<div dir="ltr"><div><div>When you transform/reproject data across UTM zones, I'll again praise the KMS Engsager Extended Mercator, which is now a part af PROJ.4 (thanks to Thomas Knudsen and Frank Warmerdam et al), and consequently in GDAL. and is relatively fast and very exact to within millimeters across multiple zones.<br>This should be default on the raster side, so far I know.<br><br>But on the vector side in QGIS, you'll need to set an environmental parameter to get exact results: OSR_USE_ETMERC = YES. This can be done in QGIS' menu Settings -> Options -> System -> and down to the Environment section - or directly in regex so it takes effect for ogr2ogr in the command prompt. Again, rasters should be OK. Otherwise put +proj=etmerc in the proj-strin.<br><br>In the high latitudes of Greenland, we had 8-900 m transformation problem across 5 zones. In ArcGIS, ESRI made a number of special UTM WGS 84 Zones 20-27 named "Complex" - just for Greenland. Which was needed, because while working with data in the northern parts, we'll often need to cross multiple zones.<br><br></div>If you are not working with data from the Arctic or Antarctic, it is probably not a substantial problem, but it might save you some meters or centimeters here and there.<br><br></div>Karl<br></div><div class="gmail_extra"><br><div class="gmail_quote">2015-06-11 5:18 GMT-02:00 ardi <span dir="ltr"><<a href="mailto:ardillasdelmonte@gmail.com" target="_blank">ardillasdelmonte@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Thanks a lot for your advice, Nicolas!<br>
<span class="HOEnZb"><font color="#888888"><br>
ardi<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
<br>
On Thu, Jun 11, 2015 at 1:09 AM, Nicolas Cadieux<br>
<<a href="mailto:nicolas.cadieux@archeotec.ca">nicolas.cadieux@archeotec.ca</a>> wrote:<br>
> Hi,<br>
> It's never a good idea to delete files that you have reprojected.<br>
> Reprojection will, in most cases slightly alter the data although that<br>
> depends on the parameters used. You could always test by reprojecting the<br>
> file back to the original projection and then by using raster math to see<br>
> the differences, if any.<br>
><br>
> If you want to save space, I suggest you compress the data using "deflate"<br>
> parameters. (You need to manually add support for bigtiff if the file is<br>
> over 4GB. ) Once you are done, you can always do raster math to make sure<br>
> data has not changed. ( original layer - new layer should give zeros).<br>
><br>
> For your second question, that is just the way it is! Since QGIS is open<br>
> source, not ever programmer puts reprojection in their code.<br>
><br>
> Cheers.<br>
><br>
> Nicolas Cadieux M.Sc.<br>
> Les Entreprises Archéotec inc.<br>
> 8548, rue Saint-Denis Montréal H2P 2H2<br>
> Téléphone: 514.381.5112 Fax: 514.381.4995<br>
> <a href="http://www.archeotec.ca" rel="noreferrer" target="_blank">www.archeotec.ca</a><br>
><br>
> On Jun 10, 2015 4:36 PM, "ardi [via OSGeo.org]" <[hidden email]> wrote:<br>
><br>
> Hi,<br>
><br>
> I've a satellite photo data set which spans across three UTM zones,<br>
> and I need to do some tasks in all three zones at once. Each image<br>
> tile from the data set is provided in two versions: one is projected<br>
> in the UTM coordinate system for the zone it actually belongs. The<br>
> other is projected over the central UTM zone, extending such central<br>
> zone.<br>
><br>
> AFAIK, QGIS can deal with layers belonging to different UTM zones,<br>
> because you see a lat/long display and layers are reprojected to<br>
> lat/long on the fly.<br>
><br>
> My question is: can I save disk space by storing the image tiles in<br>
> just one version, the version projected in its real UTM zone? Or would<br>
> I also need the versions projected into the (extended) central UTM<br>
> zone?<br>
><br>
> Just out of curiosity, what's the use of the version projected into<br>
> the extended zone? Is it because some GIS packages don't support<br>
> on-the-fly reprojection? Or why?<br>
><br>
> Thank you in advance<br>
> _______________________________________________<br>
> Qgis-user mailing list<br>
> <a<br>
> href="/user/SendEmail.jtp?type=node&node=5210266&i=0">[hidden<br>
> email]<br>
> <a href="http://lists.osgeo.org/mailman/listinfo/qgis-user" rel="noreferrer" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br>
><br>
><br>
> ________________________________<br>
> If you reply to this email, your message will be added to the discussion<br>
> below:<br>
> <a href="http://osgeo-org.1560.x6.nabble.com/Do-I-need-satellite-raster-layers-relative-to-an-extended-UTM-zone-tp5210266.html" rel="noreferrer" target="_blank">http://osgeo-org.1560.x6.nabble.com/Do-I-need-satellite-raster-layers-relative-to-an-extended-UTM-zone-tp5210266.html</a><br>
> To start a new topic under Quantum GIS - User, email<br>
> <a href="mailto:ml-node%2Bs1560n4125267h38@n6.nabble.com">ml-node+s1560n4125267h38@n6.nabble.com</a><br>
> To unsubscribe from Quantum GIS - User, <a<br>
> href="<a href="http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=4125267&code=bmljb2xhcy5jYWRpZXV4QGFyY2hlb3RlYy5jYXw0MTI1MjY3fDYzNDQ4MjQxNg==" rel="noreferrer" target="_blank">http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=4125267&code=bmljb2xhcy5jYWRpZXV4QGFyY2hlb3RlYy5jYXw0MTI1MjY3fDYzNDQ4MjQxNg==</a>;">click<br>
> here.<br>
> <a<br>
> href="<a href="http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml" rel="noreferrer" target="_blank">http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml</a>"<br>
> style="font:9px serif">NAML<br>
><br>
><br>
> ________________________________<br>
> View this message in context: Re: Do I need satellite raster layers relative<br>
> to an extended UTM zone?<br>
> Sent from the Quantum GIS - User mailing list archive at Nabble.com.<br>
><br>
</div></div><div class="HOEnZb"><div class="h5">> _______________________________________________<br>
> Qgis-user mailing list<br>
> <a href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a><br>
> <a href="http://lists.osgeo.org/mailman/listinfo/qgis-user" rel="noreferrer" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br>
_______________________________________________<br>
Qgis-user mailing list<br>
<a href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/qgis-user" rel="noreferrer" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a></div></div></blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature">Karl Brix Zinglersen<br></div>
</div>