This is a sample of some testing I've done to assess the shaded relief output after different strategies of generalization. In a ladder based generalization, the data is first generalized to an intermediate scale before being generalized again to a target scale or final product (i.e. iterative filtering). Star generalization is when the source data is generalized to be used directly at the target scale (i.e. resampling). The left 5 tiles represent a 10m DEM of the Grand Canyon area that was iteratively filtered using a low pass filter (ladder based generalization). The right 5 tiles represent the original 10m DEM that was resampled to a 30m DEM (star) and filtered (ladder). The results are then displayed at various scales to assess which is the best product.
