Underground space provides an opportunity to increase city compactness and pedestrian accessibility through the overlapping of activities in urban areas. The potential for development is not homogeneous throughout an urban area but depends on variations in environmental conditions. The Deep City project at the EPFL is currently developing a methodology to analyze the underground development potential of four resources: space, groundwater, geothermal and minerals. Rather than focusing only on the feasibility of single, predefined projects, it looks at the environmental characteristics of an urban region and develops scenarios that take advantage of synergies between resources. One of the challenges for making decisions based on the analysis is dealing with the confusion surrounding the impact that the project will have on the existing urban morphology and on the spatial experiences of the people who will frequent them. As such, evaluating underground development potential must take into consideration both the environmental and socio-economic characteristics directly related to the existing urban form. This paper lays out the recent progress of the Deep City methodology, which proposes an original analysis combining environmental characteristics with the geometric structure of the urban morphology using spatial network analysis. The proposed process is demonstrated using a case study of the city of San Antonio, Texas, in the southern United States. The preliminary results suggest that the method has the potential to bridge multiple scales from the regional to the pedestrian, but that it is yet not apparent how the results of the analysis can be turned into a normative framework for project design. The discussion will briefly address how normative rules can emerge from existing underground spatial practices, particularly for cities with no existing underground development.