[GIS] How to project a layer from WGS84 lat/long (automatic projection) to WGS 84 UTM Zone 23S in QGIS

coordinate systemqgisqgis-2.8

So I have a vector layer of polygons in WGS 84 lat/long (i.e. using the automatic projection for lat/long and WGS84 datum). I want to change that to WGS84 UTM 23S in QGIS. I tried doing like online tutorials show (right-click layer, "save as", and choose the CRS you want) but there has been zero distortions to the polygons, even for very big ones comprising tens of kilometers. In other words, it didn't work.
Can anybody help?
I'm using QGIS 2.8.2.

Somebody please help. It's suppose to be simple, but I can't get my head around this zero distortion problem. Anyone?

Best Answer

If you view the layer in QGIS with On-the-fly-reprojectionenabled (which is default if the loaded layers have different CRS). you will not see any difference.

That is the deeper sense of reprojecting: The object stays on the same spot on Earth, but the coordinates change.

If you turn off otf, you will see the layers on different places after zooming to the extent.

Related Solutions

[GIS] Basic proj4 conversion UTM grid -> longlat and back

Personally I'd use the Python bindings for the Proj4 library - http://pypi.python.org/pypi/pyproj/1.8.6

Further details at http://pyproj.googlecode.com/svn/trunk/README.html

You can pass your projection parameters directly to the projection object as follows:

PROJ_32756 = """
+proj=utm +zone=56 +south +ellps=WGS84 +datum=WGS84 +units=m +no_defs
"""
p1 = pyproj.Proj(PROJ_32756)

A full list of these strings can be found at http://spatialreference.org/ For zone 56 see http://spatialreference.org/ref/epsg/32756/proj4/

Many of these projections are already built into the library so you can just pass in the code. Below is an example of reprojecting some geojson from one projection to another. You should be able to adapt to your needs.

import pyproj
import simplejson

p1 = pyproj.Proj(init='epsg:3857')
p2 =pyproj.Proj(init='epsg:29902')

json='{"type":"Polygon","coordinates":[[[-972279.15280781,6907135.0149664],[-995516.00940234,6867999.2564914],[-946596.31130859,6844762.3998969],[-964941.19809375,6892459.1055383],[-972279.15280781,6907135.0149664]]]}'

pydata = simplejson.loads(json)
print str(pydata)
new_coords = []
for p in pydata['coordinates'][0]:
    x2, y2 = pyproj.transform(p1,p2,p[0],p[1])
    new_coords.append([x2, y2])

new_json = simplejson.dumps(dict(type="Polygon", coordinates=[new_coords], srid="29902"))
print str(new_json)

[GIS] Computing UTM Zone from lat/long point

It's not that difficult, even if you handle the zones around Svalbard and Norway. Here's an example:

ZoneNumber = floor((LongTemp + 180)/6) + 1;

if( Lat >= 56.0 && Lat < 64.0 && LongTemp >= 3.0 && LongTemp < 12.0 )
    ZoneNumber = 32;
endif
// Special zones for Svalbard
if( Lat >= 72.0 && Lat < 84.0 ) 
  if  ( LongTemp >= 0.0  && LongTemp <  9.0 ) 
    ZoneNumber = 31;
  elseif( LongTemp >= 9.0  && LongTemp < 21.0 )
    ZoneNumber = 33;
  elseif(LongTemp >= 21.0 && LongTemp < 33.0 )
    ZoneNumber = 35;
  elseif(LongTemp >= 33.0 && LongTemp < 42.0 ) 
    ZoneNumber = 37;
  endif
 endif

Convert Latitude/Longitude to UTM (attributed to Chuck Gantz).

I haven't tried this specific code, but the algorithm looks correct.

Best Answer

Related Solutions

[GIS] Basic proj4 conversion UTM grid -> longlat and back

[GIS] Computing UTM Zone from lat/long point

Related Question