|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | Re: [gislist] Global area calculations |
| Date: |
06/30/2005 07:20:01 AM |
| From: |
Quantitative Decisions |
|
|
At 03:05 PM 6/29/2005 -0700, Brian Klinkenberg wrote:
>I have a student who has a geodatabase that contains protected areas from
>around the world. She needs to calculate the areas (ha) of each of the
>areas. What means would people suggest as the best approach to take?
...
>I forgot to make something clear -- the file is in lat/long,
>so obtaining the areas directly is not possible. Standard
>'global' projections introduce too much error in the results,
>so they aren't a choice.
The crude expedient of computing the unprojected area (applying Euclidean
formulas, not spherical formulas, to the lon-lat coordinates) and then
multiplying by the cosine of the latitude of the region's centroid will
typically obtain the area to within 0.5% for most of the countries and
territories in the world (using about 111.1 km per degree to convert to
hectares). You will get better accuracy for the necessarily smaller
protected areas lying within a country.
Using *any* global equal-area projection will make *no* error relative to
the ellipsoid used to model the earth's surface (apart from any numerical
imprecision, which in many GISes will be far smaller than the error
inherent in the numerical representation of each region's boundary). The
only way you could do better would be to adopt varying region-specific
ellipsoids that (slightly) better approximate each region, perform the
necessary datum transformations, then use equal-area projections. We're
talking parts per million differences here: it's difficult to imagine a
situation where that would matter. For almost any conceivable analysis of
the size of protected regions, two or three significant digits will suffice.
Absent quantitative information about how much error is "too much," it is
not possible to recommend a "best approach."
--Bill Huber
Quantitative Decisions
www.quantdec.com
_______________________________________________
gislist mailing list
gislist@lists.geocomm.com
http://lists.geocomm.com/mailman/listinfo/gislist
_________________________________
This list is brought to you by
The GeoCommunity
http://www.geocomm.com/
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
