Other Software and Data

Hi Dominic,

'Minimum Curvature' is the other method we can use. See if it helps you.

Shweta

Howdy Dominic,

The answer to your question, in large part, depends upon your particular information needs (as Amita and Andrei have asked you about in the forum). Usually the term ‘surface’ denotes a distribution of some value in two-dimensional planar space but, much more important than the data model, are your information needs based upon the original business requirements of implementing technology in the first place. Hope that makes sense to you!

So traditionally, “surfaces” have been represented in a variety of ways in computerized systems sampling the complexity of the real-world. This graphic illustrates several of these:

Which model is chosen above should, in part, depend upon what kind of locational phenomena you are modeling, your sampling design, data collection methodology, information modeling requirements for the needed output products. Some sample spatial sampling designs:

The Triangulated Irregular Network (TIN) has long been used to represent surface morphology and, as the name suggests, TINs are made of a set of adjacent triangles with each triangle uniquely define a “section” from an infinite plane. They are joined in a triangulation (often connected via a topologic graph or what I am left to assume you mean by “vector continuity”). Each triangle has a slope (vertical angle) and an aspect (horizontal direction) and connect to each other via nodes. While there are different methods to calculate a TIN, by far the most common and widely used are two variants of the Delaunay triangulation method: standard and constrained. The constrained version is the same with the exception of how breaklines are handled. In terms of the accuracy of the surface model, because TINs have a variable point density that changes on the degree of slope, when required they can be made more accurate with the addition of additional mass points, breaklines and polygons.  

In addition, TINs are designed to specifically capture and represent typical surface features such as ridges, peaks, streamlines, etc. and some engineering quality applications (especially civil ones such as “cut and fill” procedures) require the precision coordinate storage and how slope discontinuities (such a ridgelines) are accounted for with breaklines in the data model. Typical surface analysis functions such as slope, aspect, elevations, volumetrics, vertical profiling, line-of-sight calculations, etc. are well suited to make best use of the TIN data model. Surfaces can also be represented as rasters but that is another discussion for another day…..

Cheers

Bryant