Dr Laurence Jones of CEH was one of the co-authors of a recent article that discusses how satellite data could help map biodiversity, such as assessing endangered monkey populations. He explains more about the work below:
Our new paper describes a multi-disciplinary approach using an array of emerging technologies to map biodiversity at the landscape scale. The findings are published in Nature Ecology and Evolution by an international group of researchers, including the Centre for Ecology & Hydrology. The work was led by the Kunming Institute of Zoology from China, University of East Anglia and the University of Leicester.
The study looked at ways in which an array of technologies could be used to identify how many species are alive in an area and the risks they may be exposed to. Using a combination of satellite and ground data, the team proposes that it is now feasible to map biodiversity with an accuracy that has not been previously possible.
For example, we can map multiple indicators of monkey distribution, including human activity zones as inferred from roads and settlements, direct detections from mosquito-derived iDNA, animal sound recordings, camera traps, plus detections of other species that are usually found when monkeys are present, such as other large vertebrates. This data could be used to identify areas in which monkey populations are particularly vulnerable.
“There are ten times as many satellites in operation now as there were in the 1970s. The European Copernicus satellites provide free global data every 5 days down to a resolution of 10 m. Satellite technology has undergone a massive change and has never been so accessible," co-author Heiko Balzter from the National Centre for Earth Observation at the University of Leicester explained.
"However, satellites cannot observe small animals directly. Most biodiversity is invisible to a satellite. We propose using a mix of new technology rather than a single remedy."
Laurence tells us more about how the approach could be used to map vegetation:
"As well as monitoring and modelling the distribution of animals, there is increasing potential to map vegetation types using the new satellite technology. LiDAR data can be used to map vegetation height and ground surface elevations, for example in wetlands where distance to the water table is an important control on the plant communities present. Spectral and hyper-spectral imagery can tell us a great deal about the dominant plants growing in a location, but the real opportunities lie in the continuous data updates. Repeat data over weeks and months allows us to see plants changing almost in real time, with plants coming into flower or deciduous species coming into leaf, allowing more sophisticated mapping of vegetation communities.
The technology doesn’t replace surveyors on the ground, but makes certain tasks like mapping vegetation condition over large areas much more feasible. Statistical modelling is the necessary bridge which helps integrate these new technologies and make them more than just a sum of their parts."
The paper, titled ‘Connecting Earth Observation to High-Throughput Biodiversity Data’ is published in the journal Nature Ecology and Evolution and is available online. DOI: 10.1038/s41559-017-0176.
Dr Laurence Jones
The University of Leicester issued a press release about the paper
Image above: An example of the use of Earth Observation for monitoring ecosystem services. A false colour composite of predicted abundance of Graminoids, shrubs and bryophytes representing vegetation composition on a peatland from Partial Least Squares Regression models on a hyperspectral image (University of Leicester / Dr Beth Cole).