In this study, we developed a non-invasive method to determine oxygen diffusivity (D) in food gels using an Oxydot luminescence sensor. We designed and fabricated a transparent diffusion cell in order to represent oxygen transfer into foods packaged in an 8-ounce polymeric tray. Oxydots were glued to the sides (side-dot) and bottom (bottom-dot) of the cell and filled with 1, 2, and 3% (w/v) agar gel as a model food. After deoxygenation, local oxygen concentrations in the gels were measured non-invasively at 4, 12 and 22°C. Effective oxygen diffusivities in gels (D) and water (D) were obtained after fitting experimental data to the analytical solution (data from side-dot) and the numerical solution (data from bottom-dot) to Fick's second law. Temperature had significant positive influence (P<0.05) on oxygen diffusivity estimated for different medium and analysis methods. The Dobtained from both methods were statistically different (P<0.05) at 12 and 22°C but not at 4°C. Results show that D values decreased by 72-92%, compared to D. Results also show that decreasing the temperature from 22 to 4°C reduced the D and D values by 55-60%. No significant difference (P>0.05) was found between the activation energy (E) of water and gels (1-3% w/v) for temperatures ranging from 4 to 22°C. We used a combined obstruction and hydrodynamic model to explain why D decreased as gel concentration increased. The method developed in this study can be used to study the oxygen diffusivity in foods.