Evaluation of models based on a generic infection model for controlling early blight in potatoes

Abstract

TOMCAST and two models (risk hours and critical days) based on a generic infection model (GI model) were evaluated for the timing of fungicide application to control early blight (Alternaria solani) in field experiments. A five-year experiment was conducted at AU Flakkebjerg with the starch potato cultivars Kardal (2016 & 2017), Avarna (2019 & 2020), and Allstar (2021). Barley kernels infested with Alternaria solani were used to inoculate the potatoes. The disease was assessed weekly, starting from the onset of the first symptoms. The impact of disease development was quantified using the area under the disease progress curve (AUDPC), the relative area under the disease progress curve (rAUDPC) as well as starch yield. An economic analysis was done to determine the profitability of the models. Fungicide application significantly suppressed early blight development in comparison to untreated plots. The fungicide-treated plants gave higher yields (10–35%) than the untreated plants. Nevertheless, the models reduced fungicide use by an average of 37–49%, without any yield or economic penalty. Most fungicides were saved with the TOMCAST model, but disease control and yield were generally lower than in plots treated according to the GI-based models. Overall, our study emphasizes the economic importance of early blight and the necessity of applying fungicides to control it. We also show the possibility of significantly reducing fungicide use by using decision support systems. For the first time, we demonstrate the feasibility of using GI models to time fungicide application to control early blight.