Skip to main content
Dryad

Data from: Environmental DNA metabarcoding reflects spatiotemporal patterns of fish community shifts in the Scheldt estuary

Cite this dataset

Van Driessche, Charlotte et al. (2024). Data from: Environmental DNA metabarcoding reflects spatiotemporal patterns of fish community shifts in the Scheldt estuary [Dataset]. Dryad. https://doi.org/10.5061/dryad.4qrfj6qh2

Abstract

Estuarine ecosystems face increasing anthropogenic pressures, necessitating effective monitoring methods to mitigate their impacts on the biodiversity they harbour. The use of environmental DNA (eDNA) based detection methods is increasingly recognized as a promising tool to complement other, potentially invasive monitoring techniques. Integrating such eDNA analyses into monitoring frameworks for large spatial ecosystems is still challenging and requires a deeper understanding of the scale and resolution at which eDNA patterns may offer insights in species presence and community composition space and time. The Scheldt estuary, characterized by its diverse habitats and complex currents, is one of the largest Western European tidal river systems. Until now, it remains challenging to obtain accurate information on fish communities living in and migrating through this large ecosystem, consequently confining our knowledge to specific locations. To explore the potential of eDNA-based monitoring, we simultaneously combine stow net fishing with eDNA metabarcoding, to assess the Scheldt estuary's fish communities in space and time. In total, we detected 71 fish species in the estuary using eDNA metabarcoding, partly overlapping with historic fish community data gathered at the different study locations and in contrast to only 42 species using stow net fishing during the same survey period. Community compositions found by both detection methods varied amongst sampling locations, driven by a clear correlation to the salinity gradient. Limited effects of sampling depth and tide were observed on the eDNA metabarcoding data, allowing a significant reduction of the eDNA sampling effort for future eDNA fish monitoring campaigns in this study system. Our results further demonstrate that seasonal shifts in fish species occurrence can be detected using eDNA metabarcoding. Combining eDNA metabarcoding and stow net fishing further enhances our understanding of this vital waterway’s diverse fish populations, allowing a higher resolution and more efficient monitoring strategy.

README: Data from: Environmental DNA metabarcoding reflects spatiotemporal patterns of fish community shifts in the Scheldt estuary

I. General information

Dataset title: "Data from: eDNA metabarcoding reflects spatiotemporal patterns of fish community shifts in the Scheldt estuary"

Principle investigator: Charlotte Van Driessche (charlotte.vandriessche@inbo.be)

Co-investigators: Teun Everts, Sabrina Neyrinck, David Halfmaerten, Pieter Verschelde, Jan Breine, Dries Bonte, Rein Brys

Data collection data: April, July, September 2021

Geographic location of data collection: Belgium, Flanders (Northern Region)

Keywords: biodiversity monitoring, integrated sampling, assay sensitivity, reference database, salinity, seasonal migration

Principal funders: Research Foundation Flanders grants for strategic basic research (FWO-SB, grant number 1S23822N to CVD and grant number 1S01822N to TE) and the Research Institute for Nature and Forest

As a result of this work, an updated version of the reference database used for metabarcoding was created which can be found here: https://doi.org/10.5281/zenodo.10422227.

II. Data overview and methodological information

- Datafile: VanDriessche et al. 2024_ObiTools Count Table_Riaz.xlsx

Creation data file: the 26th of October 2023

The dataset contains eight tab pages.

1. info_tab: an overview repeating this information but facilitating the use of the Excel file and explaining again the use of all different tabs

2. raw_table_sorted: sorted raw count table as provided via ObiTools

3. table_allreps_id100_all: started from raw_table_sorted, but now discarded sequences with <100% identity match with our reference database, as well as sequances with less than 500 reads across all samples.

4. table_allreps_id100_all_frac: same as tab page 3, but in relative frequencies instead of absolute counts

5. table_allreps_id100_agglomerate: started from tab page 3, but summed the counts of sequences with the same taxonomic assignment

6. table_allreps_id100_aggl_frac: same as tab page 5, but in relative frequencies instead of absolute counts

7. table_allreps_id100_agglomerated_fish: started from tab page 5, but retained only fish species and discarded samples containing <5% fish.

8. table_allreps_id100_agglomerated_fish_f: same as tab page 7, but in relative frequencies instead of absolute counts

Cells containing 'NA' in this file, refer to 'Not Available', for example when a genetic sequence was only identified to genus level and not to species level.

Across all tab pages; the same lay-out is used. Column explanations are listed below:

ASV: number of the amplicon sequence variant

family_name: latin family name

genus_name: latin genus name

species_name: latin species name

familie: Dutch translation of the family name

categorie: Dutch translation of the large taxonomic group, e.g. fish

full_name: full species name combining the information in column B, C and D

sequence: actual sequence used for comparison and match with reference database

ASV: same as column A

Remainder of the columns include identity codes for the samples. These include: the filtercode + the season and year of sampling + the name of the sampling site + low (E) or high (V) tide + replicate 1 or 2 concurring with the fishing haul + depth level of sampling (o = deep, m = middle, b = surface) + the primer assay used (Riaz or Teleo) + the technical replicate number (1, 2 or 3)

example: E2020KRW004_L21_Doel_E1O_Riaz_1

- Datafile: VanDriessche et al. 2024_ObiTools Count Table_Teleo.xlsx

Creation data file: the 19th of September 2023

The dataset contains eight tab pages.

1. info_tab: an overview repeating this information but facilitating the use of the Excel file and explaining again the use of all different tabs

2. raw_table_sorted: sorted raw count table as provided via ObiTools

3. table_allreps_id100_all: started from raw_table_sorted, but now discarded sequences with <100% identity match with our reference database, as well as sequances with less than 500 reads across all samples.

4. table_allreps_id100_all_frac: same as tab page 3, but in relative frequencies instead of absolute counts

5. table_allreps_id100_agglomerate: started from tab page 3, but summed the counts of sequences with the same taxonomic assignment

6. table_allreps_id100_aggl_frac: same as tab page 5, but in relative frequencies instead of absolute counts

7. table_allreps_id100_agglomerated_fish: started from tab page 5, but retained only fish species and discarded samples containing <5% fish.

8. table_allreps_id100_agglomerated_fish_f: same as tab page 7, but in relative frequencies instead of absolute counts

Cells containing 'NA' in this file, refer to 'Not Available', for example when a genetic sequence was only identified to genus level and not to species level.

Across all tab pages; the same lay-out is used. Column explanations are listed below:

ASV: number of the amplicon sequence variant

family_name: latin family name

genus_name: latin genus name

species_name: latin species name

familie: Dutch translation of the family name

categorie: Dutch translation of the large taxonomic group, e.g. fish

full_name: full species name combining the information in column B, C and D

sequence: actual sequence used for comparison and match with reference database

ASV: same as column A

Remainder of the columns include identity codes for the samples. These include: the filtercode + the season and year of sampling + the name of the sampling site + low (E) or high (V) tide + replicate 1 or 2 concurring with the fishing haul + depth level of sampling (o = deep, m = middle, b = surface) + the primer assay used (Riaz or Teleo) + the technical replicate number (1, 2 or 3)

example: E2020KRW004_L21_Doel_E1O_Teleo_1

- Datafile: VanDriessche et al. 2024_Raw Stow.xlsx

Creation data file: the 20th of November 2021

The dataset contains one tab page.

1. stow_net_raw: raw stow net fishing data in numbers of individuals per species as total catch from the stow nets and as reported also on the vis.inbo dataportal.

Column explanations are listed below:

Study_site: Doel, Antwerp, Steendorp or Branst

Date: Date of sampling

Tide: Low or High

Season: Spring, Summer or Autumn

Remainder of the columns include the latin species names of the species identified morphologically on the fishing vessel.

III. Sharing access and information

This dataset is licensed under a (CC0 1.0) Creative Commons Attribution Non Commercial 1.0 Generic.

This data was used generated in the Research Article (manuscript):

Van Driessche, C., Everts, T., Neyrinck, S., Halfmaerten, D., Verschelde, P., Breine, J., Bonte, D., & Brys, R. (2024) eDNA metabarcoding reflects spatiotemporal patterns of fish community shifts in the Scheldt estuary. Science of the Total Environment.

Methods

A large-scale seasonal sampling campaign combining stow net fishing and environmental DNA (eDNA) sampling was performed in the Scheldt estuary. During three seasons (spring, summer, autumn), fish communities were investigated at four different sampling sites in Belgium (Doel, Antwerp, Steendorp, Branst). Per sampling site, fishing was performed with two stow nets during low tide, and during high tide. Per fishing haul (so twice per tide), three eDNA samples were taken, once per depth level (i.e. deep, middle or surface) of the water column using a Von Dorn sampler. Water samples were analysed using eDNA metabarcoding. This dataset includes the data of the stow net fishing as well as the read counts of the eDNA metabarcoding.

Raw metabarcoding data was deposited on the NCBI’s Sequence Read Archive (SRA) under BioProject number PRJNA1065114. The bioinformatical pipeline as used on these raw read counts is available on Zenodo (https://zenodo.org/record/3731310#.Y8pdbXbMI2w). The OBITools software was used for further processing of the generated sequence data. The resulting count table as available here was used for further quality screening and cleaning, as well as for statistical analyses.

Funding

Research Foundation - Flanders, Award: 1S23822N

Research Foundation - Flanders, Award: 1S01822N