Startup Harnesses IoT, Cloud Computing to Improve Weather Forecasts

For years, predicting the weather has been a vital but imprecise science – witness how the color of the evening sky or the height of birds in flight still crops up in our conversation.

And while recent high-tech advances in forecasting have certainly benefited the United States and other industrialized countries, weather predictions have often come up short in the developing world, where populations remain more exposed and farming and infrastructure are less resilient.

Earlier in March, Cyclone Idai rammed into southern Africa, then reversed course, wandered out to sea and slammed back into the continent. Its 110 m.p.h. winds and heavy rains devastated huge swaths of Mozambique, Malawi and Zimbabwe. The United Nations  described the  weather disasterOpens a new window as one of the worst to affect the southern hemisphere.

Until recently, responsibility for weather forecasts belonged to governments, and their accuracy has relied on models fed by huge amounts of data. In the cloud computing age, though, when just about anyone can purchase cloud computing power, this is changing. New opportunities are opening up.

ClimaCell, a Boston- and Tel Aviv-based “microweather technology” startup, is one company developing a business to harness the Internet of ThingsOpens a new window for weather prediction. It envisions a vast network of interconnected devices, such as smartphones, airplanes, drones, smart vehicles and CCTV cameras, to gather data and act as weather stations.

Working under the principle that if weather affects nearly everything then nearly everything is a potential sensor, the company aims to make location-specific, short-term weather reports, with particular hope of addressing what it calls blind spots in the developing world.

ClimaCell points outOpens a new window the three main elements of weather prediction:

• Observe, or “see what the weather is doing at the moment;”
• Model, or “guess what it is about to do;”
• Disseminate, or “tell everyone about it.”

But recording atmospheric conditions with instruments ranging from thermometers and barometers to weather balloons and satellites can be complicated and hugely expensive, limiting the best equipment to wealthy countries.

The U.S. National Oceanic and Atmospheric Administration operates a series of weather satellites. Two more are under development and are estimated to cost $11 billion over their 15-year lifespans. The United States counts roughly 4,000 weather observation stations for its population of more than 300 million in its landmass of 3.7 million miles  By contrast, the Democratic Republic of Congo, with about a quarter of the United States’ landmass and population, has just five.

“Weather observations are the currency of meteorology,” says Rei Goffer, ClimaCell’s chief strategy officer and co-founder. “Without them, forecasting doesn’t work. But due to outdated thinking, this currency is in short supply, and badly distributed, often not reaching places where it would do the most good. If weather affects nearly everything, then nearly everything can be a weather sensor. ClimaCell’s global data partnerships help us reach into nearly unlimited geographies. The best part is that it can be deployed rapidly in the developing world.”

The stakes for accurate weather prediction are high in the developing world. The United Nations reportsOpens a new window that in Sub-Saharan Africa, for example, “nearly two disasters of significant proportions have been recorded every week since 2000, with around 12.5 million people per year affected.” Cyclone Idai was described as the most destructive storm ever to strike southern Africa.

Aside from such humanitarian crises, the requirements of weather prediction vary hugely between, say, harnessing wind energy in Spain and predicting flood risk in Malaysia. Enterprises and industry often miss out on the nuances they need to optimize their operational efficiency. New economy businesses, which ClimaCell describes as “the ride-sharing, on-demand, drone-flying, autonomous-vehicle producing ones” have a far greater appetite for data than traditional forecasts can provide.

Its products include HyperCast, a web-based interactive high-definition weather map, HyperCast Aviation and the Microweather API, which provides a live weather data stream that can be integrated with other software and platforms.

Its client roster so far includes Delta, JetBlue and United airlines, the New England Patriots and the Netafim agricultural technology company. It currently serves business clients, with plans to release a consumer app later this year. Other providers of weather apps include CARROT Weather and Dark Sky.

The three-year-old company saw an opportunity to tap the existing network of microwave links (point-to-point wireless connections between two cellular towers), effectively ‘seeing’ the weather by observing changes in wireless signal strength.

Shimon Elkabetz, co-founder and CEO, explains howOpens a new window they worked around the need to access data directly from multiple companies, from cellular carriers to satellite carriers and other private and governmental network operators: “The data we rely on for Virtual Sensing is not sensitive, private, or even attributable to any of our data partner’s users. With microwaves, for example, we only look at the signal strength (the energy level of the signal), rather than at the signal itself (data moving across networks). If we think of a cellular network as a highway, and the data passing through it as the cars, then we’re not looking at the cars; we’re just paying attention to how many lanes are open on the road.”

A few weeks ago, ClimaCell announced an extended flood prediction serviceOpens a new window  using its global network of 500 million virtual sensors to provide by-the-minute precipitation forecasting and send alerts 24 hours before a potential flood.

Last October, it raised $45 million in a Series B funding round led by Clearvision Ventures with participation from existing investors Canaan, Fontinalis Partners, JetBlue Technology Ventures and Square Peg Capital, and new supporters Ford Smart Mobility and Envision Ventures.

Marcy Klevorn, executive vice-president and president of Ford Motor’s mobility business, offers this preview of the automaker’s interest in the weather: “High-definition, microweather information supports multiple mobility and [autonomous vehicle] initiatives,” he says, “including route planning, the services we can offer to customers via FordPass, and sharing information via the Transportation Mobility Cloud.

“In the future,” he adds, “real-time data will allow autonomous vehicles to be routed around bad weather.”