975 Danube fish occurrence database

Title

Danube fish occurrence database

Description

The present database compiles and standardize fish occurrence datasets from federal agencies, research institutes, and conservation organizations, integrating data from sources such as the Global Biodiversity Information Facility, the Joint Danube Surveys, and the European Fish Index.  It contains 100,308 occurrence records across 113 fish species, representing 29 families and 17 orders, with a temporal range from 1856 to 2024, organized into 40 fields. In total, 473,503 entries were collected and subsequently subjected to quality checks and cleaning procedures. To facilitate data collation, formatting, and quality control, an R package, danubeoccurR , was developed, which streamlined the entire process. Additional R packages, hydrographr and specleanr, were incorporated into the workflow to aid in data manipulation and geospatial analysis.

A visualization of the spatial distribution of records is available in https://geo.igb-berlin.de/layers/geonode:danube4all_fish_occurrence_records.

Taxonomic Validation: The taxonomic names of species were verified against FishBase to ensure compliance with the most up-to-date fish taxonomy.
    
Spatial Distribution Validation: The spatial distribution of species occurrences was assessed to ensure that the recorded locations were geographically plausible and consistent with known habitats for each species. To achieve this, the dataset was first compared against environmental maps to identify and flag potential environmental outliers, and subsequently cross-checked by the data provider.
    
Temporal Validation: The temporal distribution of occurrences was examined to check for inconsistencies or improbable records.
    
Data Completeness and Consistency: The dataset was examined for missing values, duplicates, and inconsistencies in key fields such as coordinates, dates, and species names. Gaps in data were identified, and missing or inconsistent records were flagged for review and potential correction.

Data Access and Format: The dataset is available in a standard tabular format (CSV) using Darwin Core-compliant terminology to ensure compatibility with biodiversity databases. Users should refer to the metadata file for a detailed description of the column names. For convenience, a custom function named split_and_save_csv() is provided in danubeoccurR to split the occurrence dataset into independent datasets.

Geospatial Considerations: Species occurrences are georeferenced based on available locality information. Users should be aware that some records may have variable spatial precision, particularly historical occurrences. It is recommended to apply spatial filtering techniques suited to the intended analysis. For example, the coordinate uncertainty provided for records sourced from GBIF can help determine whether a record is suitable for a given analysis. Additionally, the function snap_points_on_map() in danubeoccurR allows users to manually adjust occurrence points for greater precision.

Taxonomic Standardization: Users are advised to cross-check species names with updated taxonomic databases if taxonomic revisions occur after the dataset's publication.

Data Quality and Potential Limitations: While efforts were made to standardize and clean the data, users should consider potential sources of bias, including sampling effort variations, taxonomic misidentifications, or incomplete historical records. Some records have been flagged as environmental outliers based on inconsistencies between species occurrence and expected environmental conditions. These flagged records should be reviewed carefully and may require further investigation or validation before inclusion in analyses.

 

Authors:

Species Groups

Fish

Study site

Danube

GeoNode references

GeoNode layers

• Danube4all Fish Occurrence Records

Contact

Yusdiel Torres-Cambas

Licence for data

All rights reserved. Please send a request to Yusdiel Torres-Cambas if you like to use this data. Mind our data policy: IGB Data Policy

Project

DANUBE4all Project Website

Metadata files

Data files (e.g. excel)

Machine Readable Metadata Files

FRED provides all metadata of this package in a maschine readable format. There is a pure XML file and one EML file in Ecological Metadata Language. Both files are published under the free CC BY 4.0 Licence.

• Danube_fish_occurrence_database.xml
• Danube_fish_occurrence_database.eml