Carbon neutrality does not equal climate neutrality in saltmarsh restoration

doi:10.21203/rs.3.rs-1731723/v1

Download PDF

Article

Carbon neutrality does not equal climate neutrality in saltmarsh restoration

https://doi.org/10.21203/rs.3.rs-1731723/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Saltmarsh restoration efforts often highlight high carbon burial rates as a climate mitigation opportunity. We created a 200-year managed-realignment model incorporating carbon burial, albedo change, and emissions of climate-active compounds across three successive realignment stages: mudflat, realigned saltmarsh, and mature saltmarsh. Total climatic outcomes from all forcing agents differ substantially from those derived solely from carbon burial across latitude and over time. Latitude explains a significant proportion of variation in emissions for methane, methyl bromide, and methyl chloride in mature saltmarshes and carbon burial in mature and realigned saltmarshes. The climate mitigation effects of managed realignments were significantly reduced at subtropical latitudes, and aerosol and albedo impacts had greater influence on total radiative forcing relative to carbon burial. Future land-use mitigation strategies should therefore use spatiotemporally explicit accounting of climate-impacting processes and not rely solely on carbon budgets to underpin climate mitigation strategies.

Decreasing total global radiative forcing (RF) is critical to ensuring global temperatures do not exceed pre-industrial levels by more than 2°C [1]. RF is a net change in the energy balance of the Earth system due to natural or anthropogenic modifications. Anthropogenic RF encompasses many processes of which the production of greenhouse gases (GHGs) is the most recognised [1]. The IPCC reported six additional radiative forcing components, including surface ozone concentrations, stratospheric ozone and water content, albedo, and direct and indirect aerosol effects [1]. Demonstrating the importance of non-GHG RFs on climate, total (direct and indirect) aerosol RF is roughly equivalent in magnitude and opposite in direction to that of CO₂ – although there remains a high degree of uncertainty in aerosol RF estimates despite efforts to reduce these uncertainties in recent IPCC reports [1]. The anthropogenic climate impact is therefore determined by the combined radiative effects of all forcing agents, not solely from the RF of GHGs.

Managed realignment (MR) is the (re)establishment of saltmarshes by removing or breaching coastal defences [2]. MR has attracted increased attention as a climate-mitigation strategy due to high carbon accumulation rates (CAR) in mature and realigned saltmarshes, reflecting the growing interest in carbon capture and sequestration within other "blue carbon" sinks, like mangrove forests and seagrass meadows [3] [4] [5]. However, other forcing agents – including methane (CH₄), nitrous oxide (N₂O), methyl halides (MHs), hydrogen sulphide (H₂S), and dimethyl sulphide (DMS) – are unaccounted for when estimating MR's radiative impact. To fully estimate the radiative effect of MR (and other land-use changes), evaluations must include the full range of forcing agents and their resultant RFs instead of solely considering carbon accumulation rates alone.

Previous saltmarsh evaluations consistently reported climate-impact outcomes using 100-year global warming potentials (GWP₁₀₀) [6] [7]. GWP₁₀₀ compares the 100-year integrated RFs from a single pulse emission of a given compound (e.g., methane) at time zero relative to an equal mass of CO₂ – and are reported in CO₂-equivalent emissions. For example, using the GWP₁₀₀ approach, the global average saltmarsh CH₄ flux of 0.071 ± 0.027 Tmol CH₄-C year^− 1 is equivalent to 2.27 Tmol CO₂-eq. year^− 1 [8]. Despite its widespread use, critics of the GWP-concept note that the 100-year time horizon is arbitrary, with neither climate nor policy justifications. Moreover, GWPs assume a 'pulse-emission scenario'; therefore, the concept is inappropriate when comparing the radiative effect of cumulative gases like CH₄ and N₂O, where the RF is primarily determined by their radiative efficiency, as opposed to short-lived climate pollutants for which atmospheric lifetime is the principal determinant [9]. More accurate understanding of the timing and scale of the climate effect of land use transformations, such as MR, can be better achieved through direct RF calculations.

MR has multiple benefits, from maintaining biodiversity to coastal protection; however, sites with the largest negative RF per investment must be prioritised where saltmarsh realignments are intended to yield the greatest mitigation of climate change. Regional and local factors, such as sunlight, temperature, rainfall, salinity, and the resultant evapotranspiration rate, collectively regulate the CAR and fluxes of gaseous forcing agents. For example, CH₄ production declines with increasing salinity [10] [11] [12] as formerly dominant hydrogenotrophic and acetoclastic methanogens are outcompeted by sulphate-reducing microbes [13]. Conversely, warming of freshwater sediments modifies sediment microbial communities and net metabolism, leading to greater methane emissions [14]. Saline sites may reduce methane emissions, compounding the cooling effect from C burial due to salinity-driven reductions in aerobic decomposition efficiency [10] [15] [16]. In contrast, saltmarsh-to-atmosphere fluxes of aerosol-forming compounds, like DMS and H₂S, have been reported to increase with saltmarsh salinity along with changes in the dominant reduced sulphur volatile form [17]. These local factors can be assessed in aggregate, using latitude as a proxy. Previous preliminary studies identified a peak in CAR of saltmarsh sediments at mid-latitudes (between ~ 48.5 and 58.5°N), decreasing towards the poles and the equator, and a linear decrease in CH₄ emissions with increasing latitude across China [5] [6]. However, due to a relative data paucity, no global mean estimates exist for numerous potentially important forcing agents, nor the data required to understand their latitudinal behaviour.

We conducted a meta-analysis of the existing literature for data on CAR, CH₄, N₂O, albedo, MHs, isoprene, and sulphur gases. This meta-analysis partially calculates impacts from five of the seven IPCC reported anthropogenic RF components. Insufficient data were available to calculate or incorporate the radiative impacts arising from black carbon on snow, the total surface ozone production from biogenic volatile release [18], CFCs/HFCs, or contrails, nor were we able to quantitatively account for the enhanced aerosol effects due to release of methyl iodide or other aerosol enhancing BVOCs [19].

To generate a model that would calculate the RF impacts per square metre from a coastal site directly after breaching for realignment, we assume a progression through three successive stages: mudflat, realigned saltmarsh, and mature saltmarsh. Previous studies have shown that immediately after realignment, sediments are rapidly deposited on sites, with limited biomass relative to mature marshes over the first decade (defined here as mudflat), followed by a period of similar aboveground biomass to mature marshes but with divergent biodiversity (defined here as realigned saltmarsh), which can last more than 100 years [20], followed by a fully developed, mature saltmarsh plant community. We analysed the published data for global mean fluxes of GHGs and trace gases, along with albedo and CAR for each transitional stage. We further tested the data for significant latitudinal gradients. We calculated annual, latitude-dependent fluxes for each compound, converting them into RFs over time (expressed as Watts per square metre of land transformation) to generate a 200-year model of MR climate impact (200YM), encompassing the various successional stages of salt marsh development.

Latitudinal gradients in saltmarsh forcing agents

Globally, CARs in mature saltmarshes demonstrate a shallow but significant increase with latitude (Spearman's p = 0.0004; Fig. 1a). CARs in realigned saltmarshes showed a stronger, significant positive correlation with latitude (p = 0.004; Fig. 1b), accounting for approximately half of the observed variation (R² = 0.45). A previous review found latitude explained 29.6% of observed CAR variance [5] across all saltmarshes, similar to our value of 24% in mature saltmarshes. Latitude explains a greater proportion of CAR variation in realigned sites, where sediment inputs are highest, likely attributable to local hydrogeological conditions at site transition. Indeed, many realigned saltmarshes act as coastal defence tools, with reduced elevations relative to regional sea levels, acting as regional water and sediment traps.

We consider latitude an integrated proxy for various abiotic drivers, including temperature, precipitation and growing season length. Temperature and salinity determine local and regional primary productivity (altering vegetation-determined sediment C-input) and microbial decomposition rates (affecting sediment C removal) [21]. For instance, Spartina-dominated saltmarshes – found predominantly in temperate and subtropical latitudes – demonstrate the greatest CAR compared to other species [5], with the highest CAR in our analysis found in a realigned Spartina-dominated site in the Bay of Fundy, Canada [22]. Sediment salinity broadly declines at higher latitudes due to lower evapotranspiration-precipitation ratios [23] – nonetheless, ocean currents and local geography also influence local sediment conditions [24]. The strong negative correlation observed between sediment salinity and soil organic C concentrations (where higher temperatures lead to lower sediment C) is therefore attributed to the relative metabolism of organic matter.

At lower latitudes, higher decomposition rates may offset greater autochthonous C inputs from primary productivity. However, at higher latitudes, the broadly declining temperature may slow decomposition rates beyond reductions in primary productivity, increasing the contribution of autochthonous inputs relative to C burial. Taken together, these effects can be observed in the Bay of Fundy, where high allochthonous sedimentation rates, or rapid accretion, outstrips the decomposition rates of sediment microbes, resulting in a high CAR. The world's largest tidal range is found in the Bay of Fundy (> 15 m) [25], which likely increases the site's C storage potential by creating additional 'accommodation space', or the coastal range in which a salt marsh may develop. Similar sites, though hydrologically rare, are first-rate examples of the potential for C storage when both geographic and latitudinal effects are optimised.

In mature saltmarshes, CH₄ emissions were significantly negatively correlated with latitude (Spearman's p = 0.002; Fig. 1d), the inverse of the relationship observed for CARs. This finding agrees with CH₄ latitudinal trends observed in China, attributed to latitudinal differences in temperature and changing sediment species composition [6]. Anaerobic microbial methane production in saline environments relies primarily on methylotrophic and hydrogenotrophic methanogenesis rather than acetoclastic mechanisms due to competition from sulphate-reducing bacteria [13]. However, although at the latitudinal level salinity strongly correlates with methanogenic production rates, it is less likely to be the causative factor, as the scale of the reported effect (0 to 35 ppt) is much greater than the salinity range in global seawaters (~ 33 to 38 ppt) directly influencing coastal marshes. Alternatively, efficient methane production through hydrogenotrophic methanogenesis declines at lower temperatures, meaning such methanogens have substantively reduced CH₄ production at higher latitudes [26]. We propose that reduced metabolism, due to lower temperatures causing lower efficiency of methane production due to transition from hydrogenotrophic to methylotrophic methanogens, may explain the latitudinal trend in methane flux. However, the influence of temperature on methane emissions from saltmarsh sediments is complex, in addition to modifying aerobic and anaerobic methane-oxidising microbial community transitions and their efficacy (affecting sink processes for methane in sediments), resource availability, and ebullition rates [27].

Methyl bromide (CH₃Br) and methyl chloride (CH₃Cl) are significantly negatively correlated with latitude (Spearman’s p = 0.006 and p < 0.00001, respectively; Fig. 2). MH biogenic production is greatly influenced by aboveground biomass and temperature [28] [29]. Complete removal of vegetation from saltmarshes results in MH fluxes near or below zero [30], while the effect of temperature may triple MH fluxes from equivalent biomass between 70°N to 30°N [31]. Primary productivity and temperature, therefore, are the probable cause of greater MH production at subtropical latitudes due to enhanced biomass production and faster enzymatic activity. Above 50° latitude, the data regression implies saltmarshes become a net MH sink, consistent with previous wetland research [31]. Similar sink values were reported for CH₃Cl in Abbotts Hall, England, for realigned and mature saltmarshes [32]. However, insufficient data exist for higher latitudes (> 50°N) to determine whether local or climatic factors induce an MH sink.

We compared compound fluxes for latitudinal gradients across the three realignment stages (mudflat, realigned saltmarsh, and mature saltmarsh) – only significant correlations are discussed here. As more data become available, further significant latitudinal gradients are likely to be observed. In particular, there are limited data for realigned saltmarshes and mudflats for methane and nitrous oxide fluxes above 50° latitude, as well as non-GHG radiative forcing agents at all latitudes.

Radiative Forcing Of Managed Realignment

To determine the time-dependent radiative outcomes from MR, we calculated the annual RFs (Fig. 3) – rather than GWP₁₀₀ – for each forcing agent flux over 200 years. We used randomised values from latitude-dependent Gaussian distributions generated from the published data for each realignment stage, assuming a transition of a single square metre of land. Based on published observations, the 200YM assumed ~ 10 years would be spent as a mudflat before transitioning to a realigned saltmarsh (see methods). After site flooding, an initial loss of terrestrial vegetation and rapid deposition of unconsolidated sediment occur [20]. Halophyte colonisation commences within the first 2–5 years [33], with the mudflat-like state transforming into a 'realigned saltmarsh' as more vegetation colonizes and spreads across the site. Realigned saltmarshes behave differently from their natural counterparts, persisting in this state for more than a century [34]; therefore, the 200YM assumes realigned saltmarsh will take ~ 100 years before transitioning to a mature saltmarsh. These durations and stages were chosen to reflect the broad phases of MR.

Where the RF of GHG fluxes is simple and straightforward to calculate, other, less transparent mechanisms were also included: (1) ozone depletion driven by MHs, (2) secondary CO₂ production from methane and VOC (volatile organic compound) oxidation, and (3) cloud aerosol impacts from sulphur compound oxidation. We did not include stratospheric water enhancement via methane oxidation nor enhanced aerosol, or surface ozone formation effects arising from VOCs, nitrogen-bearing compounds, or methyl iodide addition because we were unable to (a) identify credible means to estimate their impacts, (b) estimate fluxes of important trace gases from applicable landscapes (e.g., sesquiterpenes from saltmarshes), or (c) both. Figure 3 shows the model generated direct radiative impact (in fW m^− 2 or zW m^− 2 for methyl iodide and isoprene) over the 200 year realignment progression at three separate latitudes (30°, 50°, and 70°).

Long-lived forcing agents, like CO₂ (250 years) [35], CH₄ (12.4 years), N₂O (121 years), CH₃Br (0.8 years), and CH₃Cl (1 year) [36], gradually accumulate before reaching 'atmospheric equilibrium' – where annual emissions equal annual loss-to-oxidation – while VOCs, whose atmospheric lifetimes are on the order of days or weeks, such as DMS or H₂S, show distinct annual fluctuations. Nevertheless, short-lived gases are amongst the most efficacious climate forcing agents – for example, DMS produced an average annual forcing of -18.4 ± 29.7 fW m^− 2, far greater proportionally than the 200-year cumulative outcomes for CH₄ (less than − 4 fW m^− 2) or N₂O (less than + 3 fW m^− 2) at 70° latitude.

MHs (and other halocarbons) catalyse stratospheric ozone depletion: a single halogen radical, released during atmospheric MH decomposition, can destroy thousands of ozone molecules, producing negative RF or a cooling effect [36] [37]. MHs also act as efficient greenhouse gases and can be classified along with other halocarbons. In the 200YM, CH₃Br produced the greatest forcing from the net effect of ozone depletion and the greenhouse effect (0.13 fW m^− 2 at 30°) – negligible compared to most other forcing agents (Fig. 3). In contrast, secondary CO₂ production from atmospheric CH₄ oxidation by hydroxyl radicals produced a substantial RF: the 200-year RF was 13.2 ± 0.88 fW m^− 2 at 30° latitude (data not shown), surpassing its precursor's primary RF of 5.99 ± 0.24 fW m^− 2. Aerosols, generated from aerosol precursors like DMS and H₂S, scatter incoming solar radiation and act as cloud condensation nuclei, increasing cloud albedo and lengthening cloud longevity via the cloud-lifetime effect [38] [39]. Together, DMS and H₂S were the second most potent forcing agents (-32.50 ± 47.50 fW m^− 2) after C burial.

Based on our calculated outcomes, the use of MR to mitigate climate change should consider appropriate timescales and latitudes when evaluating the climate impact of a proposed site, as the net radiative effect frequently differed from C burial alone in the 200YM (Fig. 4). Notably, reduced C burial and stark changes in albedo (Fig. S1) produced a positive forcing at 30° latitude for more than 50 years. Indeed, C burial gradually increased as a proportion of the total forcing with rising latitude due to latitude-dependent declines in positive forcing agents like CH₄, CH₃Br, and CH₃Cl and the significantly greater CAR at higher latitudes overwhelm the effect of other forcing agents. Considering the majority of Chinese, European, and United States saltmarshes are situated between 25°N and 50°N, such a finding warrants a potential re-orientation of global realignment efforts towards higher latitude sites.

Local factors should also guide site selection. Although neither DMS nor H₂S currently demonstrates a latitudinal gradient, halophyte species and sediment salinity regulate their production [17] [40], thereby potentially having an outsized impact on total forcing. Albedo – the diffuse reflection of shortwave radiation relative to total incoming solar radiation – is determined by local and latitudinal factors. In comparison to pre-realignment grasslands, where approximately 25% of incoming sunlight is reflected [41], mudflats (8%) [42] and saltmarshes (20%) [43] reflect substantially less shortwave radiation, producing a pronounced positive RF. The phenomenon would be reversed, producing a cooling effect in urban-realignment sites, as the albedo of developed urban coastal land ranges from 11.2–18.7% [44]. Albedo effects are mitigated at higher latitudes due to lower incoming shortwave radiation; however, albedo impacts alone may overturn the negative RF of C burial at 30° for the first few decades of marsh transition (Fig. S1).

Differences between the total forcing and C burial alone were greatest in the earliest stages (Fig. S2). Perhaps surprisingly, despite their global climatic importance, CH₄ and N₂O did not contribute substantially to the total forcing in these systems. Though CH₄ and N₂O emissions are proportionate to the RF due to CAR in any one year in the mature stage, buried C inexorably accumulates, compounding its forcing with time. Concerns regarding annual saltmarsh CH₄ and N₂O emissions compared to annual C burial misapprehend this cumulative phenomenon. Moreover, the velocity at which the buried C becomes dominant is, in part, determined by the site latitude; for instance, it takes over 100 years to become dominant at 30°, whereas, at 70°, C burial is virtually indistinguishable from the total forcing from the outset.

The 200YM assumes that buried carbon remains dormant in saltmarsh sediments, allowing CAR to accumulate with time. However, sea-level rise modelling indicates that existing saltmarsh, prevented from landward migration by artificial defences, will decline through 'coastal squeeze' in the latter half of the 21st century [45] [46]. Once submerged, buried C may erode by wave action or oxidise to CO₂, negating its climate mitigation impact [4]. Known as 'non-permanence risk', the potential for C release is contrasted against the 'inherent permanence' of gaseous emissions, which, once emitted, can never be cancelled [47]. The lasting benefits of MR will therefore be determined by our commitment to saltmarsh preservation, for instance, through management strategies allowing saltmarsh migration with increasing sea levels.

Even if buried C remains untouched, the viability of MR as a climate-mitigation strategy appears to be dependent on latitude and local hydrogeographic factors. For example, the unusual hydrological features of the Bay of Fundy, alongside its more northerly latitude (45°N), represent an exemplary realignment site. In contrast, MR at lower latitudes (~ 30°) may result in a medium-term (25 year) increase in the site's RF, depending on the original site use and albedo. All latitudes see a net negative RF by 100 years, but this may be beyond the timeframe of benefit for many realignments that seek to address climate change rapidly.

The IPCC estimates the RF of climate change between 1750–2011 was 2.3 W m^− 2 [36]. Assuming the 50° latitude average response from the 200YM, approximately 527 Mha of coastal land needs to be realigned to negate the total anthropogenic RF after 200 years. Conservative estimates place the current mapped saltmarsh area at 5.5 Mha, with alternative estimates ranging from 2.2–40 Mha [48] [49]. If realignment efforts doubled the current global saltmarsh area evenly across the globe – a somewhat achievable target – it would result in a global cooling effect of ~-0.025 W m^− 2 after 200 years: comparable to climate forcers such as stratospheric ozone and surface albedo.

Notwithstanding MR's reduced mitigation potential, we may still wish to (re)establish saltmarshes to increase biodiversity, restore degraded ecosystems, and for coastal protection. However, MR is not likely to be a 'silver bullet' for rapid climate mitigation, though its impact can be greatly enhanced by choosing optimal regional and ecological factors, such as larger tidal ranges, greater sediment inflows, and higher latitudes. More concerning, within the broader context of climate change mitigation, saltmarshes are the most efficacious ecosystem for C burial globally [5]; therefore, other proposed land-use mitigation strategies (such as tree planting) are also unlikely to yield rapid, substantial climate benefits.

For most of the past century, mitigation strategies focused almost wholly on decarbonisation. However, as shown here, carbon neutrality is not synonymous with climate neutrality. Nine out of ten Fortune 500 companies now use the GHG protocol: incorporating seven GHGs covered in the Kyoto Protocol to address climate concerns (including CO₂, CH₄, N₂O, HFCs, PFCs, SF₈ and NF₃) [50]. Nevertheless, the seven compounds addressed by the GHG protocol neglect multiple aspects that we have identified as critical to understanding future climate outcomes, like albedo, sulphur compounds, and other biogenic trace gases (such as MHs). Furthermore, fluxes of forcing agents across various land-use types remain poorly understood. For example, several forcing agents included in the 200YM used data for mudflats and realigned sites which were first reported in this study; no prior data existed. Globally the available, published data for saltmarshes are not distributed equally across latitudes but are concentrated in few regions – for instance, the data from the southern hemisphere are exclusively from Australia. The greatest challenge is the critical need for extensive modelling efforts to quantify and understand the impacts of complex radiative actors, like methyl iodide, isoprene, and terpenes, which have individual and collective ozone depletion, smog production, and aerosol formation effects.

Data sources and collection

The systematic review was conducted according to PRISMA guidelines [1] (Fig. S3). Saltmarshes were defined as an ecosystem adjacent to the coast whose waters are either brackish or saline and are dominated by dense strands of salt-tolerant plants. Managed realignment was defined as the deliberate or inadvertent flooding of a formerly non-tidal area leading to saltmarsh development. Due to the paucity of realigned site data, created saltmarshes – where coastal land is deliberately planted to form a saltmarsh – were included under the banner of realigned sites, as they constituted an example of managed saltmarshes. Finally, mudflats or tidal flats were defined as unvegetated intertidal flats adjacent to the coast, primarily composed of sediments or muds.

Searches for relevant studies were conducted using the Web of Science database between April and July 2021, using Boolean search statements with terms relating to either a forcing agent flux or realignment stage – for example, topic = (carbon sequestration) AND (saltmarsh). For a complete listing of the search terms used, number of results, number of identified studies, and search date refer to Supplementary Information (Table S1). Studies were sought and grouped based upon known saltmarsh impacts and emissions: C burial, CH₄, N₂O, sulphur-containing compounds (DMS, H₂S, methanethiol), terpenes (isoprene, monoterpenes, sesquiterpenes), and halocarbons (CH₃Cl, CH₃Br, methyl iodide, chloroform, methyl chloroform).

Data from studies were selected for inclusion according to the following criteria:

Site must be above 30^oN or below 30^oS – to avoid conflation with mangrove forests.

Site must be either a mudflat or brackish or saline marsh.

For CARs (criteria followed by Ouyang and Lee, (2014)):

Methods for calculating CAR using C sediment concentrations and sediment depth or sediment accumulation rate (SAR) were included.

CAR must have been recorded within 75 years of the search date.

For CH₄ and N₂O emissions:

Methods detecting emissions via either flow-through, incubation, or static chambers were included.

Site must be studied for at least nine months to include seasonal emission variations.

Due to the paucity of data for sulphur-containing, halocarbon, and terpene forcing agents, the inclusion criteria were relaxed to include the maximum available data:

All measurements were included, irrespective of study duration.

All latitudes were included if sufficient evidence was provided that a site did not significantly overlap with mangrove forests.

Methods detecting emissions via either flow-through, incubation, or static chambers were included.

Overall, searches generated 5078 records, inclusive of duplications due to overlapping search terms. Following inclusion criteria, abstracts of all records returned in a search were screened by a single reviewer. If the inclusion criteria were met or relevant terms were identified, the entire paper was retrieved and assessed for eligibility. Potential studies were also identified from retrieved paper citations and were subjected to the same screening process. This process filtered the reports down to a final total of 117 (comprising 623 data points). These reports ranged from 38.04°S to 70.19°N, predominantly in the northern hemisphere (Fig. S4, Fig. S5, Fig. S6, Fig. S7). Therefore, southern latitudinal data were analysed as originating from the northern hemisphere at the same relative latitude.

The following data were extracted from reports: latitude, longitude, country, dominant halophyte species/genera, tidal range, mean annual air temperature (^°C), mean annual precipitation (mm), salinity, mean CAR or emission, and measurement method. The authors were contacted if data was not retrievable; however, no replies were received. Site coordinates were estimated from Google Maps using listed information if coordinates were not provided. CAR and methane emissions were input in g m^− 2 year^− 1; otherwise, data were input in mg m^− 2 year^− 1. Finally, data were checked against the original report to confirm veracity. For a complete listing of all data included in the study, refer to the Supplementary Information (Table S2).

Managed realignment 200-year model

All modelling utilised Python. The following packages were used: numpy [2], pandas [3], statsmodel.api [4], seaborn [5], scipy [6], statistics [7], and random [8].

To estimate the total net RF from MR, the collected data were filtered through a model simulating three realignment stages over 200 years: mudflat, realigned saltmarsh, and mature saltmarsh. The duration of the mudflat and realigned saltmarsh stages were randomly generated from a Gaussian distribution using the mean and standard deviation (SD): estimates were derived from prior realignments (mudflat = 10 ± 2.5 years; realigned saltmarsh = 75 ± 10 years) [9] [10] [11] [12]. The duration of the mature saltmarsh was equal to 200 years minus the sum of the mudflat and realigned saltmarsh stages.

The 200-year end point was selected for three primary reasons: (1) to encompass all three stages of MR, (2) computational limitations in running the model for a longer time period, and (3) to restrict the model to a time period relevant to climate change mitigation.

The model factored in emissions and land-use changes, including the RF from C burial and GHG emissions, atmospheric lifetimes of forcing agents, CO₂ production from CH₄ oxidation, halocarbon ozone depletion, sulphur-containing aerosol RF impacts, and albedo. The model runs were repeated ten times, with mean outputs providing final RF estimates for the first 200 years of MR.

Site-specific compound fluxes and atmospheric lifetimes

Within the 200-year realignment model (200YM), a random annual emission or CAR was generated using the mean and SD from Gaussian distributions. Gaussian distributions were based on published data (Table S2). If an emission demonstrated a significant latitudinal gradient, then the 95% confidence interval was calculated for a specific latitude, divided by two, and used in place of SD to generate random values. Subsequently, atmospheric compounds oxidised or were removed from the atmosphere using the half-life decay formula

$${\text{N}}_{\text{t}}={\text{N}}_{0}{\left(\frac{1}{2}\right)}^{\text{t}/{\tau }}$$

where N_t is the remaining concentration, N₀ is the atmospheric concentration, t is the given time, and τ is the emission half-life (HL). C was assumed to remain buried in the sediment and was not subject to oxidation. If not available, atmospheric half-lives were calculated from atmospheric lifetimes (Table S3)

$${\text{t}}_{1/2}=\frac{0.693}{\text{e}}$$

where t_1/2 is the atmospheric HL, and e is the e-folding time: the reciprocal of the atmospheric lifetime.

Only 20% of the annual CH₄ emission oxidised to CO₂ was included in the analysis, as that fraction constituted C not recently fixed via photosynthesis [13] [14] [15] – based on estimates from peatlands. The annual CO₂ produced then underwent HL decay for the remaining years of realignment. No other oxidation of carbon-containing compounds was calculated, as due to their short atmospheric lifetimes (< 1 year), the C released was deemed equivalent to recently fixed CO₂: equalling a net-zero radiative balance.

Radiative efficiency

To calculate RF: emissions (including CO₂) were converted to atmospheric molar concentrations in parts per billion (ppb). RF was then calculated

$${\text{R}\text{F}.}_{\text{N}}=\text{N} \times \text{R}\text{E}$$

where RF_N is the radiative forcing of an emission (W m^− 2), N is the atmospheric concentration of any individual compound (ppb), and RE is the radiative efficiency (W m^− 2 ppb^− 1) (Table S4).

Ozone depletion

Halocarbons deplete stratospheric ozone, which is known to have a cooling effect. The RF for the ozone depleted by an annual halogenic emission was calculated as

$${\text{Q}}_{\text{G}\text{H}\text{G}}= {\text{O}\text{D}\text{P}}_{\text{G}\text{H}\text{G}} \times 3.7\text{e}-13$$

$${{\text{D}}_{\text{G}\text{H}\text{G}}=\text{Q}}_{\text{G}\text{H}\text{G}} \times 2.69\text{e}+20$$

$${\text{R}\text{F}.}_{\text{N}}=\left({\text{D}}_{\text{G}\text{H}\text{G}}\times {\text{N}}_{\text{t}}\right) \times {\text{R}\text{F}.}_{\text{o}\text{z}}$$

where ODP_GHG is the ozone-depleting potential of any given compound, and 3.7e-13 is the ozone sensitivity of CFC-11 (DU g^− 1) (all ODPs are relative to CFC-11). Q_GHG is the compound-specific ozone sensitivity, while 2.69e + 20 is the atmospheric ozone concentration (DU), D_GHG is the concentration of ozone depleted, N_t is the emission at a given year, RF_oz is the radiative forcing per molecule of stratospheric ozone. See Supplementary Information (Table S5) for further information about ozone-depletion calculations.

Sulphur-containing aerosol-forming compounds

In the atmosphere, sulphur-containing compounds scatter incoming solar radiation and increase cloud longevity and albedo, causing a negative RF. To calculate their radiative efficiency, the annual change in RF assigned to sulphur from annual DMS flux (W m^− 2) was divided by the change in atmospheric molar concentration of the annual flux of DMS (ppb) [16] [17]. The radiative efficiency was then multiplied by the molar atmospheric concentration of sulphur from each sulphur-containing emission.

Albedo

Throughout realignment, the reflectance of a site changes, contributing to the overall radiative balance. To calculate this, we used the albedo

$${\text{R}}_{\text{N}}={\text{R}}_{\text{S}}(1-{\alpha })$$

where R_N is the net radiation emitted, R_S is the incoming solar shortwave radiation, and α is the albedo or reflection coefficient. For the realigned saltmarsh stage, the albedo was increased in equal increments from the mudflat (0.08) [18] to the mature saltmarsh (0.20) [19] albedo values. R_S was determined based on the modelled site's latitude, with roughly half of the incoming shortwave radiation attenuating in the atmosphere due to absorption (23%) or cloud reflectance (22%) [20]. R_N was then subtracted from the R_N for grassland (0.25) [21] at the same latitude to determine how MR had altered the radiative balance before dividing by the Earth's surface area (m²), providing the final representative square metre RF due to albedo.

Statistical analysis

All statistical analyses were conducted in R v4.0.3 [22]. The following packages were used: dunn.test [23], ggplot2 [24], ggpubr [25], plyr [26], and pgirmess [27].

One-way ANOVA analysed differences between realignment stages (explanatory variable) and CAR, CH₄, or N₂O emissions (dependent variable). Data normality was assessed with Shapiro's test, and homoscedasticity was assessed with Bartlett's test – if either was violated (α ≥ 0.05), data were log-transformed to satisfy test assumptions. Kruskal-Wallis tests were applied in cases of persistent violations of assumptions, followed by a non-parametric post hoc pairwise Dunn test where a significant effect was identified.

Pearson's correlation coefficients were calculated to assess latitudinal gradients for CAR and all emissions provided sufficient available data. Data normality was assessed with Shapiro's test. C mature saltmarsh and realigned saltmarsh, CH₄ realigned saltmarsh, and H₂S mature saltmarsh data were log-transformed to satisfy assumptions. Spearman's correlation coefficients were calculated when normality violations persisted.

Acknowledgements

The authors thank the staff of the Essex Wildlife Trust for access, support, and advice. KR, and AJD, acknowledge Natural Environment Research Council award no. NE/K01546X/1 (Do realignment sites restore microbial biodiversity-driven nutrient cycling and trace gas fluxes comparable to natural coastal ecosystems?). AJD and TMcG would also like to acknowledge Natural Environment Research Council award no. NE/J01561X/1 (A hierarchical approach to the examination of the relationship between biodiversity and ecosystem service flows across coastal margins). WRG is funded by Natural Environment Research Council award no. NE/R010846/1 (Carbon Storage in Intertidal Sediments; C-SIDE). HS acknowledges support of saltmarsh related work by the Natural Environment Research Council award no. NE/H008918/1 and a fellowship by the Hanse-Wissenschaftskolleg in Delmenhorst (Germany).

Competing interests

The authors declare no competing interests.

Masson-Delmotte, V. et al. AR6 Climate Change 2021: The Physical Science Basis. (2021).
French, P. W. Managed realignment–the developing story of a comparatively new approach to soft engineering. Estuarine, Coastal and Shelf Science. 67, 409-423 (2006).
Chmura, G. L., Anisfeld, S. C., Cahoon, D. R. & Lynch, J. C. Global carbon sequestration in tidal, saline wetland soils. Global Biogeochemical Cycles. 17, (2003).
McLeod, E. et al. A blueprint for blue carbon: toward an improved understanding of the role of vegetated coastal habitats in sequestering CO2. Frontiers in Ecology and the Environment. 9, 552-560 (2011).
Ouyang, X. & Lee, S. Y. Updated estimates of carbon accumulation rates in coastal marsh sediments. Biogeosciences. 11, 5057-5071 (2014).
Hu, M., Sardans, J., Yang, X., Peñuelas, J. & Tong, C. Patterns and environmental drivers of greenhouse gas fluxes in the coastal wetlands of China: A systematic review and synthesis. Environmental Research. 186, 109576; 10.1016/j.envres.2020.109576 (2020).
Rosentreter, J. A., Al‐Haj, A. N., Fulweiler, R. W. & Williamson, P. Methane and nitrous oxide emissions complicate coastal blue carbon assessments. Global Biogeochemical Cycles. 35, e2020GB006858; 10.1029/2020GB006858 (2021).
Al‐Haj, A. N. & Fulweiler, R. W. A synthesis of methane emissions from shallow vegetated coastal ecosystems. Global Change Biology. 26, 2988-3005 (2020).
Allen, M. R. et al. New use of global warming potentials to compare cumulative and short-lived climate pollutants. Nature Climate Change. 6, 773-776 (2016).
Bartlett, K. B., Bartlett, D. S., Harriss, R. C. & Sebacher, D. I. Methane emissions along a salt marsh salinity gradient. Biogeochemistry. 4, 183-202 (1987).
Emery, H. E. & Fulweiler, R. W. Incomplete tidal restoration may lead to persistent high CH4 emission. Ecosphere. 8, e01968; 10.1002/ecs2.1968 (2017).
Poffenbarger, H. J., Needelman, B. A. & Megonigal, J. P. Salinity influence on methane emissions from tidal marshes. Wetlands. 31, 831-842 (2011).
Conrad, R. Importance of hydrogenotrophic, acetoclastic and methylotrophic methanogenesis for methane production in terrestrial, aquatic and other anoxic environments: a mini review. Pedosphere. 30, 25-39 (2020).
Zhu, Y. et al. Disproportionate increase in freshwater methane emissions induced by experimental warming. Nature Climate Change. 10, 685-690 (2020).
Gunn, C. M. Methane emissions along a salinity gradient of a restored salt marsh in Casco Bay, Maine. Honors Theses. 175, (2016).
Setia, R. et al. Salinity effects on carbon mineralisation in soils of varying texture. Soil Biology and Biochemistry. 43, 1908-1916 (2011).
DeLaune, R. D., Devai, I. & Lindau, C. W. Flux of reduced sulfur gases along a salinity gradient in Louisiana coastal marshes. Estuarine, Coastal and Shelf Science. 54, 1003-1011 (2002).
Bai, J. O₃ concentration and its relation with BVOC emissions in a subtropical plantation. Atmosphere. 12, 711 (2021).
O'Dowd, C. D. et al. Marine aerosol formation from biogenic iodine emissions. Nature. 417, 632-636 (2002).
Garbutt, A. & Wolters, M. The natural regeneration of salt marsh on formerly reclaimed land. Applied Vegetation Science. 11, 335-344 (2008).
Loomis, M. J. & Craft, C. B. Carbon sequestration and nutrient (nitrogen, phosphorus) accumulation in river-dominated tidal marshes, Georgia, USA. Soil Science Society of America Journal. 74, 1028-1036 (2010).
Wollenberg, J. T., Ollerhead, J. & Chmura, G. L. Rapid carbon accumulation following managed realignment on the Bay of Fundy. PLoS ONE. 13, e0193930; 10.1371/journal.pone.0193930 (2018).
Shen, C., Zhang, C., Xin, P., Kong, J., & Li, L. Salt dynamics in coastal marshes: Formation of hypersaline zones. Water Resources Research. 54, 3259-3276 (2018).
Rahmstorf, S. Thermohaline circulation: The current climate. Nature. 421, 699 (2003).
Desplanque, C. & Mossman, D. J. Bay of Fundy tides. Geoscience Canada. 28; journals.lib.unb.ca/index.php/GC/article/view/4065 (2001).
Lavergne, C. et al. Temperature differently affected methanogenic pathways and microbial communities in sub-Antarctic freshwater ecosystems. Environment International. 154, 106575; 10.1016/j.envint.2021.106575 (2021).
Yvon-Durocher, G et al. Methane fluxes show consistent temperature dependence across microbial to ecosystem scales. Nature. 507, 488-491 (2014).
Rhew, R. C., Whelan, M. E. & Min, D. H. Large methyl halide emissions from south Texas salt marshes. Biogeosciences. 11, 6427-6434 (2014).
Wuosmaa, A. M. & Hager, L. P. Methyl chloride transferase: a carbocation route for biosynthesis of halometabolites. Science. 249, 160-162 (1990).
Blei, E., Heal, M. R. & Heal, K. V. Long-term CH3Br and CH3Cl flux measurements in temperate salt marshes. Biogeosciences. 7, 3657-3668 (2010).
Redeker, K. R. & Cicerone, R. J. Environmental controls over methyl halide emissions from rice paddies. Global Biogeochemical Cycles. 18, (2004).
Redeker, K. R. et al. Noninvasive analysis of the soil microbiome: Biomonitoring strategies using the volatilome, community analysis, and environment data. Advances in Ecological Research. 59, 93-132 (2018).
Virgin, D. S. S. A managed realignment in the upper Bay of Fundy: Community dynamics during salt marsh restoration over 8 years in a megatidal, ice-influenced environment. Ecological Engineering. 149, 105713; 10.1016/j.ecoleng.2020.105713 (2020).
Mossman, H. L., Davy, A. J. & Grant, A. Does managed coastal realignment create saltmarshes with 'equivalent biological characteristics' to natural reference sites? Journal of Applied Ecology. 49, 1446-1456 (2012).
Archer, D. et al. Atmospheric lifetime of fossil fuel carbon dioxide. Annual Review of Earth and Planetary Sciences. 37, 117-134 (2009).
Stocker, T. F. et al. AR5 Climate Change 2012: The Physical Science Basis. (2013).
Wuebbles, D. J., Youn, D., Patten, K., Wang, D. & Martínez-Avilés, M. Metrics for ozone and climate: Three-dimensional modeling studies of ozone depletion potentials and indirect global warming potentials in Twenty Years of Ozone Decline. 297-326 (Springer, 2009).
Carslaw, K. S. et al. A review of natural aerosol interactions and feedbacks within the Earth system. Atmospheric Chemistry and Physics. 10, 1701-1737 (2010).
Thomas, M. A. et al. Quantification of DMS aerosol-cloud-climate interactions using the ECHAM5-HAMMOZ model in a current climate scenario. Atmospheric Chemistry and Physics. 10, 7425-7438 (2010).
Wang, J. & Wang, J. Spartina alterniflora alters ecosystem DMS and CH4 emissions and their relationship along interacting tidal and vegetation gradients within a coastal salt marsh in Eastern China. Atmospheric Environment. 167, 346-359 (2017).
Climate Data Information. Albedo Measurements. http://www.climatedata.info/forcing/albedo/. (n.d.).
Guarini, J. M., Blanchard, G. F., Gros, P. & Harrison, S. J. Modelling the mud surface temperature on intertidal flats to investigate the spatio-temporal dynamics of the benthic microalgal photosynthetic capacity. Marine Ecology Progress Series. 153, 25-36 (1997).
Hughes, C. E., Kalma, J. D., Binning, P., Willgoose, G. R. & Vertzonis, M. Estimating evapotranspiration for a temperate salt marsh, Newcastle, Australia. Hydrological Processes. 15, 957-975 (2001).
Trlica, A., Hutyra, L. R., Schaaf, C. L., Erb, A. & Wang, J. A. Albedo, land cover, and daytime surface temperature variation across an urbanised landscape. Earth's Future. 5, 1084-1101 (2017).
Crosby, S. C. et al. Salt marsh persistence is threatened by predicted sea-level rise. Estuarine, Coastal and Shelf Science. 181, 93-99 (2016).
Kirwan, M. L. & Mudd, S. M. Response of saltmarsh carbon accumulation to climate change. Nature. 489, 550-553 (2012).
Kroeger, K. D., Crooks, S., Moseman-Valtierra, S. & Tang, J. Restoring tides to reduce methane emissions in impounded wetlands: A new and potent Blue Carbon climate change intervention. Scientific Reports.7, 1-12 (2017).
Mcowen, C. J. et al. A global map of saltmarshes. Biodiversity Data Journal. 5, e11764; 10.3897/BDJ.5.e11764 (2017).
Pendleton, L. et al. Estimating global "blue carbon" emissions from conversion and degradation of vegetated coastal ecosystems. PLoS ONE. 7, e43542; 10.1371/journal.pone.0043542 (2012).
Sotos, M. E. GHG Protocol Scope 2 Guidance. https://www.wri.org/research/ghg-protocol-scope-2-guidance (2015).

Methods references

Page, M. J. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 372, 105906; 10.1016/j.ijsu.2021.105906 (2021).
Harris, C. R. et al. Array programming with NumPy. Nature. 585, 357-362 (2020).
McKinney, W. et al. Data structures for statistical computing in Python in Proceedings of the 9th Python in Science Conference. 51-56 (2010).
Seabold, S. & Perktold, J. statsmodels: Econometric and statistical modelling with Python in 9th Python in Science Conference. (2010).
Waskom, M. et al. mwaskom/seaborn: v0.8.1. 10.5281/zenodo.883859 (2017).
Virtanen, P. et al. SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python. Nature Methods. 17, 261-272 (2020).
Crosta, S. Statistics 1.0.3.5. https://github.com/python/cpython/blob/3.9/Lib/statistics.py (2014).
Van Rossum, G. The Python Library Reference, release 3.8.2. (2020).
Burden, A., Garbutt, R. A., Evans, C. D., Jones, D. L. & Cooper, D. M. Carbon sequestration and biogeochemical cycling in a saltmarsh subject to coastal managed realignment. Estuarine, Coastal and Shelf Science. 120, 12-20 (2013).
Curado, G., Grewell, B. J., Figueroa, E. & Castillo, J. M. Effectiveness of the aquatic halophyte Sarcocornia perennis spp. perennis as a biotool for ecological restoration of salt marshes. Water, Air, & Soil Pollution. 225, 1-14 (2014).
Mazik, K., Smith, J. E., Leighton, A. & Elliott, M. Physical and biological development of a newly breached managed realignment site, Humber estuary, UK. Marine Pollution Bulletin. 55, 564-578 (2007).
Spencer, K. L. et al. Physicochemical changes in sediments at Orplands Farm, Essex, UK following 8 years of managed realignment. Estuarine, Coastal and Shelf Science. 76, 608-619 (2008).
Chanton, J. P. et al. Radiocarbon evidence for the substrates supporting methane formation within northern Minnesota peatlands. Geochimica et Cosmochimica Acta. 59, 3663-3668 (1995).
Garnett, M. H., Hardie, S. M. L., Murray, C. & Billett, M. F. Radiocarbon dating of methane and carbon dioxide evaded from a temperate peatland stream. Biogeochemistry. 114, 213-223 (2013).
Garnett, M. H., Hardie, S. M. & Murray, C. Radiocarbon analysis reveals that vegetation facilitates the release of old methane in a temperate raised bog. Biogeochemistry. 148, 1-17 (2020).
Thomas, M. A. et al. Quantification of DMS aerosol-cloud-climate interactions using the ECHAM5-HAMMOZ model in a current climate scenario. Atmospheric Chemistry and Physics. 10, 7425-7438 (2010).
Thomas, M. A. et al. Rate of non-linearity in DMS aerosol-cloud-climate interactions. Atmospheric Chemistry and Physics. 11, 11175-11183 (2011).
Guarini, J. M., Blanchard, G. F., Gros, P. & Harrison, S. J. Modelling the mud surface temperature on intertidal flats to investigate the spatio-temporal dynamics of the benthic microalgal photosynthetic capacity. Marine Ecology Progress Series. 153, 25-36 (1997).
Hughes, C. E., Kalma, J. D., Binning, P., Willgoose, G. R. & Vertzonis, M. Estimating evapotranspiration for a temperate salt marsh, Newcastle, Australia. Hydrological Processes. 15, 957-975 (2001).
Wild, M. et al. The global energy balance from a surface perspective. Climate Dynamics. 40, 3107-3134 (2013).
Climate Data Information. Albedo Measurements. http://www.climatedata.info/forcing/albedo/. (n.d.).
R Core Team. A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/ (2020).
Dinno, A. dunn.test: Dunn's Test of Multiple Comparisons Using Rank Sums. https://CRAN.R-project.org/package=dunn.test (2017).
Wickham, H. ggplot2: Elegant Graphics for Data Analysis. (2016).
Kassambara, A. ggpubr: 'ggplot2' Based Publication Ready Plots. https://CRAN.R-project.org/package=ggpubr (2020).
Wickham, H. plyr: The Split-Apply-Combine Strategy for Data Analysis. Journal of Statistical Software. http://www.jstatsoft.org/v40/i01/ (2011).
Giraudoux, P. pgirmess: Spatial Analysis and Data Mining for Field Ecologists. https://CRAN.R-project.org/package=pgirmess (2018).

There is NO Competing Interest.

ClimateneutralDNEcarbonneutralSupplementaryInformation.docx
Supplementary Information

Download PDF

Version 1

posted

You are reading this latest preprint version

Carbon neutrality does not equal climate neutrality in saltmarsh restoration

Status:

Version 1

Abstract

Figures

Introduction

Results

Latitudinal gradients in saltmarsh forcing agents

Discussion

Methods

Statistical analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1