[GRASS-user] Reprojection help

GRASS GRASS user
1

It's me again ... I suspect I must be stupider than most Grass users because I seem to be hitting the list with questions so often!

I have my own maps which are in lat-long format, using Australian boundaries.

I need to analyse some points relative to some centroids supplied in MapInfo format. They import fine, but they're in Lambert Conformal Conical projection.

Without being long-winded, my attempts to reproject the LCC location into a lat-long UTM using WGS84 keep failing - the utilities work but the result isn't right.

Is there something about LCC projections that make them hard to reproject?

Richard

2

Vincent,

Vincent BAIN wrote:

Hello Richard,
don't know if it can help, but for me it works, from conformal conic
Lambert to rgf93 (the official geographic coord system in France,
compliant with wgs84).

Could you send us the proj information from each of your locations, and
make sure they correspond to your project ?
  

The XY location was defined using EPSG code 20356 (Australia from Geosciences). It shows UTM projection, zone 56, AGD84 datum, and ellipsoid data.

The LCC location was "self-defined" during import - that is, I created a new location during import. The project info it created is below:
g.region -p
projection: 99 (Lambert Conformal Conic)
zone: 0
datum: towgs84=-134,-48,149,-0,-0,-0,0
ellipsoid: a=6378160 es=0.006694541854587638
north: 6784163.1212489
south: 3123142.28999373
west: 2961606.17926652
east: 6874125.97609581
nsres: 183051.04156276
ewres: 195625.98984146
rows: 20
cols: 20
cells: 400

The problem is that I don't know what EPSG code the creator of the mapset I'm trying to import was using!

I advise you to define the coordinate systems from epsg codes, the best
way -- i think -- to avoid problems defining the right ellipsoïd, datum,
and so on...

And be cautious: the coordinates read from the display are decimal...
  

<whacks forehead> I suspect you're telling me I should be using degree-minute-second format in my transformation?

Thanks, I will go back and look at how I can get the projections lined up ... I might try defining a standard Australian LCC and see if the import works; then from a standard location to a standard location might well be easier.

Richard

Bon courage,
Vincent

Le samedi 11 août 2007 à 09:32 +1000, Richard Chirgwin a écrit :
  

It's me again ... I suspect I must be stupider than most Grass users because I seem to be hitting the list with questions so often!

I have my own maps which are in lat-long format, using Australian boundaries.

I need to analyse some points relative to some centroids supplied in MapInfo format. They import fine, but they're in Lambert Conformal Conical projection.

Without being long-winded, my attempts to reproject the LCC location into a lat-long UTM using WGS84 keep failing - the utilities work but the result isn't right.

Is there something about LCC projections that make them hard to reproject?

Richard

_______________________________________________
grassuser mailing list
grassuser@grass.itc.it
http://grass.itc.it/mailman/listinfo/grassuser

3

Richard Chirgwin wrote:

It's me again ... I suspect I must be stupider than most Grass users
because I seem to be hitting the list with questions so often!

I have my own maps which are in lat-long format, using Australian
boundaries.

What ellipsoid?

I need to analyse some points relative to some centroids supplied in
MapInfo format. They import fine, but they're in Lambert Conformal
Conical projection.

Without being long-winded, my attempts to reproject the LCC location
into a lat-long UTM using WGS84 keep failing - the utilities work
but the result isn't right.

What do you mean?

Is there something about LCC projections that make them hard to
reproject?

Please let us know the projection settings of your two Locations using
"g.proj -p". Can you provide the projection info of your original data
before you imported them to GRASS? "ogrinfo -al -so" might do it.

Maciek

4

Le samedi 11 août 2007 à 17:59 +1000, Richard Chirgwin a écrit :

Vincent,

Vincent BAIN wrote:
> Hello Richard,
> don't know if it can help, but for me it works, from conformal conic
> Lambert to rgf93 (the official geographic coord system in France,
> compliant with wgs84).
>
> Could you send us the proj information from each of your locations, and
> make sure they correspond to your project ?
>
The XY location was defined using EPSG code 20356 (Australia from
Geosciences). It shows UTM projection, zone 56, AGD84 datum, and
ellipsoid data.

The LCC location was "self-defined" during import - that is, I created a
new location during import. The project info it created is below:
g.region -p
projection: 99 (Lambert Conformal Conic)
zone: 0
datum: towgs84=-134,-48,149,-0,-0,-0,0
ellipsoid: a=6378160 es=0.006694541854587638
north: 6784163.1212489
south: 3123142.28999373
west: 2961606.17926652
east: 6874125.97609581
nsres: 183051.04156276
ewres: 195625.98984146
rows: 20
cols: 20
cells: 400

The problem is that I don't know what EPSG code the creator of the
mapset I'm trying to import was using!

You mean : you have data in XY coordinate, but you don't know what coord
system it's linked to ? then it's a serious problem :slight_smile:
maybe you can try with a "standard" lambert in use in Australia

> I advise you to define the coordinate systems from epsg codes, the best
> way -- i think -- to avoid problems defining the right ellipsoïd, datum,
> and so on...
>
> And be cautious: the coordinates read from the display are decimal...
>
<whacks forehead> I suspect you're telling me I should be using
degree-minute-second format in my transformation?

I just meant sometimes we careless try to check the transformation by
comparing decimal values of coordinates to deg,min,sec values... (it's
the kind of mistake I do every day !)

Thanks, I will go back and look at how I can get the projections lined
up ... I might try defining a standard Australian LCC and see if the
import works; then from a standard location to a standard location might
well be easier.

good luck

Richard
> Bon courage,
> Vincent
>
> Le samedi 11 août 2007 à 09:32 +1000, Richard Chirgwin a écrit :
>
>> It's me again ... I suspect I must be stupider than most Grass users
>> because I seem to be hitting the list with questions so often!
>>
>> I have my own maps which are in lat-long format, using Australian
>> boundaries.
>>
>> I need to analyse some points relative to some centroids supplied in
>> MapInfo format. They import fine, but they're in Lambert Conformal
>> Conical projection.
>>
>> Without being long-winded, my attempts to reproject the LCC location
>> into a lat-long UTM using WGS84 keep failing - the utilities work but
>> the result isn't right.
>>
>> Is there something about LCC projections that make them hard to reproject?
>>
>> Richard
>>
>> _______________________________________________
>> grassuser mailing list
>> grassuser@grass.itc.it
>> http://grass.itc.it/mailman/listinfo/grassuser
>>
>>
>
>

_______________________________________________
grassuser mailing list
grassuser@grass.itc.it
http://grass.itc.it/mailman/listinfo/grassuser