SANParks staff get trained by iNat guru
iNaturalist is an invaluable resource to South African National Parks. It provides a treasure trove of visual and digital data that is available at our fingertips. At a time of budget cuts to conservation organisations leading to lower budgets available for monitoring and data collection and collation, iNaturalist is something that we should be making more use of. Other organisations already use iNaturalist as part of their operations, as the data is automatically stored, organised and accessible. E.g., the Custodians of Rare and Endangered Wildflowers programme uses iNat to conduct monitoring.
On Thursday the 28th of September, Tony Rebelo provided training to SANParks staff, from Scientific Services and Park management, on how to get the most out of iNaturalist. We learned about the many functionalities of the iNaturalist platform, including map and identification functionality. Some staff had experience with using the Android/iOS iNaturalist App, but had never explored the iNaturalist website and were not aware of the wealth of data that can be found using the filtering tool, that it is easy to create checklists for national parks on iNaturalist, that field guides can easily be accessed and created and that there are projects for IUCN Red Listed species and NEMBA listed alien species that you can join and get data from.
Tony Rebelo has made the most observations (>238,000) in Southern Africa of any user of iNaturalist, and he has made almost 300,000 identifications, putting him in the top three identifiers in Southern Africa, where there are more than 3.8 million observations.
Tony Rebelo demonstrating how to use the iNat Android app to SANParks staff at the Cape Research Centre and SANParks staff online in Kimberley, Knysna and Kruger, Augrabies Falls and Garden Route National Parks.
iNaturalist uses advanced artificial intelligence. Tony explained how good the image recognition is on iNat, stating that “for plants, artificial intelligence is better than me by far, and I think I’m pretty good”. He explained that species can only be identified automatically if iNat has more than 100 images to use for training. The “suggest an identification” function is really useful, and it can be used once you are back at your computer as well as on the app. When asked what species we thought would have the most identifications, no one got the correct answer. Some were close with choosing elephant (number four). Interestingly, the western leopard toad is in pole position. This is because there is a research project, called WLT monitoring, which is focused on individual identification, using individual markings, and monitoring of the western leopard toad population.
Although iNaturalist provides endless opportunities to benefit conservation and research initiatives, it was emphasized that it is important to check identifications in the data that you are going to use and not consider all identifications as being reliable. Tony mentioned that research grade observations only require two people to agree with an observation, which is a really low bar, as those two people could be anyone. One of the ways of checking observations that you are unsure of is by using the iNaturalist platform to ask one of the top identifiers for a species or group whether the identification is correct. You can also have more confidence in identifications made by particular experts on the platform, as information is provided on who made the observation and who identified the species or agreed with the identification.
In parks where visitors can walk around there are large volumes of data available on iNaturalist. However, in parks where visitors can’t walk around, it would be very beneficial if park staff could add a few gems to the treasure trove of images on iNaturalist and possibly add to the lists of species recorded for those parks.
iNaturalist training at the Cape Research Centre.
Participants of iNaturalist training at the Cape Research Centre
