Evaluation of Air-Assisted Spraying Technology for Pesticide Drift Reduction

Abstract

Reducing pesticide drift is essential for minimizing environmental pollution and so implementing sustainable agricultural practices. Various methods can be employed to counteract this phenomenon, one of which is air-assisted spraying. This study evaluates the impact of air assistance on spray deposition using a field sprayer designed for cereal crops. Spray quality was assessed using water-sensitive paper, allowing drift levels to be estimated not only by assessing coverage but also by measuring the average number of droplets per square centimeter on the paper surface. This method provided a more reliable evaluation of drift reduction. The results indicate that air assistance reduced drift by 40.74% in coverage and 37.55% in droplet density per square centimeter. Notably, the greatest drift reduction occurred at the point of highest coverage—1 m from the spraying area—with an average reduction of 50%. Furthermore, at this distance, an 80% drift reduction was achieved at the recommended sprayer speed of 6 km/h. These findings highlight the potential of air-assisted spraying to enhance efficiency while reducing environmental impact. This research aligns with the objectives of the European Green Deal, supporting the transition toward more sustainable pesticide application techniques and contributing to environmentally responsible crop protection practices.

1. Introduction

The application of pesticides in agriculture is essential to protect crops from pests and diseases and ensure high yields. However, one of the main challenges associated with pesticide use is spray drift, i.e., the unintended movement of sprayed droplets beyond the target area. Spray drift occurs when fine droplets are carried away by wind or air currents, leading to off-target deposition. In addition to reducing the effectiveness of pesticide application [1,2,3], it can result in the contamination of nearby ecosystems, water bodies, and non-target crops, posing environmental and public health concerns [4,5,6,7].

In response to these risks, the European Union (EU) has introduced several regulatory measures under the European Green Deal [8,9]. This framework, aimed at making the EU climate-neutral by 2050, emphasizes sustainable agricultural practices. A key component is the “Farm to Fork” strategy, which seeks to reduce the use and risk of chemical pesticides by 50% by 2030 [10,11,12]. The Biodiversity Strategy for 2030 also highlights the importance of minimizing the harmful effects of pesticides on ecosystems [13,14,15]. These policies reflect the EU’s growing concern about the environmental impact of pesticide drift.

Reducing pesticide drift ensures that more of the pesticide reaches the target area, increasing its efficiency and reducing the need for excessive chemical use. This aligns with the goal of sustainable agriculture to optimize resource use and minimize waste. Additionally, it helps protect the environment, preserve biodiversity, and as it also improves the efficiency of pesticide use, all of which contribute to more sustainable and responsible agricultural practices.

Spray drift is influenced by many factors, including weather conditions, nozzle type, droplet size, application technique, and field topography [16,17,18]. Literature shows that smaller droplets are more prone to drift, while larger droplets reduce the risk of off-target movement [1,19]. Agriculture has advanced technologies such as low-drift nozzles, air-assisted sprayers, and drift-reducing adjuvants to mitigate drift [17,20,21]. To further minimize spray drift consequences, regulatory frameworks also recommend buffer zones between treated fields and sensitive areas such as water bodies, residential zones, and conservation sites [22,23].

All these measures align with the European Green Deal’s objective of promoting environmentally friendly agricultural practices. It should also be noted that standardization measures like ISO 22866:2005, which outlines methods for measuring spray drift, help ensure safe pesticide application across EU member states [24].

Considering the above, it is clear that one of the most significant challenges in modern agriculture is balancing the need for plant protection chemicals with their environmental impact. Reducing pesticide drift is crucial to minimizing pollution, especially concerning surface water contamination [5,6,7]. Besides environmental concerns, pesticide drift also has agronomic consequences, as uneven crop coverage with the working liquid can lead to overdosed or underdosed areas, directly affecting yield quality and quantity [25,26]. Thus, evaluating spray quality can be closely linked to the degree of drift reduction. Factors such as sprayer boom stability [27,28], spraying height [29,30], nozzle specifications [31,32], and atmospheric conditions [33,34] all play critical roles. However, air-assisted spraying is particularly effective in reducing drift, making it a key approach [1,16,35].

Based on the above, the primary goal of this research was to compare drift levels when using a sprayer boom with and without air assistance.

While most studies on this subject focus on horticultural applications [36,37,38], our study used a field sprayer designed for cereal crop spraying. Additionally, we evaluated drift reduction not only quantitatively, by assessing the degree of liquid coverage, but also qualitatively, by measuring the number of droplets (nd) per unit area [39].

Both these aspects distinguish our research from previous studies and provide further insights into improving spray quality, contributing to the development of more sustainable agricultural practices that support the EU’s environmental objectives.

2. Materials and Methods

The methodology for measuring spray liquid drift in field sprayer tests is not strictly standardized. However, the literature on the subject indicates some basic principles that should be followed to ensure reliable measurements and comparable results across different experiments.

The most important source is the meta-report by Van de Zande et al. [40], which summarizes a large number of field studies (516 experiments, including 248 studies without plants, i.e., performed similarly to the experiment presented in this article) conducted in Germany and the Netherlands. This work outlines the methodology for research on spray drift in a consistent and reproducible manner, which is why we decided to base our paper on it.

2.1. Methods Used to Measure and Evaluate Spray Liquid Drift

The field tests presented in our study were conducted in strict accordance with the specified methodology. Prior to each trial, the prototype sprayer boom was positioned at a height of 0.5 m above the soil surface; this height was measured and adjusted as needed before every run. In accordance with ISO 22866 [24] and ISO 22369-2 standards [41], a reference spray application is defined as one performed with a boom height of 0.50 m above bare soil or crop canopy. This standard height is widely accepted in both scientific research and field applications, supporting its use in our experiment [40].

Drift measurements were performed until the last measurement point, where spray coverage reached zero (in this experiment, this occurred at distances of 20 or 30 m).

The driving speed of the aggregate was set to 6, 8, 10, and 12 km/h. According to the above-cited report, the key speeds are in the range of 6 to 8 km/h, which were of course considered in our experiment. Each pass was performed twice—once with air assistance activated, and once without.

Based on the above, it can be concluded that the experiment was performed in full compliance with the current standards for measuring working fluid drift in field tests.

As for the measurement of spray liquid applications, they are generally performed in two ways, i.e., using a fluorescent tracer or water-sensitive paper (WSP) [42,43,44,45]. The literature shows that both methods allow for a reliable estimation of the spraying liquid [46], and so both are widely used in field and laboratory experiments. In our experiment, water-sensitive paper was utilized, enabling the estimation of drift levels not only by assessing the degree of coverage but also by measuring the average number of drops per square centimeter of the sprayed area. This qualitative approach provided additional information on the quality of spraying [39].

The water-sensitive paper used in the presented experiment was manufactured by Syngenta and measures 76 mm in length and 26 mm in width, resulting in an area of 1976 square mm. It is important to note that these dimensions are the standard size for water-sensitive paper, regardless of the manufacturer.

2.2. Plot Layout

Figure 1 shows a scheme of the arrangement of samplers (i.e., water-sensitive papers) used in the experiment.

As can be seen, the sprayer working width was 21 m, while water-sensitive papers were arranged in three lines (A, B, and C) to obtain three measurements of the spraying liquid deposition for each run of the aggregate (i.e., three repetitions—this approach ensures that each “repetition” was performed for the same conditions).

Water-sensitive papers were mounted on wooden stakes at a height of 0.15 m above the ground. This height was selected to keep the papers as close as possible to the surface while minimizing the risk of accidental disturbance or contamination. It should be noted that the papers were placed outside the intended spray zone to capture drift rather than target deposition. Therefore, their placement was not related to crop canopy structure or plant protection requirements.

The papers were placed at distances of 1, 3, 5, 10, 20, and 30 m from the end of the sprayer boom. This choice of distances proved to be correct, as no traces were recorded on probes placed at 30 m in any of the passes.

2.3. Equipment

The experimental setup included a field sprayer unit consisting of a Claas AXOS 330 tractor (CLAAS Group, Harsewinkel, Germany) and a detachable field sprayer with an air-assisted system. The prototype of the field sprayer was developed by AGROLA Zdzisław Niegowski (Płatkownica, Poland) as part of the research project titled “A family of field sprayers with an air-assisting stream”.

The unit featured a 3 m³ sprayer tank and a hydraulic sprayer boom with a working width of 21 m, which folds on both sides during transport. The sprayer unit was equipped with two wheels with 270/95R42 tires [28]. Spraying was performed using standard flat fan nozzles (Lechler 110-03, Lechler, Metzingen, Germany), and the working pressure of the spray liquid was set at 0.3 MPa (pure water was used as working fluid).

Spraying was carried out using standard flat fan nozzles (Lechler 110-03), operating at a pressure of 0.3 MPa, with pure water used as the working fluid. According to prior research on the same air-assisted system [29], the airflow velocity measured 10 cm from the outlet was 19.36 ± 6.45 m/s. Each diffuser could be configured to operate with either one or two spray nozzles. In the single-nozzle configuration, the nozzle was mounted directly in line with the diffuser’s axis of symmetry. In the dual-nozzle setup, the nozzles were positioned symmetrically—each placed 9 cm to the left and right of the diffuser’s axis. Additionally, both nozzles were shifted 50 mm forward relative to the diffuser outlet. The spray cones from both nozzles were set parallel to each other to ensure uniform coverage.

2.4. Description and Conditions of the Experiment

The research was conducted in November 2021 on the grounds of the University of Warmia and Mazury in Olsztyn. As already mentioned, the research was conducted on a flat field with natural grass growth, without any cultivated crops. An exemplary photo from the experiment is shown in Figure 2.

The field tests were conducted on a uniform grassland surface rather than in a cereal crop field. As such, parameters commonly associated with crop canopy structure—such as canopy height or density—were not applicable in this context and were therefore not considered in the experimental design.

The experiment was conducted under appropriate spraying conditions [16], specifically at registered wind speeds ranging from 0.50 to 2.01 m/s, temperatures ranging from 7.8 to 8.9 °C, and an air humidity of 77%. Weather parameters were measured directly at the test site using a Testo 410 anemometer (Testo SE & Co. KGaA, Schwarzwald, Germany). This device features a measurement range of 0.4 to 20 m s⁻¹ for air velocity (accuracy: ±0.2 m/s + 2% of measured value, resolution: 0.1 m s⁻¹), –10 to +50 °C for temperature (accuracy: ±0.5 °C, resolution: 0.1 °C), and 0 to 100% RH for humidity (accuracy: ±2.5% RH in the range 5–95% RH, resolution: 0.1% RH). Wind direction was also observed, and spraying was only conducted when the wind blew approximately perpendicular to the sprayer boom (within ±15° deviation), ensuring consistent drift conditions. During each trial, temperature and humidity were measured before and during spraying. Although minor fluctuations occurred, each aggregate run was performed under stable conditions.

Water-sensitive papers were placed according to the arrangement shown in Figure 1 and attached to wooden stakes, as also visible in Figure 2. The papers were collected and secured immediately after drying to prevent overestimation of the sprayed area due to moisture from other sources, particularly atmospheric humidity.

It should be noted that this study focused exclusively on the assessment of spray drift beyond the intended treatment area. As such, water-sensitive papers were placed only in zones outside the target spray coverage to capture drift data.

Measurements of droplet deposition within the crop canopy were not included in this experiment, as the tests were conducted on grassland and did not involve actual crop spraying. While canopy deposition is recognized as a key parameter in evaluating application effectiveness, it falls outside the scope of the current study.

2.5. Image Analysis

Collected water-sensitive papers were digitized using an HP Deskjet 3520 flatbed scanner at a resolution of 600 dots per inch (dpi) and saved in uncompressed bitmap (*.bmp) format to preserve image fidelity for subsequent image analysis.

In the analysis of water-sensitive paper scans, previously proven methods of image processing and analysis were used [39,47,48], ensuring the reliability of the obtained results.

All image processing and analysis operations were performed using the MATLAB R2014a. (MathWorks, Natick, MA, USA) computing environment.

The first step in the analysis of the papers was image binarization, performed using the Otsu automatic thresholding method [49,50]. This method is one of the most frequently used and cited thresholding techniques [50,51,52]. The Otsu method of automatic thresholding determines the binarization threshold by minimizing the weighted sum of within-class variances of foreground and background pixels [49,51]. Its other advantage is the fact that, due to its popularity, it is also easily implemented and is not computationally demanding. As was proven in the works by Lipiński and Lipiński [47,48], the Otsu method is optimal for binarizing water-sensitive papers.

Based on the binarized images of the water-sensitive papers, the degree of coverage (i.e., the ratio of the area covered with liquid to the area of the entire paper, expressed as a percentage) was calculated for each water-sensitive paper:

$$coverage\left[\%\right]={\displaystyle \frac{covered surface}{total area}}\left[\%\right]={\displaystyle \frac{covered surface \left[{\mathrm{m}\mathrm{m}}^2\right]}{1976 \left[{\mathrm{m}\mathrm{m}}^2\right]}}\left[\%\right],$$

(1)

as well as the average number of drops per square centimeter of the paper surface:

$$coverage quantitatively\left[{\displaystyle \frac{nd}{{\mathrm{c}\mathrm{m}}^2}}\right]={\displaystyle \frac{total drops number}{total area}}\left[{\displaystyle \frac{nd}{1979 {\mathrm{m}\mathrm{m}}^2}}\right]·{19.72}^{-1}.$$

(2)

2.6. Statistical Analysis

As previously mentioned, during each aggregate run, data was collected from three distinct locations, corresponding to three replicates. Consequently, the most detailed subset of data (i.e., divided by different speeds and distances from the end of the sprayer boom) was based on n = 3. The dataset without speed differentiation was based on n = 12, while the overall dataset, encompassing all measurements, was based on n = 60.

To assess whether differences between measurement results were statistically significant, the Student’s t-test for independent samples was used. All comparisons were made at a significance level of p = 0.05, and the term “significant” in the manuscript refers to statistical significance under this threshold. Significance levels were denoted in the figures using asterisks: * p < 0.05, ** p < 0.01, and *** p < 0.001.

Prior to conducting the t-test, the assumptions of normality and homogeneity of variances were verified using the Shapiro–Wilk test and Levene’s test, respectively.

All statistical analyses were performed using MATLAB (MathWorks, Natick, MA, USA).

3. Results

The above-described experiment allowed gathering data, the analysis of which, in turn, enabled achieving the aim of the paper, i.e., to determine the influence of the air-assisting stream on the drift level in field conditions.

3.1. Evaluation of Air-Assistance in Field Spraying—Visual Results

The first two figures, i.e., Figure 3 and Figure 4 (Figure 3—without air assistance, Figure 4—with air assistance), present exemplary scans of water-sensitive papers from one of the aggregate runs. The three lines (A, B, and C) in these figures correspond to three measurements of the spraying liquid deposition for each run of the aggregate. This should be understood as each of the lines representing a separate repetition.

Presented scans were obtained at an aggregate speed equal to 6 km/h. This speed was specially selected to demonstrate exemplary scans, because 6 km/h is the recommended and in consequence the most frequently used movement speed during field spraying operations [40].

For spraying performed with air assistance, no droplet coverage was observed on the water-sensitive papers placed 20 m from the end of the sprayer boom; therefore, scans of these WSPs are not included in Figure 4.

3.2. Quantitative Assessment of Obtained Results

The next four figures (i.e., Figure 5, Figure 6, Figure 7 and Figure 8) present the results showing the impact of air-assistance on spraying quality.

Figure 5 illustrates the average coverage for NAA (No Air Assistance) and WAA (With Air Assistance) spraying for all papers. This is expressed both quantitatively, in terms of coverage, and qualitatively, as the average number of drops per square centimeter on the water-sensitive paper surface.

Figure 6 and Figure 7 show the averaged coverage obtained with and without air assistance as a function of distance, expressed by the degree of coverage (Figure 6) and by the average number of drops per square centimeter (Figure 7).

The most detailed results are presented in Figure 8, which shows the degree of coverage as a function of distance for various aggregate movement speeds, ranging from 6 to 12 km/h.

4. Discussion

4.1. Interpretation and Conclusions in the Context of Literature

A visual comparison of Figure 3 and Figure 4 clearly demonstrates the positive impact of air assistance on spray quality, as evidenced by the reduced drift beyond the target spraying area. In all three rows of water-sensitive paper, droplet coverage was consistently lower when air assistance was used. Notably, at a distance of 20 m from the end of the sprayer boom, no coverage was observed in the air-assisted scenario (WAA), whereas in the non-air-assisted condition (NAA), coverage was still recorded (Figure 3e).

This observation is further supported by quantitative results presented in Figure 5, Figure 6, Figure 7 and Figure 8, which are confirmed by statistical analysis. Specifically, the average droplet coverage beyond the designated spraying area decreased by 40.74% under air-assisted conditions (from 12.50% to 7.41%), while the average number of droplets per cm² decreased by 37.55% (from 85.78 to 53.57 nd/cm²). The consistency between these two measures confirms the reliability of the results and validates the appropriateness of using both metrics in assessing spray drift.

These findings are particularly relevant to the design of pesticide buffer zones, which aim to protect sensitive areas such as water bodies, neighboring crops, residential zones, public facilities, and natural habitats [22,53,54]. The observed drift reduction suggests that when air-assisted sprayers are used, such buffer zones could potentially be designed with reduced width, contributing to more efficient land use and pesticide management strategies.

Both metrics—droplet coverage (%) and droplet density (nd/cm²)—also showed the expected trend of decreasing deposition with increasing distance from the end of the boom. However, air assistance consistently reduced drift across all distances and at all tested speeds. Particularly noteworthy is the fact that, under WAA conditions, no deposition was observed at 20 m from the boom end, regardless of travel speed, unlike the NAA condition. These results provide strong evidence in favor of using air assistance to improve spray application safety and precision.

A more detailed analysis of coverage patterns reveals that the two measurement methods offer complementary insights. When droplet density was used as the metric, air assistance resulted in a clear reduction in deposition at all distances. In contrast, when using percentage coverage, the most pronounced difference was observed at 1 m from the sprayer boom end—approximately a 50% reduction with WAA—while differences at further distances were less distinct. This highlights the added value of using droplet count per unit area as a sensitive and informative metric, particularly in drift assessment contexts where traditional fluorescent tracer techniques are not applicable.

Based on the combined interpretation of both coverage metrics, the following additional conclusions can be drawn:

• air assistance significantly reduces drift outside the target zone, particularly at shorter distances from the end of the sprayer boom;
• although droplet diameter was not measured directly in this study, the combined analysis of coverage percentage and droplet density indicates that beyond one meter from the end of the sprayer boom, a noticeable decrease in droplet count per cm²—despite similar coverage levels regardless of air assistance—suggests that coverage at these distances was maintained by fewer, but larger, droplets;
• this observation contrasts with previous findings [17,55], which indicate that smaller droplets are more prone to drift. Our results suggest that once outside the target zone, larger droplets may actually travel farther—possibly because smaller droplets are more susceptible to deflection and turbulence, and thus disperse in directions not aligned with the main direction of movement;
• given that smaller droplets are generally more effective in penetrating crop canopies [17,56], future research should explore how air assistance might be optimally combined with nozzle selection to simultaneously reduce drift and improve deposition within the crop.

Figure 8 further illustrates that drift levels vary with travel speed. Nevertheless, air assistance consistently reduced drift across all tested speeds—an important finding that supports the broader recommendation to employ air-assisted technology in precision spraying. For instance, at a recommended operational speed of 6 km/h, drift at 1 m from the boom end was reduced by over 80%, representing more than a sixfold decrease in deposition under WAA conditions.

An additional interesting finding from Figure 8 is that at 10 km/h, WAA coverage at 1 m was among the lowest across all test conditions. While the relative decrease (75%) was slightly smaller than at 6 km/h (84%), the effect remains substantial. In contrast, at 8 km/h, the difference between WAA and NAA coverage was less pronounced, suggesting a more neutral effect of air assistance at this intermediate speed—an aspect that warrants further investigation.

4.2. Study Limitations and Future Work

Like all field-based experiments, this study is subject to certain limitations, particularly those related to environmental variability. While weather conditions can never be fully controlled, we took extensive measures to minimize their influence. Environmental parameters—especially wind—were carefully monitored throughout the trials to ensure that they did not compromise the reliability of the results.

Another limitation is the specificity of the spraying equipment used. The prototype sprayer employed in this study was designed primarily for cereal crop applications. Therefore, while the results are directly applicable to cereal fields, their generalizability to other crop types may be limited. Nevertheless, we believe that the findings can still be cautiously extended to some other row crops that are commonly treated with similar equipment, such as certain vegetable species. However, these conclusions are not applicable to fruit orchards, which require entirely different spraying techniques and equipment [36,37,38].

It is also important to note that the tests were conducted on uncropped natural grassland in order to provide consistent and repeatable conditions. As such, the interaction between airflow and crop canopy structures was not considered. In real-world cereal fields, plant canopies can influence droplet drift by altering airflow patterns and intercepting droplets. Although our simplified setup allowed for the isolation and analysis of drift behavior, future studies should validate these findings under actual crop conditions to quantify the canopy’s effect on droplet movement and deposition [40].

In terms of future research directions, combining empirical field studies with advances in computational fluid dynamics (CFD) offers a particularly promising approach. Recent developments in CFD have significantly improved the ability to simulate and predict spray drift under diverse field conditions, ranging from the nozzle scale [57,58] to full-field applications [58,59,60]. Numerous studies have demonstrated that 3D CFD models are capable of capturing the complex interactions between airflow, droplet behavior, and environmental variability.

For example, study [57] presents a validated CFD framework that accurately simulates droplet trajectories and deposition patterns under realistic spraying scenarios. Similarly, the review article [58] provides a comprehensive overview of CFD-based methodologies, emphasizing their potential to complement—or in some cases replace—traditional wind tunnel or field experiments. In addition, study [59] focuses on the development and validation of a 3D CFD model specifically tailored for air-assisted spraying, further highlighting the applicability of such tools in practical agricultural contexts.

5. Summary and Conclusions

Understanding the environmental impact of agricultural practices is critical within the Green Deal framework, particularly regarding the efficacy and safety of field sprayer use. As the agricultural sector adopts more sustainable practices, identifying and mitigating the factors contributing to spray drift becomes essential. Doing so will improve the precision of pesticide application, reduce environmental contamination, and enhance agricultural productivity while safeguarding ecological health. It can also lead to better (more optimal) use of land by reducing pesticide buffer zones.

Based on the analysis of the experiment results, it can be concluded that the key purpose of using air-assist during spraying has been achieved. The drift level, determined by both the degree of coverage and the number of drops, is reduced by an average of about 40% compared to the spray technique without air assistance. Its use reduces drift at each speed of the aggregate and at each distance from the end of the sprayer boom. It is also important that the greatest reduction in drift was achieved at the point of greatest coverage, i.e., 1 m from the sprayer boom end—a decrease by an average of 50%, and at a speed of 6 km/h—a decrease by over 80%.

Another conclusion that can be drawn from our study is that the use of water-sensitive paper, instead of a fluorescent tracer, is a method that admittedly requires more work (scanning, image processing, and analysis), but allows for obtaining additional information on the quality of the spraying thanks to which it is possible to better understand the process being studied and, consequently, better plan spraying procedures in field practice. This approach also does not require advanced laboratory equipment and is cheaper, which increases its availability for both scientists and farmers.