Businesses know they need risk management departments but the existing methodologies of identifying potential problems down the road have a massive blind spot, according to Cervest founder and CEO Iggy Bassi. Climate change could cost the US $224 billion more per year, according to an April 2019 report in journal Nature Climate Change.
The UK startup just announced a £3.7 million ($4.77 million) seed round and the launch of its real-time climate forecasting platform aimed at helping decision-makers manage risk and inform policy regarding climate change. The round was led by deep tech investor Future Positive Capital with co-investor Astanor Ventures, an impact investor focused on agrifood.
(Read Jessica Pothering’s interview with Future Positive’s Sofia Hmich here. Astanor – we’re waiting for your call!)
The company leverages AI to analyze billions of data points and forecast how changes in the climate will impact the future of entire countries down to individual landscapes, to allow businesses, governments, and land managers to adapt to climatic and extreme events.
“It’s a very focused area, using AI for climate. I think it falls outside of very traditional agtech. It was difficult to get mainstream investors but the round went relatively fast,” Bassi told AFN. “There are very few impact-driven funds that focus on climate in the institutional space. There are two in Europe and I got both to come in.
When speaking with customers– who will mainly be risk practitioners in the financial, insurance, government and CPG sectors — Bassi typically begins by asking them to point him towards their climate risk departments. Of course, most of them don’t have one, he says. But as some of the Fortune 500 companies he’s met with are identifying new patterns in their sourcing practices, he’s seeing an uptick in interest for climate services.
“When talking to a small farmer, a large government, or a large enterprise, they keep saying ‘we don’t know how to take these branches of science and put them together in a mathematically coherent way that’s operationally useful,” he says. “Climate science has not been turned into an operationally useful science.”
Cervest’s Earth Science AI platform was developed based on three years of research and development by scientists, mathematicians, developers, and engineers. It’s capable of analyzing billions of data points from a variety of sources including atmospheric science, meteorology, hydrology, and agronomy. It combines AI, imaging, machine learning, and Bayesian statistics to identify signals, or early warning signs, of extreme weather events like floods or strong winds. It can also identify fluctuations in soil health and water scarcity, according to the company.
Cervest will use the new capital to make new hires, secure new customers, and continue generating proprietary data assets.
“What we try to do is map out as much world geography as possible. If you are a large food company and you want to set up a factory in Poland, the system is pre-mapped with information about what happened to that land over the last four-to-five decades,” he explains. “You may be about to commit $30 million to $40 million to a factory in an area that has already experienced a significant change. You can go to your dashboard and you can monitor for extreme events, yield mapping, and a longterm water forecast for that region.”
Forecasts are updated in real-time through a personalized and filtered dashboard for users. The company plans to release a beta version in Q1 of next year. The company is initially focusing on the European region and is aiming to have all of Europe mapped before its beta launch. It will then continue mapping the rest of the world. It also plans to launch its yield mapping capabilities in Q2.
Cervest eventually plans to charge users on a per-signal basis for private business, with signals defined as extreme events like fire, severe wind, flood, and drought. It’s developing a different route to market for government entities. As a mission-driven business, Bassi is hoping to make it affordable.
How does Bassi know that its predictions are accurate? Sometimes the system isn’t entirely sure.
“No one can tell you 100% certainty in seven months’ time what your yield will be. We don’t talk about predictions. We talk about streaming signals every day. We have to quantify uncertainty across multiple variables. We look at accuracy all the time but different decision-makers have different needs at different times so we codify the uncertainty and we will tell them what we think based on all the known data,” he explains. “When we do our current forecasting, we go back in time and say what would we have predicted during the last five years and contrast that with what actually happened. We are constantly testing the integrity of the system.”
To date, Cervest has worked with a few big players including field trials with Syngenta throughout four countries in wheat and maize production, cookie manufacturer McVitie’s and European agricultural cooperative Lantmännen.
The climate services industry is an estimated $2.6 billion business worldwide and growing at a rate of 10% per year. It includes a number of emerging players like NOAA’s Climate Prediction Center and Australia’s Managing Climate Variability Climate Champion Program for farmers. Another climate services startup, Jupiter, has 10 current clients, according to Wired, including one of the top five US power companies and the cities of New York and Miami.
“We don’t view Cervest as being competitive to current agtech offerings but rather we would be complementary because we could feed climate signals to existing apps,” Bassi explains. “We don’t really see Climate Corp as directly competitive either given their focus on agri insurance markets. However, there are a few notable players who have emerged over the past two to three years principally in the US market. The technical capabilities and scope of services varies enormously between these players.”