Scientists have created a ground-breaking new technique for modeling and mapping the natural environment in great detail using AI’s cutting-edge capabilities.
A group of experts, led by Charlie Kirkwood of the University of Exeter, has developed a complex new method for modeling the Earth’s natural characteristics with greater accuracy and detail.
The new methodology is capable of recognizing complex features and components of the terrain much beyond the capabilities of more traditional methods, and it can use this information to create higher-quality environmental maps.
Importantly, the new method may pave the way for fresh discoveries of natural-environment linkages, which could aid in addressing some of the century’s most pressing climatic and environmental concerns. The research was published as part of a special issue on geostatistics and machine learning in the prestigious journal Mathematical Geosciences.
The process of modeling and mapping the environment is lengthy, time-consuming, and costly. Cost limits the amount of observations that can be gathered, hence filling in the gaps between these observations is essential for constructing comprehensive spatially continuous maps.
Scientists can employ a variety of data sources, such as terrain elevation data and satellite photography, to help fill in these observation gaps.
This is a fantastic example of Environmental Intelligence, the use of AI to help solve challenges in environmental science. This work is an exemplar in integrating technical knowledge of AI and machine learning with expertise in geosciences to produce new methodology that directly addresses crucial questions in mapping environmental information. The resulting methodological advances could be used to produce detailed maps of a wide variety of environmental hazards and have the potential to provide a rich source of information for both scientists and decision makers.
Professor Gavin Shaddick
Traditional modeling methods, on the other hand, rely on users manually engineering predictive features from these datasets, such as producing slope angles and curvatures from terrain elevation data in the hopes of explaining the spatial distribution of the variable being mapped.
Scientists believe, however, that there are likely to be many more intricate linkages at play in the actual environment that models based on standard manual feature-engineering procedures will simply overlook.
The study’s ground-breaking new AI technique treats environmental data extraction as an optimization issue. As a result, it can recognize and leverage relationships that might otherwise go overlooked and unutilized by humans using more traditional modeling methods.
In addition to enhancing map quality, this also opens the door for AI to uncover new relationships in the natural world, while also reducing the amount of trial-and-error experimentation in the modelling process.
Charlie Kirkwood, a postgraduate student at the University of Exeter said: “To be useful for decision making, we need our models to provide answers that are as specific as possible while also being trustworthy and that means creating accurate measures of the uncertainty associated with our estimates, which in this case are predictions at unmeasured locations.”
“Our AI approach is set within a Bayesian statistical framework which allows us to quantify these uncertainties and provide a range of uncertainty measures, including credible intervals, exceedance probabilities, and other more bespoke products that will feed directly into decision-making processes. Crucially, all this is provided whilst harnessing any available information more effectively than traditional approaches allow which you can see coming through in the detail of the map.”
Observations of stream sediment calcium concentrations from the British Geological Survey’s Geochemical Baseline Survey of the Environment (G-BASE) project were used to show the novel method.
The distribution of calcium in the environment, which is important in and of itself because it affects soil fertility, is influenced largely by geology, with different rock types containing differing quantities of calcium, but also by surface hydrological processes.
Calcium is thus a difficult use case for AI, as it must learn to recognize and use features related to both bedrock geology (e.g., different terrain textures, slope fractures) and surface hydrology (e.g. drainage, river channels).
The technology, according to the researchers, has produced a remarkably comprehensive and precise map that, despite displaying only one element calcium, depicts the geology of Britain in arguably unprecedented detail thanks to the new AI approach’s information-extracting power.
The team believes that this work represents a new dawn for environmental mapping approaches in the age of AI by merging the research talents, knowledge, and data resources of its partners, the University of Exeter, Met Office, and British Geological Survey.
Professor Gavin Shaddick, from the University of Exeter added, “This is a fantastic example of Environmental Intelligence, the use of AI to help solve challenges in environmental science. This work is an exemplar in integrating technical knowledge of AI and machine learning with expertise in geosciences to produce new methodology that directly addresses crucial questions in mapping environmental information.”
“The resulting methodological advances could be used to produce detailed maps of a wide variety of environmental hazards and have the potential to provide a rich source of information for both scientists and decision makers.”
Garry Baker, Interim Chief Digital Officer, British Geological Survey added: “This paper is an excellent demonstration of how environmental information such as the BGS geochemical database can be re-assessed via new approaches (AI spatial interpolation). It exemplifies the benefits of ongoing environmental research and how this can draw upon the extensive datasets available to everyone through the National Geoscience Data Centre and wider NERC, and UKRI data repositories.”
Dr. Kirstine Dale, Principal Fellow for Data Science at the Met Office and Co-Director of the Joint Centre for Excellence in Environmental Intelligence, spoke about the importance of this work:
“This is an important example of how data science has the potential to transform our understanding of the natural world. Critically, it highlights what can be achieved by working across disciplines, in this case bringing together mathematicians, weather specialists and computer scientists enriches our knowledge of the natural world in a way that no single discipline can.”
