'Using innovation to map habitats and land use to help us reach net zero' by Philippa Vigano, NatureScot
Land use is an important consideration in our efforts to reach net zero by 2045. How we manage our land has a huge impact on emissions because different habitats and land uses store carbon in vegetation and soils and some habitats, like peatlands, if they are looked after can even absorb carbon. A good and complete map of habitats that allows us to map extent and distribution and track change over-time is a vital tool in our response to the climate emergency.
Our users told us that they needed a complete and repeatable habitat map of Scotland that could be used to show change over time. NatureScot needed this data to feed into the Natural Capital Asset Index. Natural capital is a term for the habitats and ecosystems that provide social, environmental and economic benefits to humans Scotland has a wide range of these habitats and ecosystems - each of which makes a unique contribution to the wellbeing of those who live and work in Scotland. The Natural Capital Asset Index (NCAI) is a composite index which tracks changes in the capacity of Scotland's terrestrial ecosystems to provide benefits to people. There is more information about natural capital and the NCAI on the NatureScot website
To understand our natural capital and how it is changing we needed a comprehensive map that could be updated at regular intervals. We soon realised that we had to develop a new innovative approach to habitat mapping to deliver this data. It was not going to be possible to do this using tradition field survey, “boots on the ground” techniques. We also needed an innovation partner who had the skills needed to process big data from space (satellite data) and classify it using AI.
We collaborated with Space Intelligence an Edinburgh based satellite data company on a response to an AI for Good Challenge set by the Can Do Innovation fund “How can you use AI for good in the Climate Emergency?”
Space Intelligence used data from the Habitat Map of Scotland and worked with NatureScot to develop training data for the Artificial Intelligence, check the map outputs and make sure that the data could be used to feed into the NCAI.
The map is a 20m raster of Scotland and is classified to EUNIS Level 2.
We are developing ways to test the accuracy of the map against ground measurements and collect training data for future mapping.
The data is available to view on https://www.space-intelligence.com/scotland-landcover/
We would like to consider:
- Does anyone in The Data Lab Community you have other uses for this data?
- Does anyone have experience with developing ground measurement data and training data resources for EO/AI maps?
Philippa Vigano is Innovative Technologies Programme Manager at NatureScot, Scotland's nature agency.