International Journal of Environmental Science and Green Technology, ISSN: 3080-8693

Machine Learning Models in Ecological Data Analysis: A Comparative Review

2026-05-21T12:43:15+00:00

This article provides a comparative review of the effectiveness of machine learning models used in environmental data analysis. The aim of the study is to systematize the areas of application, advantages, limitations, and effectiveness of Linear Regression, Random Forest, Support Vector Machine, Neural Networks, and Deep Learning models in working with environmental data. Since environmental data are often multidimensional, nonlinear, spatial, and temporal, traditional statistical methods do not provide sufficient results in all cases. The results of the literature analysis prove that ensemble models such as Random Forest and XGBoost show high accuracy in many environmental forecasting tasks. While Deep Learning models are effective in analyzing complex data such as satellite imagery, biodiversity, animal movements, and time series, they require large data sets and high computational resources. In addition, the explainability of models remains an important issue. Explainable AI methods such as SHAP and LIME allow us to explain the decision-making logic of complex models. Research results show that model selection should consider explainability, data quality, computational efficiency, and ecological relevance in addition to accuracy.

Assessment of the Impact of Activated Carbon on Combustion Processes from the Standpoint of Environmental Safety of Model Condensed Systems

2026-05-21T13:23:54+00:00

This study presents an experimental investigation of the influence of activated carbon on the combustion kinetics of model condensed systems. Experiments were conducted on a laboratory-scale setup with controlled temperature and atmosphere, and all measurements were repeated to ensure statistical reliability (variance <5%). Varying the activated carbon content (0–15 wt%) led to an increase in the induction period from 15.2 s to 27.5 s, a decrease in the reaction front velocity from 12.5 mm/s to 7.8 mm/s, a reduction in peak temperature from 850 °C to 790 °C, and a decrease in total heat release from 520 kJ to 400 kJ. Kinetic behavior was modeled using second-order polynomial regression, with determination coefficients R² > 0.95, confirming excellent predictive accuracy. SEM-EDS analysis revealed a highly developed porous structure with a specific surface area of 446.8 m²/g, facilitating adsorption of reactive intermediates and stabilization of thermal effects. Gas chromatography identified major gaseous products: CO (51.5 %), CO₂ (38.7 %), H₂ (4.3 %), and N₂ (5.5 %), indicating active carbon gasification. Comprehensive data analysis, including thermogravimetry, differential scanning calorimetry, reaction-rate profiling, and cumulative enthalpy calculations, demonstrated that activated carbon effectively moderates the reaction front, controls heat release, and improves reproducibility of combustion.

Development of an Organo-Mineral Fertiliser Based on Glauconite and Vermicompost

2026-05-12T09:57:17+00:00

This study presents the development of an organo-mineral fertiliser based on vermicompost and the natural aluminosilicate mineral glauconite, as well as the evaluation of its agrochemical properties. It was shown that vermicompost enhances soil biological activity and, through organic compounds and microbiological processes, accelerates the transformation of mineral components. Glauconite acts as a source of potassium and trace elements with prolonged release and, due to its high cation-exchange capacity, ensures their retention and gradual availability. Various component ratios (4:6-7:3) were investigated, and the optimal ratio was determined to be 6:4 (glauconite:vermicompost), providing a balanced proportion of organic and mineral constituents. The proposed production technology includes grinding the mineral raw material to a fraction of 0.1-1.0 mm, mixing the components, granulation with particle formation of 5-10 mm, and subsequent mild drying at 40-60 °C. This technological scheme ensures controlled and prolonged release of nutrients. It was established that the developed organo-mineral system improves soil moisture retention capacity, reduces nutrient losses due to leaching, and stabilises the plant's nutrient supply regime.