NWS forecast last updated 3:19 am PDT Sun Sep 2, 2007:
high,ºF low,ºF
NWS CWD dev NWS CWD dev
Sun 9-2-07 73 70 +3 47 39 +8
Mon 9-3-07 72 69 +3 44 46 -2
Tue 9-4-07 70 69 +1 40 32 +8
Wed 9-5-07 64 63 +1 39 34 +5
Thu 9-6-07 65 65 0 41 35 +6
Fri 9-7-07 64 68 -4 42 36 +6
Sat 9-8-07 65 63 +2
NWS forecast last updated 8:47 pm PDT Thu Sep 13, 2007:
high,ºF low,ºF
NWS CWD dev NWS CWD dev
Fri 9-14-07 66 59 +7 31 27 +4
Sat 9-15-07 60 61 -1 34 26 +8
Sun 9-16-07 55 58 -3 36 25 +11
Mon 9-17-07 55 52 +3 39 25 +14
Tue 9-18-07 57 59 -2 40 39 +1
Wed 9-19-07 56 49 +7 34 22 +12
Thu 9-20-07 53 33 +20
================================================================
SUMMARY OF THE THREE SETS OF DATA
largest |dev|'s
high,ºF low,ºF
day NWS CWD dev NWS CWD dev
one 66 59 +7 47 39 +8
two 69 75 -4 34 26 +8
three 55 58 -3 36 25 +11
four 55 52 +3 39 25 +14
five 57 59 -2 48 39 +9
six 71 61 +10 34 22 +12
seven 53 33 +20
Some personal observations:
1) Note that in general, the more days into the future the forecast, the less accurate it is expected to be. This isn't obvious from a cursory look at the data.
2) For the worst case in the first 2 days, the temperature was 8ºF colder than forecasted. For the worst case in days 3-4, the temperature was 14ºF colder than forecasted. For the worst case in days 5-7, the temperature was 20ºF colder than forecasted.