I am tracking my kid's flight from LAX to EWR. It seems like a pretty straight West to East kind of flight but the flight path has the plane moving somewhat north through Ohio and then angling back to the southFlight path

Why wouldn't the flight just keep going East?

  • 9
    $\begingroup$ The Earth is a sphere (to a first approximation), which means that the shortest path is not what it looks like on a map. This is the actual shortest path, and you'll see it fairly approximates what the flight is actually doing (the variation is due to the fact that planes fly on straight paths between predefined waypoints which aren't designed for smooth turns for every possible routing). $\endgroup$
    – waiwai933
    Jun 2, 2014 at 2:05
  • 3
    $\begingroup$ @waiwai933 That should be an answer, not a comment. Comments are for asking for clarification and so on, not for answering the question. $\endgroup$ Jun 2, 2014 at 7:36
  • $\begingroup$ The path is a Geodesic, see this - astro.umd.edu/~richard/ASTRO340/class11_RM_2014.pdf $\endgroup$ Jun 2, 2014 at 10:32
  • $\begingroup$ @AwalGarg - the map shows a sharp corner in the path. Geodesics don't have sharp corners; there's something else going on here. $\endgroup$ Jun 2, 2014 at 23:43
  • 1
    $\begingroup$ @AwalGarg the path isn't Geodesic. Airplanes (especially in the east) are not cleared direct to their destinations, which is the only way they would be on a true great circle. This path is just a result of the points along his cleared flight plan. See my answer below for details. $\endgroup$
    – casey
    Jun 3, 2014 at 13:32

3 Answers 3


Aside from the storms causing an earlier re-routing, that jog north is to join the arrival into EWR. A common fix in NE Ohio on flight plans into EWR is the DJB VOR which is in Loarine county (I think, somewhat near Cleveland). and then routing over Slate Run and then Williamsport, PA on the way into Newark. If you are a jet coming from the west, this is literally the only way into EWR. The other two ways in are from the south joining somewhere on a line between Atlanta and Baltimore (which you can transition to from the west by going over Beckley, WV) or from the north coming in over Albany, NY. The primary reason for the strict routing starting at around the OH/PA border is the need for multiple arrival descents for both jets and turboprops into the various NY area airports (EWR, JFK, LGA, TEB, etc), the westbound departures out of NY center and the letters of agreement between Cleveland and NY centers to manage these traffic flows.

See this image: enter image description here

This represents a route of ..J29.DORET.J584.SLT.FQM3. This is not the exact route a LAX originating flight would normally get as they would end up a bit further north (going over Detroit instead of Cleveland), but they would still converge on J584.SLT.FQM3 (the portion east of the OH/PA border). No matter your position west of the OH/PA border, Cleveland center is going find a way to put you on J584 before you get very far into PA. This is why the flight couldn't just go direct to EWR and had to first fly north.


Part of the answer is that, as noted in waiwai933's comment, the earth is (roughly) a sphere, not a flat plane. This means that the shortest distance between two points is not a straight line on the map but what's called a great circle. To find this, you'd take a circle whose centre is the centre of the earth and which passes through the two points you're trying to connect. The shorter route along that circle is the great circle distance.

The intuitive reason for this is that the earth is "narrower" closer to the poles so you can gain a lot by flying towards the pole, using the "narrowness" and then flying away from the pole again. The "ideal" route from LAX to EWR is this (images from gcmap.com):

enter image description here

A much more pronounced version of this would be if your son was flying to Casablanca instead of Newark.

enter image description here

Although Los Angeles and Casablanca are at roughly the same latitude, the shortest route between them actually goes over Canada, over a thousand miles to the north of LA!

  • 5
    $\begingroup$ As far as not being a straight line on a map, that really depends on which map projection you're using. The gnomonic projection shows great circles as straight lines. On the other hand, your 'typical' (Mercator?) map isn't going to do that. $\endgroup$
    – Bob
    Jun 2, 2014 at 8:57
  • 3
    $\begingroup$ If you have a globe, an easy way to find the great circle path between two given points is to put a piece of string on the globe, one end at each endpoint, and stretch the string such that it presses against the globe; it will then trace the great circle path. Of course on a real-life flight, there will be other considerations as well, but to a first order approximation (same as the Earth being spherical) this method works well. $\endgroup$
    – user
    Jun 2, 2014 at 9:10
  • 1
    $\begingroup$ 'the earth is "narrower" closer to the poles' -- to be precise, the Mercator and other cylinder-based projections are more stretched closer to the poles. The extreme being that the single point of the north pole itself is stretched out to form the entire top edge of the map. You can "teleport" instantly anywhere you like along that line, if you're willing to fly over the pole. That's also why Greenland (near the pole) looks roughly the same size as Africa (straddling the equator) on a Mercator map, despite in fact being about 1/15 the area. $\endgroup$ Jun 2, 2014 at 10:42
  • 1
    $\begingroup$ @SteveJessop What you say about map projections is absolutely correct and very relevant but the earth really is narrower closer to the poles. You can view it as a stack of slices parallel to the equator and the radius of those slices gets smaller as you approach the poles, since the earth isn't a cylinder. In fact, the radius decreases slightly faster than it would even if the earth were a sphere, which makes it even more advantageous to go closer to the poles. $\endgroup$ Jun 2, 2014 at 17:59
  • $\begingroup$ @SteveJessop: Mercator maps elongate north-south distances toward the poles just as severely as east-west distances. The map distance between 70N and 83N would be roughly comparable to the distance between 70N and 70S or between 83N and 86N. A Mercator map which stretched from 86S to 86N would need to be more than five times as tall as one which went from 70S to 70N. $\endgroup$
    – supercat
    Jun 2, 2014 at 22:56

Actually, that flight was first routed South of its original flight path, in order to avoid a very large storm system in the mid-west. At that point, it still could not head straight to Newark, due to another storm system in Indiana.

They were originally flying a little more East, but then got routed a 'smidge' North, to avoid the system, most likely when the system didn't move as fast as they figured. As to why they kept heading North-East, instead of a more direct path from Indiana, I cannot tell.

If you (quickly) look at the flight on FlightAware.com, you can see the course corrections due to weather.

scratch the following, the TFR (temporary flight restriction) isn't until later this week.
Another factor for this flight, was that it looks like once they went North to avoid the weather in Indiana, when they came east of the weather, if they'd have gone straight towards Newark at that time, there is a NOTAM for a TFR in Western Pennsylvania. So they most likely were shunted farther North by ATC in order to avoid the TFR.

  • $\begingroup$ @Articuno Thanks. The irony, of course, is I'm the one who brought up acronyms in meta... d'oh $\endgroup$
    – CGCampbell
    Jun 2, 2014 at 3:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .