Fiddling with the carbon budget: ﬁddler crab burrowing activity increases wetland’s carbon ﬂux

Coastal wetlands store signiﬁcant amounts of carbon through sequestration. Salt marshes are also known to harbour high densities of crabs, which increase the sediment-atmosphere exchange interface through their burrowing behaviour. We hypothesized that this additional and reactive interface area could mediate gas exchange and, ultimately, could inﬂuence carbon sequestration. CO ﬂuxes were measured over patches characterized by diﬀerent densities of ﬁddler crab, , burrows within a natural salt marsh located on the coast of Massachusetts (USA). Even accounting for the importance of ecological factors such as diﬀerences in organic matter content of the soil and presence of , we demonstrated that CO release increased if local crab burrow density is considered. The increase in vertical CO ﬂuxes linked to burrow density was higher for the non-vegetated areas with respect to patches. By means of burrow casting and morphological analyses of the burrows, we could relate this diﬀerence in COﬂuxes to structural diﬀerences of the burrows themselves, which were larger and deeper in the non-vegetated areas. Our results strongly emphasize the importance of including the faunal component, and speciﬁcally the dominant burrowing species, in carbon budget assessments for vegetated coastal habitats. This study also emphasizes the critical role


INTRODUCTION
Vegetated coastal ecosystems, such as salt marshes, mangroves and seagrass beds, have 48 been increasingly recognized for their importance in storing carbon (Donato et al., 2011;49 Tomohiro & Masakazu, 2019a). Their capacity for carbon sequestration and their potential for 50 long-term carbon, or 'blue carbon', storage render these shallow water habitats valuable assets 51 to mitigating climate change (Howard et al., 2017). It has been recently estimated that, although 52 they only occupy 0.2 % of the total ocean surface, they are responsible for 50% of the total neglected in studies addressing carbon dynamics (Huhta, 2007;Kristensen et al., 2012). 64 Due to their abundance, their feeding behaviour and bioturbation activities, brachyuran

97
To close this knowledge gap, the present study was designed to assess the impact of where the soil is rich in organic matter, because it provides more reactant to the oxidation 106 reaction (Howes et al., 1985); 2) the presence of S. alterniflora would increase the CO2 flux,       After we identified high and low organic matter areas, a non-vegetated patch and a S.     Figure 3). Carbon dioxide fluxes measured within vegetated 320 patches (6.39 ± 0.47 µm m -2 s -1 ) was 1.64 times higher than fluxes recorded in non-vegetated 321 ones (3.88 ± 0.58 µm m -2 s -1 ) ( Table 1 and Figure 3).  The different levels correspond to no-burrows, 16.66 burrows m -2 (low density) and 49.98 330 burrows m -2 (high density) patches.

332
We consistently observed a significant difference in CO2 fluxes coming from patches with 333 different burrow densities, regardless of the presence of S. alterniflora and the amount of 334 organic matter present (F = 8.58, df = 1, P < 0.01, ANOVA test, Table 1 and Figure 3).

335
Statistically higher CO2 releases were recorded from the high burrow density patches with 336 respect to both the low density and no-burrow patches, showing a strong influence of such 337 biotic factor on total CO2 fluxes. There was, however, a statistical difference in fluxes between 338 the high burrow density chambers tested on non-vegetated patches and the ones measured in 339 the vegetated ones (F = 3.12, df = 2, P < 0.05, interaction factor organic matter content × 340 vegetated plots, ANOVA test, Table 1 and Figure 3). We also recorded higher CO2 fluxes from 341 the areas rich in organic matter (Table 1 and Figure 3).

342
The regression analyses between the recorded CO2 fluxes and the surface area of burrow

351
This study shows that the presence and density of fiddler crab burrows are crucial to 352 significantly increase the overall CO2 fluxes of these New England salt marshes (Table 1). In  (Figure 3), to ultimately shape the carbon budget of these ecosystems.

357
In line with our hypotheses, we recorded significantly higher CO2 fluxes within the areas 358 characterized by higher organic matter. This is coherent with previous studies, which showed

369
Across the high burrow density patches, we recorded, however, higher CO2 fluxes from the 370 areas rich in organic matter (Figure 3). We hypothesise that these lower CO2 fluxes recorded 371 for the same burrow density within the low organic matter are due to the limited amount of 372 organic matter that is to oxidize by sediment microbial communities, even when exposed to

381
There was also a significant difference in fluxes between high burrow density trials carried 382 out within non-vegetated vs S. alterniflora patches (Figure 3). A sound explanation for these 383 differences comes from our study on the burrow architecture at the different sites. We found