In this thesis, we explore various aspects related to wellbore leakage and plug and abandonment (P&A) of gas wells in British Columbia (BC) and Alberta. The objective of this thesis is two fold, first we review current practices and trends related to P&A and secondly, we investigate leakage pathways associated with reported wellbore leakage in BC.

Related to the first objective, we review available data on well architectures, industrial practices both with regards to well construction and P&A practices and highlight trends over the past few decades in British Columbia and Alberta. The data reveals a large wave of abandonments coming in the next decade and a significant increase in reported instances of surface casing vent flow, SCVF, over the past decade, which will lead to rising P&A costs. Additionally, we try to understand variability in reported instances of SCVF between operators and between provinces.

Wellbore leakage, such as SCVF, is a complex issue facing many wells as shown in the first part of the thesis. Related to the second thesis objective, we present a novel approach to modelling realistic leakage along microannulus pathways of varying thickness. We use stochastic methods to calibrate leakage pathway dimensions to the SCVF leakage rates reported in BC. Results shows that representing dry microannulus thicknesses with a lognormal distribution provides a good fit for the intermediate ranges of SVCF flow rates, but that a dry microannulus alone cannot account for all instances of wellbore leakage. Additionally, we use this model of wellbore leakage to investigate the effect of a number of operational factors on wellbore leakage. These effects on leakage have not been previously investigated and our results show that each operational effect can be significant, and why. We repeatedly highlighted that reducing wellbore leakage is often localised, requiring an impermeable barrier be placed across the cap rock but that there is an element of luck regarding success of a remediation operation. This suggests that stochastic-based models that capture the variability of the well integrity can be used to support the development of risk-based plug and abandonment P&A practices.