Methods, algorithms, and models for the creation and practical application of digital twins (3D models) of agricultural crops are presented, illustrating their condition under different levels of atmospheric CO₂ concentration, soil, and meteorological conditions. An algorithm for digital phenotyping using machine learning methods with the U²-Net architecture are proposed for segmenting plants into elements and assessing their condition. To obtain a dataset and conduct verification experiments, a prototype of a software and hardware complex has been developed that implements the process of cultivation and digital phenotyping without disturbing the microclimate inside the chamber and eliminating the subjectivity of measurements. In order to identify new data and confirm the data published in open scientific sources on the effects of CO₂ on crop growth and development, plants (ten species) were grown at different CO₂ concentrations (0.015–0.03% and 0.07–0.09%) with a 10-fold repetition. A model has been built and trained to distinguish between cases when plant segments need to be combined because they belong to the same leaf (p-value = 0.05), and when they belong to a separate leaf (p-value = 0.03). A knowledge base has been formed, including: 790 3D models of plants and data on their physiological characteristics.

   Purpose of research.  The Russian Federation and the Republic of Kazakhstan are member countries of the Kyoto Protocol, an international agreement aimed at reducing greenhouse gas emissions in order to combat climate change. This requires the formation of a model bank for automated management of the allocation of quotas for greenhouse gas emissions into the atmosphere.    Methods. Based on the analysis of carbon accounting features, a new trading methodology based on Blockchain technology and the VCG algorithm is proposed, aimed at increasing transparency, security and automation of transactions between market participants. A method and algorithm for implementing carbon emission quotas through the auction system has been developed, which ensures the openness and transparency of the distribution process.    Results. In this paper, we have modeled a digital platform for automated carbon emissions trading using Blockchain technology. A study of the existing tools and mechanisms of the carbon emissions quota market has been carried out, its main problems have been identified, such as insufficient transparency of transactions, the risk of fraud and difficulties in monitoring compliance with regulatory requirements. As part of the simulation, smart contracts were developed and tested, which automate the fulfillment of transaction conditions and transaction accounting. The use of Blockchain technology makes it possible to ensure high information security, control the course of trading and prevent unauthorized data changes. An automated system for accounting and conducting transactions increases the efficiency of quota management and reduces transaction costs.    Conclusion.  The developed prototype of the digital platform promotes the development of an environmentally responsible economy, encourages enterprises to reduce their carbon footprint, and ensures transparency and security in trading environmental quotas. As a result, the proposed model increases the confidence of market participants and contributes to the formation of a sustainable system for regulating greenhouse gas emissions.

Purpose of research: formation of a bank of models for the implementation of simulation experiments to assess and predict the values of greenhouse gas concentrations in the surface layer of the territory's atmosphere based on artificial neural networks and GIS technologies. The influence of the increased concentration of carbon dioxide in the surface layer of the atmosphere on the growth and development of agricultural plants, namely on the change in photosynthetic activity, the level of mineralization of the humus layer of the soil, which affects crop yields, is considered. The peculiarities of agricultural production determine the relevance of the creation and implementation of a new intelligent technology that will provide the opportunity to identify optimal parameters of crop production.Methods. The formation of a training sample for a neural network was carried out through numerical experiments and mathematical modeling methods. To select the best neural network topology for predicting the concentration of greenhouse gases in the territory under consideration, experiments were conducted that revealed the standard deviation and relative error. To assess the predictive abilities of the models, field experiments were conducted to measure CO2 concentrations in the surface layer of the atmosphere in the agricultural territories of the Belgorod region.Results. A software toolkit has been developed that makes it possible to visualize the dispersion and accumulation of greenhouse gases in the surface layer of the atmosphere. This makes it possible to conduct simulation experiments necessary to determine the territories that are under the influence of man-made sources. An assessment of the photosynthetic activity of plants in the selected territory was carried out, which allows us to form further recommendations for the effective use of agricultural territory aimed at increasing crop yields.Conclusion. The paradigms of neural networks were considered, experiments were conducted to identify the best topology. A software toolkit has been developed to visualize the dispersion and accumulation of greenhouse gases in the surface layer of the atmosphere for decision makers. The effects of technogenic factors on the photosynthetic apparatus of agricultural plants are analyzed, on the basis of which conclusions and practical recommendations for the cultivation of agricultural crops are formulated.

This paper presents a set of algorithms and models for the comprehensive assessment of the environmental situation in urbanised areas subject to anthropogenic impacts from industrial and transport facilities. The proposed mechanism combining fuzzy logic and neural network predictive models, with subsequent spatial visualization of the results using GIS technologies, enabling a shift from traditional component-by-component assessments to a holistic spatio-temporal characterisation of the environmental situation, with the possibility of formulating well-founded management measures. An algorithm for generating management scenarios has been developed, ensuring a closed-loop cycle: monitoring—integrated assessment—forecasting—development of environmental protection measures. The algorithm was tested using field environmental monitoring data from urban areas, collected at 82 road sections (2023–2025), for which motor vehicle traffic was identified as the primary source of anthropogenic pressure. Forty-three mathematical models were developed based on an artificial neural network, which achieved a coefficient of determination of 0.75–0.82 on a test data sample. Spatial visualisation of the results was implemented in the ArcMap 10.8 GIS system, whilst modelling was carried out in the MatLab environment. The objective of this study is to develop a set of tools for a comprehensive assessment of the environmental conditions in urbanized areas. The proposed set of tools can be expanded to include other components, however, testing is currently limited to two components (CO and noise) in a single city (Oryol, Russia).