Investigate sharp discontinuities near edges of reachable area

[dfsnow]

Investigate sharp discontinuities near edges of reachable area

[eric-mc2]

@dfsnow For some perverse reason, the plot look discontinuous or smooth depending on whether I've left or inner-joined the travel times onto the tract geometries, even though the left-over rows from the left-join shouldn't make it into the plot.

But only for IL. Haven't haven't seen discontinuities in DC metro or NY metro areas yet.

A hypothesis from Nico is that these are legit and reflect the edge-of-service at the city border

[dfsnow]

Let's test with DC and see if we see the same thing. I'm going to play around with some of the other parameters this weekend as well. I think the 3.6 km/h walk speed is probably too slow.