[Proj] +a argument different than I expected

[Michael Speth]

Hi Frank,
   I think I am confused on how the arguments should be used.  Let me 
describe what I am trying to test.  Here are the following scenarios to 
test the projections I am working on (HEALPix & rHEALPix):

* Forward Sphere
* Inverse Sphere
* Forward Ellipsoid
* Inverse Ellipsoid


Case 1 - Forward Sphere Parameters
+lon_0=45 +a=1

Case 2 - Inverse Sphere Parameters
-I +lon_0=45 +a=1

Case 3 - Forward Ellipsoid Parameters
+lon_0=45 +a=1 +e=0.5

Case 4 - Inverse Ellipsoid Parameters
-I +lon_0=45 +a=1 +e=0.5

Where I think the following:
* lon_0 is the central meridian
* a is the radius of the sphere or ellipsoid
* e is the eccentricity

Thanks for your help.

On 14/11/11 09:17, Frank Warmerdam wrote:
> On 11-11-13 11:39 AM, Michael Speth wrote:
>> Greetings,
>> I am using SVN Revision: 2113. I have found that the +a argument doesn't match
>> the value found in P->a. Here is an example:
>>
>> proj +proj=healpix -f '%.'$PERCISION'f' +lon_0=45 +a=3.0 +R_A +ellps=WGS84 -E<<EOF
>> 0 0
>> EOF
>>
>> The value of P->a is 2.996646.
>>
>> Is this correct?
> Michael,
>
> I don't really know the answer, but I'm guessing the +R_A is adjusting the
> semi major and semi minor axis a bit.  I honestly don't even know what
> you meant by the the mix of +a, +R_A and +ellps.
>
> Best regards,


-- 
Michael Speth
Research Systems Developer
Landcare Research