Climate Experiment
I ended up taking Ca-65 for my drive back to the Bay Area. It's not listed as a scenic route and it probably isn't all that unlike any other state highway in the San Joaquin Valley, but the different shades of gold and yellow in the summer-heated lumpy hills were dramatic and beautiful. It reminded me of the landscape descriptions in Red Mars, which I'm currently reading, only with field yellows instead of rock reds. Occasionally I'd pass an orchid or some kind of resort, and it'd be a dark green patch, and I do mean a patch, as if it was a different strip of material on this quilt of countryside.
In case you're interested, as iceblink was, I next took east Ca-190 to a windy country route, J37, to Ca-198. The first part of that last one is designated as scenic, but I didn't think it was as nice as the drive up to that point. I was planning to take it all the way to Ca-25, but by the time I got to the I-5 I was more in the mood for fast freeway driving, so I used that to go 45 miles north to Exit 379. I did it at about 85MPH, faster than about half the traffic and slower than the other half. That exit was county route J1, a fun, quiet, windy road that passes Mercy Hot Springs. I considered stopping for a soak, but the road appealed to me more, especially since I knew the sunset light would make it an even more pleasing experience, and dusk would make it difficult. The rest was the fairly obvious Ca-25 to US-101 to the I-880 to Ca-85 to Manaki House.
The Aircon Experiment
Before all that, I had about 30 miles of Ca-99, a flat, flat, flat open stretch of arid six-lane freeway. I realised it was the perfect opportunity to do a bit of research into the question of air conditioning via open windows.
Background: in the mid-1990s, I drove from Alice Springs in Australia's Northern Territory to Uluru, and then back up to Darwin. There are no speed limits in the Territory, so I averaged about 150km/h, a tad under 95 on the old scale. It was well over 40°C (even more over 100°F) so I eventually closed the windows and turned on the air conditioning. This seemed to affect my fuel economy dramatically, and I proved that this was true: at a constant, high speed on flat open highway, I got about 550km out of a full tank with the air conditioning off and the windows down, and only about 450km with the aircon on and the windows up. This experience has made me skeptical about what I keep hearing here in California, that opening windows creates so much drag that using air conditioner is more economical fuelwise.
The temperature yesterday was about the same, the highway was just as flat, and I happened to be driving an compact-sized Nissan, just like I did in the Territory. It was an entirely different car, of course-cars change a lot in 10 or 15 years-and since the speed limit was 70MPH, I had the cruise control set to 74, a good 30km/h less than last time I did this kind of experiment. Still, I've disproven that open windows cost more fuel than air conditioning, at least for the cars I've tried it with.
My rental car had a digital economy gauge. It read 10L/100km, occasionally floating down to 9, with the aircon on. When I turned it off, it stayed steady at 9, and even dipped down to 8 for a moment. When I used the 'max airconditioning' button, it stayed steady at 10. I turned it back off and opened all four windows. Sure enough, the drag had an effect; it was floating between 9 and 10 again; it's hard to tell if the car was using as much fuel as with the normal aircon switched on, but it definitely wasn't using as much as with the 'max aircon' setting.
I did the experiment again on another flat road at a lower speed, around 45MPH/70km/h. Again, I was using the cruise control and road was completely flat and straight, so it should be a pretty good test; I'm confident that the floating numbers on the gauge was because it was trying to display a continuous scale on a discrete display, not that the actual consumption was changing. The gauge's readings:- Aircon off/windows up: 8L/100km, occasionally floating to 9. Aircon off/windows down: still floating between 8 and 9, a little more in the 9 than before. Aircon on/windows up: mostly 9, occasionally dipping to 8.
Conclusion: I've never seen air conditioning use less fuel than having the windows down. I wouldn't call it a busted myth, since I've only tried it with two cars in not very scientific conditions (e.g. maybe the wind was playing tricks with me, although I doubt it), but in future I will more adamantly take the notion that air conditioning is more efficient than open windows with a grain of salt. It's true that open windows create drag; exactly how much probably depends on the configuration of the windows; that'd be a fun further experiment. But in the absence of other reasons, such as to filter out allergy-inducing air (a real concern in the San Joaquin Valley, especially near those pig farms-phew!), or it just being too hot (even I like aircon when it gets over 40), it makes more sense to open windows than to turn on air conditioning.
YMMV.