2

I'm deciding on whether to use a geographic projection (lat/long) or a lambert conformal conic projection for a web servicing project that will use raster, vector, and point data, for queries, generating statistics, calculating measures based on area (for example), and also modeling. Models will be mostly based on raster data, but vector to raster conversions may be used depending on the clients request.

Any suggestions? Between the two (lat/long versus LCC)? Or is there another projection you would suggest?

I thought about a UTM projection, but study areas cross multiple zones.

Czed
  • 702
  • 5
  • 15
  • Will you be using a js client like Openlayers or Google maps api? – dassouki Oct 21 '11 at 14:22
  • You might find the replies I received to this question to be helpful. – whuber Oct 21 '11 at 15:33
  • 2
    Implicitly this question assumes all work must be done in the same coordinate system. However, many "modeling" calculations are so computationally intensive that a preliminary reprojection would be an inconsequential extra burden. Is there anything that would preclude this option? – whuber Oct 21 '11 at 15:34

1 Answers1

3

You will probably want an equal area projection if you intend to do any sort of serious analysis. I've had good results with Lambert equal area projections in the past but we'd need to know more about your area of interest to recommend a specific projection.

Ian Turton
  • 81,417
  • 6
  • 84
  • 185