Interpreting data from a 'flight tracker'

If you click here, you’ll see a minute-by-minute account of a current Westjet Airlines flight from Vancouver to Toronto. I am confused by the rather large disparity between the data provided by ‘Toronto Center’ and that from Winnipeg and Minneapolis. (NB - the Toronto data begins to arrive at 10:00 PM)

Whereas earlier in the flight, things such as altitude and latitude/longitude differed only trivially between flight tracking centers, that is not the case for Toronto Center. It gives a latitude that’s over six degree further south than the other centers’, a longitude over 11 degrees further east, and an altitude that differs by as much as almost 2000 feet.

Is this normal? Have I misinterpreted things? What gives?

Thanks!

And now I see that the Toronto data is virtually in tandem with the other center (as it presumably should be). What was going on between 10:00 and 10:37? A “typo”? I would have assumed these type of things are immune to gross errors.

WAGs.

One of the ATCs needs calibration?

One of the ATCs range for accuracy is limited?

Here in the US, the data is delayed to the public.

I’m sure a Doper pilot will be along shortly to blow all this out of the water (did you really think I would say blow it out of the sky?) with cites and references.

This is professional-pilot-level guesswork, but …

ATC radar identifies a particular aircraft by its transponder code. It’s a 12-bit code so there are only 4096 possible values, and many are reserved. So no more than, say, 3500 codes are really available.

There are a lot more that 3500 aircraft flying in the US or Canadian airspace at any given time.

The trick is no code is duplicated within a center. Most times most aircraft can fly their full route using a single code. But if two aircraft fly into a center while carrying the same code, the ATC computer will detect & flag the problem, and the controller will assign a new code to one of them. The controller tells the ATC computer too associate the new code with one track & flight plan, and meanwhile the pilots dial the new code into their transponder. A few seconds later everything is resolved, and the two aircraft’s data streams (called “tracks”) are separated.
My WAG is that you were really seeing two aircraft in two adjacent centers that happened to be on the same code. And the website software probbably doesn’t do a real good job of accurately emulating the ATC code conflict software. Or the data feed from ATC to the website is simplified & doesn’t correctly cover for this possiblity.

The positional resolution of ATC enroute radar is between 2 & 5 miles depending on the specific installation, terrain, aircraft distance from antenna, etc. But the altitude is reported by the aircraft as part of the transponder response data, is reported in 100 foot increments and is accurate to within about 150’. A large altitude discrepancy would be very unlikely.

Now, what’s interesting is the distance between the two spots reported as visualized on a map.

Google Maps Note that they are not exactly “same minute” positions, but ones that matched to mappable addresses, and were “close enough”

They are clear on other sides of the great lakes.