Salt marshes are located between the marine and terrestrial systems. Because they form as sediment accumulates, they comprise a gradient of shore height with differing inundation frequencies and associated abiotic soil conditions. Along this gradient, both autochthonous vascular plant resources and allochthonous marine algal or detrital resources are available, with the availability of both varying with season and salt marsh zone. However, little is known about the importance of either resource for the soil animal food web. We investigated both spatial and temporal resource use of the soil macro- and mesofauna of the salt marsh using neutral lipid fatty acids (NLFAs). Generally, irrespective of season and zone the soil animal food web relied on carbon originating from autochthonous vascular plants and associated bacteria and fungi, with the role of bacteria generally exceeding that of fungi. However, the channelling of fungal resources consistently peaked in October, whereas seasonal changes in the channelling of plant and bacterial resources varied among salt marsh zones. Further, variations in the channelling of resources with season and zone varied among salt marsh animal species. Although being only minor, allochthonous resources of marine origin contributed to soil food web nutrition across salt marsh zones and seasons. The contribution of algae to soil food web nutrition depended on inundation frequency and season, i.e. algal productivity. Overall, the results demonstrate that the salt marsh soil fauna predominantly relies on autochthonous vascular plant resources, with the contribution of allochthonous marine resources being minor and restricted to few taxa.

Samples were taken along five transects spanning across the USM, LSM and PZ on the island of Spiekeroog (Wadden Sea National Park, Germany; 53˚45’2”- 53˚47’1”N, 7˚40’0”-7˚49’1”E) in April (spring), July (summer) and October (autumn) 2019. Within each transect, one soil core of 20 cm diameter was taken per zone and separated into two layers (0–5 and 5–10 cm depth). Cores were stored in plastic containers and kept at ambient temperature until soil fauna extraction at the University of Göttingen. Soil fauna was extracted using heat (Kempson et al. 1963); animals were extracted into a 1:1 mixture of Glycol and water. Once extracted, the animals were filtered through 45 µm gauze, flushed with water and placed into 70 % ethanol for storage at -20°C. Animals were identified under a stereomicroscope, macrofauna was identified to species, whereas mesofauna species were grouped to higher taxonomic units to gain sufficient material for lipid extraction. To achieve sufficient fauna material soil horizons and samples were in part pooled. The identification followed (Weigmann 2006), and (Schaefer 2018). Sorting was done within 2–4 weeks after extraction to minimize the loss of lipids due to placement in ethanol; prior to storage the ethanol was evaporated and the animals placed at -20°C.

Lipid extraction and allocation:

Animals were placed into 10 mL tubes and extracted as described in Haubert et al. (2004). In brief, animals were shaken in 5 ml extraction solution [chloroform/methanol/0.05 M phosphate buffer (pH 7.4), 1:2:0.8] overnight. Then, the extract was transferred to fresh tubes with additional 2.5 ml extraction solution and shaken for 1 h. Chloroform and distilled water were added (0.8 ml each), vortexed and centrifuged at 1500 rpm at 7–10 °C for 5 min. The top phase was removed, and the remaining phase was fractionated in silica columns (Chromabond® SiOH (3 ml), Machery-Nagel™, Düren, Germany) and eluted with 1.5 ml then 2 ml chloroform. Samples were then dried at 30 °C in a vacuum centrifuge before saponification with 1 ml of a sodium hydroxide–methanol solution (45 g NaOH, 150 ml CH₃OH, 150 ml distilled H₂O) at 100°C for 30 min. Followed by methylation with 2 ml HCl – methanol solution (325 ml 6.0 N HCl, 275 ml CH₃OH) at 80°C for 10 min. Finally, neutral lipids were extracted into hexane – methyl tertiary butyl ether (1:1) and washed with liquid NaOH (10.8 g NaOH, 900 ml distilled H₂O). Lipids were stored in 1.5 ml GC-vials, capped and stored at -20°C until gas chromatography.

Lipids were separated using a gas chromatograph (Clarus 500, PerkinElmer Corporation, Norwalk, USA) equipped with an Elite-5 capillary column (30 m x 0.32 mm i.d., film thickness 0.25 mm, PerkinElmer, Norwalk, USA). The analysis started with 60°C for 1 min, increased by 30°C/min to 160°C; followed by 3°C/min increase to 260°C; injection temperature was 250°C, with helium as carrier gas. Lipids were identified by retention time, based on standard mixtures composed of 37 FAMEs (Fatty Acid Methyl Esters) ranging between chain lengths of C11-C24, as well as 26 BAMEs (Bacterial Acid Methyl Esters, Sigma-Aldrich, St Louis, USA) and algal standards for 16:2ω6,9 and 16:3ω3,6,9 (Larodan AB, Solna, Sweden) (Buse et al. 2013). Lipid concentration was calculated as percentages Only lipids contributing >1 % of total were included in the analysis. Lipids were aggregated into six NLFA marker groups including algae (14:0; 16:2ω6,9; 16:3ω3,6,9 and 20:5ω3,6,9,12,15), animals (20:1ω9), bacteria (a15:0; i15:0; i16:0; 16:1ω7; i17:0 and 18:1ω7), fungi (18:2ω6,9), vascular plants (18:1ω9 and 24:0) and unspecific (13:0, 14:1, 15:0, 16:0, 17:0, 17:1, 18:0, 20:0, 20:2, 20:3ω6,9,12, 20:4ω6,9,12,15).

All values given in data table are percentages % (see above).

Program required: Microsoft Excel, or similar