Actually, I think all five factors
1. Lower heating value of winter blend fuel
2. Increased thermodynamic losses due to cold induction air/fuel
3. Increased aerodynamic drag due to increased air density
5. Sticky EBPV
have an effect. How much each one affects you depends on where you are and how you drive.
This is a highly variable factor and it is almost entirely dependent on where you operate your truck. “Winter blend” is a mixture of No. 1 (kerosene) and No.2 diesel, just as jet fuel is. Winter blend can vary from 5% No. 1 (the lame stuff we get in central Indiana) to the 66% No. 1 mix (good for 40 below) you get on the Great Plains. I suppose folks in non-coastal Alaska use jet fuel which is about 85% No.1. No.1 has lower heating value than No. 2 – 120,000 Btu/gal vs 140,000 Btu/gal and winter blend is just a pro rata of these figures. High grade Nebraska winter blend has a heating value of about 127,000 Btu/lb. If that were the only factor a Nebraska panhandle truck would see a loss of about 9.3%. A 18 MPG truck would drop to about 16.3 MPG. Trucks operating in the heart of Dixie should see no MPG drop. But we know that Great Plains trucks lose more than 9% MPG and southern trucks do experience MPG drop in the winter, so winter blend is only part of the answer.
The case for thermodynamic losses on short-run trucks is sound. School buses often run with their Cold Fronts completely zipped up because their stop-and-go operational pattern precludes them from ever really getting warm. Truly long-haul trucks always unbutton their cold fronts on long runs because the trucks get plenty warm due to road load. Class 7 and 8 trucks have staggeringly powerful (in terms of BTU/HP) cooling systems because of the prolonged high-load operation they face, so they do keep the front partially closed until the weather warms up. I think thermodynamic losses are important if you are making runs of less than 5 miles but beyond that the engines and fuel warms up to the point this factor recedes in importance as the trip gets longer.
This factor affects everyone to varying degrees. A vehicle becomes a crude propeller because it pushes air out of the way as it moves along. The vehicle moves cubic feet of air, but the work necessary to move air depends on the weight of the air moved. A cubic foot of air at 18,000 feet MSL weighs half of what air does at sea level, so an airplane get better MPG at higher altitudes. Air at 85 degrees and 40% humidity (common Midwestern summer conditions) weighs roughly 0.055 lb per cubic foot. Air at 20 degrees and 50% relative humidity weighs roughly 0.063 lb/cubic foot. That is a 14.5% difference. Your 18 MPG truck (summer) should have its winter mileage reduced to 15.4 MPG if you do a lot of Interstate cruising.
Snow and Ice
Snow, in particular, imposes rolling resistance. How much varies according to the depth and type snow. Western powder and eastern slush would have different effects on MPG. Probably more important is how many days of it you get. Denver can get heroic amounts of snow in short periods, but it usually melts off in a few days. Buffalo gets hammered day after day for months on end, so the guy in Buffalo pays a heavier MPG price than the guy in Denver while the guy in Houston doesn’t understand. Deeper snow also may require use of four-wheel drive which imposes more mechanical friction.
These trucks have a back pressure valve to enhance engine warm up. It imposes a heavy load on the engine. I noticed that no matter how clean I got my setup, the darned thing blew from October to May, so it got electrically disconnected. Bingo! I regained 1.5 MPG although the truck takes three times farther to warm up. Maybe a compromise would be to control the valve with a manual switch.
The Perfect Storm
All these factors can add up. My “Perfect Storm” scenario is a truck used for short runs in the Upper Peninsula of Michigan with a sticky EBPV. Very cold temperatures and strong winter blend at relatively low altitude and lots of snow. Such a truck may see it winter MPG drop to half or less of its summer value. The converse is also true. The guy in Houston (is it ever truly winter in Houston?) may only see a 13% drop due to cool temperature air.