The author clearly states that his analysis addresses changes in aerodynamic loss as a function of temperature only.
Many things affect mileage as a function of temperature, but they were not part of this guy’s work.
I have dabbled with a few goodies that impact efficiency on internal combustion motors.
Efficiency, and thus milage (not to mention torque produced per power stroke) greatly and measurably increases when the air is more dense, either due to increased barometric pressure, or due to temp changes
On clear winter days, when the pressure is up, makes for excellent riding (just wear lots of layers of clothes!)
As a bonus, the efficiency of combustion, a more complete burn of all the fuel molecules present, actually increases the milage.
Just to give the author his due, I suspect that a vehicle with increased sensitivity to wind pressure, ie a large cross section offered to the headwind, or a low power to weight ratio would in fact see greater changes associated with the wind drag than with the measured increase in torque applied to the drive wheel(s) on the ground.