Sediment Transport by Released Anchor Ice: Field Measurements and Digital Image Processing

• Author / Creator

  Kalke, Hayden E

• The importance of anchor ice transport of sediment on a river is significantly understudied. This study addresses the lack of data related to anchor ice release and rafting. A large sample set of anchor ice was collected in the field over the 2015-2016 and 2016-2017 winter seasons to compute a mean sediment concentration (± standard deviation) contained in anchor ice of 28.2 ± 33.2 g/L. Coarse sediment was also plucked from the top of passing anchor ice to better understand the mass and size of the largest sediment moved through this unique sediment transport process. The mean mass and major chord length of particles transported by anchor ice were 47.7 g and 3.6 cm, respectively. This indicated that a majority of the coarse sediment transported by anchor ice is gravel size. Additionally, it was found that 24% of the mass transported came from only 1.2% of the particles sampled. The sediment concentration, major chord length of coarse sediment, and mass of coarse sediment were all modelled reasonably well by a lognormal distribution. The amount of sediment transported by anchor ice can be assessed if the measured sediment concentration in anchor ice pans discussed above is combined with the quantity of anchor ice released over the course of the winter. Digital images of surface ice, both frazil and anchor ice, were collected with bridge-mounted game cameras and an unmanned aerial vehicle (UAV). Algorithms to compute the quantity of frazil and anchor ice in the flow from the images were developed. Three site-specific learning models, support vector machines, were trained to produce binary images from raw digital images that separated the total surface ice from the water. Another support vector machine model was also trained to then separate the predicted surface ice into the frazil and anchor ice components on the Peace River. These support vector machine models allowed for a variety of applications to be performed that improve our understanding of freeze-up processes. This included a spatial distribution of surface ice concentration with UAV acquired images, a time series of surface ice concentration leading to freeze-up with images acquired from bridge-mounted game cameras, and the computation of an instantaneous sediment mass flux during an anchor ice release event.

• Subjects / Keywords

  • Image processing
  • Anchor ice
  • Frazil ice
  • Machine learning
  • Support vector machine
  • River ice

• Graduation date

  Fall 2017

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R3GB1XW96

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.