DEVELOPMENT AND FIELD VALIDATION OF A BROWN PATCH WARNING MODEL FOR PERENNIAL RYEGRASS TURF

Microclimate in perennial ryegrass (Lolium perenne 'Caravelle') was mo nitored for 3 years to identify environmental conditions associated wi th brown patch (Rhizoctonia solani) outbreaks and to develop a weather -based disease-warning model. The micrometeorological parameters measu red were ambient air temperature, relative humidity, leaf wetness dura tion, precipitation, soil temperature, soil moisture, and solar radiat ion. Brown patch outbreaks were confirmed by the visual presence of fo liar R. solani mycelium. An environmental favorability index (E) was d eveloped to relate a combination of environmental conditions with brow n patch outbreaks. The initial index (E(6)) was based on relative humi dity (RH greater than or equal to 95% for greater than or equal to 8 h ; mean RH greater than or equal to 75%), leaf wetness duration (greate r than or equal to 6 h) or precipitation (greater than or equal to 12 mm), and minimum air (greater than or equal to 16 degrees C) and soil (greater than or equal to 16 degrees C) temperatures. Further analyses , however, revealed that an equally effective E was provided by a two- variable regression model (E(2)) The E(2) model is E = -21.5 + 0.15RH + 1.4T - 0.033T(2), in which RH is the mean relative humidity and Tis the minimum daily air temperature. For both E(6) and E(2) models, a th reshold value (i.e., E greater than or equal to 6) constituted a brown patch warning. Brown patch outbreaks were predicted by both models wi th 85% accuracy over a 3-year period. All major infection events were predicted. In 1993, the warning model was used to field evaluate fungi cide performance in perennial ryegrass and colonial bentgrass (Agrosti s tenuis 'Bardot'). There were equivalent levels of blighting between the warning model and a 14-day calendar-based spray schedule, but the warning schedule provided a 29% reduction in fungicide applications.