Data from: Macrofaunal diversity patterns in coastal marine sediments: Re-examining common metrics and methods
Abstract
Complex biodiversity patterns arise in marine systems due to overlapping ecological processes, including organism interactions, resource distribution, and environmental conditions. Despite the importance of documenting these patterns, describing diversity in natural ecosystems remains challenging. Here, we investigate three nearshore sub-Arctic sites to describe benthic macroinfaunal taxa and biological traits, with the ultimate aim of determining whether common diversity metrics and typical sampling efforts adequately capture community composition in these systems. First, we assess how diversity relates to sediment depth, and examine relationships among commonly used taxonomic and functional diversity indices. Second, using a power analysis, we explore how sampling effort influences the interpretation of diversity patterns in coastal systems. We report significant variation in community composition among sites, even across small spatial scales of kilometers, and find that taxonomically diverse communities do not necessarily correspond to high functional diversity. We further find that although environmental factors such as sediment depth consistently affect macroinfaunal diversity, the direction and magnitude of these relationships are site dependent. Finally, we demonstrate that typical sampling effort for coastal benthic studies may not capture macroinfaunal community composition adequately, potentially obscuring hotspots in common diversity metrics such as taxonomic or functional richness. Conversely, indices such as Simpson’s diversity may be well-suited to resource-limited studies with restricted sampling capacity. Our results highlight the importance of adopting a multi-pronged approach to biodiversity assessment and determining optimal sample sizes for a wide range of marine benthic systems, particularly in the context of biodiversity monitoring for conservation purposes.
README: Data from: Macrofaunal diversity patterns in coastal marine sediments: Re-examining common metrics and methods
https://doi.org/10.5061/dryad.76hdr7t3k
Data were collected from three intertidal soft-sediment sites around the island of Newfoundland, Canada using manual sediment push cores. Cores were sectioned into depth layers (0-2 cm, 2-5 cm, and 5-10 cm) and macrofauna were sieved through a 500 micron mesh, enumerated, and identified to the lowest possible taxonomic level. Additional sediment samples were analyzed for grain size, total organic carbon and total nitrogen, and photopigment (chlorophyll-a and phaeopigment) concentrations, to provide context on the environmental conditions at each site.
This dataset includes:
- Raw abundance data, functional trait information, and raw sediment characteristic data for each site;
- Taxonomic and functional diversity indices calculated using common methods in the field;
- Code used to produce the figures, tables, and analyses in Clinton et al. (in revision).
Description of the data and file structure
File structure
Code: This folder contains all scripts needed to produce the figures, tables, and analyses in Clinton et al. (in revision).\
Data: This folder contains raw and processed data (.csv files). RDS files generated by the power analysis are located in the sub-folder "power".
Figures: This folder contains all figures generated in the R scripts that were not included in Clinton et al. (in revision).
Description of raw data
A core x species matrix that contains raw abundances for each taxon identified at Newman's Sound (AC) and St. Paul's (SP). The column core_id follows the format: <site code><sampling date id>_<enrichment type>_<replicate id>. Note that enrichment type is unrelated to the analysis conducted in Clinton et al. (in revision).
Contains raw abundances for each taxon identified at Neddie's Harbour (NH). Note that in this file, site code is always NH for Neddie's Harbour.
A species x trait matrix, which includes trait information for each taxon. Each trait is described in detail below. Note: The traits Diet and Feeding guild are fuzzy coded (values between 0 and 1) and divided into several modalities, which are denoted in the column name using an underscore. For example, the trait diet has three modalities: Hb (Herbivore), C (Carnivore), and Dt (Detritivore). The codes used to define each modality are also outlined below:
Diet: diet_Hb = Herbivore, diet_C = Carnivore, diet_Dt = Detritivore.\
Feeding guild: _Pr = Predator, _Gr = Grazer, _Sc = Scavenger, _Ff = Filter feeder, _SD = Surface deposit-feeder, _SSD = Sub-surface deposit-feeder.\
Reworking: bd = Biodiffusor, sm = Surficial modifier, updown = Upward/downward conveyor, epi = epipelagic\
Body size: Values in the body size column are the maximum recorded size (in mm) for that taxon.\
Mobility: Value between 0 and 1 indicating the degree of mobility exhibited by the taxon within the sediment matrix. None (value: 0) = Sedentary or only moving within a fixed tube structure. Low (value: 0.3) = Limited free movement (e.g. withdraws into sediment when disturbed). Moderate (value: 0.6) = Slow movement within sediment matrix via non-permanent burrow formation. High (value: 1) = Freely mobile within sediment in a permanent, excavated burrow system.
Contains cholorphyll a and phaeopigment concentrations from sediment at each study site. Codes are defined below, following Danovaro (2010):
Site: Abbreviation for collecton site. AC = Alien Cove (Newman Sound), SP = St. Paul's, NH = Neddie's Harbour \
A750: Absorbance of the sample at 750 nm before acidification.\
A665: Absorbance of the sample at 665 nm before acidification.\
A750a: Absorbance of the sample at 750 nm after acidification.\
A665a: Absorbance of the sample at 665 nm after acidification.\
PPu: Weight (g) of the tube containing wet sediment and MgCO3.\
PPs: Weight (g) of the tube containing dry sediment (dried for 8 days in fume hood).\
Ps: Weight (g) of the dry sediment.\
CO: Optical length (in cm).\
blank_665: Reading from spectrophotometer set to 665nm with most recent blank. Note that machine was zeroed at 750nm not 665nm.\
v: Volume of the acetone extract.
Contains organic carbon (%C) and nitrogen (%N) values for each sediment depth layer at each study site. %C and %N are equivalent to mg C or N / 100 mg sample (dry weight).
Code
All code was written using R 4.1.1 and should be run in the order in which scripts are numbered (e.g., 001-x.R, 002-x.R., etc.).
000-study-site-map.R Generates components of the study site map (Fig. 1) in Clinton et al. (in revision).
- Figure outputs:
study_sites_map.tiff,NorthAmerica_inset_map.tiff
001-taxonomic-data-setup.R Script used to clean and process the raw count data.
- Inputs:
Abundances_AllTaxa_SP_NS.csv,Abundances_AllTaxa_NH.csv. - Processed data outputs:
abundances_by_core.csv,abundances_by_depthlayer.csv,abundances_by_core_for_calculation_of_indices.csv, andabundances_by_depth_layer_for_calculation_of_indices.csv.
002-taxonomic-data-analysis.R Script used to cross-reference taxon names with the World Register of Marine Species (WoRMS) database, bin taxa, calculate species accumulation curves, calculate taxonomic diversity indices for each core (all depth layers combined), and conduct univariate (Generalized Linear Models) and multivariate (ANOSIM, SIMPER, PERMANOVA) analyses related to taxonomic diversity indices.
Inputs:
abundances_by_core.csv,abundances_by_depthlayer.csv,abundances_by_core_for_calculation_of_indices.csv,abundances_by_depth_layer_for_calculation_of_indices.csv.Processed data outputs:
WoRMS_taxon_information_unbinned.csv,mean_macrofaunal_densities_by_site.csv,mean_binned_macrofaunal_densities_by_site.csv,TD_indices.csv,top10_taxa_SIMPER.rds,top10_binned_taxa_SIMPER.rds,nmds_plot_binned_composition.rds.Figure outputs:
species_accumulation_curves.tiff(Figure 4),TD_indices_boxplots.tiff,mean_shannon_by_site_effects_plot.png,mean_richness_by_site_effects_plot.png,mean_simpson_by_site_effects_plot.png,mean_pielous_by_site_effects_plot.png,nMDS_with_species_names.tiff,nMDS_with_species_names_symbols_for_dates.tiff,nMDS_with_binned_species_names_symbols_for_dates.tiff(Fig. 3a).
003-functional-data-setup.R Calculates the functional trait space (after Villeger et al. 2011) and functional diversity indices for each core. Requires the quality_funct_space_fromdist.R, plot_funct_space.R, and multidimFD.R functions found in Supplementary materials from Villeger et al. (2011).
Inputs:
abundances_by_core.csv,abundances_by_core_for_calculation_of_indices.csv,Functional_Traits_All_Macrofauna_All_Sites.csv.Processed data outputs:
Functional_Traits_All_Binned_Macrofauna_All_Sites.csv,functional_space_coordinates_all_sites.csv,functional_space_coordinates_binned_taxa_for_indices_all_sites.csv,FD_indices.csv.
004-functional-data-analysis.R Conducts univariate (GLM) analysis of the effect of site on functional indices.
Inputs:
FD_indices.csv.Figure outputs:
FD_indices_boxplots.tiff,mean_FRic_by_site_effects_plot.png,mean_FEve_by_site_effects_plot.png,mean_FDiv_by_site_effects_plot.png.
005-taxonomic-and-functional-indices-plots.R Produces nMDS plots to visualize differences in multivariate community composition among sites (Figure 3) and performs accompanying analyses (i.e., PERMANOVAs and ANOSIMs). Also produces multi-panel box plot figure of diversity indices across all cores at all sites (Fig. 5 in Clinton et al. (in revision)).
Inputs:
TD_indices.csv,FD_indices.csv,nmds_plot_binned_composition.rds.Figure outputs:
TD_and_FD_indices_boxplots.tiff(Fig. 5),nMDS_TD_indices.tiff(Fig. 3b),nMDS_FD_indices.tiff(Fig. 3c),nMDS_all_indices.tiff,nmds_plots_all_indices.tiff,nmds_plots_composition_TD_FD.tiff(full version of Fig. 3).
006-community-weighted-means.R Calculates community-weighted mean (CWM) trait values for each core, conducts paired Wilcoxon Rank Sum test between depth-layers within each site, and fits Generalized Least Squares (GLS) models to test for differences in CWM for each trait between sites.
Inputs:
Functional_Traits_All_Macrofauna_All_Sites.csv,abundances_by_core.csv,abundances_by_depthlayer.csv.Processed data outputs:
CWM_by_core.csv,CWM_by_depth_layer.csv,CWM_by_core_not_abund_weighted.csv.Figure outputs:
CWM_with_SE.tiff(Fig. 6),CWM_two_depth_layers.tiff(Fig. 8),CWM_not_weighted.tiff.
007-diversity-index-correlations.R Calculates Pearson linear correlation between each pair of taxonomic and functional diversity indices, and produces components of Figure 9 in Clinton et al. (in revision).
Inputs:
FD_indices.csv,TD_indices.csv.Figure outputs:
scatterplots_diversity_indices.tiff,correlation_plot_diversity_indices.tiff.
008-depth-distributions.R Calculates taxonomic and functional diversity indices separately for each depth layer (0-2 cm & 2-10 cm) in each core, fits Generalized Linear Mixed-effects models (GLMM) for the effect of site, depth layer, and their interaction on each index. Also calculates p-values for GLMM using Kenward-Roger approximation.
Inputs:
abundances_by_depthlayer.csv,abundances_by_depth_layer_for_calculation_of_indices.csv,functional_space_coordinates_binned_taxa_for_indices_all_sites.csv.Processed data outputs:
TD_indices_two_depth_layers.csv.Figure outputs:
mean_abundance_and_proportions_per_depth_layer.tiff(Fig. S2),abundance_and_proportions_per_depth_layer.tiff,TD_indices_boxplots_two_depth_layers.tiff,shannon_by_site_and_DL_effects_plot.png,simpson_by_site_DL_effects_plot.png,richness_by_site_DL_effects_plot.png,pielous_by_site_and_DL_effects_plot.png,FD_indices_two_depth_layers.csv,FD_indices_boxplots_two_depth_layers.tiff,TD_and_FD_indices_boxplots_two_depth_layers.tiff(Fig. 7),FRic_by_site_DL_effects_plot.png,FEve_by_site_DL_effects_plot.png,FDiv_by_site_DL_effects_plot.png.
009-rare-taxa.R Generates heat maps representing the number of rare species at each site based on different definitions of rarity. These definitions are "threshold-based" (e.g., taxa represented by no more than 5 individuals at a given site), "percentage-based" (e.g., taxa that make up < 1% of the total macrofauna at a given site), "patchiness-based" (e.g., taxa that appear in fewer than 3 cores at a given site), and combinations thereof. Produces Fig. S1 from Clinton et al. (in revision).
Inputs:
abundances_by_core.csv,abundances_by_core_for_calculation_of_indices.csv.Figure outputs:
rarity_heatmap_patchy_percentage.tiff,rarity_heatmap_patchy_percentage_binned_taxa.tiff(Fig. S1).
010-environmental-variables.R Cleans and analyzes total organic carbon, total nitrogen, phaeopigment, and chlorophyll A data.
Input:
ChlA_Phaeopigments.csv.Processed data outputs:
calculated_photopigment_concentrations.csv,sediment_CHN_2021.csv,sediment_CHN_2024_rerun.csv.
011-power-analysis.R Conducts post-hoc power analysis using random sampling of cores, and plots results.
Inputs:
TD_indices.csv,FD_indices.csv,TD_indices_two_depth_layers.csv,FD_indices_two_depth_layers.csv.Processed data outputs:
sampled-diversity-indices.rds,sampled-diversity-indices-DL.rds,sampled-diversity-indices.rds,sampled-diversity-indices-DL.rds,sampled-diversity-indices-models_1000-iterations.rds.Figure outputs:
subsets_plot_all_sites.tiff(Fig. S3),subsets_plot_all_depth_layers.tiff(Fig. S4),significance_figure.tiff(Fig. 10),significance_figure_for_PPT.tiff.
References
Danovaro R (2010) Methods for the study of deep-sea sediments, their functioning and biodiversity. CRC Press, Taylor and Francis Group, LLC.
Villéger S, Novack-Gottshall PM, Mouillot D (2011) The multidimensionality of the niche reveals functional diversity changes in benthic marine biotas across geological time: Long-term functional diversity changes. Ecology Letters 14:561–568.