Roughness Effects of Leeway Objects on Drift Properties Using Novel Data-Driven Methodology

Aiming to improve drift predictions of objects and personnel lost at sea, a novel data-driven conceptual framework using a mobile sensor platform was developed to accurately estimate the trajectory of objects in marine environments in real-time using a combination of perception-based sensing technology and deep-learning algorithms. The focus of this thesis is to address the scientific question of the effect of roughness of the drift object on the drift behavior. Surface roughness influences the aerodynamic and hydrodynamic drag acting on the surface of the object. The scope of this work can be broken down into three components. The first two components were experimental in nature and the third was validation of the experimental results using data from the United States Coast Guard (USCG). Small-scale experiments were conducted in the flume at the River Sciences Laboratory at UTSA and the drift objects used for these experiments were a rectangular box and a 3D printed sphere. There were three versions of each object with protrusions attached uniformly to increase the drift object surface area by 1.00, 1.25, and 1.50 times the original surface area for six total objects. Additionally, experiments were conducted at the large lap pool at UTSA. The objects from the flume experiments along with a larger sphere, larger box, and two boats of similar size to the flume objects were used. Validation of the experimental results was conducted using existing data on sailboats, small fishing boats, and a buoyant meat cooler. All data sets were analyzed, and correlations were made on the effects of surface roughness on the behavior of drift objects. The results demonstrate the importance of roughness effects.