Evaluation of process-driven spray retention model on ear-ly growth stage barley

The efficiency of spray application of foliar plant protection products with hydraulic nozzles on vertically oriented and hydrophobic plants at early growth stages can be very low. The spray retention by crop leaves is affected by application parameters resulting from nozzle kind, size and operating pressure as well as spray mixture physicochemical properties. When optimizing the spray application, such targets are often used to perform retention trials for comparative purpose, i.e. indoor grown monocotyledonous at two leaves stage. A typical arrangement consists in spraying few plants sufficiently spaced underneath the nozzle to avoid interference due to secondary droplets from impacts on other plants. However, retention trials turn out to ineffective for significantly discriminating between application methods and mixtures due to the high variability between trials resulting from the different droplets retained by each plant. An alternative to retention trials is to tackle spray retention with a physical approach at the droplet scale. Such tests are often performed using high speed imaging with high magnification optics to characterize droplet impacts; adhesion, rebound or shatter on small excised leaf areas and neglect, however, the overall plant architecture. The aim of this paper is to evaluate a droplet interception model connecting actual spray retention with process-driven retention models. In this study, barley plants (BBCH11) were sprayed with 2 formulations using the same nozzle. The actual spray retention was assessed by dosing a fluorescent tracer added to the sprayed mixture. The plants were placed linearly below the center of a single moving nozzle during sprayings. Each plant was reconstructed in 3D afterwards using a structured light 3D scanner and used as input for the model. A virtual nozzle was built on the base of droplet size distributions measured with high speed shadow imaging by performing an adjustment of the distribution by the method of moments. A ran-dom droplet distribution was allocated for each spraying of a barley plant. Droplet velocities were given to droplets on the basis of the droplet velocity – diameter correlation by resolving the droplet transport equations for different droplet sizes. Initial droplet positions were ran-domly given. The interception model is based on a mathematical formalism for the intercep-tion between triangles of the 3D plant and droplet directions. If the droplet impacts a leaf, the amount actually retained by the leaf was computed on the basis of the droplet impact energy and impact behavior from experiments with high speed shadow imaging. In conclusion, the interception model allowed determining the spray retention by plants and discriminating ap-plication parameters by explaining the variability resulting from various droplet size distribu-tions intercepted by single plant.