wildfires
Auto Added by WPeMatico
BreezoMeter, which powers air quality in Apple’s Weather app, launches Wildfire Tracker
BreezoMeter has been on a mission to make environmental health hazard information accessible to as many people as possible. Through its air quality index (AQI) calculations, the Israel-based company can now identify the quality of air down to a few meters in dozens of countries. A partnership with Apple to include its data into the iOS Weather app along with its own popular apps delivers those metrics to hundreds of millions of users, and an API product allows companies to tap into its data set for their own purposes.
Right on the heels of a $30 million Series C round a few weeks ago, the company is radially expanding its product from air quality into the real-time detection of wildfire perimeters with its new product, Wildfire Tracker.
The new product will take advantage of the company’s fusion of sensor data, satellite imagery and local eyewitness reports to be able to identify the edges of wildfires in real time. “People expect accurate wildfire information just as they expect accurate weather or humidity data,” Ran Korber, CEO and co-founder, said. “It has an immediate effect on their life.” He added further that BreezoMeter wants to “try to connect the dots between climate tech and human health.”
Fire danger zones will be indicated with polygonal boundaries marked in red, and as always, air quality data will be viewable in these zones and in surrounding areas.
BreezoMeter’s air quality maps can show the spread of wildfire pollution. Image Credits: BreezoMeter.
Korber emphasized that getting these perimeters accurate across dozens of countries was no easy feat. Sensors can be sparse, particularly in the forests where wildfires ignite. Meanwhile, satellite data that focuses on thermal imaging can be fooled. “We’re looking for abnormalities … many of the times you have these false positives,” Korber said. He gave an example of a large solar panel array which can look very hot with thermal sensors, but obviously isn’t a fire.
The identified fire perimeters will be available for free to consumers on BreezoMeter’s air quality map website, and will shortly come to the company’s apps as well. Later this year, these perimeters will be available from the company’s APIs for commercial customers. Korber hopes the API endpoints will give companies like car manufacturers the ability to forewarn drivers that they are approaching a conflagration.
The new feature is just a continuation of BreezoMeter’s longtime expansion of its product. “When we started, it was just air quality … and only forecasting air pollution in Israel,” Korber said. “Almost every year since then, we expanded the product portfolio to new environmental hazards.” He pointed to the addition of pollen in 2018 and the increasingly global nature of the app.
Wildfire detection is an, ahem, hot area these days for VC investors. For example, Cornea is a startup focused on helping firefighters identify and mitigate blazes, while Perimeter wants to help identify boundaries of wildfires and give explicit evacuation instructions, complete with maps. As Silicon Valley’s home state of California and much of the world increasingly become a tinderbox for fires, expect more investment and products to enter this area.
Powered by WPeMatico
a16z leads investment in Firemaps, a marketplace for home hardening against wildfires
Wildfires are burning in countries all around the world. California is dealing with some of the worst wildfires in its history (a superlative that I use essentially every year now) with the Caldor fire and others blazing in the state’s north. Meanwhile, Greece and other Mediterranean nations have been fighting fires for weeks to bring a number of massive blazes under control.
With the climate increasingly warming, millions of homes just in the United States alone are sitting in zones at high risk for wildfires. Insurance companies and governments are putting acute pressure on homeowners to invest more in defending their homes in what is typically dubbed “hardening,” or ensuring that if fires do arrive, a home has the best chance to survive and not spread the disaster further.
SF-based Firemaps has a bold vision for rapidly scaling up and solving the problem of home hardening by making a complicated and time-consuming process as simple as possible.
The company, which was founded just a few months ago (in March), sends out a crew with a drone to survey a homeowner’s house and property if it is in a high-risk fire zone. Within 20 minutes, the team will have generated a high-resolution 3D model of the property down to the centimeter. From there, hardening options are identified and bids are sent out to trade contractors to perform the work on the company’s marketplace.
Once the drone scans a house, Firemaps can create a full CAD model of the structure and the nearby property. Image Credits: Firemaps.
While early, it’s already gotten traction. In addition to hundreds of homeowners who have signed up on its website and a few dozen that have been scanned, Andrew Chen of a16z has led a $5.5 million seed round into the business (the Form D places the round sometime around April). Uber CEO Dara Khosrowshahi and Addition’s Lee Fixel also participated.
Firemaps is led by Jahan Khanna, who co-founded it along with his brother, who has a long-time background in civil engineering, and Rob Moran. Khanna was co-founder and CTO of early ridesharing startup Sidecar, where Moran joined as one of the company’s first employees. The trio spent cycles exploring how to work on climate problems, while staying focused on helping people in the here and now. “We have crossed certain thresholds [with the climate] and we need to get this problem under control,” Khanna said. “We are one part of the solution.”
Over the past few years Khanna and his brother explored opening a solar farm or a solar-powered home in California. “What was wild, whenever we talked to someone, is they said you cannot build anything in California since it will burn down,” Khanna said. “What is kind of the endgame of this?” As they explored fire hardening, they realized that millions of homeowners needed faster and cheaper options, and they needed them sooner rather than later.
While there are dozens of options to harden a home to fire, some popular options include constructing an ember-free zone within a few feet of a home, often by placing gravel made of granite on the ground, as well as ensuring that attic vents, gutters and siding are fireproof and can withstand high temperatures. These options can vary widely in cost, although some local and state governments have created reimbursement programs to allow homeowners to recoup at least some of the expenses of these improvements.
A Firemaps house in 3D model form with typical hardening options and associated prices. Image Credits: Firemaps.
The company’s business model is simple: vetted contractors pay Firemaps to be listed as an option on its platform. Khanna believes that because its drone offers a comprehensive model of a home, contractors will be able to bid for contracts without doing their own site visits. “These contractors are getting these shovel-ready projects, and their acquisition costs are basically zero,” Khanna said.
Long-term, “our operating hypothesis is that building a platform and building these models of homes is inherently valuable,” Khanna said. Right now, the company is launched in California, and the goal for the next year is to “get this model repeatable and scalable and that means doing hundreds of homes per week,” he said.
Powered by WPeMatico
OroraTech’s space-based early wildfire warnings spark $7M investment
With wildfires becoming an ever more devastating annual phenomenon, it is in the whole planet’s interest to spot them and respond as early as possible — and the best vantage point for that is space. OroraTech is a German startup building a constellation of small satellites to power a global wildfire warning system, and will be using a freshly raised €5.8 million (~$7 million) A round to kick things off.
Wildfires destroy tens of millions of acres of forest every year, causing immense harm to people and the planet in countless ways. Once they’ve grown to a certain size, they’re near impossible to stop, so the earlier they can be located and worked against, the better.
But these fires can start just about anywhere in a dried out forest hundreds of miles wide, and literally every minute and hour counts — watch towers, helicopter flights and other frequently used methods may not be fast or exact enough to effectively counteract this increasingly serious threat. Not to mention they’re expensive and often dangerous jobs for those who perform them.
OroraTech’s plan is to use a constellation of about 100 satellites equipped with custom infrared cameras to watch the entire globe (or at least the parts most likely to burst into flame) at once, reporting any fire bigger than 10 meters across within half an hour.
Image Credits: OroraTech
To start out with, the Bavarian company has used data from over a dozen satellites already in space, in order to prove out the service on the ground. But with this funding round they are set to put their own bird in the air, a shoebox-sized satellite with a custom infrared sensor that will be launched by Spire later this year. Onboard machine learning processing of this imagery simplifies the downstream process.
Fourteen more satellites are planned for launch by 2023, presumably once they’ve kicked the proverbial tires on the first one and come up with the inevitable improvements.
“In order to cover even more regions in the future and to be able to give warning earlier, we aim to launch our own specialized satellite constellation into orbit,” said CEO and co-founder Thomas Grübler in a press release. “We are therefore delighted to have renowned investors on board to support us with capital and technological know-how in implementing our plans.”
Image Credits: OroraTech
Those renowned investors consist of Findus Venture and Ananda Impact Ventures, which led the round, followed by APEX Ventures, BayernKapital, Clemens Kaiser, SpaceTec Capital and Ingo Baumann. The company was spun out of research done by the founders at TUM, which maintains an interest.
“It is absolutely remarkable what they have built up and achieved so far despite limited financial resources and we feel very proud that we are allowed to be part of this inspiring and ambitious NewSpace project,” APEX’s Wolfgang Neubert said, and indeed it’s impressive to have a leading space-based data service with little cash (it raised an undisclosed seed about a year ago) and no satellites.
It’s not the only company doing infrared imagery of the Earth’s surface; SatelliteVu recently raised money to launch its own, much smaller constellation, though it’s focused on monitoring cities and other high-interest areas, not the vast expanse of forests. And ConstellR is aimed (literally) at the farming world, monitoring fields for precision crop management.
With money in its pocket Orora can expand and start providing its improved detection services, though sadly, it likely won’t be upgrading before wildfire season hits the northern hemisphere this year.
Powered by WPeMatico
As wildfire season approaches, AI could pinpoint risky regions using satellite imagery
The U.S. has suffered from devastating wildfires over the last few years as global temperatures rise and weather patterns change, making the otherwise natural phenomenon especially unpredictable and severe. To help out, Stanford researchers have found a way to track and predict dry, at-risk areas using machine learning and satellite imagery.
Currently the way forests and scrublands are tested for susceptibility to wildfires is by manually collecting branches and foliage and testing their water content. It’s accurate and reliable, but obviously also quite labor intensive and difficult to scale.
Fortunately, other sources of data have recently become available. The European Space Agency’s Sentinel and Landsat satellites have amassed a trove of imagery of the Earth’s surface that, when carefully analyzed, could provide a secondary source for assessing wildfire risk — and one no one has to risk getting splinters for.
This isn’t the first attempt to make this kind of observation from orbital imagery, but previous efforts relied heavily on visual measurements that are “extremely site-specific,” meaning the analysis method differs greatly depending on the location. No splinters, but still hard to scale. The advance leveraged by the Stanford team is the Sentinel satellites’ “synthetic aperture radar,” which can pierce the forest canopy and image the surface below.
“One of our big breakthroughs was to look at a newer set of satellites that are using much longer wavelengths, which allows the observations to be sensitive to water much deeper into the forest canopy and be directly representative of the fuel moisture content,” said senior author of the paper, Stanford ecoydrologist Alexandra Konings, in a news release.
The team fed this new imagery, collected regularly since 2016, to a machine learning model along with the manual measurements made by the U.S. Forest Service. This lets the model “learn” what particular features of the imagery correlate with the ground-truth measurements.
They then tested the resulting AI agent (the term is employed loosely) by having it make predictions based on old data for which they already knew the answers. It was accurate, but most so in scrublands, one of the most common biomes of the American west and also one of the most susceptible to wildfires.
You can see the results of the project in this interactive map showing the model’s prediction of dryness at different periods all over the western part of the country. That’s not so much for firefighters as a validation of the approach — but the same model, given up to date data, can make predictions about the upcoming wildfire season that could help the authorities make more informed decisions about controlled burns, danger areas and safety warnings.
The researchers’ work was published in the journal Remote Sensing of Environment.
Powered by WPeMatico
Tech’s Role In Fighting Wildfires
“Burn rates” are perennially on the minds of Silicon Valley VCs and entrepreneurs, but what about the very real burn rates that ravish the surrounding California countryside, as well as other areas each year? Currently there are five states that are each battling more than 10 major wildfires, with seven million acres burned so far. Big data is already being used to understand… Read More
Powered by WPeMatico