Roadkill monitoring during the period 2003-2006 in the Kresna Gorge, SW Bulgaria

Study Extent

18-km section of the E-79 road within the boundaries of the Kresna Gorge

Sampling

The road was divided into 81 sections of various lengths. The beginning and end of each section was marked with a paint in the field. Sections were chosen as to correspond to specific biotopes or parts of the road (e.g. e beginning and the of a bridge is one section). Later one, some sections were divided into subsections (marked with letters, e.g. 18a and 18b). Within a road section, dead animals were recorded on each visit (usually every week) by a team of at least 2 observers who walked on foot along the road. Corpses recorded at a single visit were then thrown far away to avoid double counting during the next visits. The position of the animal on the road as a distance from the banquet was also recorded.

Quality Control

The originally provided dataset was inspected visually for consistency. Visualizing the sections revealed that for some sections the first or the last point had swapped latitude and longitude, or were not in the correct order. Occurrences with the following eventID's (corresponding to the road section) were manually corrected as follows: 18 (endpoint), 18b (endpoint), 20 (first point), 34 (endpoint), 35 (first point). The date at line 2976 with missing month and date was manually corrected to "2023-11-17". Month and day assumed with assumed based on its position in the original table.

Method steps

1. Taxa names were combined into a single field, regardless of the rank, which was then exported as a checklist and sent to GBIF's species look-up tool: https://www.gbif.org/tools/species-lookup. The obtained normalized checklist was then visually inspected and non-matched or taxa with wrong matches were manually corrected (e.g. the higher ranks), and then joined to the original dataset by the original taxon name.
2. Each section's start, end, and midpoint were georeferenced ad hoc using Google Earth (assumed error: 30 m). The IDs of the monitored road sections ID were treated as eventID.
3. In a GIS environment, the section's georeferenced first and endpoints were used to split the road into separate lines. The points of the latter lines (road sections) were used to create the geometry of the eventID (i.e. the footprintWKT). The midpoint was provided as the coordinates of the event.
4. Scripts and files used to produce the source are available here: https://github.com/spalp/GBIF-data-mobilization. Files: "GBIF_reshape_and_clean_Kresna.R", "Checklist_GBIF_normalized_Kresna.txt", "wktExportFinal_Kresna.xlsx".

N/A

1. Chordata
   rank: phylum
2. NA
   rank: order
3. Lacertidae
   rank: family
4. Fringillidae
   rank: family
5. Ranidae
   rank: family
6. Aves
   rank: class
7. Bufonidae
   rank: family
8. Anura
   rank: order
9. Rodentia
   rank: order
10. Passeriformes
    rank: order
11. Turdidae
    rank: family
12. Rhinolophidae
    rank: family
13. Carnivora
    rank: order
14. Colubridae
    rank: family
15. Muridae
    rank: family
16. Paridae
    rank: family
17. Mustelidae
    rank: family
18. Chiroptera
    rank: order
19. Squamata
    rank: class
20. NA
    rank: family
21. Vespertilionidae
    rank: family
22. Amphibia
    rank: class
23. Mammalia
    rank: class
24. Animalia
    rank: kingdom

Kresna Gorge, SW Bulgaria