An international team of UPV/EHU researchers has unveiled a new algorithm capable of processing satellite images and finding even very small forest fires. It is expected to become an important tool for fighting them.
Satellite monitoring of fires
The new method for detecting fires less than 250 meters away uses data from two satellites collecting optical images and data from four others measuring the high temperatures caused by fires.
Aitor Bastarrika of the UPV/EHU’s Building Heritage Research Group proposes an algorithm for global mapping of burned areas at higher resolution. The study is published in the ISPRS Journal of Photogrammetry and Remote Sensing.
Obtaining accurate and relevant information from fire-affected areas is important not only to better understand air quality, biogeochemical cycles, or climate, but also to facilitate fire control management.
A few decades ago, mapping of burned areas was based on rural studies, but since the launch of Earth observation satellites, remote sensing has become a more practical option for locating burned areas, as satellites make it easier to measure fire-affected areas both regionally and globally.
Image resolution
Image resolution is an issue for satellite-mapped areas. In fact, the resolution of global observations has been low so far. “The omission error of current products is very high: many areas that are in fact burnt are not identified as such,” said Bastarrika.
“Current systems use a pixel size of between 250 and 500 meters, so they do not detect fires of less than 250 meters. And in some ecosystems, fires of this size are very frequent.”
Using data from six different satellites, the researchers developed an algorithm to achieve higher resolution. First, they used images taken by the two Sentinel-2 constellation optical satellites: they offer good spatial resolution of 10-20 meters, but with low temporal frequency, since images of a particular location are obtained only every five days.
Second, they used MODIS (obtained from Terra and Aqua satellites) and VIIRS (obtained from Suomi NPP and NOAA-20 satellites) products that detect active fires: they detect high-temperature spots at a low spatial resolution of 375-1000 meters, but at a high frequency because they collect data every day.
Bastarrika’s proven algorithm
The algorithm developed by Bastarrika’s team uses data from two active fire detection products and uses it to train an optical imaging system to create a classification system. Then it makes predictions on both what has burned and what hasn’t.
“In addition, these forecasts were tested in 576 areas around the world. In other words, the algorithm was analyzed in all ecosystems where burnt areas are significant,” explained Bastarrika.
The algorithm developed by the team is not the only one, there are other similar ideas. However, the contributions of UPV/EHU researchers are particularly important because the algorithm is designed to be applied globally and produce medium-resolution results.
Future of mapping algorithms
“Algorithms already exist for mapping specific areas at medium resolution, but our proposal can map burnt areas across the world, does so at an acceptable resolution and is up and ready for use.”
In the future, the intent is to create advanced products with the implementation of the developed method. “Since up to now, they have been prepared to use low-resolution systems, from now on, the aim is to create products that deliver results at a medium level of resolution. Moving from low to medium resolution would make a great contribution towards identifying specific ecosystems and studying climate,” said Bastarrika.
Provided by phys.org
