[GRASS5] problems with [rsv].proj

Markus Neteler neteler at itc.it
Fri May 31 03:09:09 EDT 2002


On Thu, May 30, 2002 at 03:16:31PM -0400, rgrmill wrote:
[...] 
> GRASS5.0.0 probably should support datum changes other than nad27<->nad83. 
> That is a very fundamental capability.

Apart from the discussion when to submit datum support: I agree that
GRASS should *as soon as possible* (whatever that means) support
datum changes other than nad27<->nad83.

Find attached an ongoing small test with northern Italian sample
coordinates. The datum support already present in PROJ4 improves
the result significantly (from 25m E-W and 70m N-S shifts to 2m E-W
and 9m N-S using the cs2cs program with datum).

Find attached preliminary results. I'll post the complete results later
in PROJ4 mailing list after receiving more known points for other
parts of Italy (is anyone there who can deliver this?).

Cheers

 Markus
-------------- next part --------------
   Reprojections of sample points with and without datum transformation
   using 'proj', 'cs2cs' and a local special formula

Command lines:
cs2cs -v +proj=latlong +datum=WGS84 +to +init=epsg:26591 +towgs84=-225,-65,9

proj +proj=tmerc +lat_0=0.0 +lon_0=9.0 +k=0.999600 +x_0=1500000.0 +y_0=0.0\
     +ellps=intl +units=m

Legend:
 O: - The Lat/Long WGS84 coordinates have been obtained by GPS
    - Gauss Boaga are official cadastrial control points
    - The UTM WGS84 coordinates have been obtained by GPS (only for F:)
 C: 'cs2cs' results with datum transformation  (parameters see above)
 P: 'proj' results without datum transformation (parameters see above)
 F: Special spreadsheet formula (see below R code [1]), which converts from
    UTM to Gauss-Boaga

Table with 3 sample points:
Lat/Long (WGS84) - MONTE CIMIMIRLO        |  Gauss Boaga (Roma40, Intl ellps)
==========================================+==================================
O: 11 9 45.24184  46 4 13.11961  532.137  |  1667267.98  5104156.11   n/a
------------------------------------------+----------------------------------
C: 11d9'45.24184E 46d4'13.11961N 532.137  |  1667269.19  5104147.45 483.851
P:      "                 "        -      |  1667246.73  5104229.82   -
F: [official UTM: 667238.924 5104133.123] |  1667268     5104156      -

Lat/Long (WGS84) - MARTIGNANO             |  Gauss Boaga (Roma40, Intl ellps)
------------------------------------------+----------------------------------
O: 11 7 58.87328  46 5  8.89033  429.782  |  1664936.81  5105815.65   n/a
------------------------------------------+----------------------------------
C: 11d7'58.87328E 46d5'8.89033   429.782  |  1664938.08  5105806.98 381.467
P:      "               "          -      |  1664915.53  5105889.39   -
F: [official UTM: 664907.822 5105792.645] |  1664937     5105816      -


Lat/Long (WGS84)  - [2] see below            |  Gauss Boaga (Roma40, Intl ellps)
---------------------------------------------+----------------------------------
O: 11 07 20.625264 46 04 20.726853   263.1778|1664154.8358 5104306.7929 213.0876
C: 11d07'20.625264E 46d04'20.726853N 263.1778|1664156.37 5104298.53 214.879
P: -                                         | -
F: -                                         | -

--------------------------
[1] Formula for Northern Italy, Fuso Ovest: UTM -> Gauss-Boaga
    [From "bolletino BZ", here as R code]

NUTM <- 5104133.123
EUTM <- 667238.924
#NUTM <- 5105792.645
#EUTM <- 664907.822

a1 <-  1.00001586139729
b1 <-  0.00000816953815234324
a2 <-  1.00001587189443
b2 <-  -0.00000816372109252845
P  <- -63.43479941
Q  <-  1000060.11675565

Ngb <-  P + NUTM*a1 + EUTM*b1
Egb <-  Q + EUTM*a2 + NUTM*b2

Egb
Ngb


[2] coordinates of the GPS permanent station of the PAT (servizio geologico)


More information about the grass-dev mailing list