Since the 1970’s scientists have tapped into the power of ‘citizen scientists’ to gather data on an incredible range of subjects and in the age of ubiquitous technologies such as smart phones, apps and the internet, the potential for ‘citizens’ to both participate in and provide accurate and reliable data for use by scientists has never been greater. ‘Citizens’ can volunteer their time and join scientific expeditions, use apps on their smartphone to record data or help to train artificial intelligence algorithms used to identify plastics on a beach or cancers cells in samples - pooling their collective skills, time and effort to create a powerful new tool for use by scientists across the globe.
Citizen science - noun. Scientific work undertaken by members of the general public, often in collaboration with or under the direction of professional scientists and scientific institutions.
Recent studies on the reliability of ‘citizen science’ have shown that datasets produced by volunteer citizen scientists can match those produced by professionals. And with the development of cloud apps and data management software such as WildBook - developed by the non-profit Wild Me and research partners at the University of Illinois-Chicago, Rensselaer Polytechnic Institute, and Princeton University - scientists now have the means to obtain, curate, and analyze “Big Data”, such as that generated by citizen scientists through tourism and volunteerism.
Wildbook blends structured wildlife research with artificial intelligence, citizen science, and computer vision to speed population analysis and develop new insights to help fight extinction. In recent years it has been to identify giraffes and flapper skates, polar bears and giant sea bass, but for divers, some of the most exciting examples have been projects that aim to build up visual databases of whale sharks and manta rays.
Manta rays and algorithms
MantaMatcher is the first global online manta ray database that uses the unique pattern of spots on the belly of a manta ray to identify individuals and manage manta ray sighting and identifications worldwide. Any diver or snorkeler with a camera can take and upload a manta identification photo and researchers can also upload and organize individually identified manta rays in the regional populations they are managing. MantaMatcher promotes collaborations by cross-referencing different study sites and scientists can now examine if populations are shared between neighbouring countries, regional and long distance movements, and the lifespans of manta rays.
A sophisticated algorithm is used to match manta identification photos automatically. The algorithm works like facial recognition software for manta bellies and by automating much of the identification process, researchers can match mantas regionally, and even globally, with ease. Researchers will ultimately be able use sightings data to determine the abundance, trends, movements, and population structure of manta ray populations at different sites across the globe.
Uploads to MantaMatcher have already contributed to the better scientific understanding of these species. A paper on manta migrations in Indonesia, using data sourced from MantaMatcher, was instrumental in achieving protection for mantas across all of Indonesia's waters. Data from MantaMatcher were also used as scientific justification for Convention on Migratory Species listing of reef mantas in November 2014.
Similarly the Manta Trust’s IDtheManta programme was created to utilise the power of citizen science to aid in the research and conservation of the world's manta rays. The Manta Trust receives over 5,000 photo-ID submissions each year and, since its global roll-out in 2012, the IDtheManta Database has become the largest of its kind. Over 10,000 individual reef and oceanic mantas have been identified, through more than 100,000 photographed sightings from over 70 countries. Right from the start, these ID submissions have been manually processed and matched to mantas in the Manta Trust database by members of the Manta Team. But to grow the scale and impact of the initiative further, The Manta Trust have partnered with the University of Bristol, and the not-for-profit company, IDtheAnimal Ltd, to create a fully-automated visual biometric photo-ID software for manta rays, which will interface with the existing IDtheManta global database. The hope is to create a novel platform that streamlines the submission of ID photos, and makes the databases accessible to manta scientists and the general public around the world.
Spot my spots - whale sharks and Wildbook
Wildbook for whale sharks has to date identified over 8,936 whale sharks, with 44,348 reported sightings from 5,500 citizen scientists and 147 researchers and volunteers.
The Wildbook for Whale Sharks photo-identification library is a visual database of whale shark encounters and of individually catalogued whale sharks. The library is maintained and used by marine biologists to collect and analyze whale shark sighting data to learn more about these amazing creatures. The Wildbook uses photographs of the skin patterning behind the gills of each shark, along with any scars, to distinguish between individual animals. Cutting-edge software supports rapid identification using pattern recognition and photo management tools. Divers can assist with whale shark research by submitting photos and sighting data and the information submited will be used in mark-recapture studies to help with the global conservation of this threatened species.
Read more about Wildbook for whale sharks and how you can contribute you photos.