Conference Agenda

The use of unmanned aerial vehicles (UAVs) for pesticide application is increasing due to their operational efficiency. However, higher spraying altitudes can intensify pesticide drift, leading to potential environmental exposure and human health risks. This study developed a pesticide drift and deposition prediction model based on LC50 toxicity criteria, along with a graphical user interface (CUI) to support safer operational decision-making. A total of 80 field measurements collected during 2022–2024 were used to developed a multiple linear regression (MLR) model, including main meteorological, crop, and application variables. The model predicts drift magnitude and affected according to pesticide toxicity level. The developed model achieved an R² of 0.73 for drift prediction and an R² of 0.64 for deposition prediction. In addition, human exposure risks were assessed for UAV operators, residents, and bystanders. Estimated exposure levels were compared with predictions from international human exposure models (EFSA, BREAM2, EUROPOEM, and UK-POEM) and with scenarios involving wide-area ground spraying. These models, originally designed for conventional ground-based systems, both overestimation and underestimation when applied to UAV spraying, indicating limitations in their applicability. Overall, the proposed UAV-specific model provides a useful tool for pre-spray risk management and post-spray exposure assessment, supporting safer, informed pesticide application practices.

The structural parameters of a liquid sheet are key determinants of atomization performance, making their measurement a crucial aspect of agrochemical spray research. This study introduces an approach that combines deep learning with high-speed imaging to measure the structural parameters of liquid sheets produced by hydraulic nozzles operating in different atomization modes. A recognition model, LM-YOLO, was developed to identify the liquid sheet and its perforations. From these recognition results, a methodology was established to compute key parameters such as breakup length, sheet area, average number of perforations, and average perforation area. Prediction models for the droplet size of volume median diameter was developed based on liquid film structural parameters. The constructed model achieved a recognition accuracy of 81.0% for the liquid sheet structure of an conventional flat-fan LU nozzle and 71.3% for an IDK nozzle . And the droplet size prediction model achieved an R² value exceeding 0.8, enabling effective prediction of the atomization size parameters. The addition of a silicone adjuvant generally reduced both the breakup length and area of the liquid film. This work presents a novel method for quantifying liquid sheet structural parameters, demonstrating potential for application in precision spraying.

Spray deposition and coverage within soybean canopies remain critical challenges for achieving effective pesticide applications. This research investigated the influence of wind speed, boom travel direction, and nozzle type on droplet deposition and coverage, and airflow distributions inside soybean canopies under controlled wind-tunnel airflow. Spray deposition, analyzed using a fluorometric tracer, and coverage, quantified with water-sensitive papers, were assessed in R3-stage soybeans using XR and 3D flat fan nozzles under varying air speeds and boom travel directions. Droplet sizes were measured using a laser imaging particle sizing system. Airflow velocity and turbulence within the soybean canopy were investigated with a 3-D hot-film anemometer system. The results indicated that wind and boom direction were the main influential factors for spray coverage and deposition. The top canopy position, exposed to the highest air-turbulence intensity, received the greatest deposition, whereas the middle and bottom positions, characterized by lower turbulence, exhibited sharp declines in both deposition and coverage. The 3D nozzle provided greater coverage and deposition than the XR only under no-wind conditions. Therefore, it was essential to incorporate wind conditions and canopy structures into consideration when choosing nozzles to maximize spray penetration and achieve efficient and effective spray applications for soybeans. Funding: CNPq/Brazil.

Pulse width modulation (PWM) nozzles are claimed to improve orchard spraying by precisely controlling droplet size and flow rate, resulting in more uniform coverage, reduced chemical waste, and minimized off-target drift. In this work, Computational fluid dynamics (CFD) simulations were employed to study the dynamic orchard spraying process with the use of PWM nozzles. Model validation of the predicted on-target deposition from CFD was previously performed with dedicated orchard field trials with a variety of spraying techniques and training systems. To include canopy effects, orchards were digitized by means of 3D scanning using LiDAR synchronized with RTK-GPS, to create geometrical models of trees and leaf cover with varying leaf area density. A comparative study of canopy and ground distribution of the spray liquid was conducted for drift-reducing flat fan nozzles (IDK) under different PWM duty cycle configurations across the height, implemented on a crossflow spout sprayer (H.S.S., NL). The CFD model incorporated stochastic spray–canopy interactions due to varying leaf density and properties, further influenced by droplet velocity and size. A parametric study was performed to reveal the effect of changing droplet size distribution affected by pressure and duty cycle. Deposition was analyzed across different vertical zones of the trees.

Spray application is one of the most important methods for managing diseases and infections in field crops. However, spray application can also create undesirable side effects for humans and ecosystems, making efficient pesticide use essential. Numerous parameters influence pesticide deposition and coverage on plant surfaces, yet studying each factor individually in field trials is costly and nearly impossible. This study uses computational fluid dynamics (CFD) to investigate how droplet size distribution and spray angle from a dual‑fan nozzle affect spray deposition on winter wheat. High-resolution 3D scans of winter wheat canopies at mid and late growth stages serve as the basis for the simulations, allowing a detailed interaction simulation of spray flow in complex plant structures. Results show that finer droplets deposit more effectively on upper canopy layers, while coarser droplets—due to greater momentum—penetrate deeper and reach the lower parts of the plant and the ground. For the same spray volume, finer droplet spectra generate a higher number of droplets, resulting in improved surface coverage. The spray angle also influences deposition uniformity, and the combined selection of droplet spectrum and spray angle plays a key role in identifying the most efficient application strategy for winter wheat.

The use of unmanned aircraft systems (UAS) for the application of agricultural inputs has expanded rapidly in recent years; however, their use with plant protec-tion products remains highly limited in EU Member States due to their classifica-tion as aerial spraying under Directive 2009/128/EC. There is a clear need to ad-vance in establishing scientific evidence regarding the use of these systems, par-ticularly with respect to technical constraints, drift risk, and environmental safety.
A critical knowledge gap concerns the rotor-induced downwash generated by multirotor agricultural drones, which directly influences droplet transport, canopy penetration, spray deposition, and drift behavior.
This study evaluated the performance of the multirotor DJI Agras T50 by charac-terizing the airflow generated by its rotors and analyzing its influence on horizon-tal spray distribution under different flight configurations. Experimental trials were conducted to measure air velocity vectors under dynamic conditions. Spray distribution tests were performed following ISO 23117-2:2025 using water-sensitive papers.
Results showed that increasing forward speed significantly reduced airflow mag-nitude and negatively affected spray distribution uniformity. Lower flight speeds combined with reduced flight heights produced more stable spray profiles. The coefficient of variation analysis enabled determination of the effective working width for each configuration, providing practical guidance for operational adjust-ment and performance optimization.