Jpn. J. Appl. Ent. Zool. 34: 189-198 (1990) Forecasting mass immigration is very important for the understanding of larval outbreaks in the armyworm, P. separata. Accurate annual estimates of overwintering larval densities in warmer areas are necessary but difficult to attain; larval densities are low. I looked for correlations between meteorological factors and armyworm overwintering success, hoping to use them to forecast mass immigration years. An analysis of data from outbreak years between

more »

... 1958 and 1988 searched for common changes in monthly mean air temperature and monthly precipitation during the overwintering season (December to March) in the overwintering areas (west Japan, Hiroshima, Okayama, Shimonoseki, Miyazaki, and Kagoshima). I noted that outbreaks occurred in years in which: 1) January/February (the first half of the larval stage) weather was warmer, less rainy, than the 31-year average, 2) the aridity index (monthly precipitation divided by monthly mean air temperature) was smaller than average, and 3) weather in December and March was warm and dry without heavy rain and extreme cold. Analysis of the relationship between the armyworm developmental stages and the weather from June to October (the 2nd and 3rd generations) 1987 showed that the period from the adult to the first half of the larval stage had much less rain than the period from the second half of the larval period to the pupal stage. This characteristic was common for the overwintering generations.