I understand this type of question has been asked many times, but I'm still struggling to solve my problem.

I'm trying to import the Global Self-consistent, Hierarchical, High-resolution Geography Database (GSHHG) which I downloaded as a shapefile. Additional info (and the dataset is also available through NOAA.

After downloading and extracting the GSHHG archive, I create a new project in QGIS 2.18 and import the "GSHHS_f_L1.shp" shoreline shapefile. This file should be the full resolution (indicated by "f") version of the data containing the boundary between land and ocean, except Antarctica (indicated by "L1") and is in WGS84 (all this according to NOAA).

After importing into a blank project, the project CRS is EPSG:4326, with on the fly reprojection disabled. Which I think is as expected. If I then add the OSM Standard layer using the "QuickMapServices" plugin, OTF reprojection gets enabled but when I look at the shorelines anywhere in the world, they do not align. They are off some small amount consistently everywhere.

If instead of adding an OSM layer using that plugin, I add a layer I have in the same CRS, I still have the same issue. So, it seems to me that the GSHHG dataset is for some reason misaligned. I've tried setting that layers CRS manually to 4326, but it's still incorrect.

Example: enter image description here

Obviously, I'm still doing something wrong. How to correct this misalignment and why is this happening?

  • by how much indicatively is the misalignment? – Michael Stimson Oct 18 '18 at 3:09
  • I think you may be overinterpeting the accuracy of the data. Quote from the NOAA page: 'Accuracy: WVS Plus specifications for data layers at 1:250000 scale require that 90% of shoreline features be within 500-meter circular error of their true geographic location.' After inspection, the data seems pretty consistent with their description to me. – FSimardGIS Oct 18 '18 at 3:10
  • The error is about 500 meters everywhere in my region, so I think MichaelStimson and @FSimardGIS are on to something. But, it's still not clear to me why this and only this layer sit's in the wrong place. If gshhg were broken up by region, and reprojected using localized CRS, would this error go away? – jac Oct 18 '18 at 3:39
  • I did not read the full documentation thouroughly, but from what I saw it seems that this layer was created from multiples sources, and then stitched together in a seamless layer. The shoreline could very well be slightly misaligned in some regions because of the data source, or the process to generate the vectors from these data sources. – FSimardGIS Oct 18 '18 at 3:48
  • Why don't you take the data from openstreetmapdata.com/data/coastlines ? From the file sizes you see that the NOAA data is generalized. – AndreJ Oct 18 '18 at 12:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.