To develop a empirical model for predicting the spring flight period of overwintering Ips acuminatus adult, their density were monitored in Korean pine (Pinus koraiensis) forests at Chuncheon in 2015. The monitoring data of the beetles and temperature in the forests were used to develop the empirical predictive model based on degree-day model, and it was validated using the data from Korean pine forest at Wonju in 2018. The lower threshold temperature for flight (LTF) and a thermal requirement for the onset of flight activity of the beetles in spring were estimated. As the result, the LTF was estimated as 1.3 ℃ and 269.96 DD was required for the spring flight. The median flight date estimated by the empirical model was one day earlier than the observed flight date. Therefore, the model is suitable for predicting the spring flight of overwintering I. acuminatus.

To develop an empirical degree-day model for predicting the spring flight period of the bark beetle, Ips subelongatus Motschulsky, based on field observation, field studies were biweekly conducted in three Japanese larch (Larix kaempferi) forests in In-je, Korea from 2013 to 2014. To validate this degree-day model, we compared the model-predicted values with observed emergence data of I. subelongatus in 2015 spring at one of the sites. The flight period of over-wintering generation began on April and ended May, and flight of next generation lasted until October. The lower developmental threshold temperature (LDT) was estimated using spring emergence of I. subelongatus and field temperatures. Then a degree-day model was constructed, based on LDTs estimated from field observations data. The baseline temperature with the highest coefficient of determination was considered the LDT, and this was estimated to be 6.0℃. The explanatory power of the model was 88%. This model accurately predicted the flight of I. subelongatus in 2015 spring, as the estimated median flight dates was 1 days earlier than the corresponding observed flight date. The results of the goodness-of-fit test did not differ between observed and estimated values (ks = 0.21, P = 0.54).