Precision Application of Pesticides in Orchards Adjusting Liquid Flow Jordi Llorens Calveras, Andrew Landers and William Larzelere Dept. of Entomology Cornell University, NYSAES Geneva, NY 14456 This work supported in part by the New York Apple Research and Development Program In many orchards, canopy sprayers frequently create drift that reduces deposition efficiency, affects off-target crops, creates environmental pollution and operator and machine contamination. We are developing a system of real-time liquid flow adjustment based on canopy volume as a means to improve deposition. An array of ultrasonic sensors detects canopy size, providing information to an electronic controller which adjusts air and liquid flow. GPS can be used to determine the location of the sprayer within the field in relation to known environmental hazards such as watercourses, susceptible crops and neighbors, thus allowing air and liquid flow to be adjusted. The application of pesticides has been of concern for many years, particularly methods of reducing drift and improving deposition. There are many inter-related factors which affect spray application depending upon the target, the efficacy of the spray, the attitude of the operator, the standard of management, the weather etc. The operation of the canopy sprayer often leaves much to be desired in orchards. Most growers know that there are three factors which affect application rate: for ward speed, nozzle size and system pressure but often overlook the factors which help get the spray onto the target: airflow, liquid flow, forward speed and canopy structure. Adjusting forward speed, air and liquid flow to match the growing canopy as the season progresses is the key. Knowing how much spray has been applied is very useful for farm management purposes, to ensure every row has been sprayed and also for traceability to ensure no rows are double-dosed. We have been developing a system of real-time liquid flow adjustment based on canopy volume as a means to improve spray deposition and reduce drift. This paper describes our progress to date. Canopy sensing For vegetation detection, the sensing system is based on an array of ultrasonic sensors that are capable of detecting the tree canopy, similar to the system proposed by Solanelles and Gil in their studies (Solanelles et al., 2002; Gil et al., 2007). For this study, one array of 6 sensors was mounted on a verti- cal mast with 27 inches (70cm) between the sensors to ensure no interference between the sensor signals. With this configuration the system could detect vegetation up to a height of 11.5 feet (3.5 m). An outdoor-use and low cost ultrasonic sensor was used to sense the canopy, (model MB7052 MatBotix Inc.). To protect the electric connections a sensor body was constructed with PVC fittings and a cable connector used to protect the sensor under field conditions. The process was controlled by a DAQ (Data Acquisition) system mounted on the sprayer; an Arduino Board (Open Source I/O Interfacing, Model Arduino UNO, Arduino, Italy) has used to control the system using software developed at Cornell University. A Global Positioning System (GPS) provided location information for the sprayer. Liquid flow The canopy of fruit trees changes in size and density as the growing season progresses. The amount of spray needed to adequately cover the target is often a point of discussion, but all agree that good coverage is essential and the amount of spray required varies due to many horticultural factors such as variety, trellis design and growth stage. As many orchards have different size crops, growth stages and row widths, changing liquid flow rate to match the varying parameters is very important. In order to adjust liquid flow rate for the canopy, current practice is to change tractor forward speed, adjust sprayer pressure or change nozzles. Chang- Figure 1. Diagram of the fully automatic precision orchard sprayer. NEW YORK FRUIT QUARTERLY. VOLUME 21. NUMBER 4. WINTER 2013 7
Figure 2. Ultrasonic sensor. Manufacturer Models Calibration Degree of protection Power supply Output MaxBotix Inc. MB7052 and MB7092 Real time auto calibration ing forward speed is simplest but also affects air penetration. Changing pressure only has a minor effect on output so changing nozzles is the preferred method. Unfortunately changing nozzles is time-consuming, exposes the operator to potentially dangerous pesticide residues and is regarded as a chore by most operators. The selection of the correct nozzles to ensure good coverage is essential for precision spraying if sprays are to provide effective treatment of pests and diseases. Modern nozzle catalogues contain information on spray quality to allow growers to select the correct nozzles to match label recommendations on spray quality. Air induction nozzles reduce drift considerably and provide good coverage in the canopy, (Landers and Schupp 2001). Grower experiences of air induction (a.i.) nozzles, on both herbicide sprayers (flat fan a.i) and canopy sprayers (hollow cone a.i) confirm our trials. In an attempt to automate nozzles to provide variable flow according to the tree canopy, we conducted field trials over 3 seasons in citrus Figure 3. Left side, figure of tower sprayer with detail of mounted Lechler VarioSelect unit. Volume rate (L/min) 60 50 40 30 20 10 IP67 Beam angle 10º 3V DC to 5 V DC Analogic Max. Range 300 cm (at 5V DC ) Ultrasound frequency 10 Hz (MB7092) and 7,5 Hz (MB7052) trees located at Southern Gardens Citrus, Florida 0 30 PSI 40 PSI 50 PSI 60 PSI 30 PSI 40 PSI 50 PSI 60 PSI 30 PSI 40 PSI 50 PSI 60 PSI 30 PSI 40 PSI 50 PSI 60 PSI 30 PSI 40 PSI 50 PSI 60 PSI 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Manifolds and Pressure using a combination of Wilger COMBO-RATE nozzles with the Capstan Synchro system. This pulsing nozzle system allows for a wide range of continuous flow rates from the Wilger nozzle while holding the spray pressure at a constant 10 bar (150psi). A total of 20 Wilger nozzles and Capstan Synchro valves were installed on a Durand Wayland axial fan canopy sprayer. Wilger Capstan nozzles installed on an autonomous tractor and automatic sprayer proved to be extremely quick, reliable and allowed for individual nozzle flow-rate changes on the move, Landers et al (2012). Following on from our citrus research we are currently conducting research on apples at Cornell University to develop methods which allow adjustment of liquid flow for an orchard sprayer (Llorens et al, 2013). Adjustment of liquid flow is made via information provided by a sensing system that scans the vegetation, and by the use of one multiple array of ultrasonic sensors. In general this project aims to follow the principles of variable rate technology (VRT). We have fitted an air-assisted sprayer, John Bean Redline Tower sprayer (Durand Wayland, La Grange, GA, USA) with a Lechler VarioSelect system for proportional liquid application (Lechler USA, Franklin, TN, USA). The mounted system is based on thirteen blocks (at five different heights or manifolds) each with space for up to four nozzles. Every manifold and combination of nozzles is activated in groups by a pneumatic system mounted on the sprayer. These nozzles can be operated individually or in groups, and the system permits the use of the best combination of three nozzles. This sprayer was prepared only for applying on the right side and was equipped with three flat fan nozzles (Position A:110-01 Orange Position B: 110-015 Green and Position C: 110-02 Yellow). Every manifold and combination of nozzle was activated in groups by a pneumatic system mounted on the sprayer. Figure 4 represents the capacity of adjusting the volume rate of the VarioSelect system. Field trials Deposition on the fruit and leaves was measured by adding a determined quantity of Tartrazine into the sprayer tank. Quality of application was measured by picking leaves and fruit from specific areas of the trees. The extraction of deposited tracer on the fruit, leaves and pipe cleaners was 8 NEW YORK STATE HORTICULTURAL SOCIETY COMB_1 COMB_2 COMB_3 COMB_4 COMB_5 COMB_6 COMB_7 Figure 4. Graphic of the capacity of the orchard sprayer equipped with the Lechler VarioSelect system. The checked box represents a working manifold or nozzle. This figure represents the capacity of the system determined by rinsing them with de-ionised water and the rinsate collected and then analysed for concentration. Results were expressed in terms of deposition per area of leaf surface (micrograms/cm 2 ) and expressed by normalizing in a way that permits the comparison of results between treatments. Full details of the procedures used can be found in the reference Landers (2012).
Table 1. Spray application parameters and savings. Flow Ground Applied Pressure rate speed Volume Savings Treatment Configuration PSI GPM MPH GPA % T1 Conventional 101 3.12 2 112.89 22.7 T2 Proportional 93-126 0.68-6.27 2 87.2 Figure 5. Tower sprayer with all the electronics and nozzle blocks implemented. Results These preliminary results show that the variable application rate system on the orchard sprayer is able to attain different flow rates depending on what combination of nozzles are working. This wide range of flow rate (from 0.026 to 6.6 gpm at 60 psi, (0.1 to 25 L/min at 4 bar), permits an adjustment of the volume applied according to the canopy detected. In Table 1 we can see the basic parameters used during the testing of the system. When the sensing system detects the canopy we can save 22.7 % of the volume rate applied compared to the conventional sprayer. These savings are the result of adjusting the flow rate for each point of the application. Figure 7 shows these adjustments. This adjustment can be very important when trees are absent or when the target is very small. The conventional application (represented with the red line in Figure 7), always applies a constant amount of liquid. In terms of product deposition (yellow tracer in the trials) on the leaves, the general results show no differences in normalized deposition between the two methods (Figure 8), indicating that the volume reduction method, using information obtained with the ultrasonic sensors, performed really well. These results are very encouraging and indicate that continuing this research using new technologies will improve spray application. Discussion The ultrasonic sensors are mounted and configured for an accurate reading of vegetation in fruit trees. The electronic system is based on an Arduino board that is able to control the different sectors and nozzles of the Lechler VarioSelect system. The same system is also able to operate the position of the actuator on the air adjusting louver system as described by Landers (2011). The electronic system can register data from all the systems via a serial port operating at a maximum frequency of 2 Hz. Figure 6. Sprayer in the field test. Conclusions Precision application of pesticides in orchards provides the grower with better crop protection, less environmental pollution and better use of resources. Our field trials show that the use of mechanical and electrical techniques can significantly improve deposition and reduce drift. Normalized deposition 0.2 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 CONV PROP Apple treatments Figure 7. Graph represents the adjustment of the flow rate applied (dotted line) using information from the ultrasonic system to calculate the theoretical flow rate (blue line). We can compare the variation between the proportional application versus the continuous flow rate applied in the conventional application (red line). Figure 8. Normalized spray deposition between treatments. No statistical differences between applications. NEW YORK FRUIT QUARTERLY. VOLUME 21. NUMBER 4. WINTER 2013 9
Acknowledgements The authors wish to acknowledge funding for this project from The New York Apple Research and Development Program. We would also like to thank Lechler USA, for providing a complete VarioSelect system and Durand Wayland for supplying the sprayer. Literature Cited Gil, E., A. Escolà, J.R. Rosell, S. Planas, and L. Val. 2007. Variable rate application of plant protection products in vineyard using ultrasonic sensors. Crop Protection 26(8): 1287 1297. Landers, A.J. 2011. Improving spray deposition and reducing drift airflow adjustment is the answer. NY Fruit Quarterly 19(4):3-6. Landers, A.J. 2012. Drift from fruit sprayers-why not prevent it at source. In: Aspects of Applied Biology 114. International advances in pesticide application. Landers, A.J. and Schupp, J.R 2001. Improving deposition and reducing drift in orchards. NY Fruit Quarterly. 9(1):3-6. Landers A.J., Larzelere, W and Muise, B. 2012. Advances in autonomous pesticide application technology for orange groves In: Aspects of Applied Biology 114. International advances in pesticide application. pp. 91-98. Llorens, J., Landers, A. and Larzelere, W. 2013. Digital measurement and actuators for improving spray applications in tree and vine crops. Proc. 9th European Conference on Precision Agriculture. July 7-11, 2013. Lleida, Catalonia, Spain Solanelles, F., S. Planas, A. Escolà, and J.R. Rosell. 2002. Spray application efficiency of an electronic control system for proportional application to the canopy volume. p. 139 146. In Aspects of Applied Biology. Andrew Landers is a Senior Extension Associate who leads Cornell s spray application technology program. Jordi Llorens is a postdoctoral research fellow who works with Andrew Landers. Bill Larzelere is a technician who works with Andrew Landers. NOW YOU SEE EM NOW YOU SEE EM NOW YOU DON T When it comes to financing agriculture, we don t believe in disappearing acts. Some lenders say it s a good time to be a part of Northeast agriculture. It s true. The difference is we ll still be here when they want no part of it. Farm Credit East. When you re in agriculture for keeps, it s good to know your lender is, too. 10 NEW YORK STATE HORTICULTURAL SOCIETY
NOW YOU SEE EM NOW YOU DON T s. farmcrediteast.com 800.562.2235 NEW YORK FRUIT QUARTERLY. VOLUME 21. NUMBER 4. WINTER 2013 11
Kinz Email: Skinne@columbiatractor.com Web: www@columbiatractor.com As we bid 2013 farewell and anticipate the arrival of 2014, Crop Production Services would like to thank you for your patronage. May 2014 be safe and prosperous. Merry Christmas and Happy New Year! Distributor for: Adjuvants I Chemicals I Nutritionals Seed I Seed Treatments 140 Office Parkway Pittsford, NY 14534 (585) 586-1330 www.cpsagu.com 12 NEW YORK STATE HORTICULTURAL SOCIETY