Use of Remote Sensing and Ground Data in Comprehension of the Flooding in the Bow River Basin, Alberta

Abstract

Flooding is a devastating natural hazard throughout the world. Consequently, a flood management system is vital. Here, the aim was to investigate elements of flood management as it pertains to the Bow River in Alberta. The specific objectives included: (i) river flow forecasting at Calgary, (ii) flood extent estimation at Calgary, and (iii) river planform change detection. Analyses revealed that using a multivariable linear regression (MLR) formulated as a function of upstream gauge stations and the station of interest using antecedent flows demonstrated strong relationships (i.e., r2 = 0.93). Furthermore, the flood extent estimation gave a kappa statistic of 0.6, which is reasonable considering that the image was taken 16 days after peak flood time. Lastly, the Bow River planform change detection showed that the 2013 floods caused a higher erosion in the lower Bow River (i.e., 361.62ha) as compared to the upper Bow River area (i.e., 206.01ha).