Tapping into earth observation data for rapid disaster detection and response

In Europe, flash floods and extreme rainfall last year caused losses of €46 billion and went down as the costliest natural disaster in Germany, with many homes destroyed and lives lost. This year, fires are raging across the continent. Can technology help in predicting and mitigating disasters such as these?

Some experts certainly think so. Advances in earth observation technology, artificial intelligence (AI), edge connectivity, supercomputing, satellite communications and mobile devices promise to provide a route to early detection and disaster relief. A recent webinar hosted by Orange Silicon Valley and a number of partners looked at developments that offer hope.

Disaster risk monitoring using satellite imagery and AI

To predict and assess natural disasters, it is essential to have accurate information about the environment. Key to this is earth observation, which is the most effective instrument for the collection of geospatial intelligence data at a global level. Better sensor technology is improving yearly, with higher resolutions and advances in artificial intelligence (AI) and analytical insights.

Earth observation (EO) is poised to become a multi-billion industry this decade, with enterprises using it for numerous business cases, including tracking supply chains. EO can also be used to track and predict fast-changing weather events such as those attributed to climate change, which will impact many lives across the globe. Having a seamlessly integrated solution for low-latency AI for geolocation communications around rapid response could be a game changer.

Climate change has spiked demand for EO-based services in areas such as flood detection, drought and deforestation monitoring solutions. Search and rescue missions, for example, use hyperspectral, visible and synthetic-aperture radar (SAR) to create two- and three-dimensional reconstructions. Its advantage is that it can produce high-resolution images in adverse weather conditions, even at night, and in scenarios such as forest fires where smoke usually blocks visibility.

Turning data into insight

How can this data be turned into even more valuable insight to help tackle natural disasters? Here, Orange is working with partner Nvidia to train a supercomputer with a sophisticated model that can detect a flood, for example. To make this intelligence actionable, Orange is collaborating with start-up Balcony to provide direct action detection with AI and notification with geolocation technology. Balcony has created a modular geo-collaboration platform, which will deliver an end-to-end disaster risk monitoring workflow.

In the case of a flood disaster, for example, data is filtered using AI, and a human operator verifies and amplifies it to reduce false positives. Rescue teams can be instructed on exact emergency locations, communicating messages to victims via smartphones. The speed at which data can be scanned is highly significant. Balcony can scan around 63 square km of data in 1.2 seconds.

Communications are crucial to disaster response

Reliable communications are essential to both alert populations and organize an emergency response. Balcony has developed a spatially aware messaging service, which means engaging with people based on location so that it can see where messaging is occurring in a natural disaster, for example, and where action is required.

“Every problem we engage with is a spatial problem. It isn’t good enough to observe this – we must be able to act remotely or in person with high-resolution tools in space or time where issues are changing dynamically,” explained Andrew Schroeder, Vice President of Research and Analysis at humanitarian medical aid organization Direct Relief, at the webinar.

For example, when Kabul fell to the Taliban, many had to leave the city quickly in a hazardous situation. Direct Relief was responsible for hundreds of individuals at risk of violence. Balcony enabled Direct Relief to be fully aware of every individual working in the city and correlate them with alerts harvested from mapping, open-source intelligence and so forth. This creates a live view of people and their situations to help guide them to safety. “The idea that there is high-resolution coordination, not just to see a situation but to act on it regardless of where we are, is fundamental,” explained Schroeder.

Connecting the world

From a connectivity perspective, much of the innovation today is happening at the low earth orbit (LEO) satellite level, according to John Benko, Principal Technology Group, Orange Silicon Valley. “This really is where all the hot stuff is happening today,” he said at the webinar.

LEO satellites orbit at an altitude of 400 km to 1,500 km above the earth’s surface. Much lower than a geostationary satellite, which sits at around 35,000 km. This usually makes LEO satellites much less expensive to launch as they do not require as much power. This will be accelerated by what Elon Musk calls the three Rs – rapid, reliable rockets.

The European Union has noted the capabilities of the LEO satellite and is looking at setting up its own constellation. Dubbed The Secure Communications Initiative, the secure satellite broadband will provide high download speeds and low latency, critical for such applications as telemedicine as well as enhanced security and connectivity for governments and the military, for example.

There are also developments happening with non-terrestrial cell towers, which have the potential to connect remote regions of the world that currently don’t have any connectivity. This involves beaming down from LEO satellite constellations to smartphones instead of a dish. This can be done with reasonable latency and bandwidth, but there are considerable challenges in potential interference; for example, significantly larger antennae are required.

The power is in the data

Geospatial data can help with a range of humanitarian and crisis responses, but the biggest hurdle is still availability and accessibility.

Moving forward to make the most of these emerging technologies, there needs to be a quick way through the complex regulatory, business and technical authorities who hold the keys to the data. “There needs to be a broad commitment to access this data when it is most needed in safe privacy-controlled forms,” explains Schroeder.

The combination of AI supercomputing power at the edge of the network and targeted two-way messaging using smartphone technology at the site of disasters is accelerating the way rescue operations are carried out. Earth observation satellites are pivotal to collecting real-time intelligence for actionable insight. Orange is playing a key role in integrating these advances in rapid disaster detection and response to save lives around the globe.

To learn more about how Orange Silicon Valley enables rapid disaster detection and response using innovative technologies, click here to listen to the webinar.

Jan Howells

Jan has been writing about technology for over 22 years for magazines and web sites, including ComputerActive, IQ magazine and Signum. She has been a business correspondent on ComputerWorld in Sydney and covered the channel for Ziff-Davis in New York.