Stabilization of the Lower Jamuna River in Bangladesh—Hydraulic and Morphological Assessment

This paper presents a hydraulic and morphological analysis of the Lower Jamuna in Bangladesh with a focus on two key bifurcations that are important for stabilization of the Lower Jamuna reach. We used ground measurements, historical data, multispectral satellite images from various sources as well as numerical models. We carried out hydraulic analyses of the changes and their peculiarities, such as flow distributions at the bifurcation and hysteresis of the stage–discharge relationships. We supplemented our analysis by using numerical models to simulate discharge distribution at the bifurcations under various flow and riverbed conditions. We developed an advanced and automated satellite image processing application for the Lower Jamuna, referred to as Morphology Monitor (MoMo), using the Google Earth Engine. MoMo was found to be an effective tool for a rapid assessment and analysis of the changes in deep-channel and sandbar areas. It is also useful for monitoring and assessing riverbank and char erosion and accretion, which is important not only for morphological but also ecological impact assessment. The application can be adapted as an operational tool as well. Furthermore, we assessed the evolution of deep channels at the bifurcations based on regularly and extensively measured bathymetry data. The analysis was carried out in complement with morphological modeling, particularly for short-term prediction. In this paper we present the major findings of the analysis and discuss their implications for adaptive river management.

MixChannel: Advanced Augmentation for Multispectral Satellite Images

Usage of multispectral satellite imaging data opens vast possibilities for monitoring and quantitatively assessing properties or objects of interest on a global scale. Machine learning and computer vision (CV) approaches show themselves as promising tools for automatizing satellite image analysis. However, there are limitations in using CV for satellite data. Mainly, the crucial one is the amount of data available for model training. This paper presents a novel image augmentation approach called MixChannel that helps to address this limitation and improve the accuracy of solving segmentation and classification tasks with multispectral satellite images. The core idea is to utilize the fact that there is usually more than one image for each location in remote sensing tasks, and this extra data can be mixed to achieve the more robust performance of the trained models. The proposed approach substitutes some channels of the original training image with channels from other images of the exact location to mix auxiliary data. This augmentation technique preserves the spatial features of the original image and adds natural color variability with some probability. We also show an efficient algorithm to tune channel substitution probabilities. We report that the MixChannel image augmentation method provides a noticeable increase in performance of all the considered models in the studied forest types classification problem.