Top Navigation Bar

Article Request Page ASABE Journal Article

Effect of Air-assisted Spray Application Rate on Spray Droplet Deposition Distribution on Fruit Tree Canopies

W. Qiu, C. Sun, X. Lv, W. Ding, X. Feng


Published in Applied Engineering in Agriculture 32(6): 739-749 (doi: 10.13031/aea.32.11663). Copyright 2016 American Society of Agricultural and Biological Engineers.


Submitted for review in November 2015 as manuscript number MS 11663; approved for publication by the Machinery SystemsCommunityof ASABE in May 2016.

The authors are Wei Qiu, Lecturer, Chengda Sun, Doctoral Student,College of Engineering/Key Laboratory of Intelligent Equipment for Agriculture of Jiangsu Province, Nanjing Agricultural University, Nanjing, China; Xiaolan Lv, Vice Professor, Jiangsu Academy of Agricultural Sciences, Nanjing, China; Weimin Ding, Professor, Xuebin Feng, Lecturer,College of Engineering/Key Laboratory of Intelligent Equipment for Agriculture of Jiangsu Province, Nanjing Agricultural University, Nanjing, China. Corresponding author: Weimin Ding, College of Engineering/Key Laboratory of Intelligent Equipment for Agriculture of Jiangsu Province, Nanjing Agricultural University, Nanjing 210031, China; phone: 0086025-5860 6502; e-mail: wmding@njau.edu.cn.

Abstract.  To improve the utilization rate of pesticides and the adaptability of the application machinery to different canopies and to construct a matching relation between the application rate and the characteristics of fruit trees, this article introduces spray droplet coverage rate and deposition as the investigation indexes. Five different application rates were used with the purpose of investigating the effect of the application rate on spray droplet deposition coverage under different canopy characteristics and quantitatively describing the optimum application rates for different canopies. According to the results of the application with under air-assisted sprayer, the recommended spray application rates for the three types of canopies in the present article were 0.25, 0.20, and 0.14 L/tree in average, which is compared with the traditional manual application rate of 0.25 L/tree in average. To achieve the same level of application effect with traditional manual application, the air-assisted application rate for large and medium canopies can be reduced to 0.20 L/tree in average, this is a decrease of 1/5. Compared with the upside coverage rate of the spray droplets, the underside coverage rate of the spray droplets was more significantly affected by the reduction in the application rate. In addition, in the case of a higher spray droplet deposition and a lower dispersion degree, there would be an inconsistency between the spray droplet coverage rate and the deposition indexes. Thus, it is also necessary to measure the uniformity of the spray droplet dispersion on the leaf surface (spray droplet dispersion index) and the effective lethal area of the spray droplets, which will enable a more accurate expression of the application effect.

Keywords.Sprayer, Air-assisted, Droplet deposition, Drift, Precision agriculture.

Pest control (pesticide application) is indispensable in the fruit growing industry. Air-assisted variable-rate application makes spraying droplets penetrate deep into the canopies and increases droplet deposition coverage on leaves (Fox et al., 2008).The application method also attaches importance to the individual differences of the targets, and it is capable of adjusting the application parameters according to the target characteristics, and thus the consumption of pesticides can be effectively reduced (Cross et al., 2001a, 2001b, 2003). The air-assisted precise variable-rate application system is composed of three main parts, i.e., “pre-spray detection,” “intermediate decision-making,” and “spray adjustment;” however, current research has mostly focused on target detection, application parameter adjustment, and other related techniques (Wei and Salyani, 2004; Nuyttensa et al., 2007; Lee et al., 2010). Solanelles et al. (2006) introduced the ultrasonic sensor detection working environment, calculated an optimum application rate, and adopted a magnetic valve for the application rate control. Llorens et al. (2010) achieved variable-rate application according to the canopy width. Measurements of canopy structure with a laser plane range-finder were examined by Takafumi, and a measurement system was developed and applied successfully to actual forests. Lei (2002), as well as Zhang et al. (2009), introduced image information about crops to regulate the flow rate of the spray nozzle, thus the application rate of the pesticides was reduced. All of the above researchers have adopted sensor technology to obtain target information and then realized the adjustment of the application rate based on the target information. Lebeau et al. (2004) and Gu et al. (2011) evaluated droplet size and spray distribution from PWM nozzle. Hendrickson et al. (2001) evaluated the volumetric application accuracy of the variable-rate sprayer which consisted of pulse–width modulation (PWM) solenoids, a pressure controller, and a nozzle control system interfaced to a computer. The above research emphasized the technical realization of the variable-rate spray application of pesticides, which provided a good hardware support for variable spraying.

However, the pesticide application model, the intermediate link, and the decision-making basis is very important. As an important parameter of the pesticide application model, the application rate has always been a focus of research on precise variable-rate application (Miller et al., 2001; Pergher et al., 2003; Da Silva et al., 2006). Berk et al. (2016) summarized the application model as follows: (1) Dose expression model per ha ground (Zhu et al., 2011), (2) Leaf Wall Area based expression models (Friessleben et al., 2007), (3) Canopy volume based expression models (Gil et al., 2007; Zhu et al., 2011; Chen et al., 2012), and (4) Canopy height expression models (Anonymous, 2012). Model (1) had its own limitation as the spray pattern of constant application for per hectare could not satisfy spray requirement for tree canopies of different sizes. Canopy volume based expression model realized this problem. This model assumed tree canopies as cuboids to calculate tree canopy volume, ignore the irregular characteristics of fruit trees and gap between fruit trees. The other two spray models were modified, such as increasing leaf area index or simplify further with assumption of uniform row distances (Friessleben et al., 2007), on the basis of Model (3).

The application rate was calculated as the result of volume multiplied by a fixed application rate for per tree canopy volume. As the tree canopy structure was complex, the relationship between volume and application rate was not linear. Cross et al. (2001a) investigated the relationship between spray parameters and tree canopy size by employing three different application rates on small, medium, and large tree canopies, and pointed out that saturation phenomenon existed for small tree canopy. Therefore, on the basis of the work (Cross et al., 2001a), it is necessary to research the model relationship between application rate and tree canopy volume.

In brief, research on the target characteristics and application rate of fruit trees tends to lay a particular emphasis on the hardware design of the detection and spray application devices, but ignores the application model. Particularly, as far as the quantitative analysis of the application rate of pesticides is concerned, there is a lack of a detailed experimental and theoretical basis. With the purpose of improving the utilization rate of pesticides and the adaptability of the application machinery to different orchards, the present article conducted relevant tests to study the relationship between application rate and the canopy characteristics and determine the optimum application rates for trees with different volume characteristics, thus providing a basis for constructing the application rate decision-making model.

Materials and Methods

Spray Equipment

The 3WZ-700 self-propelled air-assisted orchard sprayer (Nanjing Agriculture University, Nanjing, China) was used, and its working process consisted of procedures like water injection, pesticide mixing, spraying, and air-assisting, etc. (fig. 1). The sprayer has a mating power of 28 kW, a working width of 4 to 5 m, a spraying height of 3 to 4 m.It has an axial air inlet, radial air outlet, and hydraulic variable transmission. The blower rotation speed can be regulated within 0 to 2,000 rpm. On each side of the sprayer there are six NH-101 nozzles (Taizhou Sunny Agricultural Machinery Co., Ltd., Taizhou, China), each with a flow rate of 1.2 L/min.

Figure 1. Orchard sprayer: (1) frame, (2) tank, (3) drive system, (4) nozzles, (5) fan, (6) spray pump, and (7) valves.

Object of Test

The test was carried out in a pear orchard at the College of Horticulture Experimental Base, Jiang Pu Farm of Nanjing Agricultural University. The object of the spray application test was Housui pear trees, which had a canopy width of 0.9 to 2.3 m, a height of 2.25 to 3 m, and a row distance × planting distance of 4 × 3 m.

The test site was divided into three areas, and three pear trees were selected in each area by their different canopy volume characteristics (large, medium, and small), i.e., a total of nine pear trees. The canopy leaf area indexes within a range of 1.8 to 4.75 were measured by LAI-2200C Plant Canopy Analyzer (LI-COR, Inc., Lincoln, Neb.). The canopy volumes were manually measured, and measurement procedure is shown in figure 2. The tree canopy was divided into two parts, for the convenience of description, it was assumed to be left and right parts, and tree volume was measured separately on each side. The tower ruler could move along the guide rail in Y axis direction. Before measurement, the maximum width of chosen tree canopy had been estimated, which was less than 2.4 m (table 1), therefore, the distance(in X axis direction) between the guide rail and the central line of the trunk L was set as 1.2 m to ensure that the tower ruler could move freely without encountering the tree canopy. Take the right part volume measurement as an example, the tower ruler started from one edge of tree canopy and moved along the guide rail to the other edge at a determined movement interval for each step; for each movement step, points on the tower ruler were chosen along Z axis direction, and the distances (in X axis direction) of each points on tower ruler to surface of tree canopy was measured, thus the distance dij (in X axis direction) between points on tree canopy and the central line of the trunk could be calculated indirectly; the procedure was repeated until the tower ruler moved across the whole tree canopy. By this way, we could figure that the right part of tree canopy volume could be assumed to be the accumulation of rectangular solids with size of (L-dij) × s × h. In this article, we defined the movement interval and height interval to be 0.1 m, which could satisfy the measurement accuracy (the parameters set-up was flexible based on tree size and measurement accuracy requirement). For the left part volume measurement, the procedure was the same. The volume of tree canopy was the sum of right and left part volume.

Figure 2. Diagram of artificial method to measure canopy volumes.
Table 1.Characterization of fruit trees used in the test.
Canopy
Type
No.Maximum Width (m)Volume
(m3)
Left Area
Density(m2/m3)
Small1.000.960.713.05
2.001.220.692.76
3.001.120.821.93
Medium4.001.631.402.32
5.001.931.632.78
6.001.711.522.75
Large7.001.982.073.65
8.002.252.262.09
9.002.212.033.93

Test Methods

Setting of Application Rate Parameters

Besides being directly related to the canopy characteristics, the application rate of the pesticides is also supposed to be closely related to the application method of the pesticides (Qiu et al., 2015). First, an experienced pesticide operator was asked to spray 30 fruit trees three times using a manual application technique (SX-LK18C, SeeSa Holding Co., Ltd., Taizhou, China), with a total application rate of 7.5 L, which meant the average spray application rate for each fruit tree was 0.25 L/tree. In another test, the 3WZ-700 air-assisted sprayer was used, and five application rates for the same 30 fruit trees were set by adjusting sprayer travel speed. These application rates were 0.08, 0.14, 0.2, 0.25, and 0.34 L/tree in average, four of which were lower than the traditional application rate. According to the preliminary tests and studies, the air-assisted orchard sprayer was operated with a blower rotation speed of 1,200 rpm, a spraying pressure of 1.0 MPa, and travel speeds of 9.7, 6.3, 4.9, 3.6, 2.5 km/h, thus achieving the corresponding test application rates requirement of 0.08, 0.14, 0.2, 0.25, and 0.34 L/tree.

Position of Sampling Points

The number and location of the sampling points were determined according to the shapes of the canopies and the density of the breach leaves on the fruit trees. In the vertical direction z, the canopies were divided into two layers, at their heights of 1.5 and 1.0 m. In the advancing direction of the application machinery y, sections a, b, and c were identified by intervals of 0.5 m in the canopies, and sections A, B, and C were also identified by intervals of 0.5 m along the air-assisting direction × in the canopies; their intersection points were selected as the sampling points, as shown in figure 3.

Figure 3. Sampling points in the canopies.

Clips were used to clamp two paper cards (size 7.6 × 7.6 cm, M&G Stationery Inc., Shanghai, China) on to the trees, which were then used to calculate the upside/underside spray droplet deposition and coverage rate of the leaves. Taking the trunk as the center, the row distance sampling line and the planting distance sampling line were laid out on the ground, and used to study the drift of the ground pesticides (fig. 4).

Figure 4. Sampling points in the ground.

Identification of Spray Droplet Coverage Rate and Deposition

Spray droplet coverage rate: Ponceau 2R solution (SSS Reagent Co., Ltd., Shanghai, China) with a mass fraction of 5% was added into the pesticide container of the sprayer. Paper cards were collected and numbered according to the sequence of sampling points, and the dried paper were removed and put into plastic bags after spraying. The bags were taken to the laboratory and an MRS-3200PU2 scanner (Microtek Technology Co., Ltd., Shanghai, China) was used to obtain scanned images which were then input into a computer image analysis system for statistical analysis. This precisely determined the spray droplet coverage density of unit area.

Spray droplet deposition measurement were as follows: Before measurement, calibration process was needed. Ponceau 2R was used as tracer in the experiment; the UV2000 ultraviolet-visible spectrophotometer (UNICO Instrument Co., Ltd., Shanghai, China) was used at a wave set of 510 nm under which the absorbance coefficient of Ponceau 2R could reach the maximum to ensure measurement accuracy; five different solution concentrations standard were chosen, and for each concentration standard, solutions were prepared with three replicates, and the corresponding absorbance coefficients were obtained directly via readings of ultraviolet-visible spectrophotometer. Table 2 shows the measurement results. The determination coefficient of the calibration curve was 0.9939, which showed a high measurement accuracy in the absorbance coefficient range from 0~2 (fig. 5). After calibration, each paper card was placed into a separate beaker and washed by a certain amount of distilled water for cleaning. The absorbance coefficients of the solutions were obtained via ultraviolet- visible spectrophotometer and the spray droplet deposition of unit are a w (µg/cm2) could be given by equation as:

   

where

V    =    the amount of water added (mL),

A    =    absorbance coefficient,

K    =    proportionality constant, which is 0.0372,

S    =    paper card area collected (cm2).

Distribution of Airflow Velocity Fields in Canopies of Different Volumes

Air velocities in the tree canopy was measured withAS865 anemograph (SMARTSENSOR, Hong Kong, China). Intersection points of plane b, planes A, B, C, and upper, lower plane were chosen as measurement points (fig. 3), and the total number of points was six. The windward side of the anemograph sensor should face the airflow direction directly, and the maximum airflow velocity was measured at each site. The operation was repeated for three times to calculate the average air velocity.

Table 2. Relation between absorbance coefficient and concentration.
Absorbance coefficient0.1030.2020.4070.4960.7880.9891.794
Concentration/(µg/mL)2.551012.5202550
Figure 5. Relation between absorbance coefficient and concentration.

Results and Discussion

Effect of Application Methods on Application Rate Selection

There is no doubt that the application rate would be determined by the properties of the pesticides themselves. However, as shown in table 3, same rate resulted in higher coverage with air-assisted application than traditional. When the same application rate of 0.25 L/tree was used, the spray droplet coverage rate and deposition on the leaves by the air-assisted application method were obviously superior to those of the traditional application method. In the cases of a larger canopy volume and a higher biomass, the advantages of the air-assisted application method can be more obvious. Compared with the traditional application method, the air-assisted application method achieved a coverage rate increase of 46.55% and a deposition increase of 73.19%. Thus, same rate of air-assisted gave higher coverage in larger canopies.

Table 4. Spray coverage distribution measured in the canopy (%).[a]
Application RateLarge CanopyMedium CanopySmall Canopy
(L/tree)UpsideUndersideUpsideUndersideUpsideUnderside
0.3431.86a30.96a43.91a33.56a49.7a38.63a
0.2531.43a26.36ab35.5ab31.97a38.93b32.65ab
0.226.95ab20.37bc26.43b26.6ab34.6bc26.39b
0.1423.57b15.26c27.5b20.38bc30.35c25.91b
0.0822.44b6.53d26.53b10.61c31.1c14.77c

    [a]    Values (means) in columns followed by the same letter do not differ significantly (Duncan’s multiple range test, P = 0.05, three replicates).

Table 3.Coverage and deposition of spray droplets on large, medium, and small canopies under different application methods (%).
Large CanopyMedium CanopySmall Canopy
Application MethodCoverage
(%)
Deposition (µg/cm2)Absorption Rate (%)Coverage
(%)
Deposition (µg/cm2)Absorption Rate (%)Coverage
(%)
Deposition (µg/cm2)Absorption Rate (%)
Air-assisted 0.20 (L/tree)23.664.3859.8626.524.4034.9230.504.4416.95
Air-assisted 0.25 (L/tree)28.94.7852.2633.744.8430.7335.795.1415.70
Traditional 0.20 (L/tree)19.722.7637.7224.053.6929.2939.84.3216.50

Effect of Application Rate on Spray Droplet Coverage Rate

Based on the application rates adopted by the traditional method, the present article tested five application rates by controlling the travel speed of the air-assisted sprayer, investigated the upside and underside coverage of the spray droplets on the leaves on three types of canopies (large, medium, and small, i.e., three fruit trees of similar size were for each type and the total number of trees was nine). As determined from the data in table 4, when using an average application rate of 0.34 L/tree, the average coverage on three types of the canopies were 31.41%, 38.73%, and 44.16%, respectively. For the application rate of 0.08 L/tree in average, the average coverage rates on the canopies of the three types were 14.49%, 18.57%, and 22.94%, respectively. The average coverage rate in the canopies gradually declined as the canopy volume increased; it also declined with a reduction in the application rate. Based on the requirements of an upside coverage rate of above 30% and an underside coverage rate of above 25% for pesticide application in orchards, and in combination with the ground losses of pesticides when sprayed by different application rates on large, medium, and small canopies (fig. 6), the recommended spray application rates for the three types of canopies under air-assisted application conditions were 0.25, 0.20, and 0.14 L/tree in average, respectively. The relation between average coverage of upside and underside of tree canopy and application rate is shown in figure 7, and by this way, the applicate rate could be determined based on the required coverage for different sizes of tree canopy.

Figure 7. Relation between coverage and application rate for different sizes of tree canopies.

Effect of Application Rate on Spray Droplet Deposition

As shown by the investigation on the effect of the application rate factor on spray droplet deposition in table 5, in the same types of canopies, the average deposition of the spray droplets on the leaves decreased with a reduction in the application rate, which was relatively consistent with the coverage data (table 4). However, when the effect of the canopy factor on the spray droplet deposition was investigated, it was found that, for the same level size tree canopy, there were no significant differences on upside of leaves under different application rates while there existed significant differences on the underside of leaves. Therefore, we treated average deposition on upside and downside leaves as criteria, and the relation between average deposition and application rate is shown in figure 8.

Table 5. Spray deposition distribution measured in canopy (µg/cm2).[a]
Application RateLarge CanopyMedium CanopySmall Canopy
(L/tree)UpsideUndersideUpsideUndersideUpsideUnderside
0.345.09a5.12a6.15a5.02a6.23a5.23a
0.254.9a4.65a4.79a4.88ab5.28a4.99a
0.24.3a4.45a4.69a4.11b5.1a3.77b
0.144.72a2.36b5.0a3.21c4.86a3.32b
0.084.48a1.65b4.64a2.17d4.92a2.1c

    [a]    Values (means) in columns followed by the same letter do not differ significantly (Duncan’s multiple range test, P = 0.05, three replicates).

The inconsistency between the spray droplet coverage and deposition indexes can be explained as follows: the canopies with different volumes caused the airflow rates to decline in different ways, so the average airflow velocities of the large, medium, and small canopies were 7.63, 5.27, and 4.46 m/s, respectively (table 6). In the air-assisting direction, the airflow velocity of section A was not significantly affected, while the airflow velocities of sections B and C declined at relatively fast rates, which was mainly because the airflow was encountering more branch and leaf barriers in the canopies in the airflow direction from outside to inside, which blocked the velocity.

The decline in the airflow velocity inevitably led to a decreased velocity of the spray droplets. Comparing with the small canopies, the spray droplets on the large canopies were susceptible to accumulation and failed to be well dispersed (fig. 9). Therefore, the large canopies were significantly different from the small ones in terms of the coverage rate of the spray droplets (table 4). Although the spray droplets failed to be well dispersed, the tracer dose observed on the paper cards was not significantly reduced, so there were no significant differences in terms of the deposition.

Figure 8. Relation between coverage and application rate for different sizes of tree canopies.
Table 6. Airflow velocity distribution in different canopies (m/s).
Large CanopyMedium CanopySmall Canopy
SectionUpperLayerLower LayerUpper LayerLower LayerUpper LayerLower Layer
A8.3310.578.6311.109.7014.43
B3.372.133.703.806.836.03
C1.600.772.332.034.134.67
Average4.434.494.895.646.898.38
Overall average4.465.277.63

In conclusion, on the one hand, the structural differences of the canopies led to different application rates; that is, the larger the canopy volume, the higher the biomass, the higher the application rate required. On the other hand, the structural differences of the canopies also caused a decline in the airflow velocity in the canopies, which resulted in the differences in the dispersion degree of the spray droplets and gave rise to relatively significant coverage rate differences among the sampling points, but the deposition differences were not very significant.

(a)(b)
Figure 9. Droplet deposition dispersion degree in different canopies: (a) well dispersed in small canopy, (b) poorly dispersed in large canopy.

Effective Lethal Area and Dispersion Indexes of Spray Droplets

Due to inconsistency between the spray droplet coverage rates and the deposition rates in tables 4 and 5, use of only coverage rate, deposition, spray droplet diameter to determine the application effect may be limited to a certain extent, as the uniformity of the spray droplet dispersion on the leaf surface also needs to be taken into account. Also, two sampling paper cards (fig. 10) were identified as examples to explain the effect of the uniformity of the spray droplet dispersion on the leaf surface on the application effect, where (fig. 10a) represents non-uniformly dispersed sampling points while (fig. 10b) represents uniformly dispersed sampling points; see measured coverage rate and deposition parameters in table7.

Each spray droplet, when dropped on the leaves, has its effective lethal area. MATLAB was adopted to extract the region of each spray droplet; the regions were then processed with dilating method, and disk structure was applied with size of 0.5, 1.0, 1.5, 2.0, and 2.5 mm, respectively; a new coverage area was obtained to approximately represent the effective lethal area. (fig. 11).

As shown by the results in figure 12, when the expansion distance d was =2.5 mm, the effective lethal coverage area on the paper card of figure 10b was greater than that on the paper card figure 10a, so adopting only the deposition and coverage data to determine the application effect may be limited to a certain extent, and new evaluation indexes need to be introduced.

(a)(b)
Figure 10. (a) Evenly distributed and (b) heterogeneously distributed sampling cards.
Table 7. Spray coverage and deposition measured in figure 9a and b sampling cards.
Figure 9a Figure 9b
Coverage rate (%)12.562.43
Deposition(µg/cm2)2.731.15

To more accurately evaluate the effect of the spray droplet deposition and coverage indexes on the application effect, the spray droplet dispersion index was introduced based on the statistics of the coverage and the deposition of the spray droplets. The spray droplet dispersion index is defined as the standard deviation of the coverage per unit area of several different sections on the sampling card, and is introduced to represent the deviation degree of the values in the group from their average value. Both paper cards figure 10a and 10b were divided into nine equal sections to calculate the coverage of each part. According to table 8, the dispersion indexes of samples figure 10a and 10b were 220.32 and 0.48, respectively. The parameter can effectively represent the uniformity of the spray droplet distribution on the paper cards.

Cards of figure 10a were binarization processing.
Cards of figure 10b were binarization processing
Figure 11. Effective lethal area after expansion of different d values.

Ground Losses

The ground sampling point collection paper cards were used to determine the Ponceau content per unit area and to further measure the ground loss of the pesticides. See the data measured under the five application rate levels in figure 6. With a reduction in the application rate, the ground losses of the spray droplets in the three types of canopies also gradually declined, and the rate of decline presented a trend of first increasing and then decreasing. In small canopies, when the application rate varied within the range of [0.34, 0.14] L/tree in average, the reduction in the ground losses was significant; for the application range of [0.14, 0.08] L/tree in average, the ground losses were at similar levels, so the ground loss of pesticides in small canopies was less under the application rate range of [0.14, 0.08] L/tree in average. In medium canopies, the ground loss varied significantly in the range of [0.34, 0.20] L/tree in average, so the ground loss of pesticides in the medium canopies was less in the application rate range of [0.20, 0.08] L/tree in average. In large canopies, the variations of ground loss were relatively flat.

Conclusion

To establish the quantitative relation between the target characteristics and application rate for canopies, traditional application method and air-assisted application method of five application rates were compared in this research. The spray droplet coverage and depositions of the sampling points were measured to construct a method for determining the optimum air-assisted application rate based on the target characteristics. Under the air-assisted application conditions, the recommended spray application rates for the three types of canopies (small, medium, and large, the corresponding volume was 0.74, 1.52, and 2.12 m3) in the present article were 0.25, 0.20, and 0.14 L/tree in average, respectively.

Figure 12. Further parameters calculated from figure 9a and 9b sampling cards.
Table 8. Discrete degree calculated from figure 9a and 9bsampling cards.
Segmented Image No.123456789
Figure 9a coverage (%)0.430.191.344.827.296.6731.2831.4535.15
Average Mc13.18
Dispersionindex (standard deviation)sc2220.32
Segmented image
Figure 9b coverage (%)2.072.223.711.582.721.851.971.641.53
Average Md2.14
Dispersion index
(standard deviation)sd2
0.48
Segmented image

Besides being related to the properties of the pesticides, the selection of the application rate is also closely related to the canopy characteristics and application method (Ashenafi et al., 2015); that is, the larger the canopy volume and the higher the biomass is, the higher the application rates required. In the case of air-assisted application, the spray droplets have relatively small diameters and are uniformly distributed, which can make full use of lethal effect of each spraying droplet and achieve a good application effect as a result of combination. Therefore, the application rate may be appropriately reduced when pesticides are applied by this method. Compared with the traditional application rate of 0.25 L/tree in average, to achieve the same level of application effect with traditional artificial application, the air-assisted application rate for large and medium canopies can be reduced to 0.20 L/tree in average, there is a decrease of 1/5.

Compared with the upside coverage rate of the spray droplets, the underside coverage rate of the spray droplets was more significantly affected by the reduction in the application rate. The canopies with different volumes, affected the rates of decrease in the airflow in different ways, and the decline in the airflow velocity results in the non-uniformity of the spray droplet dispersion, the accumulation of spray droplets in large canopies, and the low degree of dispersion. Thus, the sampling points in large canopies showed a lower coverage rate, while those in small canopies had a higher coverage rate. However, the spray droplet deposition is only related to the mass of the pesticides dropped on the paper card, and the accumulation or dispersion of pesticides does not significantly affect the spray droplet deposition, which means that the spray droplet depositions of large and small canopies were at the same level. Therefore, besides the coverage rate and deposition indexes, it is also necessary to measure the uniformity of the spray droplet dispersion on the leaf surface (spray droplet dispersion index) and the effective lethal area of the spray droplets, which will express the application effect more accurately.

Figure 6 Ground loss of liquid drift for different canopies.

Acknowledgements

The study was supported by The Natural Science Foundation of Jiangsu Province(grant no.BK20130670), Jiangxi Province 2011 Collaborative Innovation Special Funds “Co Innovation Center of the South China Mountain Orchard Intelligent Management Technology and Equipment” (Jiangxi Finance Refers to [2014] NO156)and National Natural Science Foundation of China (31301687).

References

Anonymous. (2012). Dose expression for plant protection products. EPPO Bulletin, 42(3), 409-415. http://dx.doi.org/10.1111/epp.12000

Berk, P., Hocevar, M., Stajnko, D., & Belsak, A. (2016). Development of alternative plant protection product application techniques in orchards, based on measurement sensing systems: A review. Comput. Electron. Agric., 124, 273-288. http://dx.doi.org/10.1016/j.compag.2016.04.018

Chen, Y., Zhu, H., & Ozkan, H. E. (2012). Development of a variable-rate sprayer with laser scanning sensor to synchronize spray outputs to tree structures. Trans. ASABE, 55(3), 773-781. http://dx.doi.org/10.13031/2013.41509

Cross, J. V., Walklate, P. J., Murray, R. A., & Richardson, G. M. (2001a). Spray deposits and losses in different sized apple trees from an axial fan orchard sprayer: 1. Effects of spray liquid flow rate. Crop Prot., 20(1), 13-30. http://dx.doi.org/10.1016/S0261-2194(00)00046-6

Cross, J. V., Walklate, P. J., Murray, R. A., & Richardson, G. M. (2001b). Spray deposits and losses in different sized apple trees from an axial fan orchard sprayer: 2. Effects of spray quality. Crop Prot., 20(4), 333-343. http://dx.doi.org/10.1016/S0261-2194(00)00163-0

Cross, J. V., Walklate, P. J., Murray, R. A., & Richardson, G. M. (2003). Spray deposits and losses in different sized apple trees from an axial fan orchard sprayer: 3. Effects of air volumetric flow rate. Crop Prot., 22(2), 381-394. http://dx.doi.org/10.1016/S0261-2194(02)00192-8

Da Silva, A., Sinfort, C., Tinet, C., Pierrat, D., & Huberson, S. (2006). A Lagrangian model for spray behaviour within vine canopies. J. Aerosol Sci., 37(5), 658-674. http://dx.doi.org/10.1016/j.jaerosci.2005.05.016

Duga, A. T., Ruysen, K., Dekeyser, D., Bylemans, D., Nicolai, B. M., & Verboven, P. (2015). Spray deposition profiles in pome fruit trees: Effects of sprayer design, training system and tree canopy characteristics. Crop Prot., 67, 200-213. http://dx.doi.org/10.1016/j.cropro.2014.10.016

Fox, R. D., Derksen, R. C., Zhu., H., Brazee, R. D., & Svensson, S. A. (2008). A history of air-blast sprayer development and future prospects. Trans. ASABE, 51(2), 405-410. http://dx.doi.org/10.13031/2013.24375

Friessleben, R., Rosslenbroich, H., & Elbert, A. (2006). Dose expression in plant protection product field testing in high crops: need for harmonization. Pflanzenschutz Nachrichten-Bayer-English Edition, 60(1), 85.

Gil, E., Escolŕ , A., Rosell, J. R., Planas, S., & Val, L. (2007). Variable rate application of plant protection products in vineyard using ultrasonic sensors. Crop Prot., 26(8), 1287-1297. http://dx.doi.org/10.1016/j.cropro.2006.11.003

Gu, J., Zhu, H., Ding, W., & Jeon, H. Y. (2011). Droplet size distributions of adjuvant-amended sprays from an air-assisted five-port PWM nozzle. Atomization Sprays, 21(3), 263-274. http://dx.doi.org/10.1615/AtomizSpr.2011003258

Han, S., Hendrickson, L. L., Ni, B., & Zhang, J. (2001). Modification and testing of a commercial sprayer with PWM solenoids for precision spraying. Appl. Eng. Agric., 17(5), 591-594. http://dx.doi.org/10.13031/2013.6906

Lebeau, F., El Bahir, L., Destain, M. F., Kinnaert, M., & Hanus, R. (2004). Improvement of spray deposit homogeneity using a PWM spray controller to compensate horizontal boom speed variations. Comput. Electron. Agric., 43(2), 149-161. http://dx.doi.org/10.1016/j.compag.2004.01.001

Lee, K. H., Ehsani, R., & Castle, W. S. (2010). A laser scanning system for estimating wind velocity reduction through tree windbreaks. Comput. Electron. Agric., 73(1), 1-6. http://dx.doi.org/10.1016/j.compag.2010.03.007

Lei, T. (2002). Development of a sensor-based precision herbicide application system. Comput. Electron. Agric., 36(2-3), 133-149. http://dx.doi.org/10.1016/S0168-1699(02)00097-2

Llorens, J., Gil, E., Llop, J., &Escolŕ , A. (2010). Variable rate dosing in precision viticulture: Use of electronic devices to improve application efficiency. Crop Prot., 29(3), 239-248. http://dx.doi.org/10.1016/j.cropro.2009.12.022

Miller, P. C. H., Scotford, I. M., & Walklate, P. J. (2003). Characterizing crop canopies to provide a basis for improved pesticide application. ASAE Paper No. 031094. St. Joseph, MI: ASAE.

Nuyttens, D., Baetens, K., De Schampheleire, M., & Sonck, B. (2007). Effect of nozzle type, size and pressure on spray droplet characteristics. Biosyst. Eng., 97(3), 333-345. http://dx.doi.org/10.1016/j.biosystemseng.2007.03.001

Pergher, G., Balsari, P., Cerruto, E., & Vieri, M. (2002). The relationship between vertical spray patterns from air-assisted sprayers and foliar deposits in vine canopies. Aspects Appl. Biol., 66, 323-330.

Qiu, Gu, J., Ding, W., L. Lv, C., & Sun, J. L. (2015). Experiment on control effect of different pesticide concentration using air-assisted sprayer. [in Chinese with English abstract]. Trans. Chinese Soc. Agric. Machinery, 46(1), 94-99.

Solanelles, F., Escolŕ, A., Planas, S., Rosell, J. R., Camp, F., & Grŕcia, F. (2006). An electronic control system for pesticide application proportional to the canopy width of tree crops. Biosyst. Eng., 95(4), 473-481. http://dx.doi.org/10.1016/j.biosystemseng.2006.08.004

Takafumi, T., Junko, Y., Yasuo, T.(1998). Measurement of forest canopy structure with a laser plane range-finding method development of a measurement system and applications to real forests. Agricultural and Forest Meteorology, 91:149-160. http://dx.doi.org/10.1016/S0168-1923(98)00081-1

Wei, J., & Salyani, M. (2004). Development of a laser scanner for measuring tree canopy characteristics: Phase I. Prototype development. Trans. ASAE, 47(6), 2101-2107. http://dx.doi.org/10.13031/2013.17795

Zhang, J., Cao, Z., Geng, C., & Li, W. (2009). Research on precision target spray robot in greenhouse. [in Chinese with English abstract]. Trans. Chinese Soc. Agric. Eng., 25(S2), 70-73.

Zhu, H., Altland, J., Derksen, R. C., & Krause, C. R. (2011). Optimal spray application rates for ornamental nursery liner production. HortTechnol., 21(3), 367-375.