Oil spills represent a severe environmental hazard, threatening marine and coastal ecosystems, biodiversity, and socio-economic stability. Timely and accurate detection of such incidents is critical for mitigating their ecological and economic consequences. Conventional detection techniques, including manual inspection and satellite-based observation, remain limited by high operational costs, temporal delays, and restricted spatial coverage. To overcome these limitations, this study introduces a comprehensive computer vision framework that addresses two core challenges: (i) the construction of a large-scale, high-quality synthetic oil spill dataset through mask extraction and seamless blending of oil spill regions with diverse oceanic backgrounds, and (ii) the development of a fine-tuned YOLOv11m-seg detection model trained on this enriched dataset. To further enhance the realism and spatial distinctiveness of oil spill textures, the Line Integral Convolution (LIC) is applied to estimate and visualize ocean surface flow patterns, generating coherent streamline textures that simulate the natural diffusion and transport of oil in water. The model exhibited strong generalization and precision, achieving a training accuracy exceeding IoU@0.50-0.95 to 85% over 50 epochs. Evaluation metrics confirmed its reliability, with an F1 score of 94%, precision of 94%, and recall (mAP@0.50) of 94%. These results demonstrate that the developed approach not only enhances dataset diversity but also substantially improves the accuracy and representativeness of real-time oil spill detection in marine environments.

We study the qualitative behaviour of weak solutions to a doubly nonlinear cross-diffusion system in an inhomogeneous medium with nonlinear boundary flux. The main novelty of the paper lies in the analysis of a cross-diffusive p-Laplacian system weighted by a spatially varying density in the form $\rho \left(x\right)={(1+|x\left|\right)}^n$, combined with nonlinear boundary interactions. By constructing self-similar weak solutions of Barenblatt type and employing a nonlinear separation of variables, the system is reduced to a coupled family of ordinary differential equations that characterise admissible similarity profiles. This approach allows us to identify Fujita-type critical conditions separating global solutions from finite-time blow-up and to derive explicit estimates for the solution profiles. Consequently, we establish critical thresholds driven by the interaction between cross-diffusion, degeneracy, boundary nonlinearity, and medium inhomogeneity.

A nonlinear mathematical model is developed for unsteady groundwater flow in a three-layer heterogeneous aquifer system, comprising a confined aquifer, a covering layer, and a weakly permeable barrier. The model incorporates infiltration and evaporation governed by the M.M. Krylov–S.F. Averyanov law, where evaporation intensity depends on a critical groundwater level. Governing equations are nondimensionalized and solved using the alternating direction implicit (ADI) method with quasi-linearization to treat nonlinearities. Periodic variations in precipitation and evaporation are considered, alongside variable boundary permeabilities. The approach enables realistic simulation of multi-layer aquifer dynamics under diverse climatic and hydrogeological conditions, offering a robust tool for sustainable groundwater management, drought risk assessment, and aquifer protection strategies.

Automatic classification of pollen grains in microscopic images supports melissopalynology and honey quality control but is complicated by class imbalance and morphologically similar classes. We evaluated factors affecting classification on the Cretan Pollen Dataset v1 (CPD-1; 20 classes, 4025 images) with YOLOv8s-cls, YOLO11s-cls, YOLO26s-cls, EfficientNet-B3 and a reproduced RCANet. Each configuration was trained with five seeds; hyperparameters and ensembles were selected on validation data only. Augmentation improved the macro F1-score by about 1.2 percentage points; a size-matched unbalanced control showed that this gain came from the larger training set, not from class balancing. The most difficult class, Ceratonia, was misclassified mainly as Vitis and Olea; its F1-score fell from 0.878 to 0.778 when its original training images were reduced from 26 to 10. Focal Loss and Squeeze-and-Excitation blocks did not improve Ceratonia reproducibly (changes in its F1-score from −0.03 to +0.03, not significant after Holm correction), and the YOLO architectures could not be ranked reliably. An ensemble of EfficientNet-B3 and YOLO models selected on validation data outperformed the best single model in all seeds (macro F1-score 0.9720 vs. 0.9631). Several single-run effects were not reproduced with multiple seeds, underlining the need for multi-seed evaluation and validation-only selection.