Underground subway systems provide an efficient high capacity public transportation method for commuters within dense urban areas. Passengers using underground subway systems are subjected to enclosed and crowded environments that must be controlled to maintain tolerable temperatures and ventilation. Modern subway stations and trains are cooled using conventional air conditioning, providing ac- ceptable levels of passenger comfort. However, deep and old subway systems suffer from overheating problems, because they were built before the invention of modern mechanical ventilation and air conditioning, resulting in passenger discomfort and health issues during summer conditions (Gilbey et al., 2011). These systems were not designed to handle the high numbers of passengers and train traffic in current congested cities such as London or New York. The oldest deep subway lines such as the London Underground (1863), Paris (1900), New York (1902), Berlin (1902) , and Moscow Metro (1935), commonly experience overheating problems (Griffiths, 2006). When high ambient temper- atures are combined with heat rejected from train braking and passenger traffic, the temperatures of the tunnels and platforms rise substantially above tolerable levels. The London Underground, particularly the deep sections have become uncomfortable during summer due to congestion and poorly ventilated tunnels, where during the 2006 European heat wave, temperatures as high as 47 ◦C were recorded (Griffiths, 2006). Retrofitting old deep subway systems with air conditioning is often infeasible, because the tunnels only allow enough room for trains. Furthermore, heat rejected from air-conditioning in these narrow tunnel spaces could in fact further exacerbate the overheating problems. As a result, temperatures in old subway tunnels (and surrounding ground) have increased over long term (Botelle et al., 2010). The shallow ground surrounding an old overheated subway system thus has a large potential of low enthalpy energy that can be used for low-grade heating and cooling purposes. Advances in ground source heat pump configurations makes it possible to consider extracting this geothermal energy in an efficient manner (Nicholson et al., 2014). The objective of this thesis is to investigate the use of stand-alone standard closed loop vertical Ground Source Heat Exchangers (GHE) to extract excess heat from old and deep subway tunnels. Because vertical GHEs are physically and structurally independent of the underground subway structure, they can be positioned flexibly around the tunnels. Furthermore, vertical closed loop GSHP systems are standard in urban areas because they can be easily installed, do not require large spaces, and yield good system efficiency (Kavanaugh and Rafferty, 1997). In the UK, they also qualify for subsidies under the UK Government’s Renewable Heat Incentive Scheme (DECC, 2014a). To evaluate the potential of vertical GHEs for usefully extracting excess heat from the subway tunnels, one must be able to quantify the net gains: How much useful heat can be feasibly extracted from underground tunnels over a time period ? What is the optimal GHE set-up that could maximize heat extraction ? and what are the resulting temperature drops in the underground tunnels and platforms ? This thesis presents a novel co-simulation framework designed to answer the above questions through simulation modelling. It couples a 1D model of a subway line with a 3D FEM model of vertical ground heat exchangers. The 1D subway model represents spatially averaged transient heat and air flows in the underground, while the FEM model simulates vertical close-loop GHEs next to the subway tunnels. The two models are co-simulated such that information is passed back and forth through a common temperature boundary layer at the outer tunnel walls until both models converge. As a result, the cooling effect of vertical GHEs on the underground climate is examined. In addition, the heat extraction rates of the GHEs placed next to the subway tunnels are compared with standard GHEs. Different arrangements and distances of the GHEs with respect to the tunnels are also examined to achieve the best heat extraction and cooling in the tunnels and stations simultaneously. Finally, partially insulated GHEs are investigated to provide for both heating and cooling demand, because the need to extract heat from the subway is most during the summer months, when demand from building is also for cooling. As an illustrative study, the London Underground’s Central line is selected as a representative of an old subway system that suffers from overheating and ventilation problems. The Central Line is one of the busiest line of the London Underground, and suffers from over heating problems particularly during summer conditions, where temperatures above 35 C have been recorded in some areas (Gilbey et al., 2011). Retrofitting the Central Line with vertical closed loop GHEs is investigated to examine its benefits both underground and overground: In addition to the cooling effect in the train tunnels, the heat extracted from the Central Line system is quantified against the heating demand of the surrounding buildings above the tunnels. It is shown that the GHEs enhance the passenger thermal comfort, while providing district heating to the buildings above.