GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 245-3
Presentation Time: 9:00 AM-6:30 PM

QUANTIFYING BURROW- RELATED LEAKAGE AND FAILURE OF EARTHEN DAMS AND LEVEES


BACCUS, Turner A., Geology and Geological Engineering, University of Mississippi, 120A Carrier Hall, Oxford, MS 38655 and PLATT, Brian F., Department of Geology and Geological Engineering, University of Mississippi, 120A Carrier Hall, University, MS 38677

Burrowing organisms are known to impact the integrity of human-made earthen structures, such as dams and levees. Despite this, little research has been conducted to quantify and model these impacts. Previous studies used simple tubular burrows in a homogenous soil type, however this inaccurately represents the complexity of soil properties and burrow architectures. This study aims to accurately simulate the impacts of burrows on earthen structures to construct models that visually represent and quantify the leakage that potentially could cause failure.

Models will be constructed using dimensions of burrows cast in the field at the University of Mississippi Field Station, Abbeville, Mississippi and property adjacent to Lewis Smith Lake, Alabama. Burrows were cast using plaster, expanding insulating foam, or fiber-glass resin. Burrow casts range from 95cm long and 82 cm in circumference to 10 cm long and 16.5 cm in circumference. Burrows casts will be digitized with a NextEngine Multi-stripe Laser Triangulation (MLT) scanner and sliced to obtain cross-sectional profiles. Scans will be imported to Blender to isolate individual two-dimensional profiles and these will be imported to RocScience Slide 7.0 (RS). In RS, a hypothetical levee will be modeled using parameters and characteristics representative of an actual levee, and the burrow’s two-dimensional profile will be incorporated into the external boundary. The placement, number of burrows present, and orientation of the burrows will be varied inside of so RS to simulate the effects of multiple burrowing animals of different body and population sizes. RS will also be used to create flow nets, which will be used in the analysis of the data.