Replacing or modifying the Unrath model for optimal spraying in high density orchards, by using the principles of the MABO-dosing model

In modern high density apple orchards in South Africa, the currently used pesticide dosing model (tree row volume [TRV]) together with axial fans without turrets are likely not optimal. The MABO dosing model has been developed for modern orchards in Germany and Austria. Spray deposition must be correlated with biological efficacy, i.e. pest and disease control. No biological efficacy model is available for any disease on apple. The objectives of the study were to (i) evaluate the efficacy of the MABO dosing model in high- density apple orchards in South Africa, (ii) develop a spray drift sampling and quantification protocol and (iii) develop a biological efficacy model for mancozeb and apple scab caused by Venturia inaequalis

Evaluation of the MABO dosing model was done using orchard trials along with fluorescent pigment sprays, photomicrography, and digital image analyses. Although spray drift orchard trials were not conducted due to unfavourable weather conditions, suitable spray drift collectors and a quantification method for the drift collectors were developed. This was done using a fluorescent pigment and photomacrography. The biological efficacy model was developed using apple seedlings sprayed with mancozeb, spray deposition quantification and V. inaequalis inoculations. Lesions were quantified using infrared thermal photography, image analyses, and qPCR analyses.

In 4m-row-width apple orchards, deposition parameters achieved with the MABO dosing model did not differ significantly from those of the TRV model. Deposition parameters in the orchards were furthermore not significantly affected by using a high (36000 m3/h) versus a low (28000 m3/ha) volumetric airflow rate (VAR) produced by a high profile axial fan sprayer. Evaluations in two 3.5m-row-width orchards indicated that, depending on the orchard, deposition quantity with the MABO model at a high VAR was significantly better at the top canopy position or overall deposition positions (top, middle, bottom, inside and outside canopy sampling positions) than the TRV (low and high VAR). Savings in spray costs of 40 to 28.5% (depending on spray volume) can be achieved when using the MABO dosing model versus the TRV model on a typically large scale high-density apple farm.

Black polyvinyl spray drift collectors were identified as suitable drift collectors. Spray deposition on the collectors can be quantified using macrophotography of yellow fluorescent pigment depositions.

Development of a benchmark model first showed that quantitative real-time PCR (qPCR) was a more suitable quantification method than thermal infrared imaging (TIRI) for fusi lesion quantification. TIRI overestimated the percentage of disease control. Both benchmark models showed that mancozeb yielded high levels of disease control at very low concentrations; for the qPCR benchmark model, the FPC% value of the FPC90 (90% control) corresponded to 0.15 times that of the registered mancozeb concentration in South Africa, i.e. 85% lower than the registered dosage.

The MABO model has potential for pesticide application in high-density apple orchards in South Africa; it is more user-friendly than the Unrath model and can also reduce spray costs. Based on the developed laboratory benchmark model, lower mancozeb dosages may be required for fusi control than the currently registered dosage. This will have to be evaluated further under orchard conditions.