I'm historically a user of MapInfo Professional but have recently started using FME and have been talked into now testing out a copy of QGIS.
The main problem I've come across is that the coordinate systems used in QGIS do not appear to match what is produced by the other two?
For example, setting both MapInfo Professional and FME to BNG seems to generate a CRS of: +proj=tmerc +lat_0=49 +lon_0=-2 +k=0.9996012717 +x_0=400000 +y_0=-100000 +ellps=airy +towgs84=375,-111,431,0,0,0,0 +units=m +no_defs
This appears to differ form QGIS's interpretation of BNG which is; +proj=tmerc +lat_0=49 +lon_0=-2 +k=0.9996012717 +x_0=400000 +y_0=-100000 +ellps=airy +towgs84=446.448,-125.157,542.06,0.15,0.247,0.842,-20.489 +units=m +no_defs
I'm going to admit that I have no idea what the towgs84 parameter relates to but the differences appear to cause a few meters of discrepancy when looking at features on the ground. Much as I like the simplicity of QGIS this could affect any decision to move forward with adopting the software in any capacity.
Could anybody provide me with a layman's explanation of what is going on?
Many thanks
Answer
towgs parameter is actually a 7 parameter datum shift from the CRS in question to WGS84 (EPSG:4326), because WGS84 is commonly used as a kind of a "base datum", to which all other CRSs have known 7 parameters transformations. Why the definition in QGIS is different is hard to say, but in QGIS you could easily create a custom CRS which would be defined exactly like the one in FME/Mapinfo.
The parameters are 3 translations in meters, 3 rotations usually in angle seconds and third is scale change given in ppm (in 10^-6 meter).
The parameters from software you are used to seem a bit generic (no decimal parts for translations, angles are all 0, no scale change. This system feels like a translated WGS84, and nothing else) and i would say that the ones from QGIS would probably be correct.
But these parameters only come into play if you would for example be loading GPS data to the project, which is WGS84 based of course, then the data transformed with the given 7param transformation would of course be incorrect, or better... it would be different.
No comments:
Post a Comment