One of the biggest problems in wind farms preliminary projects is the lack of a reliable topography.
Due to the tight budgets developers are often working with, it is often impossible to obtain a good topography (like the ones you can get with a LIDAR instrumented flight) or at list a decent one (as the standard field topographic surveys).
One of the possible solutions is to work with the Google Earth topography. Software like AutoCAD civil 3D makes it possible to download a cloud of points with coordinates and elevations and work with them.
The question is: how good is this info?
Unfortunately it’s impossible to answer univocally. The base grid, covering almost all the inhabited surface of the planet, has been obtained with a space mission (the NASA SRTM, Shuttle Radar Topography Mission).
The points are approximately spaced 50 meters, so it is a very rough starting point.
This base has been integrated with a “mosaic” of different DEM (digital elevation models) freely available material, so there are states in the US with a 1 meter contour line, or even cities with a very dense cloud of points (even a point every 20 centimeters).
It is currently impossible to know from where the points you are using are coming from, and if a point is “true” or interpolated. It is possible to have an impression seeing the shape of the contour lines (if you “zoom out” from the topography you will often see a pattern of squares coming from the available points).
This data is normally reasonably acceptable for a somehow preliminary project. Several commercial plug in are available to enhance the results.
For instance Plex Earth allows you to import Google Earth data (pictures and points) in different coordinates systems, importing contours in an area with any shape specified. It can also be used for preliminary volume calculation, to export objects from AutoCAD to visualize them inside Google Earth or doing the opposite (that is, importing a Google Earth KML file).