Predicting wildfire spread behavior is an extremely important task for many countries. On a small scale, it is possible to ensure constant monitoring of the natural landscape through ground means. However, on the scale of large countries, this becomes practically impossible due to remote and vast forest territories. The most promising source of data in this case that can provide global monitoring is remote sensing data. Currently, the main challenge is the development of an effective pipeline that combines geospatial data collection and the application of advanced machine learning algorithms. Most approaches focus on short-term fire spreading prediction and utilize data from unmanned aerial vehicles (UAVs) for this purpose. In this study, we address the challenge of predicting fire spread on a large scale and consider a forecasting horizon ranging from 1 to 5 days. We train a neural network model based on the MA-Net architecture to predict wildfire spread based on environmental and climate data, taking into account spatial distribution features. Estimating the importance of features is another critical issue in fire behavior prediction, so we analyze their contribution to the model’s results. According to the experimental results, the most significant features are wind direction and land cover parameters. The F1-score for the predicted burned area varies from 0.64 to 0.68 depending on the day of prediction (from 1 to 5 days). The study was conducted in northern Russian regions and shows promise for further transfer and adaptation to other regions. This geospatial data-based artificial intelligence (AI) approach can be beneficial for supporting emergency systems and facilitating rapid decision-making.

The canopy height model (CHM) is a representation of the height of the top of vegetation from the surrounding ground level. It is crucial for the extraction of various forest characteristics, for instance, timber stock estimations and forest growth measurements. There are different ways of obtaining the vegetation height, such as through ground-based observations or the interpretation of remote sensing images. The severe downside of field measurement is its cost and acquisition difficulty. Therefore, utilizing remote sensing data is, in many cases, preferable. The enormous advances in computer vision during the previous decades have provided various methods of satellite imagery analysis. In this work, we developed the canopy height evaluation workflow using only RGB and NIR (near-infrared) bands of a very high spatial resolution (investigated on WorldView-2 satellite bands). Leveraging typical data from airplane-based LiDAR (Light Detection and Ranging), we trained a deep neural network to predict the vegetation height. The provided approach is less expensive than the commonly used drone measurements, and the predictions have a higher spatial resolution (less than 5 m) than the vast majority of studies using satellite data (usually more than 30 m). The experiments, which were conducted in Russian boreal forests, demonstrated a strong correlation between the prediction and LiDAR-derived measurements. Moreover, we tested the generated CHM as a supplementary feature in the species classification task. Among different input data combinations and training approaches, we achieved the mean absolute error equal to 2.4 m using U-Net with Inception-ResNet-v2 encoder, high-resolution RGB image, near-infrared band, and ArcticDEM. The obtained results show promising opportunities for advanced forestry analysis and management. We also developed the easy-to-use open-access solution for solving these tasks based on the approaches discussed in the study cloud-free composite orthophotomap provided by Mapbox via tile-based map service.

Remote sensing of forests is a powerful tool for monitoring the biodiversity of ecosystems, maintaining general planning, and accounting for resources. Various sensors bring together heterogeneous data, and advanced machine learning methods enable their automatic handling in wide territories. Key forest properties usually under consideration in environmental studies include dominant species, tree age, height, basal area and timber stock. Being proxies of stand productivity, they can be utilized for forest carbon stock estimation to analyze forests’ status and proper climate change mitigation measures on a global scale. In this study, we aim to develop an effective machine learning-based pipeline for automatic carbon stock estimation using solely freely available and regularly updated satellite observations. We employed multispectral Sentinel-2 remote sensing data to predict forest structure characteristics and produce their detailed spatial maps. Using the Extreme Gradient Boosting (XGBoost) algorithm in classification and regression settings and management-level inventory data as reference measurements, we achieved quality of predictions of species equal to 0.75 according to the F1-score, and for stand age, height, and basal area, we achieved an accuracy of 0.75, 0.58 and 0.56, respectively, according to the R². We focused on the growing stock volume as the main proxy to estimate forest carbon stocks on the example of the stem pool. We explored two approaches: a direct approach and a hierarchical approach. The direct approach leverages the remote sensing data to create the target maps, and the hierarchical approach calculates the target forest properties using predicted inventory characteristics and conversion equations. We estimated stem carbon stock based on the same approach: from Earth observation imagery directly and using biomass and conversion factors developed for the northern regions. Thus, our study proposes an end-to-end solution for carbon stock estimations based on the complexation of inventory data at the forest stand level, Earth observation imagery, machine learning predictions and conversion equations for the region. The presented approach enables more robust and accurate large-scale assessments using limited annotated datasets.

Estimation of terrestrial carbon balance is one of the key tasks in the understanding and prognosis of climate change impacts and the development of tools and policies according to carbon mitigation and adaptation strategies. Forest ecosystems are one of the major pools of carbon stocks affected by controversial processes influencing carbon stability. Therefore, monitoring forest ecosystems is a key to proper inventory management of resources and planning their sustainable use. In this survey, we discuss which computer vision techniques are applicable to the most important aspects of forest management actions, considering the wide availability of remote sensing (RS) data of different resolutions based both on satellite and unmanned aerial vehicle (UAV) observations. Our analysis applies to the most occurring tasks such as estimation of forest areas, tree species classification, and estimation of forest resources. Through the survey, we also provide a necessary technical background with a description of suitable data sources, algorithms’ descriptions, and corresponding metrics for their evaluation. The implementation of the provided techniques into routine workflows is a significant step toward the development of systems of continuous actualization of forest data, including real-time monitoring. It is crucial for diverse purposes on both local and global scales. Among the most important are the implementation of improved forest management strategies and actions, carbon offset projects, and enhancement of the prediction accuracy of system changes under different land-use and climate scenarios.

Usage of multispectral satellite imaging data opens vast possibilities for monitoring and quantitatively assessing properties or objects of interest on a global scale. Machine learning and computer vision (CV) approaches show themselves as promising tools for automatizing satellite image analysis. However, there are limitations in using CV for satellite data. Mainly, the crucial one is the amount of data available for model training. This paper presents a novel image augmentation approach called MixChannel that helps to address this limitation and improve the accuracy of solving segmentation and classification tasks with multispectral satellite images. The core idea is to utilize the fact that there is usually more than one image for each location in remote sensing tasks, and this extra data can be mixed to achieve the more robust performance of the trained models. The proposed approach substitutes some channels of the original training image with channels from other images of the exact location to mix auxiliary data. This augmentation technique preserves the spatial features of the original image and adds natural color variability with some probability. We also show an efficient algorithm to tune channel substitution probabilities. We report that the MixChannel image augmentation method provides a noticeable increase in performance of all the considered models in the studied forest types classification problem.

Among different forest inventory problems, one of the most basic is defining dominant species. This data is crucial in forest management to determine forest category, and a cheaper remote sensing-based approach would be a useful supplement to field surveys. We used WorldView multispectral satellite imagery to address this problem as an image segmentation task dividing the image into regions with particular dominant species. Neural networks have recently become one of the most useful tools for this kind of problem, including incomplete or erroneous training labels. However, it is still challenging to distinguish between such similar patterns as different forest compositions. To handle this, we represented the multi-class forest classification problem as a hierarchical set of binary classification tasks, which allowed us to reach better results with both high- and medium-resolution satellite imagery. We also examined supplementary data such as tree height to improve the species classification results for wider tree age diversity. We conducted experiments considering six neural network architectures to find the best one for each task in the hierarchical decomposition. The proposed approach was tested on sample territories in Leningrad Oblast of Russia, for which the field-based observations were acquired and made publicly available as a single dataset. The proposed approach showed significantly better results (average F1-score 0.84) than multi-class classification (average F1-score 0.7).