香京julia种子在线播放

Ecosystems have different sensitivities to ocean warming, ocean acidification and sea-level rise (Figure 1A and section “SM3.3” of the Supplementary Materials). Interactions between drivers can be complex: additive, synergistic, or antagonistic (Crain et al., 2008). There are big gaps in multiple-drivers research (Crain et al., 2008; Riebesell and Gattuso, 2015) but experimental strategies to assess the biological ramifications of multiple drivers of global ocean change have become available (Boyd et al., 2018).

Of the four ecosystems or habitats considered here, coral reefs and Arctic biota are the most imminently threatened and will be affected to a greater degree sooner than others, with high risk that key functions will be lost globally, as identified in the 5th assessment report (AR5) of the Intergovernmental Panel on Climate Change (IPCC) (Hoegh-Guldberg et al., 2014; Pörtner et al., 2014). Coral reefs are very sensitive to ocean warming and acidification (Hoegh-Guldberg et al., 2007; Gattuso et al., 2014; Hughes et al., 2017a). They have suffered extensive losses in the past three decades due to high sea surface temperature combined with local stressors such as overfishing, destructive fishing, coastal development, and pollution. All projections indicate that the thermal conditions driving major losses will increase in frequency and exceed thresholds for the majority of reefs by 2050 (Gattuso et al., 2014; Pörtner et al., 2014). Over the 21st century under the high emissions Representative Concentration Pathway RCP8.5 scenario (van Vuuren et al., 2011), 99% of the world’s coral reefs are expected to experience annual severe bleaching due to thermal stress (van Hooidonk et al., 2016). The thermal sensitivity of coral reefs is compounded by ocean acidification (Hoegh-Guldberg et al., 2014), which diminishes coral growth and calcification (Albright et al., 2018) and can lead to increased bioerosion and vulnerability to storm damage (Andersson and Gledhill, 2013).

Arctic biota are also highly sensitive to climate change, particularly ice-associated biota that are rapidly declining in Arctic summers (Wassmann et al., 2010; Pörtner et al., 2014; Kohlbach et al., 2017). Within the Arctic, ecosystem responses vary greatly depending on ambient variability, degree of warming, and nutrient advection (Hunt et al., 2016). Warming and freshening may also impact ecosystem production by differentially increasing respiration rates and reducing nutrient supply (Duarte et al., 2012) as well as enhancing the degree of ocean acidification due to freshening (Pörtner et al., 2014). Arctic organisms that seem particularly sensitive to ocean acidification include calcifiers such as bivalves and planktonic pteropods that are key links in ocean food webs (Comeau et al., 2010; Duarte et al., 2012).

Mangroves and saltmarshes are highly sensitive to sea-level rise (Kirwan and Megonigal, 2013; Lovelock et al., 2015), particularly where coastal development and steep topography block landward migration and insufficient sediment is delivered to support accretion. A preliminary global modeling effort suggests that a 50 cm sea-level rise by 2100 would result in a loss of 46 to 59% of global coastal wetlands (up to 78% loss under 110 cm rise), but losses are sensitive to assumptions about human coastal development and may be reduced if additional tidal hydrodynamic feedbacks are included (Spencer et al., 2016). Warming and acidification are not projected to have significant direct effects on mangroves and saltmarshes, but may have positive or negative effects at local scales due to changes in species composition, phenology, productivity, and latitudinal range of distribution (Ward et al., 2016).

Temperate seagrass ecosystems are sensitive to ocean warming. For example, the thermal regime of the Mediterranean Sea already exceeds the upper thermal limit of the endemic Posidonia oceanica in some areas (Marbà and Duarte, 2010; Jordà et al., 2012). Seagrass and fleshy algae may expand in Arctic regions with warming and loss of ice cover (Krause-Jensen and Duarte, 2014). Some may benefit from carbonate chemistry changes associated with ocean acidification as their photosynthesis is CO₂-limited (Raven and Beardall, 2014) but sensitive calcifiers growing in the meadows are negatively impacted (Martin et al., 2008).

The ecosystem services considered in this study are all highly sensitive to ocean warming (Weatherdon et al., 2016; Figure 1B and section “SM3.3” of the Supplementary Materials). Global potential fisheries catches and species turnover, for instance, are projected to decrease by about 3 Mt and increase by 10%, respectively, for every 1°C of global surface warming (Cheung et al., 2016). These patterns are similar for finfish and shellfish aquaculture, as ∼90% of current finfish and shellfish mariculture production is from open-water farming where environmental conditions closely match those in the nearby ocean (Callaway et al., 2012). Shellfish fisheries and mariculture, in particular, are threatened by the combined effects of warming (Mackenzie et al., 2014), ocean acidification (Barton et al., 2012; Gazeau et al., 2013) and deoxygenation (Gobler et al., 2014). Despite possible genetic adaptation over generations (Thomsen et al., 2017), impacts on shellfish are expected to be high to very high when CO₂ concentrations exceed those projected for 2100 in the low to moderate RCP2.6 and 4.5 CO₂ emissions scenarios (Gattuso et al., 2015; Cooley et al., 2016). In addition, finfish mariculture often focuses on high trophic level species that are dependent on wild capture fisheries for feed (Troell et al., 2014) and some operations still largely rely on wild captured fish fry and juveniles (Diana, 2009). Thus, mariculture is likely to be subject to similar climatic stresses as fish stocks in the wild.

The sensitivity of coastal protection, notably wave attenuation and shoreline stabilization, to climate-related drivers differs for each ecosystem considered (Spalding et al., 2014). The cumulative impacts of increasing sea-surface temperature, ocean acidification, and non-climatic stressors such as land-based pollution reduce reefs’ ability to keep pace with sea-level rise (Yates et al., 2017). The consequences of sea-level rise on biologically structured coastal ecosystems raise concerns as these habitats are estimated to currently reduce wave height by 30 to 90% (in order of highest to lowest wave reduction: coral reefs, saltmarshes, mangroves, and seagrasses) (Fonseca and Cahalan, 1992; Duarte et al., 2013; Ferrario et al., 2014; Narayan et al., 2016). Historical global losses in coastal ecosystems [30 to 50% for mangroves since the 1940s, 29% for seagrass since 1879, 25% for saltmarshes since the 1800s (Waycott et al., 2009; Mcleod et al., 2011)] and degradation of coral reefs [30–75% since prehuman times (Pandolfi et al., 2003)] have already reduced their potential to provide ecosystem services. Projections suggest that 90% of coral reefs worldwide could be lost if warming exceeds 1.5°C (Frieler et al., 2013).

To estimate effectiveness, we first assess the potential of each measure –assumed here to be implemented at its maximum physical capacity– to bridge the gap between the high-emissions trajectory (RCP8.5, our baseline scenario) and a stringent emission-reduction scenario (RCP2.6) expected to keep mean global temperature increase below 2°C by 2100 (van Vuuren et al., 2011) (see section “SM2” of the Supplementary Materials). The differences between RCP8.5 and RCP2.6 in the year 2100 are estimated to be ∼1,400 Pg C for avoided emissions; ∼2°C for reduced sea surface warming; ∼0.25 pH units for avoided sea surface acidification; and a reduction in sea-level rise of between 0.26 and 1.1 m (Jones et al., 2013; DeConto and Pollard, 2016).

The effectiveness of the global measures is assessed in terms of maximum possible effectiveness to reduce ocean warming, ocean acidification, and sea-level rise (Figure 4A), and duration of the effect (Figure 4B). This maximum effectiveness is theoretical and almost certainly not achievable but provides the full potential of each approach. Two of the global solutions, renewable energy and alkalinization, stand out as having the highest theoretical potential for addressing all drivers (Figure 4A). This is obvious for renewable energy because of the enormous energy potential of tides, waves, ocean currents, and thermal stratification, estimated at up to 7,400 EJ year^-1 (Rogner et al., 2000; Lewis et al., 2011) and well exceeding future human energy needs. Any replacement of fossil fuels by marine renewables results in permanently avoided greenhouse gas emissions.

A similarly large and permanent intervention could be provided by large-scale alkalinization, by which CO₂ is consumed and stored either as dissolved bicarbonate and carbonate ions or as precipitated calcium carbonate, neutralizing ocean acidity. However, the feasibility and benefits of doing this must be weighed against the financial costs and environmental impacts of mining or producing vast quantities of alkaline material, distributed at global scales, and the potential biotic impacts of the trace elements or contaminants that alkalinity might contain (Renforth and Henderson, 2017).

Land-ocean hybrid methods greatly expand the mitigation potential offered by either land-based or ocean-based approaches individually. For example, the use of marine biomass for bioenergy with carbon capture and storage (BECCS) fuel eliminates limitations on terrestrial fuel capacity posed by competition for land, water, and nutrients. In turn, conversion of CO₂ from land-based biomass energy to ocean alkalinity and subsequent storage in the ocean greatly expands CO₂ storage capacity and beneficial use (via countering ocean acidification) relative to more conventional CCS approaches. However, a comprehensive understanding of the full range of options, and their costs, benefits and tradeoffs requires further research (Rau, 2014).

Albedo enhancement also has a very large potential effectiveness in moderating warming (Figure 4A), as a relatively small enhancement of the albedo of the dark ocean surface by less than 0.05 could compensate the entire GHG-driven perturbation in the Earth’s radiation balance (Crook et al., 2016; Garciadiego Ortega and Evans, 2018). However, the duration of the effect is only as long as the albedo stays high, likely to be days to months for ocean foams (Figure 4B) and, as SRM in general, it does not limit ocean acidification as atmospheric CO₂ concentration remains elevated (Tjiputra et al., 2016). Similar considerations apply to marine cloud brightening, although modeling studies indicate more limited effectiveness (Kravitz et al., 2013; Stjern et al., 2017).

Other potential solutions face physical and/or biogeochemical limitations (Figure 4A). A global deployment of iron fertilization for 100 years could sequester a maximum of ∼70 Pg C (ref. Aumont and Bopp, 2006) because other nutrient or light limitations occur when marine algae are iron-replete (Oschlies et al., 2010). Some measures demonstrate limited potential for reducing warming, acidification and sea-level rise at global scales, such as vegetation for instance. Even with very high carbon storage and avoided net emissions, the vegetation measure is constrained by the limited global area of potentially vegetated habitats, although with some scope to artificially expand that area; e.g., via seaweed aquaculture (Duarte et al., 2017; Hawken, 2017).

Local measures have a relatively low effectiveness to reduce warming, acidification, and sea-level rise at the global scale (Figure 4A). However, some have a high to very high effectiveness to moderate local ocean acidification (pollution reduction and alkalinization) and relative sea-level rise (vegetation, protection, restoring hydrology, as well as relocation and reef restoration).

The duration of the effects varies greatly between the different methods (Figure 4B). It is close to permanent for renewables as long as the infrastructure is maintained. The effects of protection are also considered permanent as long as MPAs are enforced, although future climate change will decrease their ability to provide climate mitigation and adaptation benefits (Bruno et al., 2018). The effects of vegetation can be close to permanent as long as these ecosystems are maintained or increased in the face of natural and anthropogenic pressures. In contrast, the effects of fertilization have a finite duration. Once iron fertilization is stopped, a large portion of the additional ocean carbon uptake will outgas back to the atmosphere on decadal to centennial time scales (Aumont and Bopp, 2006). By capturing and storing CO₂ for long time periods or permanently, alkalinization and hybrids methods such as conversion of CO₂ to ocean alkalinity or marine BECCS generally have long duration of the effect. In contrast, the effect of albedo enhancement and cloud brightening is short-lived (days to weeks). The loss of most benefits following abrupt termination is a characteristic of all SRM schemes (Jones et al., 2013). It is projected to increase both ocean and land temperature velocities to unprecedented speeds (Trisos et al., 2018).

Technical feasibility is evaluated by considering current technological readiness (ranging from schemes at the concept stage to schemes already deployed) and for lead time until full potential effectiveness, i.e., the time needed to reach full implementation (ranging from days to decades; see section “SM2” of the Supplementary Materials). Two local measures have the highest technical feasibility (Figure 4B): protection and restoring hydrology. Vegetation (both global and local) and renewable energy also have a high technical feasibility, closely followed by eliminating overexploitation, reducing pollution and relocation and reef restoration. Five global schemes have the lowest technical feasibility: fertilization, cloud brightening, alkalinization, albedo enhancement, and hybrid methods. The local measure assisted evolution also scores very low on this criterion. These low scores generally reflect lack of testing and deployment at scale, thus they also possess high uncertainty.

The cost effectiveness of the global and local solutions is assessed, in US$ per tonne of CO₂ emissions reduced and in US$ per hectare of surface area of implementation, respectively (Figure 4E and section “SM3.5” of the Supplementary Materials). The costs considered here are best estimates from the literature for the direct monetary costs of implementation. The non-monetary costs of implementation are considered through assessing co-benefits, disbenefits, and governability, as discussed below. Since cost effectiveness is a relative metric, it does not reflect the effectiveness of a measure to reduce changes in the drivers. For instance, cloud brightening is cost-effective despite having a moderate maximum effectiveness to moderate ocean warming, ocean acidification, and sea-level rise (Figures 2, 4A). Restoration of vegetation to increase CO₂ capture has a very low cost effectiveness but conservation of vegetation to avoid further emissions is very cost-effective. For example, conserving mangroves to avoid further CO₂ emissions is considerably cheaper than restoring mangroves to enhance CO₂ uptake [4–10 vs. 240 US$/t CO₂ (Siikamaki et al., 2012; Bayraktarov et al., 2016)]. Cloud brightening, protection, and renewable energy have the highest cost efficiency while albedo enhancement, vegetation and relocation and reef restoration have the lowest. Note that cost effectiveness generally increases over time and with increasing scale of implementation, due to learning and economies of scale, and that there is uncertainty in many of these estimates (see section “SM3.5” of the Supplementary Materials) as reflected in the low levels of confidence. This generally is a consequence of lack of economic data from testing/deployment of many of these methods at relevant scales.

Governance is the “effort to craft order, thereby to mitigate conflict, and realize mutual gains” (Williamson, 2000) amongst actors from public, private, and civil society sectors. Here, we assess the governability of global and local ocean measures in terms of the potential capability of the international community to implement them, managing associated conflicts and harnessing mutual benefits (see section “SM2.9” of the Supplementary Materials). We focus on the international dimension of decision and action to reflect the global scope of the study, despite the fact that we recognize that global and local measures do not face the same constrains for implementation – e.g., bi- or multi-lateral diplomatic issues for the former (e.g., Smit, 2014; Cinner et al., 2016; Rabitz, 2016) and local institutional and population reluctance challenges for the latter (e.g., Cinner et al., 2016).

On that basis, the governability of a scheme increases with its effectiveness (Ostrom, 2007), the predictability of its effects (Hagedorn, 2008; Ostrom, 2009), its co-benefits, the absence of disbenefits together with the presence of national-level net benefits, the presence of enabling institutions and the absence of constraining institutions, and higher normative consensus amongst relevant actors (Abbott and Snidal, 1998; Barrett, 2005). Global governability is likely to be much higher when there are national-level net benefits (i.e., national benefits outweigh the negative environmental impacts and national costs of implementation), since single nation states may then implement measures without having to rely on international cooperation (Kaul et al., 1999). This is the case for protection, vegetation as well as for relocation and reef restoration (Figure 4E and section “SM3.6” of the Supplementary Materials). Conversely, ocean-based SRM measures (cloud brightening and albedo enhancement), while being more effective in addressing drivers globally, are considered to have low governability because their implementation generally involves international cooperation to solve the free-riding dilemma with regard to global public goods (Pasztor et al., 2017). Thus nations are likely to be reluctant to unilaterally take on extra costs that may reduce their own economic competitiveness (Preston, 2013; Rabitz, 2016; Williamson and Bodle, 2016). Additionally, SRM measures entail potentially significant disbenefits and high uncertainties (Figures 4D, 5; sections “SM3.4 and SM3.4.3” of the Supplementary Materials), which further reduce their present governability. Renewable energy is in an intermediate position: renewables are becoming more economically competitive compared with fossil-fuel based energy, thereby providing national-level incentives to implementation. Taken together, the scores exhibit a fundamental tradeoff in climate policy: global measures are more effective than local ones in addressing the climate problem, but they are in general more difficult to implement due to challenges in global governance.

A principal components analysis (see section “SM4” of the Supplementary Materials) was used to reduce the eight dimensions of our assessment dataset defined by the scoring criteria to two latent dimensions that explain most of the variance in the assessment data. Three clusters of schemes emerge (Figure 6). The first one includes alkalinization at the global scale, hybrid methods, albedo enhancement, and cloud brightening, which show high potential effectiveness to reduce warming and acidification, and their impacts. However, there has been relatively little research, testing and application on such solutions, and they generally score low for technological readiness, co-benefits, lack of disbenefits, and global governability. In contrast, the second cluster includes almost all local measures (protection, reducing pollution, vegetation at the local scale, eliminating overexploitation, restoring hydrology and relocation and restoration), and is characterized by low effectiveness to reduce warming and its impacts, and moderate effectiveness to reduce ocean acidification and relative sea-level rise and their impacts. These measures are, however, technologically ready, have significant co-benefits, few disbenefits and can also help to reduce the impacts of non-climatic drivers. Renewable energy stands apart as it exhibits both high potential effectiveness and technology readiness, thus ranging in between clusters 1 and 2. The third cluster includes assisted evolution, alkalinization at the local scale and fertilization, which have low to moderate scores across most criteria assessed.

Measures which are the most technically feasible also have the highest global governability (Figure 4). They comprise protection, eliminating overexploitation, reducing pollution, vegetation, relocation and reef restoration, and renewable energy. Except for the latter, all these measures are local. Their governability is high to very high except for restoring hydrology and assisted evolution (moderate or low). Global measures such as albedo enhancement, fertilization, hybrid methods, cloud brightening, and alkalinization have a lower overall technical feasibility, partly due to lack of testing and experience, together with moderate to low governability. Yet none of these schemes do much to reduce or moderate the impacts of the climate-related drivers considered in this study (ocean warming, ocean acidification and sea-level rise).

Such conclusions highlight the need for multiple-scale and multiple-stakeholder initiatives, hence calling for improved international governance mechanisms to ensure coherency in ocean-based climate action. These governance challenges are, however, constrained by controversies on the potential solutions, which scientific investigations and policy engagement can help overcome. Controversies are mostly in the “addressing the causes of climate change” and “solar radiation management” areas of action (Figure 2). They include: the moral hazard dilemma, i.e., that development and deployment of alternative solutions might decrease effort on emission reductions (Preston, 2013; McLaren, 2016); the risk of premature lock-in of suboptimal solutions and path dependencies (Burns et al., 2016; Reynolds et al., 2016); and concerns regarding controllability and transnational effects (Williamson and Bodle, 2016). Ethical issues are also important, relating to informed consent and potential adverse impacts on countries unable to deploy such measures (Svoboda, 2012; Suarez and van Aalst, 2017; Rahman et al., 2018); and vested interests, as production and deployment of innovative measures could be a highly profitable market (Preston, 2013). Controversies related to the “protection of biota and ecosystems” and “manipulation to enhance biological and ecological adaptation” areas of action mostly arise from conflicts relating to local, national and global-scale interests, and the balance between short-term and long-term benefits and disbenefits (Cooley et al., 2016; Cormier-Salem and Panfili, 2016). Such trade-offs between “winners” and “losers” highlight the influence of social norms and values that may differ greatly between different stakeholders (Hopkins et al., 2016; Lubchenco et al., 2016). Testing the veracity of such perceptions via further research and demonstration of novel measures at relevant scales will clarify governance issues.

The global implementation or testing of renewable energy, fertilization, vegetation, eliminating overexploitation, and protection has accelerated sharply in the past two decades (Figure 3). In particular, several local measures (vegetation, protection, and eliminating overexploitation) may achieve their full potential in a few decades at their current rate of deployment.

Nevertheless, the scale of deployment for most solutions remains far below what would be necessary to effectively address climate change drivers and impacts (Figure 4B). Delivering the full potential of global measures such as renewable energy, alkalinization, and hybrid methods requires orders-of-magnitude increases in their research, testing, and deployment. Such action is considered urgent on the basis of the climatic threats to ocean sustainability (Gattuso et al., 2015), and since there are decadal lag times until full maximum effectiveness of all the global measures considered here (Figure 4B and section “SM3.1” of the Supplementary Materials). In the meantime, there will likely be significant increases in climate-related impacts on ocean ecosystems and services, which will reduce ecosystems’ ability to provide local solutions (Albright et al., 2016; Cheung et al., 2016; Cinner et al., 2016), thereby decreasing leeway for action (Gattuso et al., 2015).

It is clear that the familiar and conventional marine management strategies cannot fully counter climate change and its impacts. Accelerating research and deployment of other potential solutions will, however, challenge the capacity of science, policy, and decision-making in evaluating and deploying solutions. Defining road maps to drastically enhance action faces major constraints relating to the large uncertainties in key non-climatic variables. Thus socioeconomic conditions may flip the balance between fossil fuel markets and renewables, potentially catalyzing a rapid acceleration of the deployment of marine renewables, but not necessarily with adequate consideration of local disbenefits. There is also a need to consider a broader range of measures than those assessed here, many of which are still in their infancy and unfamiliar to marine management (e.g., large-scale seaweed aquaculture, or abiotic methods of removing or stripping CO₂ from seawater). This calls for the development of policies and funding to foster and promote research into new or emerging ocean and climate management options.