Weak effect of urbanisation on bdelloid rotifers living in lichens
Data files
Feb 20, 2024 version files 203.28 KB
-
ESM_04_-_speedy_bdello.csv
-
ESM_07_-_Speedy_COI.csv
-
ESM_08_-_Speedy_COI_data.csv
-
README.md
Abstract
Human activities have an overwhelming impact on the natural environment, leading to a deep biodiversity crisis whose effects range from genes to ecosystems. We here analysed the effect of such anthropogenic impacts on bdelloid rotifers (Rotifera: Bdelloidea), for whom these effects are poorly understood. We targeted bdelloid rotifers living in lichen patches across urbanisation gradients in Flanders and Brussels (Belgium). Urbanisation was measured as the percentage of built-up area across different spatial scales, at circles from 50 m to 3,200 m of radius around the lichen. Urbanisation effects on biodiversity were assessed on abundance, species richness, and community-weighted mean body size of bdelloid rotifers, as well as on genetic diversity of one of the most common and widespread bdelloid species, Adineta vaga. Overall, no negative effect of urbanisation was found at any diversity level and at any spatial scale. Counterintuitively, built-up area quantified at the largest spatial scale had a positive effect on abundance. These results leave open the question of whether negative effects of urbanisation are present for bdelloid rotifers, or if such effects are only visible at even larger spatial scales.
README: Weak effect of urbanisation on bdelloid rotifers living in lichens
https://doi.org/10.5061/dryad.jm63xsjjb
Here we report all the datasets and R scripts that we used to test our hypotheses about rotifer community abundance and assemblage in a context of increasing urbanisation.
Description of the data and file structure
- The main dataset, contained in the file “ESM_04-speedy_bdello.csv” describes each of the 79 samples collected for this study.
- Column 1, “lichen.patch”, uniquely defines the sampled lichen patch. It coincides with column 28, “sampleID”.
- Columns 2-22 report individual quantity of each bdelloid rotifer species. For example: in the lichen sample Gg8 only 2 individuals of Macrotrachela ehrenbergii were identified.
- Column 23, “abundance”, reports the total abundance of rotifers found in the lichen patch. For example, in the sample Gg8 only 2 rotifers were found, hence quantity is 2.
- Column 24, “richness”, informs on how many different species were found in the patch of lichen collected. For sample Gg8 the richness value is 1 as only one species was found, Macrotrachela ehrenbergii
- Column 25, “mean_size” is the community-weighted mean body size, calculated as the average of the species-specific body length for all sampled species in a community, weighted by species abundance in the community, as explained in the manuscript.
- Column 26, “Waypoint”, gives the exact GPS coordinates of the spot where the lichen was collected, unlike the coordinates, which identify the 3x3 km plot.
- Column 27, “sample” was the alpha-numeric code used to identify the tube of the sample collected for laboratory analyses.
- Column 28, “sampleID”, coincides with Column 1.
- Column 29 states when the sample was collected in DD-MM-YYYY format.
- Columns 30, “Plot”, identifies the 3x3km plot.
- Columns 31, “Subplot”, identifies the 200x200m subplot.
- Column 32, “ID”, is a number that identifies the lichen collected. The unique lichen patch would be a fusion of columns 30-31-32, i.e, “G-g-16” as seen in column 28, “SampleID”. So, for example, sample “Gg16” was sample n.16 collected in plot “G” and subplot “g”.
- Columns 33-35 are the geographical coordinates in degrees of the 3x3km plot location. Altitude, "Alt", is expressed in metres above sea level.
- Column 36, “PatchID” summarises information on the urbanisation level of the plot sampled. For example, P16SG. P= Plot, 16= ID (see column 32), S= Subplot, G= refers to urbanisation level of subplot as seen in column 38.
- Columns 37-38, “plot colour” and “subplot colour” are categorical variables that summarize the level of the urbanisation of the plot and the subplot (Red-High level, Yellow-Medium level and Green-Low level.).
- Columns 39, “Substrate” is a categorical variable that admits 2 categories: “artificial”, meaning that the lichen was collected from an artificial surface or “tree” meaning the lichen was found on a tree.
Column 40, “Substrate 2” works in a similar manner but gives more information on the substrate: i.e., genus of the tree or type of artificial surface (roof, metal, concrete etc.).
Columns 41-48, “BU50, BU100, BU200” etc. quantify the level of urbanization present in the plot as explained in the Methods section.
- Each row of the dataset contained in the excel file “ESM_07 - Speedy_COI.csv” uniquely identifies the sample, animal and patch collected and gives the sequence of the COI mitochondrial marker of each rotifer of the species Adineta vaga
- The dataset contained in the file “ESM_08 - Speedy_COI_data.csv” is identical to the file “ESM_04-speedy_bdello.csv” plus two more columns: “Haplotypes” which tells the number of haplotypes found and “Sequenced_animals” which tells the number of animals sequenced.
Code
1. The R file “ESM_03-speedy_bdello.R” allows to run the ecological analyses as they are presented in the manuscript.
2. The R file “ESM_05 - Speedy_COI.R” allows to run the analyses of genetic diversity as they are presented in the manuscript.
3. The FAS file “ESM_06 - Speedy_COI.FAS” gives the COI alignment to be used for the analyses on genetic diversity following the R script “ESM_05 - Speedy_COI.R”.
Methods
Communities of bdelloid rotifers were sampled by collecting one lichen patch in each subplot. We selected lichens of the genus Xanthoria, for which bdelloid rotifer communities have been previously studied in Europe (Fontaneto et al. 2011). Lichens of the genus Xanthoria are among the most abundant in urban and rural areas, apparently unaffected by urbanisation levels (Nekrošienė 2012). Sampling was performed in June and July 2013. Suitable Xanthoria patches could be found in all but two subplots: total sample size is thus 79 lichens and not 81.
The selection of the lichen was haphazard: the first suitable lichen patch encountered in each subplot was collected. Dry lichen thalli between 5 and 10 cm2 were cut from the substrate with a knife and kept in zip lock bags. For each lichen sample, an area of 2.5 cm2 was hydrated with distilled water in a plastic petri dish.
All active bdelloid rotifers that recovered from dormancy within the four hours following hydration in the lab were sorted and identified to species level according to Donner (1965). Previous studies on bdelloid rotifers in these lichens (Fontaneto et al. 2011) revealed that animals start recovering between 10 and 40 minutes after hydration of the sample and that no more additional bdelloid rotifers usually recover after four hours. The very few dormant stages still found in the sample that did not recover after that time were considered dead, impossible to identify at any taxonomic rank, and excluded from the analyses.
All living bdelloids were isolated, counted, and identified to species level to obtain data on (1) abundance and (2) species richness. The other descriptor of community-level diversity was (3) community-level body size (Merckx et al. 2018b), calculated as community-weighted mean body size, which is the average of the species-specific body length for all sampled species in a community, weighted by species abundance in the community. Measurements of body length for the observed species, obtained from the observed animals, are reported in Merckx et al. (2018b).
The most common and abundant species in all samples, Adineta vaga, was selected to obtain metrics of genetic diversity. DNA was extracted from each of the individuals found in a separate extraction of bdelloids from the part of each lichen patch that was not used for the morphological identifications. We amplified the barcoding fragment of the mitochondrial marker cytochrome c oxidase subunit I, COI, using Folmer primers with optimised protocols for this species (Debortoli et al. 2016). COI sequences were trimmed to a total common length of 605bp and aligned using default settings in MAFFT v7 (Katoh et al. 2019), confirming correct amino acid translation. Species identity was confirmed through online BLAST searches (Ye et al. 2006). The number of A. vaga haplotypes for each population of the species in a lichen patch was used as a metric of genetic diversity.