But Also Help in Fixing It
Some months ago, the Amazon rainforest wildfires were ablaze. Immediately after them, Australia burned. At the same as the news were filled with photos of burned forests, the countries’ decision–makers downplayed the destruction and explained it to be part of a normal natural cycle. The extent of the damage was impossible to determine from the ground, but with the help of satellite imagery it was possible to get valuable and recent information on the development of the situation in burning areas. By adding artificial intelligence to this process, satellite data and history information can be used to examine whether the damage is caused by natural changes related to the qualities of the area or by consequences of climate change.
The situation monitoring of last year’s big forest fires is just one example of cases where satellite imagery can be used. We have primarily got used to associating the unscrupulously up-to-date satellite data with natural disasters, shrinking rainforests, and diminishing glaciers. But satellite data can also be used the other way around, to restore and fix the already occurred damages in different habitats. An example of such use is forestation.
Carbon Sequestration by Increasing Forest Area
Forestation means either planting an entirely new forest on an area that has been previously in other use (afforestation) or replanting a forest that has been clear cut or otherwise depleted (reforestation). In Finland, the Forest Act requires a clear-cut forest to be renewed, but in many countries, there are no similar requirements. However, because of the reforestation obligation, afforestation has been seen as a concrete way to increase forest carbon sequestration.
Different forestation projects pertain to practically all countries on the globe. In addition to climate change, forestation can be used to fight erosion and soil degradation, to restore biodiversity and to manage flood risk.
Artificial Intelligence Finds Best Sites Fit for Forestation
However, it matters where you plant the forest. Concerning forestation projects, for example the diminishment of valuable habitats such as wetlands or meadows and the endangerment of food security have been brought up. If not handled properly, a forestation project can homogenize species diversity or cause bad forest cover growth and even disappearance of species.
Per se, there are plenty of areas fit for forestation in Finland. Some good examples are:
- field sections that are not used for farming or that are less profitable
- bottom of marshland freed from peat production
- power lines to be ground cabled
- different kinds of wastelands
Satellite data can provide a picture of the current land use to be used as the basis of the analysis. Combined with other geospatial data, it can help us to take into account for example vital habitats, protected open lands, water bank sites or other sites unfit for forestation, all the while considering regional differences.
The sites fit for forestation can be searched by combining satellite data with other data sources depicting the state of environment with the help of machine learning. The more and the better learning data the artificial intelligence has at its disposal, the more accurately it will learn to detect the suitable sites.
By incorporating weather data, forestation targets can also be analyzed on areas where for example rainfall is critical.
Staying on Path with Forestation
Forestation as a climate act divides opinions. Inadequately executed forestation attempts can be costly both from the ecological and from the economic standpoint. Questions have arisen especially regarding securing food supply. Nonetheless, forestation is a very concrete act to control the climate change, so it interests both researchers and forest owners. According to a study conducted by the Natural Resources Institute Finland, increasing forestation is justifiable from the point of view of the climate objectives of the land use sector, even if its effects are relatively small in comparison to forests’ total carbon sink.
The automatic recognition of sites suitable for forestation has been tested in the framework of Bitcomp’s development projects. The Finnish Forest Centre and the Natural Resources Institute Finland have also started a national project aiming to produce information on how Finland’s carbon sinks can be increased with forestation of marsh bases and lowly profitable or currently unused agricultural land.
Forest owners find forestation an interesting option for storing carbon dioxide. In a survey conducted by the Central Union of Agricultural Producers and Forest Owners, 49% of forest owners thought that the forestation of wasteland is a tempting option for increasing carbon storages on their own land. So, locating areas fit for forestation could also have real business potential.
Satellite Data Allows Monitoring the Success of Forestation
Satellite imagery has been used for land use evaluation for a long time, but only the modern high-resolution images produce very detailed information. Machine learning algorithms can be taught to recognize even the smallest differences for example in forest health and species composition. Therefore, it is possible to follow how well the forestation is succeeding. With satellites, the development and extent of the change can be pointed and compared to the earlier information on the area or to the ability of the area to store carbon dioxide.
With modern technology, the methods for recognizing potential forestation sites are already at our disposal. The only thing left for now is to start planting.