Approximation of temperature changes:

highest temp. 25th. at 2:00 pm = 54 degrees
assume an estimated drop from 54 to 52 degrees by 4:30 pm, just before sunset at 4:41 pm.
lowest temp 26th. morning of 26th.= 6 degrees at 6:00 am
difference in temp. = 46 degrees
difference in elapsed time = 14.5 hours
rate of drop in degrees per hour = 3.4 degrees/hour

assume 6 degrees temp at 7:30 just after sunrise at 7:20 am
highest temp. 26th. at 2:00 pm = 51 degrees
difference in temp. = 45 degrees
difference in elapsed time = 6.5 hours
rate of rise in degrees per hour = 8.2 degrees/hour

This is not exact but will give us an idea of what the temperatures were like that night.
What really happened would have depended on a number of other factors that would be impossible for me to add to the model.

25th. 2:00 pm 54
3:00 pm 54
4:00 pm 53
4:30 pm 52
5:30 pm 49
6:30 pm 45
7:30 pm 42
8:30 pm 38
9:30 pm 35
10:30 pm 32
11:30 pm 28
mid am 25 Snow would have begun falling at dew point of 26 degrees
1:00 am 21
2:00 am 18
3:00 am 15
4:00 am 11
5:00 am 08
6:00 am 06
7:00 am 06
7:30 am 14 Trace of snow on walkways in direct sun would start to melt between
7:20 am & 9:30 am as the concrete/bricks began to heat up, even though the air
temp remained below freezing,

8:30 am 22 Trace of snow on the yard in direct sun would start to melt
9:30 am 31 about 9:30 "old" snow left over from the 16th. & 17th., and dusting,
in shaded areas, would begin to melt, as air temp moved above freezing
10:00 am 35

In reality, the rate of cooling would be slowed by cloud cover. It would also
be faster to start with and slow the closer the air temp got to 6 degrees. The wind was blowing at about 11 mph/gusting to 21 mph which would have speeded up the cooling effect. Also, I think it is possible for the higher altitude air to be warmer than the ground temp, consequently, it may not have started snowing exactly when the air temp near the ground (5 feet above the ground) dropped below freezing. Or vice versa. And, we don't know the rate of the snow fall, other than it was "light" or "trace", and we don't know the duration.

The conclusion I draw from this exercise is that the light snow fall probably occured between midnight and 2:00 am, which would be earller than I had previously thought, but this seems to be more of an educated guess than just a plain guess based on possibilities. It also tells me that an intruder would have had a one to two hour window to do his crime. My view is that this murder was planned and carried out in under 30 minutes. Yet, the one or two hour window allows for a more disorganized killer. It also fits the extimated TOD.

If anyone has previously saved HOURLY data from Dec. 26, 1996, when/if it was availiable at the time, and is willing to share, we probably could be more specific.

And a special link just for Mikie:

http://www.wunderground.com/sky/ShowSky.asp?TheLat=39.900002&TheLon=-105.099998&TimeZoneName=America/Denver&Year=1996&Month=12&Day=26