A literature-based inventory of ecological interactions in Doñana National Park

Study Extent

The data collection period for papers published from Doñana National Park and extensions of protected natural areas within the Natura 2000 Network in Andalusia was extended until May 31st, 2023.

Sampling

We conducted a bibliographic compilation through an exhaustive review of over xxx published and unpublished scientific reports that provide data on interaction diversity across taxa. These reports cover a range of kingdoms, including Animalia, Plantae, Fungi, Chromista, Bacteria, Archaea, Protozoa, and Viruses. We also reviewed non-digital literature from collections at the CSIC archive. To identify relevant scientific peer-reviewed papers on biotic interactions in Doñana Natural Park, we systematically searched for the specific keyword “Doñana” and filtered the results by ecological-related categories. The interaction data extracted from the literature review was collected in a spreadsheet using a controlled vocabulary and an early validation procedure. The database structure includes 80 columns, covering the following information: (1) survey characteristics, (2) time, date and geographic location, (3) taxonomic information, (4) anatomy of the interaction (based on an ad hoc developed ontology of species interactions), and (5) interaction intensity and outcome. We collected two types of interaction data from field studies of natural populations: (i) interaction intensity data based on sampling design studies that provide information on both sampling effort and intensity, and (2) binary interaction data (presence-absence) based on studies that did not involve sampling effort, but instead relied on occurrences of interaction events.

Quality Control

The fist measure of quality control involves searching for duplicated papers by assigning a unique citation key to each scientific report. We also carefully check thesis chapters that have been published in scientific journals to avoid double recording of information. The second measure of quality control is performed during the recording of interaction data into the spreadsheet. We use controlled vocabularies in many fields and a hierarchical data entry validation for interaction description fields based on previous filtering by kingdom partners. The third measure of quality control is applied during the integration of spreadsheets into a single table in R. We have developed a SUMHAL-WP5 package for R to optimize data integration into the final database, with an adequate validation and data quality check. During this process, we check the data recorded in each column according to controlled vocabularies and predefined value ranges. We also ensure that all indispensable data for each record is filled, and taxonomy classification is accomplished according to the GBIF Backbone Taxonomy. All these quality checks are performed automatically using GitHub Actions.

Method steps

1. The bibliographic compilation involved multiple sources of information. First, we searched the ISI Web of Science (WOS) for papers containing data on ecological interactions until May 31st, 2023. The search term "Doñana" was used to retrieve papers, and we restricted the search to the categories of Environmental Sciences-Ecology, Plant Sciences, Zoology, Entomology, Marine and Freshwater Biology, Biodiversity Conservation, Agriculture, and Forestry in WOS ("Theme" as the search field). After removing papers that were clearly out of scope, we retained xx papers from WOS. An additional search on Google Scholar yielded 98 papers using the same search term. We also reviewed the list of theses published by the Doñana Biological Station since 19xx, which comprised a total of 54 works on ecological sciences. To identify duplicates, we created a citation key for each study, and special care was taken when searching for duplicated papers included in thesis works. After removing duplicates from all the sources, we obtained a joint list of xxx papers. We then screened the abstracts or, when necessary, the main text of the articles to select only those papers providing data on ecological interactions. Finally, we read the selected papers and theses in full, resulting in a final set of xx papers and xx theses that provided data for the database on ecological interactions. We also selected a restricted set of national journals that met the following criteria: (1) the surveys were conducted in Doñana National Park or extensions to protected areas in Andalusia, and (2) the aim of the studies or reports focused on biodiversity and species ecology. A total of xxx journals were selected (names), comprising xxx reports/surveys. We carefully read all the texts to extract interaction data. Interaction data extracted from literature was collected in a spreadsheet using a controlled vocabulary and an early validation procedure. The taxonomic classifications of the interacting species were verified at lower taxonomic ranks with the help of the GBIF taxonomy (https://www.gbif.org/es/species/1?root=true). The data from various databases or spreadsheets were integrated continuously using GitHub Actions to create a single CSV file. This integration process utilized the SUMHAL-WP5 package that was specifically developed in R for this purpose. In the event of an error, the corresponding author was notified and a correction was required for the proper integration of the data. Furthermore, during the integration process, the taxonomic classification of the interacting partners was automatically filled using the GBIF Backbone Taxonomy.

N/A

1. Chordata
   rank: phylum
2. Aves
   rank: class
3. Mycobacteriales
   rank: order
4. Actinomycetia
   rank: class
5. Mycobacteriaceae
   rank: family
6. Carnivora
   rank: order
7. Actinobacteriota
   rank: phylum
8. Accipitridae
   rank: family
9. Herpestidae
   rank: family
10. Bacteria
    rank: kingdom
11. Accipitriformes
    rank: order
12. Mammalia
    rank: class
13. Animalia
    rank: kingdom

Our study area primarily encompasses the Doñana National Park, which spans parts of the provinces of Huelva, Sevilla, and Cádiz, as well as extensions of protected natural areas within the Natura 2000 Network in Andalusia.