Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering)
Online ISSN : 2185-467X
ISSN-L : 2185-467X
Annual Journal of Hydraulic Engineering, JSCE, Vol.62
LOGISTIC REGRESSION ANALYSIS FOR PREDICTING GRASS INVASION INTO KINUGAWA RIVER CHANNELS
Hayata IIMURAHitoshi MIYAMOTO
Author information
JOURNAL FREE ACCESS

2018 Volume 74 Issue 4 Pages I_565-I_570

Details
Abstract
 This paper investigated causes of grass invasion into river channels and developed the prediction model by using logistic regression analysis. The river channels analyzed was located in the middle section of the Kinugawa River. The logistic regression analysis dealt with data obtained from field observations in 2015 and 2016 using a UAV measurement. The results of the regression analysis showed that the grass invaded into the channels with the factors, in order of the past vegetation history, dimensionless tractive force/height from the normal water level, and shortest distance from river streams. The accuracy of the logistic reogession model became highest in the case including the past vegetation history as a explanatory variable. The prediction result of the model indicated good ability for the grass invasion probability with 75% of accuracy rate.
Content from these authors
© 2018 Japan Society of Civil Engineers
Previous article Next article
feedback
Top