IoT solutions for halting wildfires: a laborious path through generalization
The German startup Dryad Networks, which develops IoT solutions to detect and stop fires, closed 10.5 M€ in a fundraising action in August 2022.
This fundraising has been a response to recurring mega-fires in the summer season. Year after year their number and scale are higher and higher. In Europe alone in 2022, 700,000 hectares were burned. They now affect geographical areas formerly spared, such as the United Kingdom or Sweden. The lower risk is the result of the concentration of means (planes, fire trucks) in the areas that were formerly the most susceptible to forest fires, namely the Mediterranean countries (France, Greece, Spain).
We know that technologies such as the one developed by Dryad are ready for the market. This consists of a network of hundreds of small sensors deployed throughout the forests. The sensors send the gathered data for analysis by artificial intelligence (AI). This system provides a very early warning of a fire outbreak.
Lower costs than the fires themselves
A rough calculation determined that it was cheaper to equip all of Europe’s forests (those at risk) with this type of solution, when compared to the exorbitant costs of fire tolls, for a single year. Of course, once the solution is installed, it is amortized over several years (substantial maintenance costs).
This article’s objective is to understand the various technical, financial, political, and even psychological issues that hinder the deployment of these solutions on a large scale.
How does forest fire prevention work so far?
Fire prevention is carried out by both the firefighters and the national forestry office, in addition to citizen alerts.
It consists of the following tools:
- ground patrols: patrols in vehicles equipped with binoculars and radios that roam the forests and scan for possible outbreaks of fires. They also observe the level of dryness of the soil.
- awareness-raising operations aimed at the population, visitors, and hikers (to limit the start of involuntary fires linked to cigarettes or barbecues)
- aerial reconnaissance means: helicopters and airplanes
- tactical fire ignitions, and fires deliberately started by firefighters to contain existing fire
- satellite images which, beyond being very useful in the fight against the air in progress, can indicate the areas at risk and on which preventive fires would be useful.
Needless to say that all of these tools are not sufficient, forest fires have never been that active. The described systems aim to limit the number of fire outbreaks through prevention and monitoring activities. Yet the vast majority of fires are caused by humans.
IoT solutions exist to control a fire outbreak before it becomes uncontrollable.
How can IoT seriously reinforce this architecture?
The IoT solution developed by Dryad Networks consists of a network of hundreds of sensors meshed by gateways (operating in mesh) and uploading data to the cloud. We will look at each of the stages of this solution.
The video made by Bosch and Dryad:
All sensors are powered by a solar battery. They measure temperature, humidity and air pressure, as well as several other fire-related gasses (hydrogen, carbon monoxide and dioxide, among others). They are equipped with Bosch BME688 3mm size sensors. They use the Lorawan low-powered network to communicate with the gateways.
Also powered by solar panels, the gateways ensure reliable communication between the sensors and are placed every 2 to 6 kilometers They allow two-way communication and can be updated remotely (Firmware Update Over-the-Air).
Another type of gateway is used to be installed at the exits of the forest, such as a ranger relay. It uses the cellular or LAN network (if an internet connection is available) and can be plugged into the electrical network. It strengthens the resilience of the entire surveillance network.
The “connected” forest is monitored from anywhere thanks to a cloud platform that collects, analyzes and alerts fire starts.
One could therefore think that IoT solutions are at least complementary to traditional prevention devices. No solution having been generalized in Europe to date, it is appropriate to wonder about the reasons for this slowness.
Why has there been no IoT fire prevention solution widely deployed to date?
Political, societal, and psychological reasons
The article from 2021, written by Etien Yovchev, editor-in-chief of The Recursive media, offers several explanations.
The usual inertia of the political spheres
It is politically more profitable to direct the budget towards tangible measures rather than prevention. Acquiring a new canadair of several fire engines is something visible, so everyone can understand. Logic dictates that to fight fires, we focus on firefighters to fight fires, rather than embarking on technological experiments, even if they have been verified.
We will therefore blame here the lack of anticipation of the decision-making spheres.
Resistance to change in all organizations
It is difficult to introduce (from above, i.e. via legislation) new tools to major public service organizations. The implementation of an IoT solution would involve an overhaul of the operation of fire prevention for firefighters.
If the implementation fails, the consequences can be disastrous.
Barriers in the adoption of any new technology
Like any technological innovation, IoT solutions follow an adoption curve that can skate before the majority shifts to early adopters.
Commercial and economic reasons
A Spanish company’s case study, specializing in developing IoT solutions, provides us with some other answers. Dating from 2017 and recounting an experience with a large Spanish national forest park, she explains to us that the technology was already mature at that time, the problem turned out to be commercial:
(…) hidden costs came from the number of sensors multiplied by the density and extension, plus the lack of an answer to who and how would install and maintain those sensors in steep and inaccessible places. The sensor system wasn’t a problem. Communication technology wasn’t a problem. IoT wasn’t a problem and it was delivering a proper technology solution, I dare to say.
But there was no business case behind it according to the available budget.
We can however consider that the aggravation and generalization of forest fires on the globe modify the economic deal.
Are we able to deploy this solution on a very large scale for the upcoming summer?
Like any IoT solution with a (very) large number of terminals, the question arises of the real scalability of the project. We are emerging from a major semiconductor crisis, which has impacted all economic sectors. Part of the automotive sector has even been shut down, and the rise in prices has become irreconcilable with market prices.
Supply chain uncertainties
China’s zero covid policy (being relaxed at the time of writing), rising commodity prices and geopolitical instability in Ukraine and Asia have not finished shaking up supply chains and price volatility. The cost of maritime transport remains very high, and it is difficult to envisage producing so many sensors on European or American soil while keeping the solution budget at reasonable prices.
The considerable amount of components (we are talking about deployment of a large part of the forests of Europe here) raises serious doubts about the ability to obtain them all within a few months (for summer 2024 for example). Added to this is the time for assembly, verification, testing and installation.
For an operation like this — deploying thousands of sensors over several million square kilometers of forests — one can imagine the challenge in terms of human resources.
IoT architecture involves a large number of different disciplines which must work in perfect coordination: hardware, software, connectivity, security, cloud, and data; in addition to all the traditional support functions of a technology company (design, sales, marketing, administrative functions, legal support).
The specificity of the deployment of this forest fire prevention architecture lies in its high density of sensors, to be installed manually on the trees, while progressing in the forest, mainly on foot.
One can imagine the use of large numbers of volunteers, but who will have to be trained and supervised?
The crucial stage of a large-scale test is still missing, which will prove the efficiency of the system and give a precise idea of the budget for its deployment, then its annual maintenance.
In the short term, uncertain control of costs and deadlines
The difficulty in controlling costs and manufacturing times on the one hand and deployment on the other hand, is undoubtedly the main reason for the lack of mass adoption, to date, of the IoT fire prevention systems.
There is also a question of the durability of sensors, for instance, in storms (with falling trees and branches), the ability of mini solar panels to capture the sun’s rays through treetops which change by nature. How many sensors will need to be replaced each year?
In the medium term, an inevitable deployment
Megafires will, unfortunately, continue to be a growing problem in Europe. They hardly impact transport and tourism, which was not the case a few years ago.
The plausible restoration of production chains, or their restructuring, and the growing adoption of IoT solutions by public services and local authorities should overcome most of the obstacles mentioned, but how long will this take?
As public organizations are reluctant to make the first move, the first massive large-scale deployment will likely enter private forestry, opening the path for other entities.
About me // French Marketer who specialises in B2B tech and IoT. With the background of lobbying and over 10 years experience as creative director, I help IoT start-ups achieving stable growth, meaningful branding and long-lasting demand generation. More on www.dxm-agency.com