This talk will explore how QGIS can be used in academic research not only to visualize existing spatial data, but to synthesize heterogeneous datasets and construct new spatial knowledge.

Presented from the perspective of a design student increasingly working with spatial analysis as a research method, the talk brings together two projects that approach the limits of spatial data from opposite directions. One that asks what can be made visible by combining imperfect spatial traces. Using population, infrastructure, terrain, hazards, and telecom data, the project uses standard GIS operations synthetically to construct new suitability and connectivity indices for Seoul—spatial variables that did not previously exist as individual datasets.

A second study on satellite imagery reverses the question: what has the data itself already made invisible? By examining information loss across different urban morphologies and image resolutions, the project treats spatial data itself as an object of investigation, not just as an input to analysis.

Together, the case studies highlight QGIS as an academic research environment for inference, synthesis, and critical inquiry that is capable of generating new datasets and interpretations while also raising questions about data credibility, uncertainty, and the limits of what spatial analysis can make legible.