Russian Scientists Use Digital Technology to Protect Far Eastern Forests
Forests cover 48% of Russia’s territory. Although the overall number of wildfires has declined, fires in some regions still reach catastrophic proportions. This is especially true in the taiga, where even a small blaze can spread rapidly and engulf vast areas.
Scientists have long been searching for effective ways to monitor forests. Researchers at Pacific National University (PNU) have developed a smart monitoring system powered by artificial intelligence. It predicts where wildfires may start and how they could spread, while also detecting biological threats.
Staying One Step Ahead
The taiga has no cellular towers, so the researchers first developed autonomous relay stations. These self-powered transmission units, which can run on solar panels, provide connectivity in areas without conventional communications infrastructure. A drone sends the photos and video it collects to a relay station, which forwards the data to a satellite. The satellite then relays it to Pacific National University. During early tests, the drone could operate up to 30 kilometers from its launch point, where the base station was located. That range has since increased to 50 kilometers. The team plans to extend it to 70 kilometers, bringing a forest area of 770,000 hectares within the system’s coverage.
At a central data-processing hub, artificial intelligence analyzes the incoming information. The software has been trained on thousands of wildfire examples and can identify patterns involving dry conditions, high temperatures, wind direction and the speed at which flames might spread. Instead of simply detecting a forest that is already burning, the system predicts where a fire could break out in the near future. That gives emergency services time to prepare and deploy personnel and equipment before a blaze takes hold.
Russia already operates the ISDM-Rosleskhoz remote forest-fire monitoring system, which uses satellite imagery, including data from Russian Kanopus-V satellites, along with ground-based and aerial monitoring to detect fire outbreaks. PNU’s new system adds a different capability: rather than merely detecting fires, it aims to anticipate them.
Mapping High-Risk Areas
In 2025, Russia recorded 6,800 forest fires covering 4.3 million hectares. In 2024, there were 8,900 fires across 8.3 million hectares. Despite significant progress in reducing the number of fires, the threat remains serious. Digital technologies can help prevent a fire before it starts or detect it in its earliest stages, when it is still easier to extinguish. Researchers at Penza State University have developed an AI-powered software system that predicts where forest fires are likely to occur. It automatically processes terabytes of data from satellites, ground weather stations and unmanned aerial vehicles (UAVs). AI evaluates weather conditions, vegetation health, historical records and many other factors, including expert assessments, to identify areas at high risk of ignition before a fire breaks out.
The East Siberian State University of Technology and Management has also developed a machine-learning program to forecast wildfires. It collects and processes satellite data, including archived Earth images from the MODIS satellite, to identify thermal anomalies and map potential fire outbreaks. In paralle, the Moscow Aviation Institute has developed a system that uses neural networks to recognize smoke and distinguish it from other plumes, such as steam from a combined heat and power plant. It determines the coordinates of a fire and sends them to Russia’s Ministry of Emergency Situations. All processing takes place onboard the drone.
Multiple Systems, One Shared Goal
Russia also has a nationwide remote monitoring and management system called Lesokhranitel (Forest Sentinel). The platform connects more than 3,300 cameras across over 70 regions. In areas under its surveillance, the number of fires has fallen by at least 20%. Another system, Lesnoy Dozor (Forest Watch), also monitors forests for the earliest signs of ignition. In addition, forestry agencies in nearly every Russian region have UAVs, with the national fleet exceeding 2,000. More than 5,100 additional drones are scheduled for procurement by the end of 2030. Drones cost less, can be deployed faster and are more accessible in remote parts of Siberia and the Russian Far East.
Russia is shifting from responding to fires after they break out toward proactively managing wildfire risks. PNU’s project is one of the more comprehensive examples of this approach, combining AI, robotics and autonomous communications. Such platforms are likely to be particularly useful in regions that are difficult to reach by road. For Russia’s IT sector, the technology represents a promising area at the intersection of environmental protection, artificial intelligence and digital infrastructure.
Earlier fire detection can help prevent flames from reaching populated areas, give emergency crews more time to respond and reduce the damage caused by wildfires. Autonomous communications systems could also serve purposes beyond firefighting, helping keep people safe in areas without reliable network coverage, from search-and-rescue expeditions to industrial site monitoring.
