Environmental Monitoring Commons^™

An Engineering Geological Anatomy Of The Padma River Bank Failure And Erosion, 2018: A Case Study Of Naria Bank Section, Bangladesh, Mir Fazlul Karim, Nazrul I. Khandaker, Shahidul Haque, M. Zillur Rahman, Muhammad Qumrul Hassan, Asm Maksud Kamal, Masud Ahmed, Belal A. Sayeed

Publications and Research

The Naria town of Bangladesh is developed on the right bank of the Padma River. The bank is an old natural levee of Meghna River. The Holocene-Recent geology of Naria is actively dominated by the fluvial processes of Ganges-Brahmaputra-Meghna River system where the deltaic sediments are characterized as unconsolidated fine sand and silt, covered by thin veneer of clayey silt and loam. The annual volume of water discharge and flow dynamics are dependent on the intensity of the rainfall, runoff and the length of dry winter. Excessive river bank erosion, channel avulsion, renewed submergence of floodplains, and formation of natural …

Development Of Digital Bathymetry Maps For Lakes Azuei And Enriquillo Using Sonar And Remote Sensing Techniques, Michael Piasecki, Mahrokh Moknatian, Fred Moshary, Jorge Gonzalez

Publications and Research

This article presents an improved algorithm for optimization and development of a digital bathymetric model (DBM) for Lake Azuei (LA) (Haiti) and Lake Enriquillo (LE) (Dominican Republic) using the ANUDEM method. Both sonar data and contour lines of the lakes’ layout extracted using Landsat imagery were compiled for bathymetry development. We show that the performance of the ANUDEM method was strongly dependent on the density and irregularity of the spatial distribution of the data. Changing the resolution of the output grids and deriving auxiliary topographically corrected contours improved the ANUDEM performance and minimized the systematic errors of the method. Statistical …

Observational Time Series For Lakes Azuei And Enriquillo: Surface Area, Volume, And Elevation, Mahrokh Moknatian, Michael Piasecki

Publications and Research

In this report, we present historical time series of surface area, volume, and elevation for lakes Azuei (Haiti) and Enriquillo (Dominican Republic). The intention is to present a history of the lakes’ levels for both bodies of water as derived from Landsat imagery that is augmented by reports and narratives that reach further back in time. We also summarize lake level time series data collected and developed through various other efforts and compare these data sets to our time series. The time series contains 45 years’ worth of data ranging from 1972 to 2017 which we developed from Landsat imagery …

Towards Improving Accuracy And Interpretability Of Deep Learning Based On Satellite Image Classification, Yamile Patino Vargas

Dissertations and Theses

ABSTRACT

The study of satellite images provides a way to monitor changes in the surface of the Earth and the atmosphere. Convolutional Neural Networks (CNN) have shown accurate results in solving practical problems in multiple fields. Some of the more recognized fields using CNNs are satellite imagery processing, medicine, communication, transportation, and computer vision. Despite the success of CNNs, there remains a need to explain the network predictions further and understand what the network is determining as valuable information.

There are several frameworks and methodologies developed to explain how CNNs predict outputs and what their internal representations are [1, 4, …