BioSense Institute

Accurate wheat yield prediction is essential for ensuring food security and sustainable resource management under the increasing challenges of climate change. This study investigates the integration of unmanned aerial vehicle (UAV)-based multispectral imaging and machine learning (ML) techniques to improve yield forecasting in European wheat cultivars. Field experiments were conducted on 400 sub-plots with varying NPK fertilization regimes and five wheat varieties, monitored across six phenological stages during the 2023 growing season in Vojvodina, Serbia. A DJI Phantom 4 Multispectral UAV collected high-resolution imagery, from which 65 vegetation indices were computed. Using PyCaret’s automated ML framework, 25 regression algorithms were evaluated for yield prediction. Ensemble models, particularly Random Forest, Extra Trees, Gradient Boosting, and LightGBM, consistently outperformed linear and kernel-based approaches. The highest prediction accuracy was achieved with the Random Forest Regressor during full flowering (BBCH 65–69), yielding an R2 of 0.952 and an RMSE of 0.44 t/ha. Results highlight the temporal dynamics of model performance, with optimal predictions occurring during reproductive stages. The findings confirm that UAV-derived multispectral data, coupled with ensemble machine learning, provide a non-invasive, accurate, and computationally efficient method for yield forecasting. This framework has significant potential for supporting precision agriculture, enabling real-time decision-making, and enhancing the resilience of wheat production systems.

Pure SnO2 nanofibers were synthesized via an electrospinning method and subsequently calcined at 550 °C to investigate the structure–property relationship governing H2S gas sensing performance. X-Ray diffraction confirmed the formation of the crystalline rutile-type SnO2. FE-SEM and TEM methods revealed a hierarchically porous morphology with fiber diameters ranging from 70 to 160 nm. BET measurements indicated a high specific surface area of 75 m2/g, consistent with the observed porous architecture. Gas sensing measurements toward H2S revealed a pronounced response value of 25 at 200 °C with the response time of 23 s, both superior to those recorded for acetone, ethanol, and hydrogen. The enhanced sensitivity and dynamic response are attributed to the large surface area and interconnected porous network of the nanofibers, which provide the abundant active sites and facilitate efficient gas diffusion.

Two new species groups within the avidus-nigritarsis lineage of the hoverfly genus Merodon Meigen are here defined and assessed, i.e., the Merodon caudatus and Merodon ottomanus groups. Within the M. caudatus group, two species are recognised:Merodon caudatus Sack and a new species, Merodon crispotarsus sp. nov. Within the M. ottomanus group, an extensive examination of material from different collections revealed the presence of four new species, Merodon auriolus sp. nov., Merodon paeninsula sp. nov., Merodon projectus sp. nov., and Merodon rostrum sp. nov., apart from Merodon ottomanus Hurkmans. The genetic characterisation of species from the M. ottomanus group was performed through Maximum Parsimony (MP) and Maximum Likelihood (ML) analyses of the 5' end of mitochondrial COI gene sequences, and the existence of the five above-mentioned independent taxa was supported. In addition, we confirmed the validity of M. auriolus sp. nov., M. paeninsula sp. nov., and M. ottomanus by using an integrative taxonomic approach including wing shape differences. Diagnoses, keys for both species groups, and the species they consist of, as well as distribution maps for all studied species, are provided.

ABSTRACT Global food security is challenged by population growth and limited agricultural land. Intensifying existing cropland use is crucial for increasing agricultural production. Mapping cropping intensity, defined by annual crop cycles and commonly classified as single, double, or triple cropping, is vital for food production modeling. While most studies focus on regions with higher cropping intenisity, this study targets Vojvodina in northern Serbia, where double cropping is less prevalent. We investigated the impact of different vegetation indices (VIs) on detecting double cropping using machine learning (ML) and Sentinel-2 imagery with collected ground truth data over two years with contrasting weather conditions: one dry and one with above-average rainfall. Our approach improves existing methods by integrating VIs not previously examined in related studies. Alongside NDVI, indices such as CVI, VARI, and ExG significantly improved model performance. In the dry year (2022), overall accuracy reached 95.80%, with an F1-score of 91.19% for classifying double cropping, while in the wet year (2023), the accuracy was 93.56% and the F1-score 84.96%. This approach simplifies data requirements while maintaining high accuracy, making it applicable to regions with similar geographical characteristics. Additionally, this study fills a key knowledge gap on the extent and distribution of this practice in Serbia.

This study provides an assessment of BTEX compounds—benzene, toluene, ethylbenzene, and xylene isomers—in urban precipitation collected in the city of Novi Sad, Republic of Serbia, during autumn and winter 2024, analyzed by gas chromatography-mass spectrometry (GC-MS). By combining chemical analysis with meteorological observations and HYSPLIT backward trajectory modeling, the study considers the mechanisms of BTEX removal from the atmosphere via wet scavenging and highlights the role of local weather conditions and long-range atmospheric transport in pollutant concentrations. During the early observation period (September to late November), average concentrations were 0.45 µg/L benzene, 3.45 µg/L ethylbenzene, 4.0 µg/L p-xylene, 2.31 µg/L o-xylene, and 1.32 µg/L toluene. These values sharply dropped to near-zero levels in December for benzene, ethylbenzene, and xylenes, while toluene persisted at 1.12 µg/L. A pronounced toluene spike exceeding 6 µg/L on 28 November was likely driven by transboundary air mass transport from Central Europe, as confirmed by trajectory modeling. The environmental risks posed by BTEX deposition, especially from toluene and xylenes, underline the need for regulatory frameworks to include precipitation as a pathway for pollutant deposition. It should be clarified that the identified risk primarily concerns aquatic organisms, due to the potential for BTEX infiltration into surface waters and subsequent ecotoxicological impacts. Incorporating such monitoring into EU policies can improve protection of air, water, and ecosystems.