Geography Reference
In-Depth Information
through the overlaying of such layers. Layers are analogous to the separate plates used to
print the various colours of topographic maps, since each colour represents a different class
of phenomena. In Google Earth, imagery constitutes a patchwork base layer, and all other
features are superimposed on it. Linear features such as roads and rivers, and point features
such as landmarks, present no problem to this approach. However, when area features such
as census tracts are coloured by variables such as population density, the result is a layer that
obscures all lower layers, including the base mapping. In such cases it is helpful to make
overlaid layers partially transparent.
It is impossible to determine the location of any feature on the surface of the Earth exactly,
since any method of measurement has inherent uncertainties. It follows therefore that no
two layers will fit precisely, unless one has been derived from the other or has somehow
inherited the same positional errors. Misfits of as much as 20 m are common with the layers
supplied by Google, such as base imagery and roads (Figure 2.2), and larger errors will tend
to occur when data are imported from other sources, perhaps because of issues over the
definition of latitude and longitude (see previous section), or because of poorly measured
positions. In such cases the normal fix is to zoom out so that misfits are not bothersome,
but this may be problematic in very detailed applications.
Figure 2.2
An example of Google Earth misregistration. The roads layer is displaced roughly 20 m
to the west with respect to the base image. The small superimposed high-resolution image was
georegistered using control points surveyed with GPS, and matches the roads layer but not the
Google Earth base image
Search WWH ::
Custom Search