Those pictures actually look better than I was expecting. Of course, the "straight into the Arduino" one is perfect, but I was expecting to see much longer rise times at the end of the long wire.
I'm a little surprised at the overshoot in the yellow traces. With just a resistor acting as a pullup, I would've expected a rounded off trace approaching the 5 volt limit. Something must be putting energy into the cable to cause the overshoot? Do you have the cable all coiled up, perhaps still on the cable spool? I wonder if it's acting as an inductor, that could be where the extra energy is coming from to overshoot in both the positive and negative directions. Does it look the same when the wire is stretched out? (Perhaps without the inductive kick the edges get rounded off more?)
The timing of the pulses hasn't changed much, they just got delayed a little bit. The timing still looks like it within specifications.
It's been over 35 years since my university electronics courses, and I don't often deal with the intricacies of a long transmission line. Hopefully someone with some more experience here will see something in these pictures.
Rockwallaby makes some good points about being near the power lines. Is that where they are while you were making these pictures? If not, was it working properly when you were taking them?