Controlling Ipomoea species in soybean fields with a leaf age development model

Summary To control invasive morning glory ( Ipomoea ) species in multiple soybean fields of an Agricultural Cooperative Corporation, the effects of sequential herbicide applications based on a soybean leaf age development model were studied. Because the error between the estimated and observed number of days for a certain soybean leaf stage was 0.4–1.9 (RMSE of 0.11–0.56), the model was approved as practical for determining the time of applications of herbicides. In the fields where bentazon and glufosinate were applied at the 2- and 5-leaf soybean stages, respectively, the per centages of area infested by Ipomoea coccinea or I. lacunosa tended to decrease in comparison with those where the herbicide applications were delayed or not performed. The effects of sequential herbicide applications in two successive years to reduce the infested area were unclear, because 52–81% of the total area of the soybean fields was not treated at the scheduled time due to bad weather. Ipomoea lacunosa was more difficult to control than I. coccinea because it grew rapidly and were missed easily at the application time of herbicides. In particular, there was a greater percentage of area infested by morning glory species, in comparison to the previous year, with increase in delay of application of bentazon. Therefore, timely control of bentazon at the 2-leaf stage of soybean is important for control of morning glory.