Off-site (i.e. desktop) quantification of wetlands can be greatly enhanced using fine-scale digital elevation models (DEMs).  This best occurs when the wetland extent is controlled by distinct topographic changes and when those changes are captured by the DEMs.  Light Detection and Ranging (LIDAR) is an excellent source of dense elevation observations that can be used to create fine-scale DEMs.  However the standard triangular irregular network (TIN) processing of the data results in a needlessly rough surface.  Whether the roughness is due to the inclusion of some vegetative returns in the bare earth data, or if it is from bare earth returns in a hummocky area, the result is that topographic breaks along the edges of wetlands are obscured.  Smoothing algorithms can squelch surface roughness but simultaneously mute the topographic changes of interest.  Non-TIN processing of the LIDAR data can be employed to create DEMs that honor landscape breaks along edges of wetlands while smoothing the more random topography of hummocky areas.  Presented here is a discussion of how non-TIN processing assist in the visualization of depressional wetlands in a pine flatwoods system.