satellite new

The data and intelligence satellites provide are being leveraged on a wider scale within financial services in areas such as financial inclusion and climate risk. Anita Hawser reports.

When most of us think of space, we usually think of astronauts, or exploration of the moon or Mars. But with thousands of satellites now orbiting the earth, these ‘eyes in the sky’ provide vital information, such as weather predictions, climate and environmental monitoring, communications, positioning, navigation and timing, and earth imaging. Several industries – scientists and researchers, farmers, mining and construction, banking and insurance – rely on these for a growing range of needs. There are communication satellites, weather satellites, Global Positioning System (GPS) satellites and earth observation satellites.

But the real revolution with satellites has come in the past five years with smaller satellites, says Beau Legeer, director of imagery and remote sensing at Esri, a provider of geographic information system (GIS) software, location intelligence and mapping. “As there are hundreds of these smaller satellites, we can map the Earth every day at higher frequency, which bigger satellites can’t do,” he explains. Electro-optical and radar satellites can even measure things like methane.

One of the unique advantages of satellite data is its trustworthiness. In Ukraine, for example, satellite imagery proved vital in helping experts refute Russian claims that bodies of civilians shown lying on the streets in Bucha, a city in Ukraine’s Kyiv Oblast region, were not killed during Russian occupation of the city. “As the saying goes, ‘A picture paints a thousand words’,” says Mr Legeer. “Right now, satellite imagery has a high level of trust, which will only grow with the revolution in smaller satellites. You can’t interfere with the picture.”

While satellites have been around for more than 50 years, it is only now that the data and intelligence they provide are being leveraged on a wider scale within financial services. Some of the earliest adopters of satellite data in financial services were hedge funds, which used it to count cars in Walmart car parks so that they could get ahead of corporate earnings news.

Insurance companies, too, are ahead of the game, says Alex Martonik, industry practice lead, business resilience, financial services and insurance at Esri, and are using geospatial data to predict weather patterns or the severity of wildfires. “They are also using it for change detection and damage assessment hours after a natural disaster strikes, so they can more quickly process insurance claims,” he adds.

Emerging use cases

Satellite data from free and open sources, such as the EU’s Earth observation programme Copernicus’s Sentinel satellites, updates up to every five days, while ‘commercial costed’ satellite data updates 12 times a day, says David Taverner, senior director at Caribou Space, which uses space tech for sustainable development. “This speed enables stakeholders to react quickly with the latest information available to them to validate a newly issued insurance claim following a disaster.”

But with the reduction in the cost of launching and building satellites, other use cases for satellite data in financial services are now starting to emerge. One such use case, says Mr Taverner, is financial inclusion in developing countries. The challenge in developing countries is the lack of data to help inform decision-making, he explains.

He continues: “No credit profiles exist, particularly in rural areas, so alternative methods are being sought to assess creditworthiness using satellite information. By accessing quality satellite data on potential customers and risks, the requirement for finance providers to do on-the-ground assessments and portfolio monitoring can be reduced, resulting in cost savings for finance providers.”

Another advantage, says Mr Taverner, is that many use cases in finance can be addressed using Earth observation and Global Navigation Satellite System (GNSS) data that is available for free from satellite constellations such as Copernicus and US GPS. “But satellite data is not a silver bullet on its own,” he says. “It is only effective at improving financial inclusion when combined with other more traditional technologies such as mobile.”

Satellite data also needs to be integrated into software that converts it into insights and recommendations, which he says requires teams with software engineering, machine learning and data science skills.

Space-derived data is even more illuminating, when it’s combined with other data – such as psychometric, demographics, financial and agronomic – to generate credit scores for customers without traceable financial histories. “Take the example of a farmer,” says Mr Taverner. “The farm boundaries can be captured using GNSS and, based on that data, historical farm performance information can be reviewed using a diverse set of space data, such as soil moisture, water extents and heights.”

This information can then be used, alongside historical track record of yields, to assess credit or investment risks of allocating capital to the agricultural sector. In May 2021, ICICI Bank become one of the first banks in India to use satellite data to measure land, irrigation and crop patterns, which, when combined with demographic and financial data, allows the bank to make faster lending decisions for farmers.

Greater transparency

Another big advantage is that satellite constellations cover the entire globe, enabling vast and remote areas and conflict regions to be monitored more cost effectively than using ground teams, aeroplanes or drones. This has allowed companies like London-based OilX to “count the world’s oil, one barrel at a time”, bringing greater transparency to an otherwise opaque industry. “In the past, I used to call someone at a port via phone to look out the window and tell us how many tankers are in the port today,” says Florian Thaler, CEO, and co-founder of OilX. “But using satellite derived data means we don’t need to be in all these places now.”

OilX uses data from Copernicus Sentinel all-weather, day-and-night radar satellites, developed under the European Space Agency Business Applications programme with Aresys, a spin-off from the Politecnico di Milan. Artificial intelligence (AI) algorithms are then applied to transform low resolution satellite data into high resolution, allowing OilX to detect oil storage tanks anywhere in the world, even in countries such as China, Iran and Saudi Arabia, where reliable information on oil stocks is hard to come by.

“We can even detect tanks through clouds,” says Mr Thaler. “And radar data can show us the movement of the tank’s lid up and down, which we can use to create a cylinder volume calculation that is very accurate.” OilX then packages that information and sells it to hedge funds, investment banks, oil majors and commodity traders to help them better gauge movements in oil prices.

“By combining Earth observation and AI, you can do things that were unthinkable in the past,” says Mr Thaler. “But we keep a razor-sharp focus on a single sector. The difficult part is to bring the power of satellite data and find applications to where it is truly useful. And for that, you need satellite, AI and domain expertise.”

ESG application

For the past five years, the World Wildlife Fund (WWF) has used geospatial data and satellite imagery to better understand and observe what is happening to conservation sites, such as protected areas, trying to define not just where the change is, but who is behind it. “Ideally it would be great to generate ground data from everywhere, but that is not practical,” says David Patterson, head of conservation intelligence at WWF. “The only way we can collect data at a global scale, regularly, is by using satellites.”

The only way we can collect data at a global scale, regularly, is by using satellites

David Patterson

Mr Patterson sees satellite data as just an additional dataset in the environmental, social and governance (ESG) workflow. He says: “What’s changed for us in the last five years is there’s been a huge shift in financial companies’ interest in geospatial approaches. Before it was quite niche – maybe we can find out some specific insight for a specific site. But now the approach is being mainstreamed because the financial community is trying to understand biodiversity.”

Investors are turning to geospatial and satellite data to get access to high frequency impact data on climate and nature risks, data they only used to be able to get annually in corporate disclosure reports, says Marianne Haahr, executive director at Green Digital Finance Alliance. “This means asset managers and asset owners can engage or exclude parts of their portfolio based on observed – rather than reported – company behaviour,” she explains. “It can also enable automated proof of impact reporting on green bonds or other sustainability-linked instruments.”

For the past two years, Swiss private bank Lombard Odier has used remote sensing and satellite data to assess nature-related risks and opportunities and make judgements about the physical impact of climate change on companies it invests in and their supply chains. For the bank, it is a way of not only identifying companies in its investment portfolios that are most at risk from climate change, but also identifying those that are more likely to come out on top.

“Spatial and remote sensing data allows us to have an unbiased view,” explains Laura García Vélez, a quantitative analyst at Lombard Odier. “We can look at the physical impacts of climate change, which are location specific. How susceptible is a company to wildfire risks? That depends on their location. When there’s an event like a hurricane, we can issue a real-time alert to our investment teams telling them which companies are in its path.”

We need to reconcile what the remote sensing and geospatial information tells us with what companies are self-reporting

Laura García Vélez

It is also a way for the private bank to check whether the information reported by companies on their environmental impacts matches what the satellite imagery is showing, which can be challenging, says Ms García Vélez. “There are often gaps in geospatial databases. Some may have good disclosure of company facilities in North America and Europe, for example, but data sometimes lacks when it comes to other geographies. We need to reconcile what the remote sensing and geospatial information tells us with what companies are self-reporting.”

If the data captured by remote sensing or satellites does not match what the company has reported, the bank is then able to engage in a more constructive dialogue and encourage them to do better in terms of environmental stewardship. “We are starting to have these conversations with companies around forest management issues,” she says.

Location-specific data

Currently, Earth observation data is mainly applied for licensed industries, such as mining, oil and gas, and physical assets. “Investors have access to asset geolocation to overlay a location with coordinates with satellite data to assess, for example, deforestation impacts of a portfolio company,” Ms Haahr explains. But for non-licensed industries, such as agriculture, asset geolocation is more difficult to access, she says, especially for the first mile of a value chain; in that case, investors cannot make use of satellite data.

For deforestation risk monitoring, Ms Haahr says the frequency and granularity provided by open-source data sets from Sentinel-1, Sentinel-2 and Landsat satellites, which enable almost-daily updates, is sufficient. But this is not always the case for other types of risks, such as monitoring facility-level methane and carbon dioxide emissions. For that, Ms Haahr says a few ESG data providers leverage GOSAT (an oceanographic satellite designed to measure sea surface heights), which detects atmospheric methane by its absorption of radiation in the shortwave infrared.

When it comes to ESG, Mr Patterson says companies should report where their operations are, so it is possible to compare what they are reporting with what is happening. “We need transparency and accountability,” he says. But satellite data is not perfect for everything, he adds. “With biodiversity, we’re not so much interested in the number of species in a given area; what is more important is understanding the ecosystem’s health and how that’s impacted by commercial operations over time. That’s very difficult to track and requires a combination of ground and Earth observation data and often results in proxy indicators.”

But Alex Money, founder and CEO of Oxford Earth Observation – which uses remote sensing, machine learning and climate science to identify and measure the sustainability risk exposure of real assets – believes satellite data could be transformational, both in the operational sense, for example assessing how companies that rely on water – such as mining, food and beverage producers – to operate will be impacted by water stress, and on the regulatory side. “Disclosing, explaining and benchmarking – this is where location-specific data becomes a bigger part of the regulation story, which will make this data much more important,” he says.

Spatial finance

The Spatial Finance Initiative, established by the Alan Turing Institute, Satellite Applications Catapult and the Oxford Sustainable Finance Programme, has even coined the term ‘spatial finance’ to describe the emerging phenomenon of “integrating geospatial data and analysis into financial theory and practice”. In its 2021 Spatial Finance Report, it writes: “Spatial finance provides an opportunity to enhance transparency within the financial system.”

But it also points out that many spatial finance applications today remain largely underexploited, through lack of awareness, analytical skills and asset-level data tying physical assets to ownership structure. This makes it difficult to leverage satellite data for risk assessment. “Access to satellite data is not the challenge, but designing and managing the AI to analyse the data and translate it into investor relevant [insights] is a challenge,” says Ms Haahr. “Many asset managers do not have in-house capabilities in their ESG departments to leverage raw satellite data.”

Mr Money says having this data be free and accessible will be important for innovation so companies can create a commercially viable model on top of it. “The input data is basically free, but then what do you say about it? How confident can you be about it? How validated are the predictions you make? That is where the focus is coming,” he says.

PLEASE ENTER YOUR DETAILS TO WATCH THIS VIDEO

All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker



For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Join our community

The Banker on Twitter