further anthropogenic inputs in vast amounts of particulate matter will also influence visibility and contacts among mesopelagic
organisms, similar to what might be expected from the improper disposal of tailings from deep-sea mining operations (e.g., Drazen et al., 2020).
The biological pump exports roughly 10 Pg C/yr from the upper ocean to depth over the entire globe (e.g., Siegel et al., 2014; DeVries and Weber, 2017; Boyd et al., 2019). If this natural export flux decays with depth following the so-called Martin curve (Martin et al., 1987; Buesseler et al., 2007, 2020), the natural flux of organic matter that arrives at 2,000 meters averaged over the entire world ocean will be ~1.2 Pg C/yr. For the above scaling analysis for sequestering 0.1 Gt CO2/yr, the purposeful input of seaweed will increase the global delivery of organic matter by nearly 5 percent. An input of 1 Gt CO2/yr (0.27 Pg C/yr), as suggested in the X-Prize competition,3 will have a substantive impact on the natural inputs of organic matter delivered to a horizon of 2,000 meters (increasing the global flux of organic matter at that depth by ~25 percent). These alterations will surely alter mesopelagic and deep-sea food webs by adding foreign biomass with potentially different food qualities. The decomposition of the added biomass will lower oxygen levels and increase acidity and nutrient levels, leading to increased deoxygenation, acidification, and eutrophication. Further, the nature of the purposeful inputs will likely be highly heterogeneous in space and likely intermittent in time because it seems difficult to ensure that the inputs of organic matter will be or could be dispersed uniformly at depth, especially given that the infrastructure required to cultivate macrophyte biomass carbon to CDR-relevant scales will be localized to a few port cities. This concentration of organic matter inputs will greatly increase the local-scale environmental impacts of the purposeful additions of organic matter. Further, there remains a great deal of uncertainty in the fate of the dissolved organic matter produced as seaweed grows. Coordinated research in all
of these areas is needed to better assess the impacts of seaweed cultivation and sequestration as a viable CDR strategy.
Further, the growing of large-scale crops of macrophyte biomass will surely affect euphotic zone ecosystems. The farm will divert ambient nutrients that drive the upper ocean ecosystem into cropped biomass, likely reducing rates of phytoplankton NPP and thereby in turn reducing the fluxes of C export from the surface ocean into the ocean interior and decreasing the flow of energy into higher trophic levels that support fisheries and other valued marine resources. These effects may increase the need for more ocean CDR to offset these losses. This may be offset somewhat because macrophyte biomass typically has higher carbon-to-nutrient concentrations (either phosphorus or nitrogen) than typical organic matter concentrations in plankton-dominated ecosystems (e.g., Rao and Indusekhar, 1987). Understanding the ecological and environmental impacts of large-scale cultivation of seaweeds is a critical area where research is sorely needed. Another issue to resolve is the fact that seaweed cultivation will likely introduce nonnative species to ecosystems where the farming occurs because nearshore macrophyte species will need to be farmed in offshore biomes. It is also likely that cultivars will need to be selected that can maximize biomass production in low-nutrient environments. The introduction of nonnative species may have detrimental ecological impacts and legal implications that complicate the permitting processes.
Huge, robust structures will be required to farm enough biomass carbon to make seaweed cultivation a viable CDR strategy. Much work is needed to engineer robust systems that will likely need to be situated in deeper ocean waters than present-day seaweed farms, typically in shallow waters. Again, ARPA-e’s MARINER program is conducting important work developing prototypes for these systems (ARPA-e, 2021a). The size of these structures suggests that there should be concerns about the risks of entanglement with whales and other air-breathing vertebrates, displacement of fishing and other ecosystem services, and the hazards to navigation created by the structures needed for seaweed cultivation. Further, efficient ship systems that minimize their carbon emissions need to be developed to manage cultivation and sequestration systems.4
Monitoring and Verification
Every CDR strategy will need a monitoring and verification program to prove its veracity so that the end-to-end costs, benefits, and environmental impacts can be quantified. For seaweed cultivation, these assessments will need to be conducted on both local (e.g., farm, injection location, etc.) as well as on global scales. Local-scale efficacy could be assessed via coupled observations and modeling similar to that of a standard oceanographic process study. The goal will be to conduct fieldwork in contrasting sites and farm types so that detailed process numerical models can be developed, tested, and applied to other sites. Developing systems to achieve this goal should follow the planning for major oceanographic process studies focused on the biological carbon pump (e.g., Siegel et al., 2016). This will require substantive resources. Using the EXPORTS program as a basis, a month-long assessment of carbon export pathways costed out at roughly $115M (including ship time, scientists, staffing, sample and data analyses, logistics, etc.).
Monitoring and verification on global scales will be harder. Numerical modeling, informed by observations, will be one critical component. For example, aerial optical observations, both satellite and drone based, would be very useful for mapping farm biomass and productivity for seaweed taxa that form canopies (Bell et al., 2020; Cavanaugh et al., 2021) and may be useful for assessing the displaced natural productivity of the farm. Future satellite missions, such as NASA’s upcoming
4 For an example of a recent prototype, see https://arpa-e.energy.gov/technologies/projects/autonomous-tow-vessels.
Plankton, Aerosol, Cloud and ocean Ecosystems5 and Surface Biology and Geology6 missions, will provide hyperspectral imagery on a variety of spatiotemporal scales that will be useful for mapping seaweed canopy biomass and productivity (Bell et al., 2015a,b, 2020). Existing global biogeochemical monitoring systems, such as the emerging Biogeochemical (BGC) Argo array,7 could also be useful for this task by assessing levels of dissolved oxygen utilization over time, especially on local spatial scales near where the biomass has been conveyed. However, relying on the BGC Argo array to assess impacts on global scales may be problematic. For example, the global sum of present-day global dissolved oxygen utilization levels corresponds to roughly 1,500 Pg C with an uncertainty of about 200 Pg C (Carter et al., 2021). Thus, detecting even a 1-Gt CO2 (0.27-Pg C) change in global inventories due to purposeful injections of cultivated seaweed biomass will require severe constraints on the required accuracy and precision of oxygen (O2) measurements and the timescales required to see these changes. For example, an input of 1 Gt CO2/yr would represent less than a 1:2,000 reduction in global mean oxygen concentrations, potentially requiring detailed analyses and many years detecting these small global-scale changes. Further separating already-committed anthropogenic changes expected (e.g., Moore et al., 2018) from the purposeful CDR changes may be very difficult to assess. Thus, numerical modeling informed by local-scale process studies may be the best way to assess the impacts of seaweed cultivation and sequestration. Note also that these models probably need to better account for higher trophic levels because these processes are generally left out of most Earth system models (e.g., Bonan and Doney, 2018).
The costs needed to construct, operate, and maintain the farm infrastructure will be substantial. Scaling seaweed cultivation to useful CDR levels will require immense farm structures (7.3 million hectares in extent for sequestering 0.1 Gt CO2/yr) to be built, maintained, and operated. Large power sources will be required in this process as well to transport and maintain these facilities. To this end, offshore seaweed farms may involve combining renewable energy sources. There are two current seaweed farm pilot projects offshore Belgium and the Netherlands that are using existing offshore renewable energy farms as their foundation. In essence, the farms are colocated between turbines or solar plants. One project is colocated with a seafloor wind farm (the Norther Wind Farm), and the other is colocated with the North Sea Farmers offshore floating solar farms (United). The goal of these pilot projects is to move forward with almost complete automation of the growth and harvesting of seaweed by using some of the renewable power generated in situ. Results of these pilot projects are slated for 2022 (Durakovic, 2020).8
Lessons learned from these and other pilot studies will be important to quantify life-cycle cost analyses that will be needed for determining the net benefits of seaweed cultivation to CDR.
There are a wide variety of co-benefits created by seaweed cultivation CDR. Building, operating, and maintaining farms will provide a huge enhancement to the blue economy. Seaweed cultivation CDR could provide jobs and livelihoods for many. The quality of these jobs matters to people: for example, in a study of attitudes about seaweed cultivation generally, interviewees in Scotland were wary of large-scale internationally owned seaweed farms. They have witnessed a pattern in
the salmon farming industry where individual operators were bought out by international owners, which was perceived to reduce benefits and jobs. They emphasized the need for social innovation that would provide decent pay for all involved and benefits for the community more broadly, and were more accepting of locally based, cooperative development (Billing et al., 2021).
Seaweed cultivation also has potential ecological co-benefits. On local scales, seaweed farming, particularly in coastal waters, could act to reduce the effects of anthropogenically driven acidification on shellfish aquaculture farms and could help reduce the effects of deoxygenation and eutrophication created by these farms (e.g., Neori et al., 2004; Xiao et al., 2017). Other co-benefits with potential greenhouse gas reductions include the additions of macrophyte biomass to animal feeds that could reduce methane emissions (e.g., Maia et al., 2016)
The legal framework for ocean CDR is discussed in Chapter 2. Many of the international and domestic laws discussed in that chapter could apply to seaweed cultivation.
With respect to international law, Webb et al. (2021) concluded that seaweed cultivation undertaken for the purpose of CDR is likely to be considered a form of “geoengineering” under the Convention on Biological Diversity (CBD). As discussed in Chapter 2, in 2010, the parties to the CBD adopted a nonbinding decision recommending that countries take a “precautionary approach” and avoid geoengineering activities that could affect biodiversity.9 The decision provided an exemption for “small scale scientific research studies . . . conducted in a controlled setting” that “are justified by the need to gather specific scientific data and . . . subject to a thorough prior assessment.”10 However, because the decision is not legally binding and merely offers guidance on the conduct of geoengineering activities, countries could conduct or authorize other projects that do not meet the specific requirements (Webb et al., 2021).
Seaweed cultivation projects could also implicate provisions of the United Nations Convention on the Law of the Sea (UNCLOS). Parties to UNCLOS must “take all necessary measures” to prevent and control pollution resulting from, among other things, the introduction of alien species to “a particular part of the marine environment” where they “may cause significant and harmful changes.”11 Prior to conducting or authorizing seaweed cultivation projects involving the growing of nonnative species, parties to UNCLOS would need to conduct a risk assessment, consult with other potentially affected countries, and take other steps to minimize any adverse effects (Webb et al., 2021).
It is unclear whether the Convention on the Prevention of Marine Pollution by Dumping of Wastes and Other Matter (London Convention) and associated Protocol (London Protocol) would apply to seaweed cultivation projects. There is an open question as to whether the sinking of cultivated seaweed could constitute “dumping” for the purposes of the London Convention and Protocol (Webb et al., 2021). In theory, “dumping” could also occur if growth-stimulating materials were added to ocean waters to fertilize seaweed crops, and/or waste products (e.g., nets and lines) from farms were disposed of in the water (Webb et al., 2021).
9 Report of the Conference of the Parties to the Convention on Biological Diversity on the Work of its Tenth Meeting, Decision X/33 on Biodiversity and Climate Change, Oct. 29, 2010 (hereinafter Decision X/33). See also Report of the Conference of the Parties to the Convention on Biological Diversity on the Work of its Eleventh Meeting, Decision XI/20 on Climate-related Geoengineering, Dec. 5, 2012 (hereinafter Decision XI/20); Report of the Conference of the Parties to the Convention on Biological Diversity on the Work of its Thirteenth Meeting, Decision XIII/4, Dec. 10, 2016 (hereinafter Decision XIII/4).
10 Para. 8(w), Decision X/33. Affirmed in Para. 1, Decision XI/20 & Preamble, Decision XIII/4.
11 Art. 196, United Nations Convention on the Law of the Sea, 1833 U.N.T.S. 397, Dec. 10, 1982.
Webb et al. (2021) examined the application of domestic law to seaweed cultivation projects. The applicable law will depend on precisely where and how projects are conducted. State, and in some cases local, law would apply to projects undertaken in state waters (i.e., generally within 3 nautical miles of the coast). There has been no comprehensive analysis of all applicable state and local laws. However, Webb et al., (2021) found that at least three states—Alaska, California, and Maine—have laws requiring permits or other approval for seaweed cultivation projects. Webb et al. (2021) noted that some other states have more general aquaculture permitting laws that could apply to seaweed cultivation. However, some only provide for the issuance of permits for shellfish or finfish farming and do not anticipate the permitting of seaweed cultivation. Federal permits (e.g., from the U.S. Army Corps of Engineers) may also be required for some seaweed cultivation projects in state waters (e.g., those involving the placement of structures in the water). State and federal agencies must generally consult with affected Native American tribes prior to issuing permits (Webb et al., 2021).
Federal law would apply to seaweed cultivation projects undertaken in federal waters (i.e., generally 3 to 200 nautical miles from the coast). Webb et al. (2021) reviewed the potentially applicable federal laws. Although there is no federal permitting regime for seaweed cultivation, projects that require use of the seabed (e.g., to anchor structures or lines) may require a seabed lease or other authorization (e.g., under the Outer Continental Shelf Lands Act). Permits would also be required under the Marine Protection, Research, and Sanctuaries Act (MPRSA) for projects involving the dumping of materials, including nets and lines, into ocean waters. It is uncertain whether the sinking of cultivated seaweed would constitute dumping and thus require a permit. Projects affecting other marine species or ecosystems may be subject to additional requirements, for example, under the Endangered Species Act (ESA) and Marine Mammal Protection Act (MMPA).
Domestic law would have limited application to seaweed cultivation projects outside U.S. waters (i.e., more than 200 nautical miles from shore). Where U.S. citizens or vessels are involved, the ESA and MMPA could apply. Dumping from U.S. vessels, or other vessels that were loaded in the United States, may also be subject to domestic permitting requirements under the MPRSA.
Community engagement will be required for effective governance, especially when activities affect coastal populations and livelihoods (see Chapter 2). Community engagement will also be critical in capturing potential co-benefits.
A research agenda for assessing whether seaweed cultivation and sequestration is CDR worthy will require an assessment of the components compiled in Figure 5.2. These are
- Improve existing technologies that enable the cost-effective, large-scale farming and harvesting of seaweed biomass;
- Create and assess the means to convey large amounts of harvested biomass to depth in the ocean interior or to the seafloor without large losses of carbon;
- Understand the long-term fates of seaweed carbon (i.e., both biomass at depth and DOC released during growth) and use this understanding to develop numerical models of the fates of seaweed carbon in the environment;
TABLE 5.1 CDR Potential of Seaweed Cultivation
What is known about the system (low, mostly theoretical, few in situ experiments; medium, lab and some fieldwork, few carbon dioxide removal (CDR) publications; high, multiple in situ studies, growing body of literature)
Science of macrophyte biology and ecology is mature; many mariculture facilities are in place globally. Less is known about the fate of macrophyte organic carbon and methods for transport to deep ocean or sediments.
What is the confidence level that this approach will remove atmospheric CO2 and lead to net increase in ocean carbon storage (low, medium, high)
The growth and sequestration of seaweed crops should lead to net CDR. Uncertainties about how much existing net primary production (NPP) and C export downstream would be reduced due to large-scale farming.
Will it remove CO2 durably away from surface ocean and atmosphere (low, <10 years; medium, >10 years and <100 years; high, >100 years), and what is the confidence (low, medium, high)
Dependent on whether the sequestered biomass is conveyed to appropriate sites (e.g., deep ocean with slow return time of waters to surface ocean).
What is the potential scalability at some future date with global-scale implementation (low, <0.1 Gt CO2/ yr; medium, >0.1 Gt CO2/yr and <1.0 Gt CO2/yr; high, >1.0 Gt CO2/yr), and what is the confidence level (low, medium, high)
Potential C removal >0.1 Gt CO2/yr and <1.0 Gt CO2/ yr
Farms need to be many million hectares, which creates many logistic and cost issues. Uncertainties about nutrient availability and durability of sequestration, seasonality will limit sites, etc.
Intended and unintended undesirable consequences at scale (unknown, low, medium, high), and what is the confidence level (low, medium, high)
Environmental impacts are potentially detrimental especially on local scales where seaweeds are farmed (i.e., nutrient removal due to farming will reduce NPP, C export, and trophic transfers) and in the deep ocean where the biomass is sequestered (leading to increases in acidification, hypoxia, eutrophication, and organic carbon inputs). The scale and nature of these impacts are highly uncertain.
Encompass use conflicts, governance-readiness, opportunities for livelihoods, etc.
|Possibility for jobs and livelihoods in seaweed cultivation; potential conflicts with other marine uses. Downstream effects from displaced nutrients will need to be considered.|
How significant are the co-benefits as compared to the main goal of CDR and how confident is that assessment
Placing cultivation facilities near fish or shellfish aquaculture facilities could help alleviate environmental damages from these activities. Bio-fuels also possible.
|Cost of scale-up
Estimated costs in dollars per metric ton CO2 for future deployment at scale; does not include all of monitoring and verification costs needed for smaller deployments during R&D phases Low, <$50/t CO2; medium, ~$100/t CO2; high, >>$150/t CO2 and confidence in estimate (low, medium, high)
Costs should be less than $100/t CO2. No direct energy used to fix CO2.
|Cost and challenges of carbon accounting
Relative cost and scientific challenge associated with transparent and quantifiable carbon tracking (low, medium, high)
The amount of harvested and sequestered carbon will be known. However, an accounting of the carbon cycle impacts of the displaced nutrients will be required (additionality).
|Cost of environmental monitoring
Need to track impacts beyond carbon cycle on marine ecosystems (low, medium, high)
All CDR will require monitoring for intended and unintended consequences both locally and downstream of CDR site, and these monitoring costs may be a substantial fraction of overall costs during R&D and demonstration-scale field projects.
|Additional resources needed
Relative low, medium, high to primary costs of scale-up
Farms will require large amounts of ocean (many million hectares) to achieve CDR at scale.
- Build and test a demonstration-scale system for seaweed cultivation and sequestration CDR that in principle can be scaled up to 0.1-Gt CO2/yr levels and deploy these systems in diverse oceanographic settings;
- Validate and monitor the CDR performance of the demonstration-scale seaweed farming and sequestration systems on local scales;
- Evaluate the environmental impacts of large-scale seaweed farming and sequestration systems both in the upper ocean where the farming occurs and at depth where the seaweed is transported for sequestration;
- Understand better the legal framework required for seaweed-based CDR and the socioeconomic factors that would affect coastal communities and Indigenous groups; and
- Document “best practices” and perform spatial planning exercises to assess the best places for conducting seaweed cultivation CDR.
The first research component should build upon the successes achieved and challenges identified by ARPA-e’s MARINER program. ARPA-e has made a substantial investment in MARINER (>$30M over 3 years), and work in this area needs to continue so that development-scale facilities can be developed. Given the investments made to date, this work can be done quickly, with development-scale (≥1 km2; ≥100 hectare) farms in place in the next decade. At CDR scale, many million hectares of farm facilities need to be established. Hence, seaweed cultivation systems will very likely need to be engineered to operate efficiently at ocean depths of hundreds to thousands of meters. This engineering expertise exists within the oil and gas industries. Research should also include a focus on increasing the farm’s C yield and reducing crop durations via the selection of appropriate phenotypes and the deployment of specific apparatuses (e.g., AU, vertically profiling farms, etc.) that will enhance nutrient concentrations in the farm. Improvements in the abilities to monitor and model crop growth as a function of environmental conditions are also needed. Further, efficient ocean transportation systems that minimize their C emissions and appropriate harvesting techniques need to be developed so that the CDR gained is not lost due to transport and to protect those organisms that forage or inhabit the farms. Last, permitting has been a challenge in developing and testing pilot-scale farms in the MARINER program and will likely be a challenge for further research. Addressing permitting challenges is essential to enable researchers to develop and test new farming and harvesting technologies.
Harvested biomass needs to be conveyed to depth without losing C biomass in the surface ocean and deposited in durable oceanic reservoirs. There has been little research conducted in this area. Demonstration-scale engineering studies need to be conducted, illustrating that the harvested biomass can be conveyed mechanically to a durable reservoir in the ocean with minimal losses, either in the water column or on the seafloor. Freely sinking biomass has also been suggested, and the decomposition rates of freely sinking biomass need to be fully quantified to assess whether this conveyance strategy would be successful. Research budgets in this area should be comparatively modest.
Assessments of the long-term fates of seaweed biomass and its by-products (cf., both biomass at depth and released DOC during growth) are required to develop predictive models. This could be accomplished via a set of biomass and dissolved organic matter decomposition experiments conducted in situ, in mesocosms, or in laboratory settings. The results of this need to be synthesized into a numerical model of seaweed fates. Challenges are to conduct these experiments on timescales that are relevant to seaweed cultivation CDR—years to decades. This work also needs to lead to predictive numerical models where farm, harvest, and sequestration scenarios can be explored.
Demonstration-scale systems for seaweed cultivation and sequestration aimed to be scaled to 0.1-Gt CO2/yr levels need to be developed and tested. This work will build from the first two components in this research plan and logically are supported after clear paths emerge from these two elements. This will answer the question whether seaweed cultivation and sequestration is a viable CDR strategy. The demonstration-scale system will also be useful for validating its CDR performance as well as assessing its environmental impacts (see below). Many of the same environmental and social impacts of the research for farming apply to the implementation of a demonstration-scale system. These goals would be most readily achieved if these systems were deployed in diverse ocean settings.
The CDR performance of the demonstration-scale seaweed farming and sequestration system will need to be monitored on local spatiotemporal scales. This will require both process oceanographic field studies (similar to the recent NASA EXPORTS campaign; Siegel et al., 2021b) and selected sensor arrays embedded in the farm and conveyance infrastructure. These data will address the system performance of a demonstration-scale system. Regional-scale numerical modeling of ocean circulation and mixing coupled with ecological and biogeochemical modules will also be a big part of the validation and monitoring of these systems, which in turn will require data from the process studies and system models on macrophyte fates and influences of the farms and harvesting on the environment (discussed previously).
One would also need to evaluate the environmental impacts of large-scale seaweed farming and sequestration both in the upper ocean where the farming occurs and at depth in the water column or the seafloor where the seaweed is conveyed for sequestration. This work would need to include the downstream impacts of displaced nutrients on ecosystems and ecosystem services. Fieldwork would also be needed to achieve the previous task (validating and monitoring CDR performance). Hence, portions of these tasks could be done simultaneously. Additional measurements would be needed to understand the effects of seaweed sequestration on the macrofaunal communities in the water column and on the seafloor.
Research on the social/economic factors and governance for seaweed cultivation CDR is also required. The legal framework for seaweed-based CDR is murky, and many questions remain unanswered. Further, many socioeconomic factors affect coastal communities and Indigenous groups that need to be considered. Additionally, an understanding of public perceptions and whether there is a social license to conduct this work is required.
Last, the above components need to be synthesized into a “best practices” manual, and spatial planning exercises need to be performed to assess the best places for conducting seaweed cultivation
CDR. This synthesis of the emerging state of knowledge on seaweed cultivation CDR needs to be completed before one should consider implementing these technologies at scale.
In summary, seaweed cultivation and sequestration could be a compelling ocean CDR strategy (see Table 5.2). There is a good understanding of the underlying biology, ecology, and biogeochemistry of macrophytes and their cultivation, although many advances are needed to grow seaweed biomass to meet CDR requirements. In principle, it should work (i.e., reduce atmospheric CO2), but there is a large degree of uncertainty about how much productivity and C export it would displace from planktonic ecosystems and the durability of the sequestered carbon if not properly conveyed to an appropriate site. Scaling to CDR-worthy levels (≥0.1 Gt CO2/yr) will be difficult due to the large amount of farming area required. However, much has been learned already in the MARINER program, and there should be recognition of the many marine engineering accomplishments made by the global oil and gas industries to date. The costs should be less than $100/metric ton CO2; assuming that the MARINER’s cost target for growing macrophyte biomass is met ($75/metric ton CO2), the other costs (e.g., conveyance, monitoring, etc.) should be smaller. Research needs to be continued to help ensure that this cost target is achieved. The energy expenditures should be small relative to some other CDR strategies because solar energy can fix CO2 into organic matter. On the other hand, several potentially detrimental environmental factors exist where farming occurs and where the biomass is sequestered. The scale of these impacts is highly uncertain at this time. There are both positive and negative social impacts from CDR via seaweed cultivation and sequestration. If conducted at scale, it will enhance the blue economy, which will benefit both coastal communities and many marine industries. There may also be several co-benefits from placing farms adjacent to other uses (e.g., fish farming, etc.), which may help mitigate some environmental damages conducted by aquaculture facilities. On the negative side, the vast farms represent hazards to navigation, and they may displace fishing and other uses via the placing of farms or the reduction in planktonic productivity and trophic exchanges due to the large-scale cultivation of seaweed biomass.
TABLE 5.2 Research and Development Needs: Seaweed Cultivation and Sequestration
|No.||Recommended Research||Question Answered||Environmental Impact of Research||Social Impacts of Research||Estimated Research Budget ($M/yr)||Time Frame (years)|
|5.1||Technologies for efficient large-scale farming and harvesting of seaweed biomass||Can we build efficient demonstration-scale farms to grow and harvest seaweed biomass that have the potential to be CDR worthy?||Moderate on local scales. Demonstration-scale farms will affect local ecosystems, reducing ambient nutrients, net primary productivity, C export, etc. Farming of nonnative cultivars would be an environmental risk. Farms could also reduce the effects of acidification in the upper ocean.||Concerns with permitting due to macrofaunal entanglements, introduction of nonnative cultivars, hazards to navigation, displacing fishing effort, etc.||15 based upon MARINER funding as a starting point||10|
|5.2||Engineering studies focused on conveying harvested biomass to a durable oceanic reservoir with minimal losses of carbon||How do we convey large amounts of seaweed biomass to depth or seafloor with minimal losses?||Minimal for engineering testing||Minimal for engineering testing||2||10|
|5.3||Assessment of long-term fates of seaweed biomass and byproducts||Can we predict the long-term fates of seaweed carbon?||Minimal||Minimal||5||5|
|5.4||Implement and deploy a demonstration-scale seaweed cultivation and sequestration system||Can a system be built that will scale to CDR-worthy scales?||Moderate on local scales, affecting upper ocean and conveyance depths||Moderate—Concerns with permitting due to macrofaunal entanglements, introduction of nonnative cultivars, hazards to navigation, displacing fishing effort, etc.||10||10 starting 5 years from now|
|5.5||Validate and monitor the CDR performance of a demonstration-scale seaweed cultivation and sequestration system||Is seaweed cultivation and sequestration a viable CDR strategy and can we monitor its performance?||Minimal||Minimal||5||10 starting 5 years from now (done in concert with Task 5.6)|
|5.6||Evaluate the environmental impacts of large-scale seaweed farming and sequestration||What are the environmental impacts of seaweed sequestration?||Minimal||Minimal||4||10|
|5.7||Research on the socioeconomic factors and governance||Would seaweed cultivation CDR affect communities and stakeholders?||N/A||N/A||1||2|
|5.8||Document “best practices” and perform spatial planning exercises||How should seaweed cultivation be conducted and where?||N/A||N/A||1||2|
NOTE: Bold type identifies priorities for taking the next step to advance understanding of seaweed cultivation as an ocean CDR approach.
This page intentionally left blank.