INTRODUCTION
Once again its time for the Old Farmer's Almanac to hit the streets and tout how good their weather forecasts have been for the past two-hundred or more years of relying on their "secret formula".  In their latest press release they claim 80 percent accuracy. But how often does anyone go back to scrutinize an old forecast to see if it "verified".  I have found that people often don't even remember last week's forecast let alone one that was printed a year ago.

But, here's what I found by looking back at the regional weather forecasts from last November's Western Edition of the Old Farmer's Almanac (Yankee Publishing, Dublin NH,  2002).  The weather forecast section of the Almanac is divided into 16 regions; with Region 16 encompassing the southern three-fourths of California.  Each month has a summary of the forecast average temperature and rainfall for the month and how much above or below normal the forecast is.   The actual forecasts from the 2002 Old Farmer's Almanac (OFA) for Region 16 (Fig 1.)

Fig.1 
(click to enlarge images)

DATA
To evaluate the veracity of the forecasts each monthly forecast was compared to the actual temperature and rainfall by month for the United States Climate Divisions generated by the NOAA-CIRES Climate Diagnostics Center US Climate Divisions Plotting Page.  Data for climate divisions was chosen because of the large number of sites that are used in determining the division average, thus eliminating the bias of a single station.

U.S. Climate Division Precipitation (Nov. 2002 - Apr. 2003) (click to enlarge images)

U.S. Climate Division Temperatures (November 2002 - October 2003) (click to enlarge images)

Nov 2002	Dec 2002	Jan 2003	Feb 2003	Mar 2003	Apr 2003
May 2003	Jun 2003	Jul 2003	Aug 2003	Sep 2003	Oct 2003

ANALYSIS
The analyses in Table 1 and 2 (below) compare the monthly forecasts for Region 16 from the 2002 Old Farmer's Almanac with the actual observed monthly climate division anomalies for temperature and precipitation.  Because of the negligible rainfall in the California in the summer months, precipitation for the months May through September was not examined.  Furthermore, if either the forecasts or the observed data were geographically split then separate analyses were done for each geographic area.

The observed data has been color-coded to reflect each of three categories.  Forecasts which were essentially correct are blue.  Those which had the right sign (i.e., above normal observed when above was forecast or below normal observed when below was forecast) but where the quantity was incorrect are coded in yellow.  And forecasts which had the wrong sign are coded red.

SUMMARY
For California, Region 16, the precipitation forecast (Table 1)  was only correct for one month (February 2003) and that was only in Southern California.  Consequently of the 6 months of rainfall data that was evaluated the Old Farmer's Almanac was correct just 8 percent of the time and had the right sign but incorrect quantity another 8 percent, meaning that it was wrong about 84 percent of the time.

The temperatures forecasts for region 16 (Table 2) were correct for 5 of the 12 months or about 42% of the cases. If climatology had been used it would have been correct 58% of the time.  The OFA was wrong 58% of the time, with the wrong sign in half the those cases and the right sign but wrong amount in the other half.

Table 1.  Precipitation Analysis Region 16 - California

Table 2.  Temperature Analysis Region 16 - California

	Farmers Almanac	Observed
Nov-02	-1̊  NW, +1̊  SE	+2̊  to 4̊ NW, +2̊  to 4̊  SE
Dec-02	Avg. North, +5̊  South	+2̊  to 3̊  N, Avg. S
Jan-03	+4̊	+4 to 6
Feb-03	+1̊  East, +3̊  West	Avg
Mar-03	-1̊	+2̊  to 3̊
Apr-03	Avg. East, +5̊  West	-3̊  to 5̊

May-03	Avg. East, +3̊  West	+1̊  to 3 ̊ E, Avg. W
Jun-03	-3̊  East, +3̊  West	+2̊  to 3̊  E, Avg to +2̊  West
Jul-03	+3̊	+2̊  to 4̊
Aug-03	+4̊	Avg to +2̊
Sep-03	+4̊	Avg to +4̊
Oct-03	+5̊	+3̊  to 5̊