Abstract

Knowledge of iceberg location and size is important for safety reasons as well as for understanding many geophysical and biological processes. This dissertation analyzes large tabular icebergs in the Southern Ocean using the SeaWinds scatterometer. SeaWinds is a spaceborne radar designed to measure the microwave backscatter from the Earth's surface. Using resolution-enhancement techniques, backscatter measurements are processed into backscatter images in which icebergs can be observed. An iceberg detection methodology is formalized using daily scatterometer images. Radar profiles from common Antarctic scatterers are quantified and an iceberg detection methodology is formalized using daily scatterometer images. Iceberg positions are determined in real-time and a time-series of iceberg positions is maintained in an Antarctic iceberg database. Using the Antarctic iceberg database, characteristic iceberg motion trends are identified. Iceberg detection and tracking is demonstrated through real-time operational support of the 2005, 2008, and 2009 National Science Foundation Antarctic cruises. To supplement iceberg position reports, I develop multiple algorithms to estimate iceberg size and rotational orientation from backscatter images and from raw backscatter measurements. Estimates derived from SeaWinds images are found to be more accurate. Using iceberg size parameters in conjunction with Newton's equations of motion and forcing profiles (e.g., ocean and air currents), I also develop an iceberg motion model to predict the translational and rotational motion of large tabular icebergs. To improve model results, a Kalman filter is used to incorporate actual iceberg measurements into the motion model, and statistics from the Kalman filter are used to evaluate model performance. Simulated iceberg motion is found to best coincide with observed iceberg motion in regions where slower iceberg drift speeds are observed. The model is less accurate at high speeds. The iceberg motion model is inverted to produce estimates of ocean currents given observations of iceberg size and motion. Multiple ocean current estimates are combined using reconstruction techniques and compared with numerically-derived ocean currents from the Ocean Circulation and Climate Advanced Modeling (OCCAM) project. It is found that reconstructed ocean currents coincide with OCCAM currents in regions where observed iceberg motion is not extreme. Also, reconstructed ocean currents coincide more with OCCAM currents that have been averaged over multiple years than with monthly-reported values.

Degree

PhD

College and Department

Ira A. Fulton College of Engineering and Technology; Electrical and Computer Engineering

Rights

http://lib.byu.edu/about/copyright/