AI Unleashes an Explorative Lens on Wildlife Photography to Uncover Climate Change Clues

An innovative imaging device may be utilized to train artificial intelligence systems to identify indicators of how the Earth is reacting to escalating extremes. 

The system has been developed by a consortium of global scientists, comprising researchers from the University of Edinburgh, University College London [UCL], UMass Amherst, iNaturalist, and the Massachusetts Institute of Technology [MIT]. The aim is that INQUIRE, a new imaging instrument, may assist in the creation of wildlife analysis algorithms and enhance the efficiency of detailed evaluations based on content. 

Targeting content from citizen science and photo-sharing platforms, the tool has the potential to scan publicly accessible images and perform millions of analyses, extracting visual records for indicators of climate change documented in real-time. Such data might encompass alterations in dietary habits of specific species, overall population health, and emergent interactions and relationships among various animal and plant species. 

‘The multitude of wildlife images uploaded online daily provides researchers with critical insights into the distribution of diverse species on our planet. However, merely identifying the species within a photo merely scratches the surface,’ stated Dr. Oisin Mac Aodha, Reader in Machine Learning. ‘These visuals potentially represent a vastly rich source of information that remains predominantly untapped. Ability to swiftly and precisely sift through this wealth of data could provide crucial insights into how species adapt to multifarious challenges like climate change.’

‘The diligent curation of data, focusing on capturing genuine examples of scientific inquiries across various fields in ecology and environmental science, has proven essential in broadening our understanding of the current capacities of existing AI methodologies in these potentially significant research areas,’ added Dr. Sarah Beery, Assistant Professor at MIT. ‘It has also highlighted deficiencies in present research that we can now seek to resolve, especially concerning intricate compositional queries, specialized jargon, and the nuanced distinctions that differentiate categories of relevance for our collaborators.’

Further climate change and net zero topics: 

Contracts finalized on Teesside East Coast Cluster carbon capture and storage

Building decarbonisation efforts will fail without comprehensive carbon accounting

Europe’s most state-of-the-art weather satellite becomes operational

Image credits: Geranimo via Unsplash