Arcenty I. Bredikhin, Andrey V. Melnikov

Yugra State University, Yugra Research Institute of Information Technologies

Khanty-Mansi Autonomous Okrug-Yugra has a large area of forest territories (more than 90% of the total area of the district). And forest vegetation naturally dies sooner or later, as a result of which carbon dioxide is released into the atmosphere from organic matter and global warming increases.
In order to prevent an increase in global temperature, it is necessary to estimate the carbon stock in the form of the amount of vegetation biomass.
One of the ways to assess vegetation biomass is to create carbon maps using of the Earth remote sensing methods. The use of satellite images obtained with remote sensing methods and methods of their processing will make it possible to obtain a map of the district with full coverage of the entire territory, and the use of machine learning models will make it possible to develop a model with which it will be possible to obtain a carbon map of the district with a given accuracy.
This paper provides an overview of the decisions of foreign scientists over the past 5 years in the field of remote sensing aimed at creating carbon maps. Based on this review, a research program has been proposed that will allow us to develop an approach that allows us to obtain a digital carbon map of the KhMAO with the necessary accuracy.

remote sensing, satellite image, carbon, vegetation biomass, machine learning