The Natural History Museum’s (NHM’s) “living laboratory” garden has collected more than 11 million environmental data records in its first year, feeding readings from more than 50 sensors into the Museum’s Amazon Web Services (AWS)-powered Data Ecosystem.
The sensors in the Nature Discovery Garden continuously capture temperature and humidity to track microclimate variation, underwater acoustic recordings from the pond, birdsong and ambient urban noise such as traffic.
During June’s heatwave, soil sensors in the garden’s woodland areas recorded temperatures 10°C cooler than the paved areas of the Darwin Centre Courtyard, giving researchers evidence for how trees, shade, soil moisture and permeable surfaces can help cities adapt to rising temperatures.
The readings feed a Data Ecosystem built on AWS cloud technologies, which collects, processes and shares millions of observations from sensors, imagery and other scientific sources. In practice, that means an internet of things (IoT) data pipeline in which devices publish telemetry to a message broker such as AWS IoT Core, streamed through a service such as Kinesis into an S3-based data lake, then processed, catalogued and analysed with tools such as Athena and SageMaker – including machine learning models that classify birdsong and other acoustic recordings.
Ed Baker (pictured), acoustic biology researcher at the Natural History Museum, said: “We call the Museum gardens a living laboratory, and the sensor network and AWS Data Ecosystem have made this a reality. By collecting millions of data points, from temperature and humidity to birdsong and underwater sound, we can build a far richer picture of how urban nature responds to various pressures and becomes more resilient to future environmental challenges.”
The museum is now expanding the network into its Evolution Garden, an area with higher visitor footfall and a contrasting planting style, to study how visitor activity and habitat design influence biodiversity.
Long-term monitoring since 1995 has recorded more than 3,500 species in the gardens, with a further 90 identified since they reopened in 2024, including the emperor moth and the small red-eyed damselfly.
The project is part of a wider pattern of deployments of environmental sensor networks allied to cloud data platforms. In Chicago, the Array of Things deployed hundreds of urban sensing nodes publishing open data through a cloud portal; in Germany, the Ammod automated multisensor stations and the Nature 4.0 project combine networked sensors with machine learning for biodiversity monitoring. The NHM reports that its Data Ecosystem has scaled to accommodate a 200% increase in use since the network was switched on.
Hilary Tam, sustainability leader for EMEA at AWS, said: “The Natural History Museum is showing what’s possible when science and cloud technology come together. As the sensor network grows, the AWS-powered Data Ecosystem is able to scale to enable researchers to connect millions of environmental observations, revealing patterns that would otherwise remain invisible.”
For enterprise IT leaders, the project is a reminder that the sensor-to-data-lake-to-machine-learning architecture is now cheap and standard enough to run in a museum garden, and so increasingly available for smart-city and infrastructure monitoring. But the hard part is not collection – it’s curation.
As Computer Weekly has reported, poor-quality data can derail environmental analysis, and the same rigour around sustainable IT and its environmental footprint applies to any organisation turning sensor telemetry into decisions.
