Which mammals can be identified from camera traps and crowdsourced photographs?

Abstract

While museum voucher specimens continue to be the standard for species identifications, biodiversity data are increasingly represented by photographic records from camera traps and amateur naturalists. Some species are easily recognized in these pictures, others are impossible to distinguish. Here we quantify the extent to which 335 terrestrial nonvolant North American mammals can be identified in typical photographs, with and without considering species range maps. We evaluated all pairwise comparisons of species and judged, based on professional opinion, whether they are visually distinguishable in typical pictures from camera traps or the iNaturalist crowdsourced platform on a 4-point scale: (1) always, (2) usually, (3) rarely, or (4) never. Most (96.5%) of the 55,944 pairwise comparisons were ranked as always or usually distinguishable in a photograph, leaving exactly 2,000 pairs of species that can rarely or never be distinguished from typical pictures, primarily within clades such as shrews and small-bodied rodents. Accounting for a species geographic range eliminates many problematic comparisons, such that the average number of difficult or impossible-to-distinguish species pairs from any location was 7.3 when considering all species, or 0.37 when considering only those typically surveyed with camera traps. The greatest diversity of difficult-to-distinguish species was in Arizona and New Mexico, with 57 difficult pairs of species, suggesting the problem scales with overall species diversity. Our results show which species are most readily differentiated by photographic data and which taxa should be identified only to higher taxonomic levels (e.g., genus). Our results are relevant to ecologists, as well as those using artificial intelligence to identify species in photographs, but also serve as a reminder that continued study of mammals through museum vouchers is critical since it is the only way to accurately identify many smaller species, provides a wealth of data unattainable from photographs, and constrains photographic records via accurate range maps. Ongoing specimen voucher collection, in addition to photographs, will become even more important as species ranges change, and photographic evidence alone will not be sufficient to document these dynamics for many species.

Efficacy of machine learning image classification for automated occupancy‐based monitoring

“…We selected three model species that were of management interest and differed in size, phenotypic similarity to sympatric species, and densities (Appendix S2). Our low incidence of errors with single review aligned with research demonstrating that our target species tended to “always” or “usually” be distinguishable in images (Kays et al., 2022).…”

“…Manual image classifications by an expert observer (“Gold Standard;” Gomez Villa et al., 2017) or two independent observers (Cove et al., 2021; Tabak et al., 2019) are time consuming, but have been employed to generate research‐grade data. Manual image classification accuracy varies among species but tends to be lower for species that are small or sympatric with phenotypically similar species (Gooliaff & Hodges, 2018; Kays et al., 2022; Zett et al., 2022). We selected three model species that were of management interest and differed in size, phenotypic similarity to sympatric species, and densities (Appendix S2).…”

The potential of iNaturalist for bee conservation research—A study case in a southern Brazilian metropolis

“…This represents a major obstacle for conducting reliable biodiversity research without specimen examination. Identification is not a simple task for any organism, even for mammals, many smaller species could not be accurately identified by photographs (Kays et al, 2022). The poor quality of some photographs could hamper identification as verified by Rosa et al (2022).…”

Image‐based analyses from an online repository provide rich information on long‐term changes in morphology and human perceptions of rhinos

“…antlers, horns or tusks), photographs of these animals can facilitate analyses of morphological changes in response to changes in human behaviour or environmental conditions. Many medium‐sized and small mammals should also be considered for image repositories, given that they can generally be identified to a species level from photographs (Kays et al, 2022), though they are unlikely to be as prominently featured in artistic representations. Groups such as voles, shrews and mice, which are difficult to differentiate between in photographs, are likely to be least suitable. Include associated information with each image—for effective analysis, we suggest all images should have associated information recorded, including their location, date of production, species identification and original context.…”