You're going to get what you are going to get. It's not so much that the weather is cold but on how you drive in it. Had to go to Baltimore from McLean the other day. OAT was about 37 °F. It's a couple of miles from my garage to the beltway (with its heavy but not too heavy as USG is shut down) traffic, then up 95 and into Baltimore traffic to the destination (Johns Hopkins). 49 miles, 331 Wh/mi. Parked on the roof of the parking garage and a couple of hours later left from there to the Baltimore Super Charger. That's 5.7 miles (including the missed turns) and temperature had dropped to 36 °F. Consumption for that little leg was 503 Whr/mi. Not so good. From the charger I was right back on 95 and ran home to McLean logging 47.5 mi at 306 Whr/mi. Now that's more like it. So why so much variability? Here are a couple of things to think about.
1. A big factor in the consumption numbers you see in the Trip data is that when you first start out everything the car has used since you woke it up is logged in the first 10th mile. Thus if you, as I did at Baltimore, set the car to be toasty warm when you get to it 10 - 15 minutes later, all the energy the heater used whilst waiting for you to arrive gets assigned to the first 10th mile of the drive. If you look at the trip data as you go you will see Wh/mi numbers that are often well over 1000 in the first tenth and staying high in the hundreds for the first mile or two then gradually tapering off as that initial bolus averages down.
2. The motors are very inefficient at low speeds. To get starting torque that magnetic field is whipping past that stator to develop the large currents needed for the large starting toque. Power dissipated goes as the square of the current. Looking at data gathered by ABRP it appears that whereas the car will deliver a mile for 320 Whr at around 40 mph (the sweet spot) it takes 395 at 20 mph and 600 at 10. Thus if the early part of your drive involves stop and go in city traffic you mileage is going to be pretty poor.
3. If the battery has gotten cold soaked because you have left the car out overnight or parked it exposed to wind on a cold day the battery's performance will be limited. In moderately cold weather (30's) it appears that the only effect is that there is a limitation on regeneration. But this can be an appreciable effect on mileage. Whenever you use the brakes the heat they dissipate is lost to the surrounding air whereas with regeneration a substantial portion of it is recovered. In more extreme weather the battery heater may come on. It is reported to draw as much as 4 kW. That's quite a bit of energy per unit time and clearly will effect mileage. The strategy with respect to this is to keep the car in a heated garage or, if you are on travel consider charging the car (at a super charger) in the morning prior to departure as opposed to in the evening when you arrive with the idea being that some of the charging current goes to warm the battery rather than charging it. I'm not hopeful that this is going to help much. The only supercharge for which I have data indicates 99.7% efficiency. Home charges, by comparison, run 87 - 91% efficient meaning more of the charging power goes for production of heat but, of course, some of that is dissipated in the transistors in the OBC - not in the battery and the charging power level is lower.
Summary of the first 3 items: Longer trips at average speeds near 40 mpH are going to show better mileage that shorter trips at lower speed or with lots of stop and go.
4. Wind is a huge factor. Driving 65 mpH into a 20 mpH headwind requires the energy per mile that driving 85 mpH does and that energy increases, theoretically, as the square of the airspeed (speed + headwind) and, according to the ABRP data, to the 2.52 power.
5. Load increases bearing forces, deformation of the tires and deformation of the roadway. These result in increased power consumption.
6. Road surface. It should be obvious that it takes more power to plow through mud or snow than to zip along smooth, dry concrete.
7. How much of the trip is up and down hill. Even if you leave from sea level and arrive at sea level you will use more energy if you go through hilly terrain than if the whole trip is at sea level. This is because energy recovery from regeneration isn't 100% efficient.
You are trying to predict what's going to happen on your long trip to Florida presumably using ABRP or something like it and for now that is about the best you can do. But you must tell ABRP what you consumption is, whether it is going to be raining on the day you go and what kinds of winds you will encounter. Clearly there are questions as to what to enter for the first number and it is impossible to predict the weather weeks in advance.
What I would suggest is that you get some data on how you drive your car and the best way I know of to do that is TeslaFi. After letting it log data for you over a few dozen trips you will have a pretty good picture of what to expect under broad categories of driving conditons e.g. urban, suburban, freeway. This will give you better numbers to put into ABRP and give you a plan to take (or send) to the car. But monitor the available information from the car as you go a prepare to modify the plan according as to what that information tells you. You are actually having to think more like the pilot of an airplane. He can't just check his fuel and pull over anywhere when he thinks it's getting low. He has to always be sure he has enough to get to an airport (and under IFR flight rules to an alternate airport and then half and hour beyond that). You must always be able to get to a Super Charger (and perhaps one beyond the one you are planning on or at least a ChaDeMo). I think this may be upsetting to some but it seems sort of fun to me.
Note that the use of TeslaFi has implications on range as it is constantly pinging your car for status information which keeps it awake and consuming power. You can configure it to sleep to minimize this and, of course, on shore power it doen't matter.
