A Web Based Solution to Track Trawl Vessel Activities Over Pipelines in Norwegian Continental Shelf

Abstract

Vessel Activities such as trawling and anchoring represent a risk to offshore marine structures such as pipelines, subsea structures, cables and platforms. Third party interference is a major contributor to the damage and failure statistics for subsea pipelines. Detecting such activity at an early stage, increases the probability of introducing cost efficient mitigation measures before costly repairs are necessary. The main goal of this study is to develop an interactive web-based solution to track and monitor trawl vessel activities in the Norwegian Continental Shelf which can be used for assessing integrity of pipelines. Vessels share their location and identity via the Universal Shipborne Automatic Identification System (AIS) over a 24-hour period, refreshing under different time intervals. Hence, there are billions of data points and terabytes of data to feed into our computer systems. Making sense of them poses many challenges, of which the main challenge is to identify the type of the fishing vessel. This problem is important because, identifying the vessel type forms the preliminary in recognizing trawling activities. Trawl patterns have shown to change over time and sometimes also because of a new pipeline being installed. The detailed information about the trawl activity is essential to have an accurate assessment of where to inspect and where to implement corrective intervention, based on up to date trawling intensity and equipment used. The main contribution of this thesis is to implement a machine learning approach to identify the type of fishing vessels and provide a web based solution to perform detailed analysis of trawl vessels activities over the pipelines for a chosen area of interest.