Although full-scale training exercises remain the gold standard for preparing emergency services for extreme events, simulations have the potential to become a practical alternative by providing a cost-effective means to test diverse rescue scenarios in controlled environments. Building simulations with qualitative data can increase the realism of the scenarios and the behavior of the agents. This paper aims to describe how ethnographic data have been used to build and refine a conceptual model. We focus on modeling decisions in which qualitative analysis played the most pivotal role. To build the conceptual model, we applied bricolage methods to ethnographic data was collected through interviews and observation of full-scale underground mining exercises. The model includes goals, organizational structure, and organizational culture (such as safety and efficiency among the rescue service, emergency medical service, and mining companies), and frames agent decision-making processes within theories of situation awareness and routine dynamics to frame agent decision-making structures. The purpose of the model is to be used for interactive exploration and training purposes of stakeholders in situation awareness and decision-making. Some of the main challenges of working with qualitative data arise from the need to choose suitable levels of abstraction and levels of detail, as well as compensate for occasional gaps in the data.