Contours are polylines that are used to connect locations of equal value and show how values change across a surface model. Most of the time, elevation values are associated with contours, but any continuous data can be used, for instance: precipitation, pollution or even atmospheric pressure.
When talking about contours it is not uncommon to hear about Engineering Contours versus Cartographic Contours. The question becomes what is the difference between the two?
LIDAR data is an inherently random point cloud where each point really represents a sphere since the determination of each point contains error. The random point structure means that airborne LIDAR data generally lacks continuous edge information. Edges can be cleaned up by supplementing a point cloud surface model with breaklines. The inherent error in each point, however, can cause two points sitting next to one another and representing a feature of the same elevation to have two different values. However small the difference, if located on either side of a contour interval, the difference can be enough to show a different representation in the contours. Engineering contours (Figure 1) are those which directly reflect the surface model from which they are generated. Since engineering contours are accurate to the surface model they are often very jagged and contain many isolated contours. Smoothing techniques can be run on the surface or resulting contour file to generate cartographic contours (Figure 2), which are often considered “pretty” with smooth curves, much like were seen in the days when cartographers used to hand draw the contour lines on maps to outline areas of equal elevation.