This recent cold spell has once again shown that people are often confusing windchill with an actual temperature.. but these are two entirely different things. Just this morning for example, the headline in the Free Press mentioned how temperatures in Winnipeg, with the wind chill, would approach -45 this morning. In fact, the temperature forecast was -34C which is a lot different than a temperature of -45C, which is different than a windchill of -45.
Temperature is a measure of how much heat energy a parcel of air holds. The more heat energy, the higher the temperature. Windchill on the other hand is an index that estimates the RATE of energy loss of an air parcel given its original heat energy AND a prevailing wind that helps to accelerate the amount of energy leaving the air parcel (i.e. cooling rate). So, temperature is the amount of heat energy an air parcel has.. windchill is the rate of cooling of that air parcel.
Look at it this way... You have a bucket of water at +5C. You put it outside, where the air temperature is a constant +2C, but a wind of 50 km/h is producing a windchill of -5. Will that bucket of water freeze? Of course not. The water will only cool down to the ambient air temperature.. +2C. It can't get any colder than its environment unless its cooled by some other method (e.g. refrigeration) The windchill of -5 merely implies that the water will cool down to +2C at a rate equivalent to that if the water was outside at -5C with no wind. So instead of taking one hour to cool down to +2C, it will only take 10 minutes thanks to that 50 km/h wind. But the water temperature will never go below +2c, no matter how extreme the wind or windchill is. That's the effect of windchill. It describes an equivalent RATE OF COOLING, not an actual temperature. So if it's -20C with a -40 windchill, your car will likely still start if it's not plugged in because the battery will never get below -20C. On the other hand, if it's -40C with no windchill, your car will not start if it's not plugged in because the battery will cool off to -40C (given it's outside long enough).
Before 2001, the Prairies used a windchill index that was given in watts/sq metre (e.g 1800 W/m2) This was a more scientifically valid measure of what windchill actually was.. a rate of cooling, expressed in the amount of energy loss in watts per square metre of area. However in 2001, the windchill was re-calculated and standardized for all of North America to the windchill index we use today, which was mainly based on the more common "equivalent temperature" index being used in Southern Ontario and the United States. A survey conducted at that time showed that most people found the watts/sq metre unit "too technical" a term to fully understand.