Global dataset of soil organic carbon in tidal marshes

Tidal marshes store large amounts of organic carbon in their soils. Field data quantifying soil organic carbon (SOC) stocks provide an important resource for researchers, natural resource managers, and policy-makers working towards the protection, restoration, and valuation of these ecosystems. We collated a global dataset of tidal marsh soil organic carbon (MarSOC) from 99 studies that includes location, soil depth, site name, dry bulk density, SOC, and/or soil organic matter (SOM). The MarSOC dataset includes 17,454 data points from 2,329 unique locations, and 29 countries. We generated a general transfer function for the conversion of SOM to SOC. Using this data we estimated a median (± median absolute deviation) value of 79.2 ± 38.1 Mg SOC ha−1 in the top 30 cm and 231 ± 134 Mg SOC ha−1 in the top 1 m of tidal marsh soils globally. This data can serve as a basis for future work, and may contribute to incorporation of tidal marsh ecosystems into climate change mitigation and adaptation strategies and policies.


I.
Methods for unpublished data

Site description and sampling
In May-July 2017, three replicate soil cores were collected from three salt marshes in three estuaries in West Wales, United Kingdom.All marshes were similar in terms of elevation, extent, geomorphology (i.e., estuarine marshes), position within the estuary (i.e., situated in the mid part of the estuary, sheltered from the ocean) and land use (i.e., light to moderately grazed).All cores were extracted at similar elevations relative to mean sea level (i.e., 1.4-2.0m above mean sea level, obtained using a Leica differential GPS).Replicate cores were collected within tens of metres of each other.The cores were sampled by manual percussion and rotation using PVC pipes and following the guidelines in the Blue Carbon Manual 1 .
Compression during coring was assessed by measuring the length of the core protruding from the soil surface inside and outside the core 1 .All results reported in this study refer to the decompressed depths (cm).Upon extraction, cores were sealed at both ends, transported to the laboratory and stored at 4°C until processing.

Laboratory analysis
Cores were sliced at 2 cm-thick resolution for their entire length.Each slice was dried at 60°C until constant dry weight (DW) to estimate dry bulk density, homogenized and divided into two sub-samples.One subsample was analysed for soil grain-size, using a Malvern Particle Sizer 2000, after organic matter digestion with hydrogen peroxide (conducted at the Geography Science Laboratories of the Department of Geography, University of Cambridge).Another sub-sample was ground in an agate mill and analysed for: organic carbon, loss on ignition, was acidified with 4% HCl to remove inorganic carbon, centrifuged (3400 rpm during 5 min), and the supernatant with acid residues carefully removed by pipette, avoiding resuspension.
The sample was then washed with Milli-Q water, centrifuged and the supernatant again removed.The residual samples were re-dried (60°C until constant weight) and encapsulated in tin capsules.Corg was analysed using a Costech Elemental Analyzer interfaced to a Thermo-Finnegan Delta V Isotope RatioMass Spectrometer at UH-Hilo Analytical Laboratory.
The accuracy of the analysis of the Standard reference material NIST 8704 (Buffalo River Sediment) was ≤1%.The Corg content (%) is reported for the bulk (pre-acidified) samples.

Site description and sampling
The three unpublished cores "OS MCD1", "OS MCD2", and "OS MCD3" were sampled on Torrens Island in an estuary near Adelaide, Australia.Core locations were recorded with a GPS.The soil sampling procedure follows that presented by the study Gorham et al., 2021 2 .Briefly, the soil cores were sampled by manually hammering PVC pipes with a 7.5 cm inner diameter and a length of 1.5 m into tidal marsh soil.Compaction was minimised by hammering softly and rotating the corer often.The difference in surface soil elevation on the inside and outside of the core was measured and recorded to assume linear downcore soil compaction 3 .The samples were kept at 4°C until further analysis.

Laboratory analysis
Back in the laboratory, the cores were opened lengthwise and sliced at 1 cm intervals.Bulk density was calculated after drying the soil samples at 70°C.Soil samples were then milled to a fine powder.The subsequent soil sample analysis procedure was the following: digestion of carbonates with 4% HCl, centrifugation, decanting the supernatant, rinsing with milliQ water, centrifugation, decanting, and drying at 70°C prior to encapsulation.The organic carbon was then measured using an elemental analyzer coupled with a mass spectrometer.

Fig. S1
Fig. S1 Soil organic carbon density (kg m -3 ) from this dataset (grey triangles) and from the Coastal Carbon Research Coordination Network (CCRCN) in blue circles, binned into four horizon depths, according to the sample layer mid-point: 0-15, 15-30, 30-50, 50-100 cm.Values in the purple diamonds correspond to the median for each horizon bin, and error bars are the absolute deviation of the median.

Fig. S2
Fig.S2One plot for each study located in Figure4of the manuscript, with data points with both soil organic matter and soil organic carbon values, used to calculate a study-specific conversion equation for SOM to SOC (solid black line, with prediction intervals in grey).
TableS2Analysis of variance table comparing the study-specific models to the general model, which corresponds to the quadratic equation relationship between soil organic matter and soil organic carbon, presented in FigureS2.