Carbon capture and storage (CCS) presents a short-term option for significantly reducing the amount of carbon dioxide (CO2) released into the atmosphere and mitigating the effects of climate change. To this end, National Grid initiated the COOLTRANS research programme to consider the pipeline transportation of high pressure dense phase CO2. Part of this work involved the development of a mathematical model for predicting the near-field dispersion of pure CO2 following the venting, puncture or rupture of such a pipeline. This article describes the application of this model to the simulation of punctures in buried pipelines, and specifically three scenarios – a puncture at the side, at the base and at the top of the pipeline. Such scenarios following human interference with the pipeline are the most common type of pipeline failure and form an important part of the quantitative risk analysis (QRA) required in the development of such pipelines for CCS. In each scenario, a idealised crater is modelled, dispersing CO2 into dry air. In two of the experiments, an idealisation of a naturally formed crater is used. In the third, the idealisation is based on the pre-formed crater. We present the steady state flow in each scenario and, using Lagrangian particle tracking techniques, give estimates on the amount of solid CO2 deposited in the crater. In the case of the side puncture, experimental data above the crater are available and the model qualitatively and quantitatively predicts the nature of the flow in this case. The validated steady state flows at the top of the crater presented here for these three common scenarios provide the basis for developing robust source conditions for use in computational fluid dynamics (CFD) studies of far-field dispersion and for use with pragmatic QRA models, as well as representing a significant step towards modelling full-scale ruptures of CCS pipelines.