Data from: Soil microarthropod communities along salt marsh transects of the Wadden Sea are predominantly structured by niche differentiation
Data files
Apr 21, 2026 version files 53.75 KB
-
005_Appendix_Table_1_Collembola.xlsx
17.29 KB
-
006_Appendix_Table_2_Oribatida.xlsx
14.73 KB
-
007__Appendix_Table_3_Mesostigmata.xlsx
16.02 KB
-
README.md
5.70 KB
Abstract
The relative importance of deterministic versus stochastic processes in community assembly is widely discussed in ecology. Stochastic models explain community assembly by ignoring the niches of species, whereas deterministic models assume the niche concept to be key for understanding the assembly of species. Whether stochastic or deterministic processes dominate in shaping the community composition of soil animals is gaining increased attention. Here, we investigated the spatial heterogeneity of soil microarthropod communities (Collembola, Oribatida, Mesostigmata) along a salt marsh transect from the upper salt marsh (USM) to the lower salt marsh (LSM) to the pioneer zone (PZ) on three islands, i.e., Norderney, Spiekeroog, and Wangerooge, in the Wadden Sea of Germany. We hypothesized that microarthropod communities are predominantly structured by niche-based processes and change in a deterministic way from the USM to the LSM to the PZ, whereas microarthropod communities differ little between the three islands of similar history and position in the Wadden Sea. Supporting our hypothesis, Oribatida and Mesostigmata communities differed more strongly between the three zones than between the islands, indicating that environmental factors changing along the salt marsh gradient select for specific mite communities in a deterministic way. Collembola communities also differed between the three zones,s but also between the islands, indicating that they are structured by both environmental filtering as well as stochasticity. However, even the differences in mite and Collembola communities between the islands may - at least in part - be explained by environmental filtering, e,.g. by differences in nutrient input from the Ems river estuarine. Overall, the results indicate that stochasticity plays only a limited role in structuring microarthropod communities in the dynamic marine – terrestrial boundary of salt marshes, pointing to the importance of niche differentiation and environmental filtering.
Dataset DOI: 10.5061/dryad.xd2547dw7
Description of the data and file structure
Appendix Table 1 Collembola
List of all Collembola collected along the salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone) of the three islands (Norderney, Wangerooge, Spiekeroog)
Appendix Table 2 Oribatida
List of all Oribatida collected along the salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone) of the three islands (Norderney, Wangerooge, Spiekeroog)
Appendix Table 3 Mesostigmata
List of all Mesostigmata collected along the salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone) of the three islands (Norderney, Wangerooge, Spiekeroog)
Files and variables
File: 005_Appendix_Table_1_Collembola.xlsx
Description: List of all Collembola collected along the salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone) of the three islands (Norderney, Wangerooge, Spiekeroog)
Variables
- Insel: three islands (Norderney, Wangerooge, Spiekeroog)
- Transekt: salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone)
File: 006_Appendix_Table_2_Oribatida.xlsx
Description: List of all Oribatida collected along the salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone) of the three islands (Norderney, Wangerooge, Spiekeroog)
Variables
- Insel: three islands (Norderney, Wangerooge, Spiekeroog)
- Transekt: salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone)
File: 007__Appendix_Table_3_Mesostigmata.xlsx
Description: List of all Mesostigmata collected along the salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone) of the three islands (Norderney, Wangerooge, Spiekeroog)
Variables
- Insel: three islands (Norderney, Wangerooge, Spiekeroog)
- Transekt: salt marsh transect (USM = upper salt marsh, LSM = lower salt marsh, PZ = pioneer zone)
Code/software
All statistical analyses were carried out in R 4.3.1 (R Core Team, 2023). Rarefaction analysis was performed using the iNEXT() function in the ‘iNEXT’ package (Hsieh et al. 2024) for the three islands and the three zones to evaluate if sample effort was adequate for reliable species richness estimation. Prior to the analyses, animal species that occurred in one sample only were excluded. The analyses were performed for Oribatida, Collembola, and Mesostigmata separately. Differences in density and diversity between the islands and salt marsh zones were inspected by analysis of variance (ANOVA) with the fixed factors “Island” and “Salt marsh zone” using the aov() function in the ‘stats’ package. Prior to ANOVAs, data were log-transformed to improve homogeneity of variance.
Shifts in community composition along the salt marsh transects and across the islands were investigated by abundance-based pairwise dissimilarity using the beta.pair.abund() function in the ‘betapart’ package (Baselga 2013, 2017; Baselga et al. 2023). The overall abundance-based pairwise dissimilarity (βbc), measured as the Bray–Curtis index, was further partitioned into two additive components: (i) turnover component (= balanced variation in abundance, βbc.bal), whereby individuals of species in one site are substituted by the same number of individuals of different species in another site; and (ii) nestedness component (= abundance gradients, βbc.gra), whereby individuals are lost from one site to another, that is, βbc = βbc.bal + βbc.gra (Baselga 2013, 2017). To delineate the degree of dissimilarity between soil microarthropod assemblages, cluster heat maps based on the UPGMA dendrograms of βbc, βbc.ba,l and βbc.gra matrices were produced using the heatmaply() function in the ‘heatmaply’ package (Galili et al. 2018). The analyses were performed for all three taxa together, but also separately for Collembola, Oribatida, and Mesostigmata.
We performed variance partitioning analysis to investigate how much of the variation is explained by the independent factors “Island” (Norderney, Spiekeroog, Wangerooge) and “Salt marsh zone” (USM, LSM, PZ) using the varpart() function in the ‘vegan’ package (Oksanen et al. 2022). Subsequently, we used Redundancy Analysis (RDA) with 999 permutations to test if the effects of the two factors “island” and “salt marsh zone” were significant.
Shifts in community composition across the islands and along the salt marsh transects were investigated by Non-metric Multidimensional Scaling (NMDS) using the metaMDS() function in the ‘vegan’ package (Oksanen et al. 2022). This resulted in a dataset with two dimensions for Oribatida, three dimensions for Collembola, and four dimensions for Mesostigmata (dimcheckMDS() function in the ‘goeveg’ package; Goral & Schellenberg, 2021). The values of the NMDS axes were used as dependent variables in MANOVA (manova() function in the ‘stats’ package) with “Island” and “Salt marsh zone” as fixed factors. To depict species overlap between zones and islands, Venn diagrams were used (ggVennDiagram() function in the ‘ggVennDiagram’ package; Gao et al. 2021).
Access information
Other publicly accessible locations of the data:
- No
Data was derived from the following sources:
- No