The current study presents experiments for the initial stages of fluidization induced by a localized fluid injection. The process was studied by recording high-speed videos in a 2-D-region far from the boundaries using Planar Laser Induced Fluorescence and Refractive Index Matching. The experimental setup allowed for several parameters to be systematically studied including particle sizes, initial bed heights, and injection port diameters. The results show that the critical flow rate required for fluidization is primarily dependent on the initial height of the granular bed. However, this dependence is not a linear relation but progressively plateaus for larger heights. Conversely, the diameter of the chimney relates only slightly to the injection port diameter and significantly more to the diameter of the particles. (c) are average of the video sequence; (d), (e), and (f) are standard deviations of the video sequence. (a) and (d) are of a static bed, (b) and (e) show a bed with a fluidized cavity and pictures (c) and (f) show a chimney. Figure (d) is solid black indicating no movement and resulting zero standard deviation. Image processing done by ImageJ. 37 Solid symbols are for the injection port of D 5 10 mm and open symbols are for D 5 40 mm. Symbols are as follow: Circles, 3 mm particles; squares, 5 mm; triangles, 7 mm; hexagon, 10 mm. [Color figure can be viewed at wileyonlinelibrary.com]