Wildlife camera monitoring of a Norwegian alpine carnivore guild 2011-2020
Citation
Rød-Eriksen L, Vang R (2023). Wildlife camera monitoring of a Norwegian alpine carnivore guild 2011-2020. Norwegian Institute for Nature Research. Sampling event dataset https://doi.org/10.15468/rtt7mf accessed via GBIF.org on 2024-12-05.Description
Daily aggregated wildlife camera trap data from alpine regions in south-central Norway from 2011 to 2020. The data stems from several projects which have aimed to document the occurrence of Arctic fox in relation to its most common competitors and predators (red fox, wolverine, golden eagle and white-tailed eagle).Sampling Description
Study Extent
Sampling was carried out from March to May in each mountain region in each year. The region of Dovrefjell-Sunndalsfjella (62°23′48″N 9°10′23″E; 18 cameras) only had sampling for the first three years (2011-2013), whereas Sylan-Kjølifjell-Forollhogna (62°40′45″N 10°47′30″E; 7 cameras) was sampled from 2011 to 2019. The other two areas, Børgefjell (65°11′00″N 13°54′00″E; 17 cameras) and Blåfjella-Skjækerfjella-Lierne (64°13′00″N 13°06′33″E; 10 cameras), were sampled for the whole study period 2011-2020.Sampling
Camera traps were set to a 5 minute time lapse interval. Memory cards and batteries were checked by field personnel twice during the sampling period per year (once after three weeks, and then again after 4-5 weeks when the camera was deactivated). Memory cards were sent by mail to, or shared electronically with, the project owner. Images were then copied to a central data server with scheduled backups at NINA. Image processing was carried out by NINA personnel and students attached to related research projects.Quality Control
Prior to the initial deployment of cameras, NINA developed a standard protocol for camera trap studies. The protocol included sections on how to prepare for the field work (camera configuration, equipment list, geographic information, handling of carcasses etc.) and how to correctly place the cameras (height above ground, angling towards carcass, direction of camera to avoid direct sunlight etc.) for best possible data quality. The protocol also included clear instructions on how to share the images with the project owner. Default equipment were either replaced immediately (if possible) or during the camera checks. During image processing, quality control was performed by doing double processing by two different people of randomly selected parts of the data. Any decrepancies were assessed and the data set corrected. The complete dataset was controlled again through an automated script during compilation and preparation for open publication.Method steps
- The images stored in the central data server were processed by first applying a script (using the R language) to extract image metadata (filename, date/time, camera serial number, recorded temperature etc.). The resulting text files (CSV) were formatted to a predefined template used to manually process the images. Image processing included recording of both empty and non-empty images, where non-empty images were classified to observed species, as well as registering potential failed pictures (technical failures, snow/ice on camera lens or other failure that made species identification impossible). In addition, the status of the carcass was recorded as either present (available) or absent (consumed). All image metadata were then compiled and stored in a central SQLite database. During compilation, the quality of the data set was verified through checking that each metadata file was formatted uniformly, that data was recorded for the correct parameter, standardization of nomenclature, and verifying correct association between observations and geographical location. The final data set was then aggregated to daily detections/non-detections of each target species and formatted to fit the Darwin Core standard.
Additional info
Parts of this data set have been utilized in Rød-Eriksen et al. (2022): https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2656.13875Taxonomic Coverages
All positive camera trap observations were identified to species. Both the blue and white morph of the Arctic fox was detected, but aggregated to detection/non-detection of the species. Eagle observations were reviewed by experts for correct species classification.
-
Haliaeetus albicillacommon name: White-tailed eagle rank: species
-
Aquila chrysaetoscommon name: Golden eagle rank: species
-
Gulo gulocommon name: Wolverine rank: species
-
Vulpes lagopuscommon name: Arctic fox rank: species
-
Vulpes vulpescommon name: Red fox rank: species
Geographic Coverages
The data has been collected within four major mountain areas in south-central Norway: Børgefjell (65°11′00″N 13°54′00″E), Blåfjella-Skjækerfjella-Lierne (64°13′00″N 13°06′33″E), Sylan-Kjølifjell-Forollhogna (62°40′45″N 10°47′30″E) and Dovrefjell-Sunndalsfjella (62°23′48″N 9°10′23″E).
Bibliographic Citations
Contacts
Lars Rød-Eriksenoriginator
position: Researcher
Norwegian Institute for Nature Research
NO
email: lars.rod-eriksen@nina.no
userId: http://orcid.org/0000-0003-0482-4384
Roald Vang
metadata author
position: Section Manager
Norwegian Institute for Nature Research
Trondheim
NO
email: rv@nina.no
userId: http://orcid.org/0000-0002-4006-8689
Lars Rød-Eriksen
metadata author
position: Researcher
Norwegian Institute for Nature Research
NO
email: lars.rod-eriksen@nina.no
userId: http://orcid.org/0000-0003-0482-4384
Lars Rød-Eriksen
administrative point of contact
position: Researcher
Norwegian Institute for Nature Research
NO
email: lars.rod-eriksen@nina.no
userId: http://orcid.org/0000-0003-0482-4384
Nina Elisabeth Eide
administrative point of contact
position: Senior Researcher
Norwegian Institute for Nature Research
NO
email: nina.eide@nina.no
userId: http://orcid.org/0000-0002-7645-3300