Affiliation:
1. Civil and Construction Engineering, Brigham Young University, Provo, UT 84602, USA
Abstract
Identifying low-head dams (LHDs) and creating an inventory is a priority, as fatalities continue to occur at these structures. Because obstruction inventories do not specifically identify LHDs and they are not assigned a hazard classification, there is no official inventory of LHDs; a multi-agency taskforce is creating one now by identifying LHDs using Google Earth Pro (GE Pro). The purpose of this paper is to assess whether a machine learning approach can accelerate the creation of the national inventory. We implemented a machine learning approach to use a high-resolution remote sensing data with a Convolutional Neural Network (CNN) architecture. The model achieved 76% accuracy in identifying LHDs (true positives) and 95% accuracy identifying Non-low-head-dams (true negatives) on the validation set. We deployed the trained model for the National Hydrologic Geospatial Fabric (Hydrofabric) flowlines in the Provo River watershed. The results showed a high number of false positives and low accuracy due to the mismatch between Hydrofabric flowlines and actual waterways. We recommend improving the accuracies of the Hydrofabric waterway tracing algorithms to increase the percentage of correctly classified LHDs.
Funder
Kenneth and Ruth Wright Family Foundation
Subject
Water Science and Technology,Aquatic Science,Geography, Planning and Development,Biochemistry
Reference32 articles.
1. What we know (and don’t know) about low-head dams;Tschantz;J. Dam. Saf.,2014
2. Department of Defense (2017). Rules and regulations. Fed. Regist., 82.
3. Low-head dam impacts on habitat and the functional composition of fish communities;Smith;River Res. Appl.,2017
4. Backward flow velocities of submerged hydraulic jumps;Leutheusser;J. Hydraul. Eng.,2001
5. Submerged hydraulic jump remediation at low-head dams: Partial width deflector design;Hotchkiss;J. Hydraul. Eng.,2018