Performance of a weather-based forecast system for chemical control of coffee leaf rust

Abstract

Coffee leaf rust (CLR), caused by Hemileia vastatrix, is a major disease in coffee fields worldwide. In Brazil, fungicides are extensively used for disease control; nevertheless, disease outbreaks can occur if applications are not properly timed. In this study, we evaluated a CLR forecast system (FS) for scheduling fungicide foliar sprays in seven field trials (conducted in five different sites). Two trials were carried out in 2015–16 season and five in the 2016–17 season. The CLR-FS uses the cumulative infection rate based on daily infection rates, accumulated since the beginning of each season, considering minimum air temperature and average relative humidity. The treatments were based on soil drench followed by foliar sprays: T1 - control; T2 - two or three sprays according to the regional calendar of each site; T3 - FS with more conservative thresholds (MC) with three sprays; T4 - FS with less conservative thresholds (LC) with three sprays; T5 - FS LC with two sprays; T6 - same as T3, without soil drench. We assessed CLR incidence monthly and estimated the area under the disease progress curve (AUDPC) for comparison of treatments. In six experiments, the sprays adjusted by the FS showed lower AUDPC than T2. However, all treatments controlled CLR appropriately in seven experiments. In four experiments, treatment T5, with only two foliar sprays, was as efficient as the other treatments with three applications. The CLR-FS used in a less conservative threshold could improve chemical control of CLR; however, regional calibrations may be necessary.