Estimation of Rice Yield Loss Using a Simple Linear Regression Model for Bacterial Blight Disease

Abstract

Field experiment was carried out in hot and humid summer (Transplanted Aus) season to realize the yield loss of a susceptible rice variety Purbachi inoculated with bacterial blight (BB). Treatments consist of BB inoculations at different crop growth stages like maximum tillering (MT), panicle initiation (PI), booting (Bt), flowering and heading stages differently including a control (no BB inoculation). Disease severity index (DSI) was measured at 14 days after inoculation (DAI) and harvest. Data on 1000-grain-weight and yield was recorded at harvest. Significant variation on DSI was observed among different BB inoculated crop growth stages. MT, PI and Boot stage inoculations showed similar (DSI 7.1-8.0) but higher DSI than flowering and heading stages inoculation (3.2-5.3) even control (0.00) at 14 DAI. However, all the treatments showed similar DSI 9.0 at harvest. Bacterial blight can affect the grain weight to some extent although it was insignificant among the treatments (0.1-4.5%). DSI showed negative correlation with 1000-grain weight (r=-0.77*) and similarly with the yield (r=-0.97**). The yield ranged from 2.4-3.4 t/ha among the treatments. The yield loss was observed 5.8-30.4% in the BB inoculated treatments. MT, PI and Boot stages inoculation affected the yield much resulting 21-30.4% yield loss. It could be concluded that a susceptible variety can be affected with significant yield loss up to 30.4% with severe outbreak of B B. A simple regression equation = 4.09-0.211X( = Yield, X = BB severity score) is suggested for the prediction of yield loss in susceptible variety in summer season. Bangladesh Rice j. 2019, 23(1): 73-79