• Camera settings and biome influence the accuracy of citizen science approaches to camera trap image classification

      Egna, Nicole; O'Connor, David; Stacy-Dawes, Jenna; Tobler, Mathias W.; Pilfold, Nicholas W.; Neilson, Kristin; Simmons, Brooke; Davis, Elizabeth Oneita; Bowler, Mark; Fennessy, Julian; et al. (2020)
      Scientists are increasingly using volunteer efforts of citizen scientists to classify images captured by motion-activated trail cameras. The rising popularity of citizen science reflects its potential to engage the public in conservation science and accelerate processing of the large volume of images generated by trail cameras. While image classification accuracy by citizen scientists can vary across species, the influence of other factors on accuracy is poorly understood. Inaccuracy diminishes the value of citizen science derived data and prompts the need for specific best-practice protocols to decrease error. We compare the accuracy between three programs that use crowdsourced citizen scientists to process images online: Snapshot Serengeti, Wildwatch Kenya, and AmazonCam Tambopata. We hypothesized that habitat type and camera settings would influence accuracy. To evaluate these factors, each photograph was circulated to multiple volunteers. All volunteer classifications were aggregated to a single best answer for each photograph using a plurality algorithm. Subsequently, a subset of these images underwent expert review and were compared to the citizen scientist results. Classification errors were categorized by the nature of the error (e.g., false species or false empty), and reason for the false classification (e.g., misidentification). Our results show that Snapshot Serengeti had the highest accuracy (97.9%), followed by AmazonCam Tambopata (93.5%), then Wildwatch Kenya (83.4%). Error type was influenced by habitat, with false empty images more prevalent in open-grassy habitat (27%) compared to woodlands (10%). For medium to large animal surveys across all habitat types, our results suggest that to significantly improve accuracy in crowdsourced projects, researchers should use a trail camera set up protocol with a burst of three consecutive photographs, a short field of view, and determine camera sensitivity settings based on in situ testing. Accuracy level comparisons such as this study can improve reliability of future citizen science projects, and subsequently encourage the increased use of such data.
    • Elephants, ivory, and trade

      Wasser, Samuel; Poole, Joyce; Lee, Phyllis; Lindsay, Keith; Dobson, Andrew; Hart, John; Douglas-Hamilton, Iain; Wittemyer, George; Granli, Petter; Morgan, Bethan J.; et al. (2010)
      ...Tanzania and Zambia (1 5, 1 6) are exploit- ing this restricted moratorium in their peti- tions. Approval requires demonstration that their elephant populations are secure, law enforcement is effective, and sales will not be detrimental to elephants....
    • Information sharing for gorilla conservation: a workshop in Ruhija.

      Imong, I.; Abwe, Ekwoge E.; Ikfuingei, R.; Onononga, J.R.; Makaga, L. (2012)
      Gorilla conservationists and researchers working on the ground at different sites often face the challenge of accessing valuable yet unpublished information about ongoing projects outside their immediate locality, and sharing experiences on their respective projects. Poor information sharing among field workers means that those planning or carrying out projects at one site may not be able to learn from the experiences of others who might have implemented similar projects at other sites.
    • Landscape-level changes to large mammal space use in response to a pastoralist incursion

      Masiaine, Symon; Pilfold, Nicholas W.; Moll, Remington J.; O'Connor, David; Larpei, Lexson; Stacy-Dawes, Jenna; Ruppert, Kirstie; Glikman, Jenny A.; Roloff, Gary; Montgomery, Robert A. (2021)
      Pastoralists and their livestock have long competed with wildlife over access to grazing on shared rangelands. In the dynamic 21st century however, the configuration and quality of these rangelands is changing rapidly. Climate change processes, human range expansion, and the fragmentation and degradation of rangeland habitat have increased competition between pastoralist livestock and wildlife. Interactions of this type are particularly apparent in East Africa, and perhaps most obvious in northern Kenya. In 2017, following months of intense drought, a pastoralist incursion of a protected area (Loisaba Conservancy) occurred in Laikipia County, Kenya. An estimated 40,000 livestock were herded onto the conservancy by armed pastoralists where the cattle were grazed for approximately three months. Using 53 camera trap sites across the 226 km2 conservancy, we quantified spatial patterns in site visitation rates (via spatially-explicit, temporally-dynamic Bayesian models) for seven species of large mammalian herbivores in the three-month period directly before, during, and after the incursion. We detected significant changes in space use of all large mammalian herbivores during the incursion. Furthermore, these patterns did not return to their pre-incursion state in the three-month period after the pastoralists and their livestock left the conservancy. Thus, in addition to reduced site vitiation rates for these large mammalian herbivores, we also detected considerable displacement in response to the livestock incursion. Our results illustrate that pastoralist incursions can cause large-scale disruptions of wildlife space use, supporting the notion that livestock can competitively exclude large mammalian herbivores from grazing access. We discuss the implications of this research for applied management decisions designed to alleviate competition among wildlife and pastoralist livestock for the benefit of wildlife conservation and pastoralist well-being.