We’ve just published the first ever Scotland-wide high resolution habitat maps, for free
Space Intelligence, in partnership with NatureScot, has produced a groundbreaking new mapping system for Scotland. The country’s first-ever nationwide high resolution habitat maps also provide insight into how Scotland’s ‘Natural Capital’ is changing over time. Our maps can help landowners and policy makers with their land-use decision making, including working out where to restore habitats. Restoration in ways suggested by our map will capture carbon in the most efficient way and tackle the climate emergency, as well as improve biodiversity and the amenity value of our landscape. Until now, government agencies and land managers have lacked accurate, up-to-date landcover information to be able to do this.
Following the success of this project, NatureScot will use the Space Intelligence maps to improve the accuracy of Scotland’s annual Natural Capital Asset Index assessment, and operate under the Scottish Government’s National Performance Framework. The Space Intelligence system has already been centralised in the AI strategy for Scotland, and is a demonstration of Edinburgh’s rapidly evolving status as the Data Capital of Europe.
The Space Intelligence system is now available online, for anyone to use, for free: www.space-intelligence.com/scotland-landcover
A first for Scotland: High-resolution nationwide mapping
Our team developed the landcover mapping system by partnering with Scotland’s Nature Agency – NatureScot – to create a system to create consistent, repeatable nationwide landcover maps for Scotland for 2019 and 2020. We are thrilled that these maps are now complete.
These are the first complete high-resolution nationwide habitat maps of Scotland. By high resolution, we mean both spatially (each map features nearly a billion pixels, each 0.04 hectares in size), and in terms of classes, with the maps separating every patch of Scotland into one of 22 different classes.
High-accuracy AI system
In developing the maps we worked very closely with NatureScot and their habitat specialists to get a better understanding of Scotland’s habitat species and landcover types, and how they are different depending on the region within Scotland. We collected 584,500 high quality data samples across Scotland for 22 types of landcover. We then used our advanced cloud-based AI platform to analyse tens of thousands of satellite images over Scotland, from a range of different types of sensors, in order to produce the maps.
An independent assessment of accuracy suggested pixels are placed in the correct class over 90% of the time, an excellent result given the number of classes considered. The workflow we have established means that updated versions of the map can be easily made in future, enabling future landcover changes to be mapped every year, or more frequently if needed.
Detail of the map around Loch Rannoch, Scottish Highlands, showing the Black Wood of Rannoch to the south of the loch, forest to the north, and steep hills and moorland all around. This is compared to a single season satellite image, showing that the detailed differences in classes and forest types are hard to see by eye, but picked up by Space Intelligence’s advanced artificial intelligence, which looks at each habitat patch hundreds of times using different satellites throughout the year.
Insights into landcover change
As well as the 2019 and 2020 maps, we have produced a change map showing how the landscape has changed over this 12 months period. This will enable Scotland’s government bodies, charities, and individuals, to track how their landscape is changing.
Planning for Net Zero and biodiversity gain
These maps are crucial for understanding the current state of the environment and how it has changed over time. However, stakeholders are interested in enhancing land use decision-making for the future, and deciding where to restore habitats. Restoration will capture carbon and tackle the climate emergency, as well as improve the amenity value of our landscape. With this in mind we developed a set of ‘heat maps’ in partnership with NatureScot which can help stakeholders to explore the best areas for habitat restoration based on a number of different factors.
This technology is still in development with our partners and so the heat maps should be treated as preliminary. Nevertheless, these layers, which range from carbon storage through to affordability, can be explored along with the maps described above, on our new platform www.space-intelligence.com/scotland-landcover
Two example heat maps showing competing factors that could be considered in prioritising habitat restoration: the opportunity for sequestering carbon, and the accessibility of the site.
Impact on the climate emergency
Our system project helps land owners and policy-makers address the climate emergency, since some types of vegetation and ways of using the land store more carbon and are more beneficial to biodiversity than others. That is, by planning to increase certain landcover types, we can lock away more carbon, and enhance biodiversity. However until this project was funded, government agencies and land managers have lacked accurate, up-to-date information to be able to do this. Moreover, the data from this system will also be used by NatureScot to increase the accuracy of Scotland’s annual Natural Capital Asset Index assessment.
Innovative partnership and funding
This mapping system, called SLAM-MAP, was developed thanks to funding from the Scottish Government through the CAN DO Innovation Challenge Fund managed by Scottish Enterprise, plus direct funding from NatureScot themselves.
We have had fantastic support from our partners on this project, NatureScot, and in particular Philippa Vigano and Tom McKenna in the Innovative Technologies team. We’re also very grateful to Scottish Enterprise – particularly Maria Lorente – and the Unlocking Ambition fund for helping us to grow over the past year.
Ed Mitchard, our Chief Scientific Officer and Co-Founder, said:
Mapping Scotland’s landcover at this high resolution and to 22 different classes was a significant challenge for our team for a number of reasons. Firstly, Scotland’s habitats are diverse and often exist in small patches and along subtle gradients, needing us to integrate multiple types of satellite data in our advanced AI system to perform the classification. Secondly, Scotland is one of the cloudiest countries in the world, meaning we had to go through tens of thousands of images to obtain multiple cloud-free views of each habitat patch, as well as bring in other satellite data types that can see through clouds. And finally, applying our detailed mapping to a whole country involved scaling our cloud-based technology to deal with datasets in the billions rather than tens of millions of pixels. We are delighted to have successfully overcome these challenges, producing highly accurate maps, and look forward to taking our new experience and technological developments to the next challenge.
Murray Collins, our Chief Executive Officer and Co-Founder, said:
We are deploying operational systems that help us address the twin environmental challenges of our age: the climate emergency and biodiversity crisis. It is thrilling to see how our work has been centralised in the AI Strategy for Scotland; and we have just received the exciting news that SLAM-MAP has been selected as the winning story for the ‘Our Planet’ event at DataFest in October this year. We look forward to working with our partners to establish Scotland as a global leader in the monitoring of Natural Capital, through the development of AI applied to big data from space.