This topic will discuss some issues concerning the
projections of the surface datasets and the analysis performed on these
datasets.
Some short definitions
Projection - The two-dimensional representation of the three-dimensional space.
Coordinate System - a reference system for measurements defined by the projection
Geographic Coordinate System - measures locations in degrees - latitude and longitude. Since latitude and
longitude are angular measurements they are not suitable for measuring distances. The
major parameter of a Geographic Coordinate System is its datum.
Projected Coordinate System - uses a projection to transform the latitude and longitude
to X and Y coordinates and makes the linear measurements more accurate. Each projected
coordinate system is based on a Geographic Coordinate System.
Spatial Domain - the range and precision of coordinates that can be stored in a
feature dataset.
Spatial Reference - contains information for the coordinate system and spatial domain
extent for a feature dataset.
Coordinate systems and 3D analysis.
The Geographic Coordinate System provides a way to store common coordinates for locations anywhere in the
world. Due to this fact it is used in many areas (Location based services, navigation, etc.).
If however we want to measure distances and areas on data in a GCS we are facing an obvious problem - the units
of measure of a GCS are actually angles - a distance of 2.5 Decimal Degrees does not mean much, an area of 1.5 "
Square Decimal Degrees" (if such term existed) means even less.
One can argue that using GCS we can calculate distances and areas on the Spheroid and the results
will be in meaningful distance/area units (meters, feet, etc..) and more accurate than the results
derived from projected data. This might be true, but only on large scale (continental) data where the projected
data will be more distorted by the single projection used to represent it in Cartesian coordinates.
If we take into consideration the geographic extent of the surface data that is normally used for 3D analysis,
we can conclude that an appropriately selected projection for the location of the data will give us better results.
Based on the discussion above, the functions of ET Surface work as follows:
All functions preserve the Spatial Reference of the input data source unless an option is available and
the user selects it. The assumption is that if the user keeps a dataset in certain projection he has reasons for that,
and all the products of this data set must be in the same projection.
Functions in the TIN Surface Analysis and Raster Surface Analysis groups will only work
on surfaces in Projected coordinate systems.
All other functions work on surface data in any Geographic and Projected Coordinate System.
Where there is more than one input dataset and they have different coordinate systems, the function will
try to reproject on the fly the relevant dataset.
In order to get correct results for Slope, Volume and 3D Area the Z units should be the same as the units
of the spatial reference of the data.