Conference PaperPDF Available

A combined experimental and modelling study of dust drift during wheat sowing

Authors:
  • Flanders Research Institute for Agriculture Fisheries and Food (ILVO)
  • Independent Researcher
Aspects of Applied Biology 147, 2022
International Advances in Pesticide Application
297
A combined experimental and modelling study of
dust drift during wheat sowing
By D NUYTTENS1, R GODAERT1, M A ADAMU2, N JONES3, A C CHAPPLE3,
B JENE3, B SORNIN3 and P VERBOVEN2
1Flanders Research Institute for Agriculture, Fisheries and Food (ILVO), Belgium
2BIOSYST-MeBioS, KU Leuven, Belgium
3Seed Treatment Industry Stewardship Steering Committee (STISSC) on behalf of its members:
Syngenta Crop Protection AG, Bayer AG, DuPont Crop Protection, BASF SE and Nufarm
GmbH & Co. KG1
Corresponding Author Email: david.nuyttens@ilvo.vlaanderen.be
Summary
Although often considered as a safe pesticide application method, treated seeds can pose
environmental risks, e.g. when abraded particles are expelled during seed drilling. The
ILVO-KU Leuven consortium has tried to shed more light on the phenomenon using an
integrated experimental and modelling approach, despite the availability of eld dust drift
data being scarce. In collaboration with STISSC, an intensive campaign of eld dust drift
experiments was performed in 20202021, both for mechanical and pneumatic wheat
sowing. Dust drift measurements included horizontal and vertical gauze netting collectors.
In addition, detailed information about the drilling machinery (geometry, dust emission rates,
air ow rates), the wheat seeds (Heubach: dust and a.i. values) and the physicochemical
dust characteristics (size, shape, density, internal porosity, a.i. content) were collected. The
gathered data was used to rene and validate the computational uid dynamics (CFD) dust
drift model previously developed by KU Leuven (Adamu et al., 2022).
The results show that the mean Heubach values for the wheat seed samples were a factor
10 smaller than the internationally accepted maximum permissible of 4.0 g.100 kgseeds
-1,
indicating a good seed quality. The dust emission rates of the mechanical and pneumatic
seeder were similar (c. 2.46 g.100 kgseeds
-1), although the dust originating from the latter
was slightly ner.
The CFD model simulations explored the eect of wind speed, drill type, dust emission
sources and dust properties in comparison of seed drilling of other crops species.
Both the experimental and modelling results are useful for risk assessment purposes and
will help to better understand the dust drift phenomenon.
Key words: Cereal drill, dust emission, CFD, particle deposition, pesticides
Introduction
Although often considered as a safer and more sustainable alternative to spraying
(Nuyttens et al., 2013), some insecticide seed treatment products were taken o the market after
1As represented for the purposes of the academic research agreement by Syngenta Crop Protection AG,
Schwarzwaldallee 215, 4058 Basel, Switzerland.
298
the observed acute bee mortality. Also for fungicide seed treatments it has become more dicult
to keep them available or to get new ones approved (Foqué et al., 2020). Dressing seeds with
pesticide coatings is a widespread and eective way to control pests and diseases, using smaller
doses with potentially less harmful side eects than spraying. But this technique also brings
disadvantages: residues of systemic pesticides can be present in guttated water, plant pollen and
nectar of seed-dressed plants, but most importantly, abraded seed particles (small dust particles)
can be emitted to the environment during sowing (Nuyttens et al., 2013). The nal destination
of dust can be water bodies, other elds in the neighborhood or environmental vegetation where
insects, especially honeybees, can be aected by the dust particles (Tapparo et al., 2012; Krupke
et al., 2012)pesticide use, and climate change or some combination of these factors, and managed
honey bees, Apis mellifera, are part of worldwide pollinator declines. Here we exposed honey bee
colonies during three brood generations to sub-lethal doses of a widely used pesticide, imidacloprid,
and then subsequently challenged newly emerged bees with the gut parasite, Nosema spp. The
pesticide dosages used were below levels demonstrated to cause eects on longevity or foraging
in adult honey bees. Nosema infections increased signicantly in the bees from pesticide-treated
hives when compared to bees from control hives demonstrating an indirect eect of pesticides on
pathogen growth in honey bees. We clearly demonstrate an increase in pathogen growth within
individual bees reared in colonies exposed to one of the most widely used pesticides worldwide,
imidacloprid, at below levels considered harmful to bees. The nding that individual bees with
undetectable levels of the target pesticide, after being reared in a sub-lethal pesticide environment
within the colony, had higher Nosema is signicant. Interactions between pesticides and pathogens
could be a major contributor to increased mortality of honey bee colonies, including colony collapse
disorder, and other pollinator declines worldwide ©2012 The Author(s).
Seed (coating) dust is generated due to abrasion by seed-seed and/or seed-wall interactions during
transportation and drilling of seeds. Dust particles come in dierent sizes, shapes, and densities
and can contain a range of concentrations of active ingredients (a.i.) (Foqué et al., 2017a,b, 2020)
that can have a negative impact on the environment (Devarrewaere et al., 2015; Nuyttens et al.,
2013); however, have directed attention to the emission of abraded pesticide-coated seed particles
to the environment during sowing. This phenomenon of drift of pesticide dust can lead to pesticide
contamination of air, water and other natural resources in crop-growing areas. This review article
presents the state of the art of the phenomenon of dust emission and drift from pesticide seed
dressing during sowing and its consequences. Firstly, pesticide seed treatment is dened and its
pros and cons are set out, with the focus on dust, dust emission and dust drift from pesticide-coated
seed. The factors aecting emission of pesticide dust (e.g. seed treatment quality, seed drilling
technology and environmental conditions.
In comparison to spray drift, the dust drift phenomenon is far less studied and documented. For
wheat (and other species), little eld data is available and conducting experiments to generate
more data is expensive. Both the agricultural sector, the phytopharmacy and government
agencies would benet greatly from more dust drift data and a better dust drift risk assessment tool.
That is why within the framework of the STISSC II project, dust characterization and dust drift
experiments combined with CFD (Computations Fluid Dynamics) modeling were set up for both a
mechanical and a pneumatic wheat drill. Because of the many factors involved and the complexity
of the dust, there is not yet a reliable dust drift model that can be used as a risk assessment tool.
A CFD model developed by KU Leuven for maize seed and sugar beet dust (Devarrewaere et al.,
2015, 2016, 2018; Gruyters et al., 2020; Adamu et al., 2022) already showed promising results to
predict dust drift outcomes. This model has now been expanded to wheat seed dust using the eld
trial data. To optimize the model for this type of dust, the physicochemical properties of the dust,
the machine characteristics and emission rates were characterized with dierent measurement
techniques.
299
Materials and Methods
Two extensive dust drift measuring campaigns were conducted -one with a pneumatic drill and
one with a mechanical drill- measuring downwind dust deposits using gauze nets and Petri dishes.
Seed treatment quality, seeder characteristics and dust emission rates were determined. A complete
physicochemical characterization of the abraded dust particles was performed. All collected data
were used as input and for validation and renement of an existing CFD dust drift model.
Treated seeds, seed quality, seeders and trial sites
Winter wheat seeds coated with the commercial product Landor CT were used. Seeds were
treated with 200 mL formulation per 100 kg seeds. Formulation contained 25 g.L-1 Fludioxonil,
20 g.L-1 Difenoconazol and 5 g.L-1 Tebuconazol. In this study, Fludioxonil was considered. The
abrasion resistance of the seeds was measured at ILVO using the Heubach test as described by
Zwertvaegher et al. (2016) according to the protocol described by the ESA STAT Dust Working
Group (2011). Three replications, each with 100 g of treated seeds, were performed. Heubach a.i.
tests were performed only on seeds used at the Emstek trial by Innovative Environmental Services
Ltd (Witterswil, CH).
Wheat dust emission rate and eld drift trials were done in November 2020 in Emstek (DE) with
a pneumatic seeder (Horsch Express 3 TD, 3 m working width, 20 rows) and in March 2021 in
Wallsbüll (DE) with a mechanical seeder (Maschio Gaspardo Dama, 3 m working width, 24 rows)
(Fig. 1). The selected elds were screened for residues of Fludioxonil and were conrmed to be
free of residues.
Fig. 1. Pneumatic seeder Horsch Express 3 TD (left) and mechanical seeder Maschio Gaspardo Dama
(right).
Stationary dust generation and emission rate trials
The dust emission rates of both stationary seeders were measured using Bayer’s high eciency
cyclone setup, described in Chapple et al. (2014). With the pneumatic seeder, 147.9 kg of seeds
passed through the capture system and cyclone (Fig. 2, left). With the mechanical seeder 198.8
kg were processed (Fig. 2, right). The amount of dust collected was weighed and used for further
physicochemical characterisation.
Physicochemical dust characterisation
The particle size distribution, sphericity, porosity and density of the dust generated by both
seeders were determined using wet laser diraction, X-ray micro-CT scanning, sonic sieving and
gas pycnometry, as described in previous work (Devarrewaere et al., 2015; Foqué et al., 2017a,b).
300
Fig. 2. Stationary dust generation and emission rate trials with the pneumatic (left) and the mechanical seeder (right).
Fig. 3. Dust drift eld trial layout.
301
Total dust samples were separated in size classes (viz. 10–20 μm, 20–32 μm, 32–53 μm, 53–75
μm, 75–106 μm, 106–125 μm, 125–180 μm, 180–250 μm, 250–355 μm, 355–500 μm, 500–1000
μm and 1000–2000 μm) via sonic sieving. Fludioxonil content of the dierent size fractions
was determined by the Walloon Agricultural Research Centre (Gembloux, BE) using Ultra High
Performance Liquid Chromatography with Diode Array Detection (UHPLC-DAD). Depending
on the amount of dust per size fraction, some fractions needed to be merged for successful density
and a.i content measurements.
Dust drift eld trials and collectors
Dust drift trials were set up on bare soil, in a at and open area to avoid localized turbulent wind
conditions, as illustrated in Fig. 3. Winter wheat seeds were sown at a target sowing rate of 240
kgseeds.ha-1. Both trial sites were large enough to consist of a central sowing area of ≥ 1 ha (≥ 200 m
length by ≥ 50 m width) and a downwind sampling area (≥ 30 m width) parallel to the sowing area.
To set up the dust collectors, the edge of the sowing area was situated at the rst drilled row + 0.3 m.
The sampling area was divided into six subplots (A–F) with a distance ≥ 8 m between subplots.
Subplots A, C and E contained vertical gauze nets (3D dust collectors, c. 3.5 m wide × 2 m high)
at 3 m from the edge of the sowing area. On each of the three gauze nets, ve squares of 0.5 m ×
0.5 m were marked with their centres at 0.65 m height from the soil surface and a distance of 0.15
m between the squares. The gauze nets were wetted with glycerol prior to sowing to enhance dust
adsorption.
Subplots B, D and F contained Petri-dishes (2D dust collectors, 135 mm diameter) on the bare
soil lled with 75 ml of glycerol. Each of these subplots consisted of four lines of ve Petri-dishes
parallel to the sowing direction put at distances of 1 m, 3 m, 10 m and 20 m from the edge of the
sowing area. The distance between the Petri-dishes within the same line was 1 m.
At least 60 mins after the completion of sowing, the marked squares were cut from the gauze nets
and all Petri-dishes were closed and collected to avoid cross-contamination.
At both trial sites, three dust drift trials were performed on three dierent days. During each
sowing operation, the wind speed and direction and the air temperature were continuously
monitored using an upwind and a downwind weather station. In the upwind weather station, three
3D ultrasonic anemometers were mounted at 1 m, 2 m and 5 m height measuring at 1 Hz. In the
downwind weather station, three 3D ultrasonic anemometers were mounted at 1 m, 2 m and 3 m
height measuring at 5 Hz.
CFD model
The Computational Fluid Dynamics (CFD) dust drift model continuously under development by
KU Leuven (Devarrewaere et al., 2015, 2016, 2018; Gruyters et al., 2020; Adamu et al., 2022)
was used to simulate the six sowing events and to predict downwind dust drift values. The collected
machine characteristics, emission rates and dust characteristics are used as an input for the model.
Measured and predicted dust drift values are compared to validate and further rene the dust drift
model.
Results and Discussion
Seed quality
The Mean Heubach value (HV) for the wheat seed samples was 0.413 ± 0.134 g.100 kgseeds
-1 for the
Emstek trial and 0.533 ± 0.07 g.100 kgseeds
-1 for the Wallsbüll trial which is in line with the results
from 59 other wheat samples treated with Landor in 2019 as reported by Kämpfer et al. (2021).
These values are lower than previous results reported for other wheat samples (Zwertvaegher et
al., 2016) and well below the internationally accepted maximum permissible HV for wheat of
4.0 g.100 kgseeds
-1 as stated by ESTA (2022). This shows the good seed quality of Landor treated
302
wheat seeds in terms of dust abrasion during sowing. For a sowing rate of 240 kg.ha-1, these values
correspond with 0.991 g.ha-1 and 1.279 g.ha-1. Although originating from the same seed batch, the
slightly higher HV for the Wallsbüll seeds can probably be explained by the fact that these seeds
were stored four months longer. The mean Heubach AI value 0.0049 g.100 kgseeds
-1.
Dust emission rates
Stationary measured dust emission rates were as good as identical for both machines, i.e. 2.46 g.100
kgseeds
-1 for the pneumatic seeder and 2.47 g.100 kgseeds
-1 for the mechanical seeder, corresponding
to an emission rate of about 5.9 g.ha-1. Because of the complexity of these measurements, few
dust emission rate data is available. The values measured are clearly higher than emission rates
measured for beet seeders ranging from 0.02 to 1.23 g.ha-1 for dierent machine types because of
the lower beet sowing rate (c. 4 kg.ha-1) (Foqué et al., 2020). For maize and using a single sowing
element, Foqué et al. (2017c) estimated a dust emission rate of 7.89 g.ha-1.
Physicochemical dust characteristics
Particle size distribution (PSD)
Fig. 4 shows the volumetric PSD determined by wet laser diraction of the dust collected from the
pneumatic and mechanical seeder. In addition, the overall wheat particle size distribution is shown
based on six independent wheat dust samples as reported by Foqué et al. (2017a). Overall, the
three curves correspond well with Dv0.5 values2 of 409 µm, 542 µm and 409 µm for the pneumatic,
mechanical and overall wheat dust, respectively. Dust from the pneumatic seeder contained more
smaller particles (< 150 µm) compared to the mechanical seeder while the mechanical seeder dust
contained more bigger particles (> 1000 µm). For other species, Foqué et al. (2017a) found Dv0.5
ranging from 103 µm (rapeseed) up to 1038 µm (barley).
The PSD obtained with sonic sieving is shown in Table 1. Values are expressed as mass percentages
per particle size class. Results conrm the slightly ner dust from the pneumatic seeder compared
to the mechanical seeder. In the Emstek dust, 71.2% of all dust particles had a diameter smaller
than 355 µm. On the other hand, in the Wallsbull dust there was only 63.0% smaller than 355 µm.
Fig. 4. Volumetric PSD of the wheat dust collected from the pneumatic and mechanical seeder and the
overall wheat particle size distribution (Foqué et al., 2017a).
2D[v0.5] = Particle diameter below which smaller particles constitute 50% of the total volume.
303
Density
Gas pycnometry measurements conrmed the trend generally observed in previous studies (Foqué
et al., 2017b) that smaller particles have a greater (apparent) density (Table 1). On the other hand,
density values were higher than for previous wheat samples (Foqué et al., 2017b) which might be
caused by a dierent formulation.
Chemical content
Table 1 shows the Fludioxonil concentration for the dierent size classes of both the pneumatic
and mechanical dust. Measured Fludioxonil concentrations (ranging from 0.15% to 0.58%) were
low and generally lower than a.i. concentrations of previous wheat dust samples (Foqué et al.,
2017b) because of the low Fludioxonil concentration (25 g.L-1) in the seed dressing formulation. The
results conrm earlier ndings of Foqué et al. (2017b, 2018) that for wheat, the relation between
between a.i. content and particle size can generally be described by a peak-like function. Here,
highest a.i. content was found in the particle size range from 125–250 µm.
Sphericity and porosity
Micro-CT was used to explore the sphericity and porosity of the dust particles. Sphericity changed
with particle size (Fig. 5, left), expressed as a function of equivalent diameter. A subset of smaller
particles (< 50 µm) was more spherical (sphericity ~1), while larger particles were more elongated
cylindrical (brush hairs from the wheat grains) or plate-like abraded parts of the seed. In both latter
cases, the sphericity drops below 0.5.
Fig. 5 (right) plots the porosity as a function of equivalent diameter of the particles. Two distinct
types of particles were observed. The rst set of particles were relatively small (< 80 µm) with very
high X-ray attenuation, likely because they are composed of anorganic elements with high density
and zero porosity. The second set was composed of organic material with medium X-ray attenuation
and varying degrees of porosity. The porosity of these particles increased linearly with particle size
from less than 10% for small particles (< 50 µm) up to 50% for the larger particles (> 200 µm).
Apparent density and porosity values (as a function of particle size) will be used to calculate
envelop density values which will be included in the CFD modelling together with the sphericity
information.
Fig. 5. Sphericity of pneumatic dust particles (left) and porosity (right) as a function of particle equivalent
diameter.
Dust eld drift trials
Targeted meteorological conditions for the eld trials were a wind speed (at 2 m height) between
1.5 and 5 m.s-1 and a wind direction within a ± 30° deviation from the orthogonal direction to the
sowing area. Average wind velocities, deviations from the orthogonal wind direction and temperature
for the six trials are shown in Table 2. For the Emstek trials (trial 1–3), both wind speed and wind
direction were well within the predetermined limits. This was not always the case for the Wallsbüll
trials where wind velocity was on the lower end in trial 5 and both wind speed and direction were
unstable and deviating in trial 6.
304
Table 1. Dust properties of threated wheat seeds collected from the pneumatic
(Emstek) and the mechanical seeder (Wallsbüll)
Pneumatic seeder dust Mechanical seeder dust
Size range
[µm]
Mass
fraction [%]
Apparent
density
[g.cm-3]
a.i.
content
[%]
Mass
fraction
[%]
Apparent
density
[g.cm-3]
a.i. content
[%]
< 20 0.2
2.55
0.15
0.1
/
/
20–32 0.0 0.0 /
32–53 1.4 0.4 /
53–75 4.0 0.29 0.2 /
75–106 6.9 0.35 1.7 0.39
106–125 19.2 1.84 0.29 10.6 2.01 0.44
125–180 16.4 1.68 0.43 18.6 1.61 0.55
180–250 14.2 0.43 19.3 0.58
250–355 8.8 1.70 0.39 12.1 1.64 0.55
355–500 7.4 0.23 10.2 0.44
500–1000 12.8 1.52 0.16 13.8 1.56 0.30
> 1000 8.6 0.17 12.5 0.28
At the moment of writing this publication, dust drift analyses are still ongoing. Complete results
will be presented at the conference.
Table 2. Average (± SD) wind speed, deviations from the orthogonal wind direction (Δ wind dir.)
and temperature during the 6 dust drift trials (trial 13: Emstek, trial 46: Wallsbüll)
Trial 1 Trial 2 Trial 3 Trial 4 Trial 5 Trial 6
Wind speed (m.s-1)4.6 ± 1.0 3.1 ± 0.7 2.6 ± 0.5 4.6 ± 0.9 1.4 ± 0.5 2.0 ± 0.7
Δ wind dir. (°) 7.4 ± 9.2 -5.8 ± 10.8 -11.8 ± 11.2 -28.4 ± 14.2 13.1 ± 27.6 61.3 ± 28.8
Temperature (°C) 11.5 ± 0.2 9.4 ± 0.2 13.8 ± 0.1 7.3 ± 0.3 2.5 ± 0.6 7.4 ± 0.4
CFD model
CAD models of the drills and tractors used in the eld were developed and input to the CFD
simulation framework that was described by Adamu et al. (2022). The actual wind conditions of
Table 2 were simulated, resolving the wake ow generated around the machine while accounting for
the speed of the tractor. The measured dust particle properties given in Table 1 and Fig. 5 were input
at corresponding emission source points from the machines, with their respective emission rates.
The consequent movement of several thousands of dust particles were calculated in the turbulent
airow around the machine to the point where they deposit on the ground downwind of the sowing
area. From the distribution of dust deposits on the ground, drift curves were derived. Simulations
are planned and will be benchmarked to the eld trial results and used to better understand the eld
trial observations and improve the predictive capacity of the model. Comparison is also made to
the results for sugar beet drilling presented by Adamu et al. (2022), considering dierences in dust
characteristics and drilling types.
Conclusion
Compared to spray drift, the dust drift phenomenon is far less studied and understood. The
physicochemical wheat dust properties and dust emission rates gathered in this study are vital for
305
the development of predictive dust drift models. Little eld drift data is available and generating
more eld data is expensive. The presented data for wheat drilling, using two dierent seeding
techniques, extends the current dust drift database and helps to better understand the complex dust
drift phenomenon. The data is used as input and to rene and validate the CFD dust drift model. This
model allows a better and quicker assessment of dierent scenario’s (such as varying wind speeds
and direction, seed species, treatment, drilling machinery) with a lower cost than conducting more
eld studies and can thus serve as a less expensive risk assessment tool. Still, the model depends
on the quality of the input data like the dust characteristics and the emission rates.
Acknowledgements
The authors gratefully acknowledge the nancial support of STISSC (Seed Treatment Industry
Stewardship Committee).
References
Adamu M A, Nuyttens D, Godaert R, Jones N, Chapple A, Jene B, Nicolai B, Sornin B,
Verboven P. 2022. CFD dust drift simulation from sugar beet seed drilling: explicit modelling of
machine types. Aspects of Applied Biology 147, International Advances in Pesticide Application,
pp. 87–94.
Chapple A C, Vrbka L, Friessleben R, Schnier H-F, Cantoni A, Arnold A C. 2014. A novel
technical solution to minimize seed dust during the sowing process of maize using vacuum based
equipment: principles and an estimate of eciency. Aspects of Applied Biology 122, International
Advances in Pesticide Application, pp. 119–124.
Devarrewaere W, Foqué D, Heimbach U, Cantre D, Nicolaï B, Nuyttens D, Verboven P.
2015. Quantitative 3D Shape Description of Dust Particles from Treated Seeds by Means of X-ray
Micro-CT. Environmental Science & Technology 49:7310–7318. DOI: 10.1021/acs.est.5b02250.
Devarrewaere W, Janssen S, Foqué D. 2016. CFD modelling to simulate dust emissions from
pneumatic seed drills. Aspects of Applied Biology 132, International Advances in Pesticide
Application, pp. 227–234.
Devarrewaere W, Foqué D, Nicolai B, Nuyttens D, Verboven P. 2018. Eulerian-Lagrangian CFD
modelling of pesticide dust emissions from maize planters. Atmospheric Environment 184:304314.
https://doi.org/10.1016/j.atmosenv.2018.04.051.
ESA STAT Dust Working Group. 2011. Physical method - assessment of free oating dust and
abrasion particles of treated seeds as a parameter of the quality of treated seeds - Heubach test,
Version 1.0.
ESTA. 2022. Esta Dust reference values Heubach test method. Retrieved from https://euroseeds.
eu/esta-the-european-seed-treatment-assurance-industry-scheme/dust-reference-values-heubach-
test-method/.
Foqué D, Devarrewaere W, Verboven P, Nuyttens D. 2017a. Characteristics of dust particles
abraded from pesticide treated seeds: 2. Density, porosity and chemical content. Pest Management
Science 73:1322–1333. https://doi.org/10.1002/ps.4524.
Foqué D, Zwertvaegher I K A, Devarrewaere W, Verboven P, Nuyttens D. 2017b. Characteristics
of dust particles abraded from pesticide treated seeds: 1. Size distribution using dierent measuring
techniques. Pest Management Science 73:1310‒1321. https://doi.org/10.1002/ps.4526.
Foqué D, Beck B, Devarrewaere W, Verboven P, Nuyttens D. 2017c. Comparing dierent
techniques to assess the risk of dust drift from pesticide-coated seeds. Pest Management Science
73(9):19081920. https://doi.org/10.1002/ps.4557.
306
Foqué D, Biocca M, Pochi D, Pulcini P, Nuyttens D. 2018. Comparing two dust drift mitigation
strategies to the outcomes of a conventional vacuum based precision drill using in-eld validated
indoor static tests. Aspects of Applied Biology 137, International Advances in Pesticide Application,
pp. 285292.
Foqué D, Devarrewaere W, Verboven P, Butler Ellis C M, Jones N, Chapple A C, Jene B,
Sornin B, Nuyttens D. 2020. Dust drift during sowing of sugar beet: Part 1 – Dust properties and
sowing equipment. Aspects of Applied Biology 144, International Advances in Pesticide Application,
pp. 165174.
Gruyters W, Foqué D, Devarrewaere W, Nuyttens D, Jones N, Chapple A C, Jene B, Nicolai B,
Sornin B, Verboven P. 2020. Dust drift during sowing of sugar beet : Part 2 – Predictive modelling
of the fate of dust particles using CFD. Aspects of Applied Biology 144, International Advances in
Pesticide Application, pp. 247253.
Kämpfer C, Weimar-Bosse C, Ahrens K, Hölscher T, Hilscher E, Stähler M, Rautmann D.
2021. Study on dust abrasion of cereal seeds treated in 2019 and active ingredient content in the
dust. Journal für Kulturpanzen 73:2131.
Krupke C H, Hunt G J, Eitzer B D, Andino G, Given K. 2012. Multiple routes of pesticide
-exposure for honey bees living near agricultural elds. PLoS ONE 7:e29268.
Nuyttens D, Devarrewaere W, Verboven P, Foqué D. 2013. Pesticide-laden dust emission and
drift from treated seeds during seed drilling: a review. Pest Management Science 69:564‒575.
Tapparo A, Marton D, Giorio C, Zanella A, Solda L, Marzaro M, Vivan L, Girolami V.
2012. Assessment of the environmental exposure of honeybees to particulate matter containing
neonicotinoid insecticides coming from corn coated seeds. Environmental Science & Technology
46:2592–2599.
Zwertvaegher I K A, Foqué D, Devarrewaere W, Verboven P, Nuyttens D. 2016. Assessment
of the abrasion potential of pesticide-treated seeds using the Heubach test. International Journal
of Pest Management 62(4):348–359. https://doi.org/10.1080/09670874.2016.1206993.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Abraded particles from seeds that are treated with plant protection agents can be expelled during seed drilling, potentially leading to environmental and health risks. The ultimate goal of a current collaborative effort of industry and researchers is to develop a validated simulation model capable of predicting the effect of seed species, environmental conditions and seed drill design and settings on dust drift in field conditions. The model is formulated using computational fluid dynamics (CFD) with sub-models and inputs based on experimental characterization (see Part 1). Simulations were performed to compare with dust deposition profiles measured in the field for pneumatic and mechanical drilling of sugar beet seeds. Analysis of the results allowed to understand better the effects of different factors and dust properties. The results so far identified important effects of the emission source position, the emission rate and the wind velocity on the drift profiles. For the first time, the contribution of different dust particle size fractions to deposition profiles is detailed.
Article
Full-text available
Although seed treatment is considered as a safe pesticide application method, treated seeds can pose environmental risks when abraded pesticide-laden seed particles are expelled during sowing. Shortly after relating bee killing incidents to neonicotinoid-laden dust particles expelled by vacuum based precision maize drills, the dust drift topic received increasingly more attention from different research groups. CREA-IT and ILVO have both developed devices to mitigate the adverse effects during sowing with vacuum based precision drills. Both mitigation strategies were compared to a conventional drill and to each other using the validated indoor dust drift setup of CREA-IT and two type of seeds i.e. seeds coated with thiacloprid and seeds coated with a tracer. In addition, Heubach values and physicochemical properties of the seeds used were measured. Ground and airborne drift deposition results confirmed the important drift reduction capacities of both technologies with drift reduction values ranging from 44% up to 90% compared with the conventional technique with air deflectors. These results confirm that this static method is capable to assess the efficiency of drift reduction devices in a relatively fast and efficient way compared with field trials.
Article
Full-text available
Background: Although considered as a save pesticide application method, treated seeds can pose environmental risks when abraded pesticide-laden seed particles are expelled during sowing. This dust drift risk is clearly linked with the seed coating quality. Seed coating quality is traditionally assessed with Heubach dustmeters and guidelines are established in terms of 'Heubach value'. This technique may, however, not take all drift sensitive particles in to account. In this study, results of the Heubach test are compared to two alternative setups: mechanical sieving and individual sowing element. Results: The abrasion potential assessed with the Heubach dustmeter was much lower than the total dust fraction generated by mechanical sieving and the individual sowing element. The amount of dust produced and its particle size distribution of the both other techniques were comparable. Conclusion: It looks like the Heubach dust meter underestimates the risk of dust drift. Using one of the alternative methods might be a more appropriate way to assess the abrasion potential of seeds. Due to the low investment cost required, mechanical sieving seems a good approach for non-specialized labs. The individual sowing element setup is the most realistic simulation of the in-field dust drift generation but requires a higher initial investment. Therefore, this setup is most suitable for specialized labs and is recommended for further research in this area.
Article
Full-text available
During sowing of pesticide-treated seeds, pesticide-laden dust and abraded seed particles may be emitted to the environment, possibly leading to environmental contamination and posing health risks. In many countries there is currently no legislation concerning the acceptable amount of dust of treated seeds. This study aimed to gain insight in the abrasion potential of available pesticide-treated seeds and its associated factors. The abrasion potential of 45 seed samples of 7 different species [viz. sugar beet, oat, barley, wheat, spelt, pea, and maize] was determined using the Heubach test and amounts of dust were expressed as g 100 kgseeds-1, g 100,000 seeds-1, and g ha-1. The abrasion potential fell generally within the boundaries of maximum permissible values adopted by different countries. Species, seed treatment company, number of AI’s and combination of AI’s had significant effects on the abrasion potential, whereas little or no effect of agitation and conservation was found. However, species were situated differently with respect to each other depending on the unit in which the abrasion potential was expressed. A standard unit that takes into account the species’ seed rate is suggested to give the fairest assessment of dust drift risk and would allow international comparison. Link to this article: http://dx.doi.org/10.1080/09670874.2016.1206993
Article
Full-text available
Airborne drift of dust from pesticide-treated seed to non-target organisms has become a major environmental concern in recent years. Predicting the dispersion of this type of dust is challenging because dust particles are very irregular in size, shape, density, internal porosity and chemical composition. In this work, a CFD Lagrangian particle tracking model that accounts for the irregular properties of these particles was developed. Dust properties were measured experimentally and implemented in the CFD model. Simulated dust deposition patterns agreed well with wind tunnel data. Simulations also demonstrated the importance of micro-CT to account for particle shape and porosity. The model can be used to simulate realistic seeding scenarios, indicating the influence of dust properties, seeder design and wind conditions. The modeling approach presented in this work can also be applied to the simulation of other types of dilute particle flows.
Article
Full-text available
Crop seeds are often treated with pesticides before planting. Pesticide-laden dust particles can be abraded from the seed coating during planting and expelled into the environment, damaging non-target organisms. Drift of these dust particles depends on their size, shape and density. In this work, we used X-ray micro-CT to examine the size, shape (sphericity) and porosity of dust particles from treated seeds of various crops. The dust properties quantified in this work were very variable in different crops. This variability may be a result of seed morphology, seed batch, treatment composition, treatment technology, seed cleaning or an interaction of these factors. The intra-particle porosity of seed treatment dust particles varied from 0.02 to 0.51 according to the crop and generally increased with particle size. Calculated settling velocities demonstrated that accounting for particle shape and porosity is important in drift studies. For example, the settling velocity of dust particles with an equivalent diameter of 200 µm may vary between 0.1 and 1.2 m s-1, depending on their shape and density. Our analysis shows that in a wind velocity of 5 m s-1, such particles ejected at 1 m height may travel between 4 and 50 m from the source before settling. While micro-CT is a valuable tool to characterize dust particles, the current image processing methodology limits the number of particles that can be analyzed.
Article
An Eulerian-Lagrangian 3D computational fluid dynamics (CFD) model of pesticide dust drift from precision vacuum planters in field conditions was developed. Tractor and planter models were positioned in an atmospheric computational domain, representing the field and its edges. Physicochemical properties of dust abraded from maize seeds (particle size, shape, porosity, density, a.i. content), dust emission rates and exhaust air velocity values at the planter fan outlets were measured experimentally and implemented in the model. The wind profile, the airflow pattern around the machines and the dust dispersion were computed. Various maize sowing scenarios with different wind conditions, dust properties, planter designs and vacuum pressures were simulated. Dust particle trajectories were calculated by means of Lagrangian particle tracking, considering nonspherical particle drag, gravity and turbulent dispersion. The dust dispersion model was previously validated with wind tunnel data. In this study, simulated pesticide concentrations in the air and on the soil in the different sowing scenarios were compared and discussed. The model predictions were similar to experimental literature data in terms of concentrations and drift distance. Pesticide exposure levels to bees during flight and foraging were estimated from the simulated concentrations. The proposed CFD model can be used in risk assessment studies and in the evaluation of dust drift mitigation measures.
Article
Background: Apart from particle size, drift of abraded seed particles during sowing is mainly affected by two other physical properties viz. particle shape and envelope density. The impact of these abraded seed particles on the environment is highly dependable on their active ingredient content. In this study, the envelope density and chemical content of dust abraded from seeds was determined as a function of particle size for six seed species. Results: Envelope density and active ingredient content both change as a of function of particle size. Important differences in these physicochemical properties were observed between the six species. Functions were fitted to the collected data to describe the physicochemical properties as a function of particle size. Conclusion: The gathered physicochemical information is essential for the CFD based dust drift prediction models and can be useful for other prediction models as well as for the ongoing risk assessment of active ingredients used for seed treatment on ecosystems and ecosystem-services. In addition, the results can help to better understand the dust drift phenomenon and to develop mitigation strategies.
Article
Background: Particle size is one of the most important properties affecting the driftability and behavior of dust particles abraded from pesticide dressed seeds during sowing. Three particle sizing techniques were used determine the particle size distribution of dust abraded from seeds from six different species. Results: Important differences in dust particle size distribution between species were observed with the finest dust for rapeseed and the coarsest dust for barley. Wet laser diffraction and sonic sieving particle size results correlated well while micro-CT is able to deliver 3D-information and additional physical particle properties (shape, porosity) CONCLUSION: All particle sizing techniques have their (dis)advantages and none of them is able to perfectly describe the real size distribution of non-spherical particles. The gathered particle size information can be used in dust drift prediction models, risk assessment tools and will help to better understand the dust drift phenomenon.