Maybe you misunderstood - I suggest you run your report again at the higher altitudes. This will be informative, and cost you nothing. This site is the best there is in terms of simulating your reception environment. You may not go to 100', but more simulation data will be informative and add another dimension to the results.
Once you exhaust your simulation options, you have to start collecting experimental data, or locate someone with experience in siting antennas in your neighborhood. That's the situation that everyone is in with OTA. At some point you will have to assume some risk and buy some equipment to get real data for your location.
I gave you a link to an antenna that is likely to work for WBRA or WCYB, but not both at the same time. I don't think there's any reasonable way to get both 3 and 5 from your locale without two such antennas and an A/B switch, or two tuners, or a rotator.
For UHF, AFAIK the Antennas Direct DB8e with both panels pointed in the same direction is the most sensitive widely available antenna ... or maybe the Super G is better ...
http://www.summitsource.com/antennac...le-p-7648.html
With the DB8e you could try pointing each panel in a different direction so you'd get 31 and 32 on one panel and maybe 18 and 17 on the other ... but negative NMs are getting really dicey. 2-edge stations should be very sensitive to height, since they have to diffract twice over the horizon. Certainly the simulation does not characterize the diffraction edge, just like it does not include local obstructions like trees and buildings.
You can also stack properly separated UHF (or VHF) yagi antennas - google it and you'll find some info.
You will need a little luck, I think ... but I would take it as a challenge. Good luck!