I can suggest the app ABRP (a better route planner) as a good tool for estimating battery energy needed for the mountain route segments. The elevation changes, assuming the roads are known to whatever mapping this app uses, are included in the app's estimate. The variables of the EV make and model, battery degradation, usual speeds, ambient temps, rain-or-snow -- these are all settings in this app, for the particular trip being planned.

Before I knew about the app ABRP, I had used my desktop PC and Google Map to check my planned highway/road routes for total amounts of elevation gains and total descents. The bicycle option in GoogleMap, being sure to keep the route on the highway (not some alternate paths), shows the elevation totals for the up and for the down. I used basic physics to calculate the battery energy in kWh to lift the Bolt 1000 meters, assuming the Bolt plus some passengers and cargo as a total 2000 kg. Raising a 2000 kg weight 1000 m increases its potential energy about 5.5 kWh and to do that work the Bolt needs to deliver that energy at the tire-road interface. Assuming about 20 percent losses for the dc-to-AC conversion, motor losses, and some more drivetrain loss, I added a guesstimate of 20 percent. This means 1.2 x 5.5 = 6.6 kWh of battery energy was my estimate just for the raising by 1000 m. I would add that to the flat-road energy needed to drive the km of distance. This 6.6 kWh for 1000 m gain is not far from EVphoto's 1 kWh for each 500 foot gain.

Initially I was then estimating the "credit" for the total descents by thinking that the descent was giving back the potential energy, but with some losses and so the energy going back into the battery could be guesstimated as 20 percent less than the basic 5.5 kWh, or an amount 4.4 kWh per 1000 m of descent. Recently I realized that during the descent this assumed 20 percent loss would be reasonable only if the descent was so steep and the speeds such that the driver's screen actually shows some kW of power going into the battery. Most likely the slope and speed (at least at highway speeds) would be such that the forward travel power is more than this "descent power", so the driver's screen still would show that power is going from the battery toward the drivetrain. In that circumstance, the descent energy that the Bolt (plus cargo) is giving back is all going toward reducing the energy needed to drive the Bolt along the road. My observation then is that the Bolt's mass is directly contributing to pushing the car forward on the downward slope, and so reduces the power needed from the battery, by that full amount. Therefore I really should have given the descents the full credit at a value of 5.5 kWh per 1000 m of descent. I realize that there are variables like actual weight of Bolt plus its load, and the road surface, but these estimates did seem to be reasonable, even when I was still doing the descent's credit with my first (false) guesstimate for the descents.

This physics exercise was interesting even without much ability to test my guesstimating, and it did give me the comfort for my mountain highway trip planning in my first years of Bolt driving. Now, I would just use the app ABRP and with experience on my usual highway routes, that provides enough confidence. Also, in recognition that my usual mountain highway routes do not have DCFCs closely spaced, I use a strategy of arriving at a DCFC with enough left to reach the next DCFC station.