On 2/27/07, <b class="gmail_sendername">Jose Gomez-Dans</b> <<a href="mailto:jgomezdans@gmail.com">jgomezdans@gmail.com</a>> wrote:<div><span class="gmail_quote"></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi Guillaume,<br>I just now realised your messages in the GDAL mailing list regarding<br>the reprojection of MODIS, and the 21km offset problem. I work in<br>UTM30N/European Datum 50, and have found the same problem. Using
<br>Markus' scripts, the problem is still there (in fact, the problem<br>arises in the first warp, from MODIS sinusoidal into WGS84/LatLong).<br>Have you found a way around this?<br></blockquote></div><br>For me the problem is solved using Markus' trick. Here is a simplified version :
<br>1. warp to sphere :<br>gdalwarp -t_srs "+proj=latlong +ellps=sphere" src_dataset dst_dataset1<br>2. force projection to WGS84 :<br>gdal_translate -a_srs "EPSG:4326" dst_dataset1 dst_dataset2<br>3. warp to the expected final projection :
<br>gdalwarp -t_srs EPSG:NNNNN dst_dataset2 dst_dataset3<br><br>A+,<br>Guillaume.<br>