Assessment of pesticide application quality with a manual sprayer in spinach Received for publication: 30 January,2016. Accepted for publication: 30 November, 2016. Doi: 10.15446/agron.colomb.v34n3.54453 1 Department of Civil and Agricultural Engineering, Faculty of Engineering, Universidad Nacional de Colombia. Bogota (Colombia). bcastilloh@unal.edu.co Agronomía Colombiana 34(3),346-354 2016 Assessment of pesticide application quality with a manual sprayer in spinach Evaluación de la calidad de aplicación de una pulverizadora de acción manual en espinaca Juan David Sesquile1 and Bernardo Castillo1 ABSTRACT RESUMEN Two experiments were conducted to evaluate the performance of three hollow cone nozzles (TeeJet® TX 800050 VK, TeeJet® TXA 8004 VK and TX-Royal Cóndor®) and one f lat fan nozzle (TeeJet® XR 8004 VS) with two manual application techniques on a crop of spinach (Spinacia oleracea L.). In order to assess the quality of the application of pesticides, WSP (water sensitive paper) collectors and f luorescent tracer Tinopal® CBS-X were used. In one of the trials, percentage of tracer retained by the leaf surfaces was also determined. In this study, and based on the methodology of collectors, it was observed that the technique of applying two passes with the TX-Royal Condor® nozzle could be recommended for the application of pesticides with a hand-operated sprayer in the spinach crop. However, this was not corroborated by the f luorescent tracer technique. En esta investigación se evaluaron tres boquillas de cono hueco (TeeJet® TX 800050 VK, TeeJet® TXA 8004 VK y TX-Royal Cóndor®) y una boquilla de abanico plano (TeeJet® XR 8004 VS), con dos técnicas de aplicación con el objetivo de mejorar la calidad de aplicación de plaguicidas que se realiza en los cultivos de espinaca (Spinacia oleracea L.). Se realizaron dos ensayos de campo y se emplearon dos técnicas de evaluación de las aplicaciones: El método de los colectores de papel hidro- sensible y el método del trazador f luorescente con Tinopal® CBS-X, el cual se usó para determinar los depósitos de trazador sobre las hojas en uno de los ensayos. En este estudio, y con base en la metodología de colectores se observó que la técnica de aplicación de dos pases con la boquilla TX-Royal Cóndor® podría recomendarse para la aplicación de plaguicidas con una pulverizadora de accionamiento manual en el cultivo de la espinaca. Sin embargo, esto no fue corroborado con la técnica del trazador f luorescente. Key words: collectors, nozzles, spraying techniques, Tinopal® CBS-X, water sensitive paper. Palabras clave: colectores, boquillas, técnicas de pulverización, Tinopal® CBS-X, papel hidrosensible. Normally, the optimum characteristics of pesticide ap- plication are established using parameters such as: DV0.1, DV0.5, DV0.9, which represent the distribution of the droplet diameters such that droplets with a diameter smaller than DV0.1, DV0.5, and DV0.9 compose 10, 50 and 90% of the total liquid volume, respectively; droplet density per unit area and coverage percentage (Teixeira, 2010). For example, when using artificial collectors such as: WSP or PVC sheets, a coverage of 20% is proposed as the minimum acceptable level (Magdalena, 2004), although it is believed that a 30% coverage ensures a satisfactory control (Holownicki et al., 2002), or in a specific case, with PVC sheets, that a 40% coverage would be considered good since 95% mortality of california red scale in state 1 was obtained with S. Ultra-Fine oil (Castillo, 2005). Lately, it has been stated that a coverage percentage between 20-50% can be considered as good. Below 20%, treatment is considered insufficient, above 50 to 80% in excess, and for over 80% as overdose (Mangado et al., 2013). Introduction The need to make proper use of pesticides has led agri- culture to the implementation of appropriate application techniques for crop and pest requirements (Foqué and Nyuttens, 2011). This has encouraged a number of studies on methods for evaluating the application of pesticides. Probably the most popular method is the water sensitive paper collectors (Hoffmann and Hewitt, 2005), since it is a technique that provides accurate information (Cunha et al., 2011), is easy to implement in the field and economical. Nevertheless, the retention capacity of the droplets by the leaves is different from the retention capacity of water sensi- tive paper collectors (Holownicki et al., 2002). Therefore the method of f luorescent tracers is acquiring a great interest today because it is a practical and accurate method that allows quantitative and qualitative assesses of the amount of pesticide that was deposited on plants after spraying (Palladini et al., 2005). 347Sesquile and Castillo: Assessment of pesticide application quality with a manual sprayer in spinach Moreover, when natural collectors, such as leaves, are used to evaluate f luorescent tracer deposits, classes have been handled with qualitative visual ratings in relation to the amount of coverage, as pointed out Firveda et al. (2002). The use of f luorescent tracers to quantify the deposits has been a common methodology in recent years to establish the quality of applications and models of biological ef- fectiveness (Holownicki et al. 2002; Derksen et al., 2007; Fritz, et al., 2011). The objective of this study was to evaluate the quality of spray application of a knapsack sprayer of manual action in a crop of spinach, and for this two methods were used: the first one with water sensitive paper collectors and the second one with a f luorescent tracer. Materials and Methods Two experiments were conducted in 2011, at the Agri- cultural Center Marengo (CAM) located in Mosquera, Cundinamarca (Colombia). In an area of 790 m2, eight experimental units were established, with a sowing density of 22 plants per m2. Spinach (Spinacia oleracea L.) Hybrid 424 Grendell was planted. The sprays were made with a knapsack sprayer of manual action of local manufacture, since this type of equipment is suitable for undersized and hilly areas, which use is no- torious of Colombian small farmers (Matthews, 2008). The equipment used consists of a tank of 20 L, with a plunger pump and copper spray extension. The nozzles used are described in table 1. High discharge nozzles were chosen to obtain a higher coverage and higher density of drops. Additionally, for the same discharge f lat fan nozzle and a cone hollow are compared. Also, low discharge nozzles (both hollow cone), were selected in order to save time and water maintaining efficiency, also it was made possible to extend the range of application rates. Now, as the width of the bed was 90 cm and all the nozzles had the same spray angle (80°), then the operator held nozzle height between 50 and 55 cm over the plants of spinach. In order to avoid unfavorable weather conditions, such as high temperature, low relative humidity and strong winds, applications were carried out in the morning between 8 a.m. and 10 a.m. for both field trials. Measurements of air temperature, relative humidity of the air and wind speed were not made. Experiment 1 The first experiment was carried out from February to April 2011, and the experimental design was a randomized complete block with a factorial arrangement of 4 nozzles for 2 application techniques for 2 leaf surfaces (upper side and underside), with 5 repetitions. Application techniques consisted of a pass or run applying the pesticide on a bed, and the two passes consisted of a run outward and a return applying the pesticide on the same bed. TABLE 1. Characteristics of the nozzles used. Nozzles Type Discharge (L min-1) Filter mesh size2 Bar 4 Bar Low discharge TeeJet® TXA 800050VK Hollow cone 0.16* 0.22* 100* TX-Royal Cóndor® Hollow cone n.d. 0.33° 100 High discharge TeeJet® TXA 8004VK Hollow cone 1.29* 1.82* 50* TeeJet® XR8004VS Flat fan 1.29* 1.82* 50* *: taken from Catalog 51-ES (Spraying Systems Co.®, 2011). °: obtained in field trials with the equipment without movement. n.d.: no data. The crop was affected by downy mildew disease, which is caused by the fungus Peronospora farinosa, and its action threshold was the mere presence of the disease (Flórez and Segura, 2010). Therefore, the fungicide Metalaxyl-M 4% + Mancozeb 64% was selected, with a dose of 250 g of fungicide per 100 L of water. To assess the quality of application two qualitative meth- ods were used: WSP collectors and a f luorescent tracer Tinopal® CBS-X (with a dose of 10 g L-1) was used, since this does not present a toxicological risk to humans (Bier- man et al., 1998) and is soluble in water (Derksen et al., 2007), this technique allowed to use spinach leaves as collectors (Fig. 1). To perform the image analysis of collectors, field samples (5 WSP collectors 3.5 cm x 2.6 cm by nozzle, technique and leaf surface, and 5 spinach leaves by nozzle, technique and leaf surface) were taken to a dark room, where photographs were taken with experimental setup: Two light sources and a (Samsung PL150) digital camera with a resolution of 600 dpi, attached to a tripod (Firveda et al., 2002),. Then with the ImageJ 1.45S program, the area and diameter of each of the stains was determined. To determine the diameter of the drop (DDrop) that generated the stain on the WSP collector, the equation of “DropletScan™” program was used, which was presented by Hoffmann and Hewitt (2005), and that is: 348 Agron. Colomb. 34(3) 2016 DDrop = Diameter of stain (1) 1,6333 + (0,0009 × diameter of stain) Subsequently the following data were determined: DV0.5 (volume median diameter in μm) of drops and stains on leaves, density of drops and stains, and the coverage per- centage for collectors and leaves. Experiment 2 It was conducted from July to September 2011, and the experimental design was randomized complete block with a factorial arrangement of 4 nozzles for 2 application techniques (which were mentioned previously), for the 2 surfaces of the leaf with 6 replicates. This crop was attacked by leafminer (Liriomyza huidob- rensis Blanchard), with a percentage of infestation of 35%, which is below the threshold action, which is 50% of the leaves with eggs or galleries (Flórez and Segura, 2010). However, it was considered necessary to perform control so as not to compromise crop yield. Insecticide (14.1% thiamethoxam + 10.6% Lambda-cyhalothrin was used; with a dose of 100 cm3 ha-1. upper side and they were four cuttings of 1 cm diameter to each one. Then, to remove the tracer deposited in the four cuttings of each leaf, one side of the cutting was washed for one minute, using a clamp and a Pasteur pipette with 10 mL of washing solution. The washing solution obtained was stored in jars lined with foil, and placed in polystyrene fridge boxes to 6°C, to be transported to the Laboratory of Biotechnology of the National University of Colombia, Bogotá. Regarding the samples on the underside, were selected 3 of the 6 leaves that had been photographed on the underside, and the process is repeated. Three mL of each of the washing solutions were taken and decanted for 45 minutes in test tubes. The absorbance of each solution was then quantified with a spectrophotom- eter (SmartSpect™ Plus, Bio-Rad). The absorbance values of the replicates were averaged and corrected by the action of photodegradation. By using a calibration curve the tracer concentration in the solution was calculated, and employing the equation 2 (Fritz et al., 2011), deposits were calculated. DSample = Cwashing sol. × Vwash (2) Acollector Where Dsample: Deposit in the sample (L cm-2) Cwashing sol: Concentration of the washing solution (ppm) Vwash: 10 mL Acollector: 3.1416 cm2 Meanwhile, the percentage of tracer retained by the spinach leaf surfaces after sprayer was calculated as the difference (in percent) between the tracer doses (μg cm-2) applied and the amount of tracer deposited in the leaves. The results were analyzed by multi-factor ANOVA (GLM model), in order to study the differences between factor levels LSD method was used for a confidence level of 95%. A simple linear regression was also performed to relate the covering variables between artificial and natural collectors. Results and discussion For nozzles 3 and 4, the use of artificial collectors proved to be inconvenient. Since for both application techniques (one and two passes), all the upper leaf collectors got coverage level percentage above 60%, which did not allow to assess correctly the diameter and number of impacts, they had many overlapping stains. Authors like Fox et al. (2001), Fox FIGURE 1. Collectors. A: Spinach leaf with tracer. B: WSP card. To assess the quality of applications, WSP collectors and the qualitative method of f luorescent tracer were used as indicated in Expt. 1. In addition, tracer deposits on both sides of the leaf were evaluated with 3 replications. Tracer concentration was 10 g of Tinopal. CBS-X per liter of water. Sampling by the upper side leaf, was as follows: they were selected 3 of the 6 leaves that had been photographed by the A B 349Sesquile and Castillo: Assessment of pesticide application quality with a manual sprayer in spinach et al. (2003) and Cunha et al. (2011) reported similar results. As for the classification of droplet sizes of nozzles, only collectors that were located on the underside of the leaves were considered. According to the methodology of “British Crop Protection Council (BCPC)”, used by Nuyttens et al. (2007), to classify droplet sizes of the nozzles, assessment of the droplets size of the nozzles was made (Tab. 2). The average data of field sprayer calibrations are also presented in table 2. The TX-Royal Cóndor and TXA 800050 VK nozzles had a medium and fine droplet size respectively in Expt. 1. But for Expt. 2, the droplet sizes were very fine and very fine, respectively. This behavior can be attributed to the efficient operation of the knapsack sprayer, since these devices have a quite variable application pressure (Kromann et al., 2008). Also the operator affects the application, as this person regulate simultaneously: the speed, the height and position of the spray extension, and in accordance with Rodriguez et al. (1994) and Beltrán et al. (1994), factors such as training, fatigue, the operator’s mood have significant inf luence on the quality of spray application. Also, it is worthy mention that droplets mea- sured are those impacted on WSP collector, therefore, the observations correspond to a portion of the total droplets sprayed from the nozzle. Moreover, to meet the GLM model hypothesis, it was necessary to perform logarithmic transformation of DV0,5 parameter drops and stains density. It was required to transform the data by using the cube root of the vari- ables: drops and stains, coverage percentage of leaves and collectors. Experiment 1 Table 3 shows that factors: nozzle, application technique and leaf surface have significant inf luence on variables DV0,5 of stains, density stains and the coverage percentage of leaves and collectors. The factor application technique does not affect the DV0,5 of the drops (for this variable the factor leaf surface was not taken into account because only the collectors were considered on the underside). In figure 2, it is observed that in upper side of leaves mean DV0,5 stain sizes are significantly larger than the sizes of DV0,5 on the underside of the leaves, for all nozzles. Mean- while, Figure 3 shows that the coverage percentage of the upper side of leaf is significantly higher than the coverage percentage on the underside of leaves, similar results were reported for Derksen et al. (2001), Olivet (2009), Foqué and Nyuttens (2011). Moreover, it is observed that the higher application rate, the higher coverage is obtained, which is consistent with what was reported by Derksen et al. (2001) and Holownicki et al. (2002). According to Figure 3, notice that in upper side of leaves high discharge nozzles produce a medium coverage rate significantly higher than the low discharge nozzles do, and in turn the nozzle 2 has TABLE 2. Results of the calibration of the backpack sprayer manually operated. Nozzle Nozzle 1 Nozzle 2 Nozzle 3 Nozzle 4 TXA800050VK TX-Royal Cóndor TXA8004VK XR8004VS Application technique 1 pass 2 passes 1 pass 2 passes 1 pass 2 passes 1 pass 2 passes Pressure (Bar) 3.8 - 4.1 3.8 - 4.1 3.8 - 4.1 3.8 - 4.1 2.0 - 2.4 2.0 - 2.4 2.0 - 2.4 2.0 - 2.4 Experiment 1 Droplet size Fine Medium Medium Coarse Application rate (L ha-1) 105.9 187 131 272.7 523 951.1 530.4 963 Speed (km h-1) 1.709 1.818 1.709 1.782 1.69 1.8 1.71 1.806 Experiment 2 Droplet size Very fine Fine Medium Coarse Application rate (L ha-1) 101.1 188.9 151.1 272.2 550.2 1,014.4 537.8 1,025.6 Speed (km h-1) 1.756 1.765 1.731 1.765 1.718 1.769 1.711 1.743 TABLE 3. Statistical significance of the main effects on the dependent variables of the experiment 1. The confidence level is 95%. Factors DV0,5 Drops Density of drops Coverage of collectors DV0,5 Stains Density of stains Coverage of leaves F P F P F P F P F P F P Nozzle 5.28 0.005 3.99 0.017 28.50 0.000 59.69 0.000 10.39 0.000 41.53 0.000 Technique 3.13 0.087 8.52 0.007 22.69 0.000 5.35 0.024 6.62 0.013 12.16 0.001 Leaf surface No data No data 219.93 0.000 151.12 0.000 59.13 0.000 201.52 0.000 Repet. 1.27 0.268 0.26 0.614 0.23 0.633 0.05 0.826 0.43 0.515 0.21 0.647 350 Agron. Colomb. 34(3) 2016 Lo g 1 0 D v 5 0 Nozzle 2.2 2.6 3.4 3.0 3.8 0 1 2 3 4 6 UndersideUpper side R oo t ( % c ov er ag e) Nozzle 0 1 3 2 4 0 1 2 3 4 6 UndersideUpper side R oo t ( % c ov er ag e) Nozzle 0 1.5 1.0 0.5 2.5 2.0 3.0 0 1 2 3 4 6 2 Passes1 Pass R oo t ( % c ov er ag e) Nozzle 0 1 4 3 2 5 0 1 2 3 4 6 UndersideUpper side R oo t ( % c ov er ag e) Nozzle 1.3 2.8 2.3 1.8 3.8 3.3 4.3 0 1 2 3 4 6 2 Passes1 Pass FIGURE 2. Interaction between the nozzle and leaf surface factors for DV0,5 of stains on crop leaves in experiment 1. FIGURE 3. Interaction between the nozzle and leaf surface factors for the coverage percentage of leaves in experiment 1. FIGURE 4. Interaction between the nozzle and application technique fac- tors for the coverage percentage of leaves in experiment 1. FIGURE 5. Interaction between the nozzle and application technique fac- tors for the coverage percentage in WSP collectors in trial 1. FIGURE 6. Interaction between the nozzle and leaf surface factors for the coverage percentage in WSP collectors in trial 1. a significantly greater percentage than the medium cover- age of nozzle 1. Hollow cone and f lat fan nozzles of high discharge do not differ in the size of the impacts and the coverage reached. All this, when crop leaves are used as collectors. Figure 4 shows that only the nozzle 3, for two passes, has a significantly higher coverage percentage than when a single pass is applied. In figure 5 it is noted that the collectors of WSP placed on the upper side of leaves have a significantly greater cover- age percentage than the collectors that were located on the underside of leaves, which coincides with that reported by Sumner et al. (2000), Magdalena (2004), and Holownicki et al. (2002). There is a strong agreement between the behavior of WSP collectors and the leaves, in connection with the coverage, if figures 3 and 5 are compared, and 4 and 6. 351Sesquile and Castillo: Assessment of pesticide application quality with a manual sprayer in spinach Besides, in figure 6, it is noted that the nozzles 2 and 3 differ significantly between application techniques, by implementing two passes the coverage percentage is greater than when only one pass is made, demonstrating that ap- plication volume rate is more important than nozzle type or the droplet size, in order to achieve a better target covering. Moreover, in figures 3 and 5, it is observed that, on the underside of the leaves of spinach, higher coverage rates are achieved with medium and large droplet nozzles. This indicates that under field conditions and sparse foliage, me- dium and large droplet nozzles achieve a possible adequate minimum deposits on underside of leaves. Regarding of parameters used to evaluate the quality of the application, the coverage percentage proved to be the most reliable and consistent, coinciding with those reported by Salyani and Fox (1999) and Holownicki et al. (2002). Experiment 2 In table 4 it is observed that factors nozzle, application techniques and leaf surface significantly inf luence DV0,5 of stains, density of stains and coverage percentage on the leaves and collectors. Meanwhile, the nozzle factor signi- ficantly inf luences the DV0,5 of drops. Figure 7 shows that for two passes, the medium density of stains is significantly higher than when only one pass is applied. Thus, the increase in the application rate causes an increase in the number of stains per unit area, particularly nozzle 2 (273 L ha-1), which is consistent with that reported by Rodríguez et al. (1994). With regard to figures 8 and 9, it is seen, again, as in experiment 1, the nozzles 2 and 3, for two passes, have a mean coverage percent greater than when applied one pass, giving preference to the use of larger volumes of application. Meanwhile, to the nozzles 1 and 4 there were no significant differences in coverage for this increase in volume, possibly due to the small size of the droplets in the nozzle 1 and conversely, in the case of the nozzle 4, the oversized drops could generate runoff losses, these results are similar to those reported by Koch and Weisser (1998) and Salyani et al. (2013). Thus, in cases where collectors presented coverages higher than 80%, runoff signals were observed on the leaves, where the drops were accumulating on the leaf, to the point of generating deposits scurrying by leaf veins toward the ground, and therefore it was never formed a film of liquid to wet the entire surface of the leaf. Again, there is no difference between hollow cone and f lat fan nozzles of the same discharge regarding droplet sizes, spot sizes and coverages in natural and artificial collectors as in experiment 1. It is surprising not find smaller sizes and higher densities of drops and stains for more coverage TABLE 4. Statistical significance of the main effects on the dependent variables of the experiment 2. The confidence level is 95%. Factors DV0,5 Drops Density of drops Coverage of collectors DV0,5 Stains Density of stains Coverage of leaves F P F P F P F P F P F P Nozzle 11.77 0.000 1.09 0.365 27.99 0.000 115.73 0.000 3.91 0.012 42.77 0.000 Technique 2.55 0.12 0.01 0.921 7.87 0.006 12.09 0.001 5.38 0.023 18.31 0.000 Leaf surface No data No data 168.26 0.000 164.92 0.000 132.59 0.000 297.71 0.000 Repet. 0.28 0.6 0.59 0.448 0.33 0.567 2.81 0.098 0.75 0.388 0.76 0.386 FIGURE 7. Interaction between the nozzle and application technique fac- tors for spot density in experiment 2. R oo t ( N o. o f s ta in s/ cm 2 ) Nozzle 2.7 3.9 3.5 3.1 4.7 4.3 5.1 0 1 2 3 4 6 2 Passes1 Pass R oo t ( N o. o f s ta in s/ cm 2 ) Nozzle 0.6 1.5 1.2 0.9 2.1 1.8 2.4 0 1 2 3 4 6 2 Passes1 Pass FIGURE 8. Interaction between nozzle and application technique factors for the coverage percentage of leaves in experiment 2. 352 Agron. Colomb. 34(3) 2016 (except Fig. 6, two passes) on artificial and natural collec- tors as claimed by the theory of the turbulence action in the hollow cone nozzles. On the other hand, it was determined that exists a linear relationship between the coverage percentage on the leaves and the coverage percentage on the artificial collectors, and for that an analysis of simple linear regression was used, which confirmed that between these two parameters there was a significant linear relationship (P=0.000), the degree of data adjustment R2 was 0.679 in experiment 1, and 0.693 in experiment 2. This relationship is consistent with Saly- ani andWhitney (1988), and indicates that the estimation made by a coverage method can be obtained with the other. However, the coverage percentages on the leaves were lower compared with the coverage percentages on the collectors, which can be explained by the phenomenon of retention of the drops which is a very complex process, where the peculiarities of a surface affect behavior retention of the drops and the formation of deposit (Bertola, 2008). Trough quantitative fluorescent tracer technique, in table 5) the percentage of tracer retained by the leaves after spraying is presented, and it is clear the lower deposition obtained in the underside of leaves compared with the upper side of the leaves in accordance with that obtained in natural and artificial collectors. Now, if the upper and underside retention percentages are added for each nozzle and technique, table 5 shows that the application of a second pass does not achieve a gain in the percentage of tracer retained by the leaves. This behavior is probably due to, in the case of high dis- charge nozzles, part of the volume applied in the double pass is lost by runoff, which may indicate that the retention capacity of leaves of spinach is limited (Koch and Weisser, 1998; Panneton et al., 2000). Also, it was found that low discharge nozzles with fine or medium droplets reached same percentage of deposits or something more compared to high discharge nozzles; this is in disagreement with the analysis derived from the collectors. The exact explanation for this behavior is unknown, but it is certainly related with the real retention R oo t ( % c ov er ag e) Nozzle 1,2 2,7 2,2 1,7 3,7 3,2 4,2 0 1 2 3 4 6 2 Passes1 Pass FIGURA 9. Interaction between the nozzle factors and application tech- niques for the coverage percentage in WSP collectors in experiment 2. TABLE 5. Average volume tracer deposited spinach leaves. Nozzle Technique Leaf surfaces Application rate (L ha-1) Tracer dose applied (μg cm-2) Deposit recovered (μg cm-2) Tracer retained (%) surfaces add 1 1 pass Upper side 101.1 10.11 4.5 (46.4) 45,0 68,3 Underside 2.4 (2.0) 23,3 2 passes Upper side 188.88 18.888 6.9 (3.3) 36,8 53,6 Underside 3.2 (37.6) 16,8 2 1 pass Upper side 151.11 15.111 8.3 (22.3) 54,8 79,3 Underside 3.7 (57.6) 24,5 2 passes Upper side 272.22 27.222 13.1 (28.7) 48,3 67,8 Underside 5.3 (7.3) 19,5 3 1 pass Upper side 550.22 55.022 32.5 (17.2) 59,1 64,5 Underside 3 (9.2) 5,4 2 passes Upper side 1,014.44 101.444 39.6 (37.1) 39,1 56,9 Underside 18 (27.1) 17,8 4 1 pass Upper side 537.78 53.778 25.4 (34.4) 47,3 75,7 Underside 15.3 (25.9) 28,4 2 passes Upper side 1,025.56 102.556 28.5 (30.6) 27,8 42,3 Underside 14.9 (33.5) 14,5 Value in parenthesis is the coefficient of variation of deposit recovered. 353Sesquile and Castillo: Assessment of pesticide application quality with a manual sprayer in spinach capacity of droplets of different sizes. Although the use of fluorescent tracer to assess the amount of deposited product was quite wasteful methodology, it seems to be the most accurate technique for establishing the actual formation of the deposit which should determine the effectiveness of the pesticide in controlling pests. Conclusions The increase in application volume, as a result of the application of the second pass, generated a significant increase in the coverage percentage achieved by the hollow cone nozzles TX-Royal Condor and TeeJet TXA8004VK. Therefore, due to the smaller amount of water needed to prepare the mixture (273 L ha-1 of application volume) for the technique with two passes of the TX-Royal Condor nozzle, it could be suggested as having the best charac- teristics for the application with a hand-operated back sprayer, compared to the application volume (951 L ha-1) that the equipment would use with theTeeJet TXA8004VK nozzle. The above according to the results obtained with the artificial and natural collectors. However, this pos- sibility is not verified with the information obtained with the quantitative methodology of the f luorescent tracer. Although it is not known the exact causes of this situation, it may be advisable to prefer the results achieved with this type of technique when establishing real recommenda- tions on the best nozzle combination, manual spraying equipment and application technique. Literature cited Bertola, V. 2008. Some applications of controlled drop deposition on solid surfaces. Recent Patents Mech. Eng. 1(3), 167-174. Doi: 10.2174/2212797610801030167 Bierman, E.P.B., D.H. Brouwer, and J.J. Van-Hemmen. 1998. Imple- mentation and evaluation of the f luorescent tracer technique in greenhouse exposure studies. Ann. Occup. Hyg. 42(7), 467-475. Doi: 10.1093/annhyg/42.7.467 Castillo, B. 2005. Modelización de la pulverización de aceites mine- rales y su relación con el control del piojo rojo de california. PhD. thesis. Departamento de Mecanización y Tecnología Agraria, Universidad Politécnica de Valencia, Valencia, Spain. Cunha, M., C. Carvalho, and A. Marcal. 2011. Assessing the ability of image processing software to analyse spray quality on water- sensitive papers used as artificial targets. Biosyst. Eng. 111(1), 11-23. Doi: 10.1016/j.biosystemseng.2011.10.002 Derksen, R.C., S.A. Miller, H.E. Ozkan, and R.D. Fox. 2001. Spray de- position characteristics on tomatoes and disease management as inf luenced by droplet size, spray volume, and air-assistance. Proc. ASAE Annual International Meeting, Paper No. 011120. Doi: 10.13031/2013.7351 Derksen, R.C., H. Zhu, R.D. Fox, R.D. Brazee, and C.R. Krause. 2007. Coverage and drift produced by air induction and conventional hydraulic nozzless used for orchard applications. Trans. AS- ABE 50(5), 1493-1501. Doi: 10.13031/2013.23941 Firveda, M.C., A. Canatalogo-Jr., H.H. Ramos, A.C.L. Lino, and I.M. Corrêa. 2002. Using imagery analysis software for evaluation of spraying coverage. Bragantia 61(1), 305-310. Doi: 10.1590/ S0006-87052002000300013 Flórez, R. and M. Segura. 2010. Espinaca (Spinacia oleracea L.) producción y manejo poscosecha. Corredor Tecnológico Agro- industrial, Cámara de Comercio de Bogotá,  Bogotá, Colombia. Foqué, D. and D. Nuyttens. 2011. Effects of nozzle type and spray angle on spray deposition in ivy pot plants. Pest Manag. Sci. 67(2), 199-208. Doi: 10.1002/ps.2051 Fox, R.D., M. Salyani, J.A. Cooper, and R.D. Brazee. 2001. Spot size comparisons on oil and water sensitive paper. Appl. Eng. Agric. 17(2), 131-136. Doi: 10.13031/2013.5454 Fox, R. D., R.C. Derksen, J.A. Cooper, C.R. Krause, and H.E. Ozkan. 2003. Visual and image system measurement of spray deposits using water-sensitive paper. Appl. Eng. Agric. 19(5), 549-552. Fritz, B.K., W.C. Hoffmann, P. Jank, B. Devisetty, and S.W. Dean. 2011. A f luorescent tracer method for evaluating spray trans- port and fate of field and laboratory spray applications. ASTM Int. 8(3), 1-9. Hoffmann, W.C. and A.J. Hewitt. 2005. Comparison of three imag- ing systems for water-sensitive papers. Appl. Eng. Agric. 21(6), 961-964. Doi: 10.13031/2013.20026 Holownicki, R., G. Doruchowski, W. Swiechowski, and P. Jaeken. 2002. Methods of evaluation of spray deposit and coverage on artificial targets. EJPAU 5(1), 10. Kromann, P., D. Leon, A. Taipe, J.L. Andrade-Piedra, and G. Forbes. 2008. Comparison of two strategies for use of translaminar and contact fungicide in the control of potato late blight in the highland tropics of Ecuador. Crop Prot. 27(7), 1098-1104. Doi: 10.1016/j.cropro.2008.01.006 Koch, H. and P. Weisser. 1998. Quantitative aspects of spray f luid retention. IUPAC, 2D 024, IUPAC, 3.-7.8. 1998, London, U.K. Magdalena, J.C. 2004. Efecto de la utilización de pulverizadores de f lujo transversal e hidroneumático tradicional sobre la calidad de los tratamientos fitosanitarios en manzanos. PhD. thesis. Departamento de Mecanización y Tecnología Agraria, Uni- versidad Politécnica de Valencia, Valencia, Spain. Mangado, J., S. Arazuri, P. Arnal, C. Jarén, and A. López. 2013. Measuring the accuracy of a pesticide treatment by an im- age analyzer. Procedia Technol. 8, 498-502. Doi: 10.1016/j. protcy.2013.11.066 Matthews, G.A. 2008. Attitudes and behaviours regarding use of crop protection products - A survey of more than 8500 smallholders in 26 countries. Crop Prot. 27(3-5), 834-846. Doi: 10.1016/j.cropro.2007.10.013. Nuyttens, D., K. Baetens, M. De Schampheleire, and B. Sonck. 2007. Effect of nozzle type, size and pressure on spray droplet characteristics. Biosyst. Eng. 97(3), 333-345. Doi: 10.1016/j. biosystemseng.2007.03.001 Olivet, J.J. 2009. Optimización de los tratamientos fitosanitarios en cultivos de invernadero en Uruguay. PhD. thesis. Depar- tamento de Mecanización y Tecnología Agraria, Universidad Politécnica de Valencia, Valencia, Spain. 354 Agron. Colomb. 34(3) 2016 Palladini, L.A., C.G. Raetano, and E.D. Velini. 2005. Choice of tracers for the evaluation of spray deposits. Sci. Agric. 62(5), 440-445. Doi: 10.1590/S0103-90162005000500005 Panneton, B., A. Bélanger, C. Vincent, M. Piché, and M. Khe- lifi. 2000. Effects of water volume rates on spray deposi- t ion a nd cont rol of t a r n ished pla nt bug [Hem ipter a : Miridae] in strawberry crops. Phytoprotection 81(3), 115. Doi: 10.7202/706205ar Rodriguez, J.A., O.F. Suarez, and F.R. Leiva. 1994. Evaluación de técnicas de aplicación de plaguicidas en un cultivo de clavel (Dianthus caryophyllus L.) en la sabana de Bogotá. Agron. Colomb. 11(2), 175-183. Salyani, M. and J.D. Whitney. 1988. Evaluation of methodologies for field studies of spray deposition. Trans. ASAE 31(2), 390-395. Doi: 10.13031/2013.30720 Salyani, M. and R.D. Fox. 1999. Evaluation of spray quality by oil- and water-sensitive papers. Trans. ASAE. 42(1), 37-43. Doi: 10.13031/2013.13206 Salyani, M., H. Zhu, R.D. Sweeb, and N. Pai. 2013. Assessment of spray distribution with water-sensitive paper. Agric. Eng. Int. CIGR J. 15(2), 101-111. Sumner, H.R., G.A. Herzog, P.E. Sumner, M. Bader, and B.G. Mul- linix. 2000. Chemical application equipment for improved deposition in cotton. J. Cotton Sci. 4(1), 19-27. Spraying Systems Co. 2011. TeeJet, Catálogo 51-ES. En línea: http:// www.teejet.com/media/427750/cat51_spanish.pdf; consulted: February, 2014. Teixeira, M. 2010. Estudio de la población de gotas de pulverización. pp. 67-75. In: Magdalena, J. (ed.). Tecnología de aplicación de agroquímicos. Programa Iberoamericano de Ciencia y Tec- nología para el Desarrollo (CYTED), Buenos Aires, Argentina.