In this article, two criteria are suggested to inspect whether a model of geospatial information diffusion is valid to fill in the missing observation in a gap unit. The first criterion serves for inspecting a gap unit in the study area, from which a sample is taken to train the model. The second criterion serves for a gap unit outside the area but nearby. The standard deviation of the trained model plays a key role in the former. The valid radiance determined by the maximum distance between the units of the sample point passed the test and the center of the study area plays an important role in the latter.