Evaluation of the agdisp ground boom spray drift model

Abstract

AGDISP is a wellestablished spray drift model that has been validated for aerial spraying of forests Recently a prototypical ground boom option has been added to AGDISP This was evaluated in the current study by collecting data from spray trials over a grass sward using a ground boom sprayer and representative application parameters Spray solutions were made up of water sticker adjuvant and a metal cation which was changed for each spray application Deposition from spray drift was measured by analyses of the cation deposits on artificial targets (plastic tapes) placed on the grass surface Measured deposition was compared with profiles calculated using AGDISP AGDISP overpredicted deposition from spray drift by a factor of 35100 outside the spray block Possible reasons for these discrepancies are given Options are to improve measured deposition and the algorithms for the deposition on the downwind swath of the spray block and evaporation of droplets

Comparison of the mechanistic AGDISP ground boom spray model with experimental data

“…The ratio of model to measured decreased from an average of 12.4 (by reanalysis of these data) to 7.8 (to 63%) for the New Zealand data (analysis of the results presented in Woodward et al (2008)), while for the Canadian data the average ratio reduced from 3.9 to 1.5 (to 38%) and the Belgian data from 2.9 to 2.2 (to 76%).…”

“…The New Zealand trials (Woodward et al 2008) reported four treatments using XR11003 and 04 flat fan nozzles at 150 kpa and 300 kpa at a target application rate of 200 litres/ha at 0.5 m above the ground of 0.1 m high grass with a ground speed of 2 m/s. One replicate of four 5 m wide, 20 m long swaths was carried out per treatment.…”

“…Despite the model calibrating well with Spray Drift Task Force data from the United States the ground model did not calibrate well, overestimating drift with New Zealand data by 3.5 to 100 times (Woodward et al 2008), Canadian data an average of 3.9 times (Connell et al 2009), and Belgium data an average of 2.9 times for the flat fan data (R.B. Barton, unpublished data).…”

“…Specific changes that impact the ground model are the improved evaporation of small droplets and the correction of a programming error that affected the magnitude of air entrainment generated by ground booms (Schou 2011). However, further development is required and this paper reports on comparisons of AGDISP 8.25 with field trial drift data from New Zealand (Zabkiewicz et al 2008), Belgium (Nuyttens 2007;Nuyttens et al 2007aNuyttens et al , 2010Nuyttens et al , 2011 and Canada (Wolf & Caldwell 2001) and discusses potential areas for improvement. Each of these trials used different methodologies as did the Spray Drift Task Force data used to calibrate AGDISP.…”

An automated nozzle controller for self-propelled sprayers

“…Unique to AgDRIFT® is the inclusion of aircraft wake influences on drift (Bilanin et al, 1989). AgDRIFT® contains a tier 1 (a developed regression model) model for ground applications; however Woodward et al (2008) found through in-field testing that AgDRIFT® overpredicted out-of-swath deposition by a factor from 3.5-100. Developments to revise the tier 1 prediction model with a more analytical approach to provide greater accuracy are ongoing (Teske et al 2001, Teske et al, 2004.…”