Abstract: To address the challenges in quantitatively characterizing the spatial distribution features and filling volume of overlying rock fractures during gangue slurry backfilling, this study systematically investigates the migration patterns and fracture development mechanisms in overburden structures within mining areas. A quantitative computational model was developed to evaluate cavity and fracture spaces in both caved and fractured zones, elucidating the regulatory effects of key parameters including mining height, rock fragmentation coefficient, compaction zone length, and working face advancement length on fracture space evolution. The results show that the volume of cavities and crack spaces in the non-compacted zone of the collapse zone is about 2.5 times greater than that in the compacted zone and about 7.5 times greater than that in the crack zone. Through a series of multi-scale gangue slurry injection experiments, we analyzed the flow and diffusion dynamics within the caved zone. Furthermore, we established a theoretical predictive model for gangue slurry filling volumes applicable to both low-level grouting and adjacent grouting methods. Field validations demonstrate that gangue slurry exhibits excellent flow and diffusion characteristics in the caved zone, with model predictions showing strong agreement with actual filling volumes. The proposed framework provides reliable theoretical guidance for gangue slurry backfilling engineering applications.