R.E. Mickle, E.G. Kettela, J. Regniere, S. Foster
A series of research sprays to evaluate new products were planned and conducted utilizing strategies to optimize on-target foliar deposit and minimize off-target drift. In support of these trials, BioSIM, a phenology model for plant and budworm development, was used for predicting optimal spray timing. Planning, staging and preliminary analyses of the aerial sprays was coordinated through GypsES, a GIS-based expert system. Survey flights along roads in the vicinity of the research blocks produced accuracies of 10-24m between aircraft DGPS positions and scanned (200 dots/in) NTSC 1:50000 georeferenced maps of the area. Comparisons of flight lines with NTDB road data increased the accuracy to 6-7m. Bracketing the spray, real-time profiles at the spray block were telemetered back from an AIMMS-10 meteorological package to a base station located 100km away. These profiles were used as input to a spray-optimization model using the spray-fate model, AgDISP. Optimized spray strategies involving multiple line spraying were then forwarded to the spray pilot who was on route to the block for incorporation into the DGPS guidance systems. Post spray, one second ASM data were used to predict deposit potential at the top of the canopy based upon the actual spray program and compared against on-site deposit measurements. For sprays to plantations, modeled deposit and drop density showed significant correlation with field measurement. Predicted block-average deposit for different application strategies of Mimic gave the same results as block-averaged measured deposit. For natural regenerated forests, the comparisons were inconclusive. The optimization spray on small blocks (width = 270m) for emission drop size distributions with DV0.5 of 100um predicted 56% more in-block deposit over conventional equal track-space strategies. Deposit variation was significantly improved from a COV of 0.52 to 0.2 for the optimization strategy. The majority of the improvement was due to predicted deposit along the upwind side of the block. Average droplet density (drop diameter > D LD50 ) across the block from the optimized spray was predicted to improve by nearly a factor of 2 with a COV of 0. 14 compared to 0.58 for the conventional spray. A similar result was predicted when the emission drop size distribution was reduced to a DV0.5 of 50um. Average predicted deposit from the optimized strategy increased by 80% with COVs of 0.07 compared to 0.55 for conventional sprays.