CEE graduate students and faculty,
For this Friday March 27th, we will have Scott Hamshaw presenting his PhD Dissertation Proposal titled:
"Characterization and prediction of suspended sediment flux using Bayesian and artificial neural networks at multiple scales".
This weeks CEE Graduate Seminar is at Kalkin 004 from 12:50 pm - 1:40 pm.
New computational tools are developed to characterize and predict suspended sediment flux over varying temporal scales in order to inform watershed management decisions. Fine sediments are recognized as an important diffuse source pollutant in surface waters due to their impacts on aquatic ecology and role in the transfer and fate of substances and nutrients such as phosphorous. In addition, suspended sediment transport is an immensely complex process that lends itself to analysis with non-linear, data-driven methods. In this work, high temporal-resolution sediment monitoring from turbidity sensors provide a robust data set to develop and test applicability of artificial neural networks (ANNs) to predict and forecast fluvial sediment flux at different time scales. ANNs are utilized for advanced temporal analysis of storm-event sediment dynamics and for predicting in ungauged or data poor catchments. These networks, will be combined with new methods for measuring the contribution of streambank erosion (i.e., unmanned aerial system (UAS) and terrestrial laser scanner (TLS)) to help improve estimates of sediment loading. The overall sediment budget framework is designed using Bayesian network analysis to leverage the data sets and better estimate uncertainty.