ansell
commented
5 years ago Two tickets on helpdesk referencing this:
- 29340
- 29341
The other is for another record in dr376:
https://biocache.ala.org.au/occurrences/c69255a5-55bc-4d9d-8b68-382b2616c4a7
Open ansell opened 5 years ago
ansell
commented
5 years ago Two tickets on helpdesk referencing this:
The other is for another record in dr376:
https://biocache.ala.org.au/occurrences/c69255a5-55bc-4d9d-8b68-382b2616c4a7
djtfmartin
commented
5 years ago On first look, I think it's reprojection of AGD84 to WGS84 (EPSG:4326), and GDA94 to WGS84. This is expected behaviour.
ansell
commented
5 years ago Thanks, I knew I was missing something but didn't click that it was that.
ansell
commented
5 years ago A further query from the support ticket has been made, as it doesn't appear that AGD84 to WGS84 explains this issue by a few orders of magnitude. There may be something else occurring.
That does explain it to some extent, however A conversion from AGD84 to WGS84 (or to the EPSG:4326 projection) in that area would be a movement of approximately 207 meters toward the North East. The figure you have given by rounding moves the point around 26,000 meters to the North West. This is not encompassed by the uncertainty of 10,000 meters. It does place the point around 26,200 meters from where it was reported. A more accurate placement would be still around -25, 135.24 with an uncertainty of 10,000 (actually the MANIS calculator gives an uncertainty for that verbatim location plus the geodetic conversion of 1,499.6 meters). The uncertainty inherent in the Geodetic Datum Conversion (~207 meters) is less that the precision of the verbatim coordinate of .01 degrees (~1,700 meters or so at this latitude)
charvolant
commented
4 years ago Also, coordinate conversion on functional test data seems for be failing for datums supplied as AMG66 or GDA94. Location-1.csv.zip
The following record, (
AD 97731494indr376) doesn't look like it is sensitive, but thedecimalLongitudeon it has been modified by the processor.https://biocache.ala.org.au/occurrences/8b621b42-7829-4b52-b618-98945b5598ef
The original line in the
dr376dump that it was loaded from is the following, which includes25asdecimalLatitude,135.4asdecimalLongitude,"25,00,,S;135,24,,E"asverbatimCoordinates, and10000ascoordinateUncertaintyInMeters:The record in Cassandra appears correct for
decimalLongitudebut notdecimalLongitude_p:Running
biocache process-single 8b621b42-7829-4b52-b618-98945b5598efin verbose logging mode shows the following which I can't find any further clues in as whendecimalLongitudefirst appears on the processed record it shows up as135.0: