Connect the Dots across the 180 degree line of Longitude

01-01-2013 11:15 AM
New Contributor
I have thousands of storm observations downloaded from NOAA's web site as points (each point is an observation/measurement/location).  When I try to convert those storms points to storm tracks using either the Points to Line tool in ArcMap or the ET Geowizards Points to Polyline tool, any storm track that crosses the 180 degree line ends up circling the wrong way around the world.  I've attached an image.

I imagine this is because when these tools connect the dots they move East to West by consecutive X/Y positions.  When one point is on one side of the 180 degree line at say... 177 and the other is in the Western Hemisphere at -179, rather then taking the shortest path to connect the dots, the path goes to Longitude 176, 175, 173...2...0..-1...-100...-179.

I've tried numerous methods to stop this.  What worked for some of the points was to reproject the data using a UTM system, then calculate Lat and Lon in Meters, export the attribute table, and import that new table displaying X and Y using the newly calculated positions.  But that only works for some of the data points (the points that fall within the UTM zone I selected). 

Is there either a global projection that does not use negative values at all, or is there a way to tell ArcMap to follow different rules when connecting the dots?

It seems that even NOAA hasn't solved this.  If you download the same data sets as polyline shapefiles (which they offer) no storms cross the 180 deegree line of longitude.  There is an obvious gap in their data.... the same gap seen in my image.

Perhaps some sort of routing is the answer?

***Update.  It seems that if you right click on the Points to Line tool and select Properties, go to Parameters, then you do have the option to set a direction value.  However these options are greyed out for me and I am unable to do anything but view the Parameters listed in the Pane.
0 Kudos
1 Reply
Regular Contributor II
you cannot ever draw a line across a projection interuption.
If your projection is interupted at 180° and -180°, the only existing path between them crosses 0°

The UTM solution is on the right track, but you might try another projection:
an equatorial stereographic azmuthal might work, centered at 179°.

Another solution is to shift all your locations to a 0 - 360 system, construct your lines,
and split the resulting tracks at 180.
Then shift it all back to the -180 to 180 coordinate range.
0 Kudos