Weathering the Storm
BY : Nayantara Bhat
Understanding the specialized and high-tech field of meteorological technology
Anyone who’s ever made weekend plans that depend on good weather knows that forecasting is both a blessing and a curse. While it’s comforting to not be entirely blind to meteorological changes, it’s often the case that forecasts are either vague or completely incorrect. That’s where meteorological technology comes in.
Weather technology is a small but rapidly-growing field. Most startups in this space are business-to-business (B2B) as opposed to consumer-facing. A notable player is the Climate Corporation [acquired by Monsanto in 2013], an agriculture-focused company that examines weather and other field data to help farmers withstand the increasingly volatile weather conditions caused by climate change.
Such insights could have tangible impacts on many other essential industries, including, but not limited to, renewable energy generation, aviation, shipping, drone operation, and by extension–the individual.
According to Chinese University of Hong Kong Professor Gabriel Lau, the technologies used in the field range from supercomputers running numerical models to project changes in the earth’s climate, to advanced communication devices disseminating information across the globe, to the deployment of large-scale satellites and other observational instruments.
“The use of technology has advanced very rapidly in the weather and climate field over the recent years,” says Lau. Remote-sensing techniques and new methods of viewing dynamic atmospheric events, which happen too fast for cameras to capture, are contributing to a higher-accuracy understanding of the field.
Better monitoring and prediction models have become all the more essential with the growing urgency of climate change, necessitating the use of high-speed computing and data archive facilities, and conscientious long-term data collection and management.
However, progress in this field isn’t equally distributed. Developing countries are lacking in the infrastructure that is crucial to collecting all this data. That said, they are now realizing the importance of forecasting and monitoring the climate and communicating the information with the public. Lau believes that if developing economies are given sufficient access to information collected by other entities, they will be able to address the climate issues they are facing more effectively.
“Advances in communication and Internet technologies have contributed to improved data access,” says Lau. “Weather and climate operations and research in Japan, China, Taiwan, Singapore, India and South Korea have attained high standards in recent years.
With all of this technology being put to use, the hope is that industries and individuals will be able to more efficiently gather the information needed to take action against climate change. Boston-based ClimaCell is one startup that is trying to make weather less of a wild card. Founded by a trio of Israeli Air Force veterans–Shimon Elkabetz, Rei Goffer, and Itai Zlotnik–the startup is already making waves in aviation and other industries, recently carrying out a limited launch of their consumer app.
Jumpstart interviewed ClimaCell on technology, industry, and the real-world applications of their solution.
What is ClimaCell’s founding story?
As is often the case with startups, ClimaCell was founded to answer a real-life need our founders experienced. In our case, it was the need to have weather forecasts you can actually rely on, given the massive operational and safety implications of being surprised by weather conditions during high-stakes operations. Once they started working on ClimaCell, they saw how broad the need was, from industries to emerging economies.
Can you describe your technology?
ClimaCell is pioneering a ‘Weather of Things’ approach: the combination of unique data from the connected world and proprietary, high-resolution models to make sense of that data.
On the data side, ClimaCell takes all of the regular meteorological data used today, and adds in hundreds of millions of data points from the connected world. What was previously just a cellular tower, vehicle, aircraft or a satellite dish, is now receiving new life as a weather sensor.
This data is then analyzed by the ClimaCell Bespoke Atmospheric Model, which analyzes the data more regularly (every few minutes vs once an hour) and at far higher resolutions.
How do these systems provide hyperlocal weather information?
Existing meteorological data leaves many blindspots. Satellites, for example, are extremely expensive to deploy, even in developed countries, and don’t pick up what’s happening at ground and rooftop levels.
By tapping into a variety of data types, such as the changes in signal strength sent between wireless towers due to precipitation, this new form of data is processed into weather data and ingested by ClimaCell’s models.
What information does ClimaCell provide to users?
The result of this new kind of data and modelling is MicroWeather: hyper-accurate, specific, and customizable information. Customers can choose which weather parameters they care about (rain, snow, sleet, temperature, humidity, wind, thunder), the areas they care about (at a street-by-street level), and the timeframe they care about (current conditions, short-term forecasts, longer-term forecasts, or historical information).
How easy or difficult has it been to get access to the requisite data sources you need for your software?
The data sources we use from the connected world (wireless signals, microwave links, airplanes, drones) are all part of existing infrastructure. It is the rare win-win situation, as it gives the companies a secondary use for their data–which is all anonymized, network-level information–while providing us with the data richness and diversity we need to vastly improve weather forecasting.
Who are the intended users of this platform?
There is a broad need for accurate weather forecasting since weather has such huge impact across sectors. Our customers fall into four categories: existing industries such as aviation, transportation, and construction; new economy and on-demand companies; developing countries, where there is little to no traditional forecasting infrastructure; and, finally, we are launching a consumer app in the coming weeks.
What kind of industries would benefit from this software?
One third of the global economy is comprised of weather-sensitive industries and each needs different kinds of weather information. Renewable energy generation companies, for example, are focused on wind and solar parameters to anticipate supply, while drone operators may care about precipitation and thunder to plan routes and ensure safety. There is yet another need in developing countries, as many have little, if any, forecasting infrastructure, impairing planning on everything from agriculture to flood relief.
What are the implications of ClimaCell’s technology when it comes to climate change research and conservation efforts?
ClimaCell is focused on weather forecasting. That said, as weather conditions become more volatile and unpredictable due to climate change, the need for accurate, specific weather forecasts continues to rise.
How do you deal with infrastructure limitations in developing countries?
This situation exists in many emerging economies, but let’s take India as an example:
As of now, weather forecasting in India is very inaccurate. ClimaCell’s technology uses the connected infrastructure that already exists to predict the weather.
In the two pictures (above), the one on the left is showing one government weather station in Mumbai, which is at the airport. The picture on the right shows ClimaCell’s sensors. Since traditional forecasting infrastructure is extremely expensive, many emerging economies have few satellites, radars and weather stations, if any. However, they are part of the connected world. Therefore, ClimaCell can use them to predict the weather.
[The image on the top left is taken from the India Meteorological Department (IMD) website, and might not represent all the IMD weather stations in that region. Still, the availability of real-time observations in that region is very limited. The image on the right is an anonymized depiction of our virtual sensors. Locations of the actual devices were randomized, but it is true to the density of our sensors in that region.]
How will better weather forecasting affect our daily lives?
Weather impacts everyone’s lives, as we are constantly making weather-related decisions. Should I take my dog for a walk now or will I get stuck in a downpour? Should I cancel a camping trip due to a vague ‘30% chance of rain tomorrow’ forecast?
Of course, ClimaCell’s forecasts, which are in use by companies and emerging economies, affect people’s lives behind-the-scenes, everyday. This is true for an airline being able to avoid delays by knowing the exact time and location of a lightning storm, a roofing company that is able to maximize work time by knowing exactly when to expect rain, and a developing country getting flooding information that allows them to evacuate residents in advance.