California Climate, PDO, LOD, and Sunspot Departure

Length of Day Compared to Temperature

Above: Length of Day Compared to Global Temperature since 1890

Below is a paper by former California State Climatologist, Jim Goodridge. Jim  has been quietly working on a  number of rainfall and climate projects both for himself, and as a consultant for the California Department of Water Resources.

In this essay, Jim looks at many factors that can be linked to California’s climate. and also into something that hasn’t gotten a lot of discussion – Length of Day. It seems there may be some correlation to surface temperatures and the Pacific Decadal Oscillation. Jim also has looked at accumulated departure of solar irradiance and of sunspot numbers.

I present his essay here without comment, for the purpose of discussion. My only contribution, just for the sake of curiosity, was to provide the graph above that compares Length of Day to Global Temperature for the last 100 years.

For those that wish to distribute this essay, here is a link to it in Adobe Acrobat PDF form:

Persistence in California Weather – J. Goodridge (PDF 893k)


Persistence in California Weather Patterns

Jim Goodridge

 State Climatologist (Retired)



The evidence for a major climate shift since the mid 1970s is quite real.  California indices of rainfall and temperature have both shown an increasing trend since 1975.

Physical changes in Earth weather systems that accompany the 1975 weather trend changes include the Pacific Decadal Oscillation (PDO) index, a 1975 change in the Atmospheric Angular Momentum (AAM) index and a 1940 increase in solar irradiance.

The PDO index is a measure of the East West sea surface temperature difference in the North Pacific Ocean. The AAM index is a measure of the ratio of East West vs. North-South winds on the planet that affect the earth rotation rate.

Solar irradiance has been monitored from satellites for three sunspot cycles. The sunspot numbers and solar irradiance were shown to be highly correlated. Since sunspot numbers have been increasing since 1940 the irradiance must also be increasing.

The several occurrences of 100-year storms on the American River in the last 50 years are suggestive of climate fluctuation. The search for a rhythm of the records is the object of this essay.

The rainfall record at Sacramento is used to illustrate how a clear view of trend can be obscured by a large number of data points. Rainfall has been measured at Sacramento since 1849. The measurements were originally compiled for the Smithsonian Institution. They were continued by the Army Signal Service and continue to this day with the National Weather Service (NWS). The records used here are tabulated as water year (October to September) totals. These records are used without any adjustment for possible changes due to the many station moves or other factors like changes of gage orifice height above ground surface.

Rainfall at Sacramento, CA

The long-term rainfall trends were clearly apparent when plotted as accumulated departure from average. The Sacramento rain record consists of a relatively wet period from 1850 to 1910 and a dry period from 1910 to 1936. These are not readily visible in annual data plot. An increase in rainfall trend followed the sever drought of 1977. This is quite apparent in the accumulated departure from average plot.

Accumulated Departure from Average Rain at Sacramento, CA

A temperature index was prepared for California by averaging all of the 47 station NWS records with complete data for the 1900 to 2006 period. These records were not examined for in -homogerities. The major factor in air temperature variation in coastal California is the influence of SST.  Another factor is urban waste heat.

107 year California Rainfall Trend

Accumulated Departure from Average for 47 California Detrended Temperature Records

One of the key climate variation indicators for California is the Pacific Decadal Oscillation Index (PDO). This is calculated monthly at the University of Washington. Their web site is . This indexes the East-West sea surface temperature (SST) difference in the North Pacific Ocean, North of 20° North latitude. The PDO index started in 1900 with SST measurements compiled from ship reports. It was more carefully calculated during the cold war era when accurate SST measurements were used to monitor the location of under sea vessels (submarines). Modern SST measurements are still based on ship reports but are now supplemented with satellite measurements.

Accumulated Departure from Average PDO

The accumulated departure from average plot of the PDO index has a peak in about 1944 as well as during the current period of heating starting in 1975. The rising limb of this PDO index represents a warming period of SST on the West Coast coast, where as a declining trend represents a time of cooling. In general, cooling on the US Pacific Coast occurs when coastal North winds induce offshore coriolis force that result in off shore winds that lower tides that induce cold-water upwelling. This reflects the delicate hydrostatic balance of the oceans. Warming SST is associated with South winds that induce an onshore coriolis force with higher tides, which suppress cold-water upwelling.

The average annual air temperature at California stations for 1948 to 1997 was plotted against the average annual SST at 35°N, 125°W.  It was found to correlate with well (r^2 over 0.7) with SST at coastal stations. This indicates that up to 70 percent of annual average air temperature variation is similar to the average annual variation in SST at 35°N, 125°W. This ocean station was selected because of a great coherence the SST if ocean stations in the North-South California Current. The air temperature index was de-trended (by 2°F per century). This was to remove the urban affect so plots of the air temperature index and PDO index could be compared. Both plots show peaks in about 1944 and low values at the time of the major climate reversal of 1975.

California Air Temperature -vs- Sea Temperature

A rainfall index for California was compiled by averaging 100 rainfall records that were complete for the period from 1900 to 2006. This period was chosen to correspond with the period of the PDO index.  The early part of the rain index doesn’t correspond closely with the PDO index, but the PDO index peak of 1944 is clearly noticeable in the rainfall index. A low point in the PDO index seems to have preceded the major drought of 1976 and 1977. Most of these records were from the National Weather Service. Some were from other water Agencies such as the City of San Francisco or the Marin Municipal Water District.  None of these records were checked for homogeneity

The Atmospheric Angular Momentum index (AAM) is a factor used in numeric weather forecasting. The AAM index is a measure of the ratio of East West vs. North South winds on the planet. The AAM index accounts for about 98 percent of the variation in Earth’s length of day variation (LOD) (measured in milliseconds per day). LOD variation is measured daily by the Naval Observatory and has been monitored since 1640 at England’s Greenwich Observatory. The annual average of the daily length of day is shorter after 1975. This reflects a greater North-South component in the worldwide wind patterns as well as a major change in California’s rain and temperature trends.

Atmospheric Angular Momentum and Length of Day

The sun was once considered to be constant in its heat output, hence the term “Solar Constant”. Recent observations suggest that the sun is a variable star as are sun-sized stars. Observations of solar irradiance have been made with great precision from orbiting satellites since 1978. These observations indicate that the solar irradiance varies with the historic sunspot numbers. Using this relationship, 500 years of solar irradiance is easily inferred.

Variations of Length of Day

Sunspot numbers since 1900 were plotted as accumulated departure from average in order to compare them with weather variables. The rainfall, temperature, PDO and LOD indices were compared with the sunspot numbers.  The sunspot number index indicates a declining trend for the 1900 to 1940 period and an increase from 1940 to 2005. The eleven-year cycle is clearly visible. An increase in irradiance since 1940 is plainly indicated.

Solar Variance

Accumulated Departure from Average Susnpot Numbers

The accumulated departure from average California rainfall index indicates a declining rainfall from 1915 to 1934 and a general increase in rainfall corresponding to the major climate shift of 1975. This is suggestive of a 35-year lag time in solar influence, if that relationship is real.

A comparison of the accumulated departure from average of the California temperature and the Pacific Decadal Oscillation Index (PDO) indices indicate both peaking about 1943 and generally declining until the major climate shift of 1975. Again, this suggests a 35-year lag time in solar influence.

The variation in the length of day (LOD) index shows little similarity with the accumulated departure from average of the sunspot numbers before 1930, but the climate shift of 1975 is clearly noted, 35 years after the 1940 sunspot minimum.

Evidence for climate variation is inferred from the sunspot numbers. The “Solar Constant” sunspot relationship clearly suggests a long-range historic view of solar irradiance from 1500. The solar irradiance has been clearly increasing since 1940. The Maunder Minimum of sunspot numbers from 1660 to 1710 was clearly a time of worldwide cold temperatures. The year 1816 was known as the year without a summer. These historic cold periods are clearly visible on the accumulated departure from average graph of inferred solar irradiance derived from sunspot numbers. There have been published reports of solar influence on the surface changes of Mars resulting from increasing solar irradiance. See William K. Hartman, Traveler’s Guide to Mars, Workman Publishers, 2003 page 406.

Accumulated Departure from Average Solar Irradiance

Another reference to sunspot activity is:

“Unusual activity of the Sun during recent decades compared to the previous 11,000 years”

Nature, Vol. 431, No. 7012, pp. 1084 – 1087, 28 October 2004.

By S.K. Solanki, I. G. Usoskin, B. Kromer, M. Schüssler, and J. Beer

The abstract of the Solanki et al paper reports the current occurrence of highest sunspot numbers of the last 11,400 years. They also point out that the “solar variability is unlikely to have been the dominant cause of the strong warming during the past three decades”. The conclusion of “unlikely cause” is not in agreement with the recently established irradiance-sunspot number relationship. If a relationship of the last three sunspot cycles applies equally to the last three sunspot cycles then it must be assumed that it applies to the last 11,400 years. It would appear that whereas the direct solar irradiance may not be significant in itself the modifications to ocean and atmospheric currents might reflect some solar influence.

The preservation and protection of temperature measuring sites needs to be considered. The future of temperature measurement in urban regions is fast being modified by urban growth. Waste heat and environmental changes are underway at the present time. A question is; at what distance from a waste heat source is the temperature measurements being modified and how much?  A monitoring program for temperatures in areas free of land use change and waste heat is much needed.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
October 8, 2007 6:13 am

The only problem with comparing anything to average temp, how reliable is the CRU?

Cr Bruce McCallum.
October 8, 2007 6:20 am

I have been studying Climate Change for over 50 years,and my conclusions correspond so close to yours it is hair raising,these climate changes have been swinging from Global Warming to Global Cooling at intervals very close to your calculations,and it has been occurring for Millions of years,and I am not a Scientist or any kind of whizkid,I am an ordinary Motor Mechanic,using only the Common Sense given to me by the Good Lord,I thank you for this very informative paper,Best Regards,Bruce McCallum.

October 8, 2007 7:09 am

In the Solar Cycle plot, the red “annual” line wouldappear to have more points than there are years. Is this curve more along the lines of monthly/weekly values of the last 365.25 days of data?

Stan Needham
October 8, 2007 8:44 pm

My, my; the interesting things you learn on blogs never ceases to amaze me. I never knew there was such a thing as deviation in the length of a day.

Stan Needham
October 8, 2007 8:50 pm

And, slightly OT, but how cool is this?

Robert Hawkins
October 9, 2007 5:16 am

Does all this mean we wont all be roasting 12 months out of the year and lining up to buy Soylent Green in 2022 ??

October 9, 2007 7:19 pm

When I was a kid (40 years ago) generally speaking, weather in Central California was warmer and less active. In particular, the old stereotype about spring and fall having the truly warm weather (I’m near coastal in location) and the summer having the coastal stratus and onshore with a few good interior outbreaks, and then, a somewhat mild winter with probably about 10 or 11 storm systems. Very average rainfall. Dry from April until November.
It would seem that slowly but surely, after a minor hiatus in the late 70s and 80s, climatic winter has lengthened, especially at the tail end. Especially since the mid 90s. Summers don’t seem to have quite as many interior outbreaks and more onshore push. Falls are far more variable, with about one in four being OK and the rest early and unseasonably wet. Winters and even Springs seem to have more serious cold cP outbreaks. For example, after the low elevation snow event of 1976, there was absolutely nada through the rest of the 70s, through the 80s and into the mid 90s. We did have that really bad (dry) freeze late Dec ’90 – early Jan ’91. But that was about it. Then suddenly, after the ’97-’98 El Nino, everything changed and it seems that the capability for there to be more frequent serious low elevation snow events, and other cold episodes, was unleashed. We had the sea level snow event on Dec 20, 1998, and then a bad freeze into the New Year. Low snow in Feb 2000. I think there was one in winter 2003 – 04 but may be mistaken on that one. Very early mountain snow in the high country Oct 2004 and a couple good low elevation snow events (to sea level) Feb – Mar 2005. Repeat of that Feb – Mar 2006. Then, although a dry winter overall, 2006 – 2007 obviously was disasterous in terms of cold, especially the record breaking Jan 2007 Great Freeze.
Bottom line, weather has gone downhill over the past 10 years. It is colder, more biased toward abnormally high rainfall, and generally more “northern.” I’ll stick my neck out and call it a climate shift, and not the type that Al Gore wants to talk about.

October 9, 2007 10:38 pm

I noted with interest the plot of the atmospheric angular momentum index, and the description of it defining the relationship between east/west winds and north/south winds. The index shows a dramatic drop since the 1970s, with the implication that north/south winds have increased significantly during that time.
If I have interpreted this correctly, this has some interesting ramifications. Increased north/south winds imply more mixing of tropical, temperate, and polar regions, and this got me thinking again about something I have wondered about for a while — to wit:
Since the radiative power is proportional to the fourth power of absolute temperature, for the same mean temperature, the less the variation of temperature (e.g. between the tropics and the polar regions), the less total power is radiated away. Put another way, for the same power radiated away, the mean temperature must be higher if the variation is lower.
A simple example illustrates the concept and gives an idea of the order of magnitude of the effect. Split the earth into two zones of equal area: tropical and non-tropical. The tropical area has a temperature of 300K (~27C), and the non-tropical area has a temperature of 280K (~7C). If the mixing were improved (as by increased north/south winds) so that the tropical temperature were reduced to 299K and the non-tropical temperature were increased to 281K, the net radiated power would be reduced by a fraction of a percent.
This would cause the total temperature to increase until the net radiative power again balanced the incoming solar power. In this very simple example, the mean temperature would have to increase a full 0.1K to radiate this same amount of power with a spread of 18K instead of 20K. (I was surprised when I crunched the numbers how big this value was.)
Given that by all accounts we are seeing this reduced differential between tropical and polar temperatures of about this magnitude, I wonder how much of the increase in “global mean temperature” could be ascribed to this effect.

November 4, 2007 9:27 am

[…] Guest Weblog: CO2 variation by Jim Goodridge, former California State Climatologist 4 11 2007 Moderators note: This is reposted from an essay that Jim sent to me. Jim was the State Climatologist for California for many years, and still does consilting work for the Californiua Dept. of Water Resources on rainfall studies. Jim has a command of simplicity in presentation, as you’ll read below. Jim’s presentation lends some insight into why CO2 increases lag temperature in historical and proxy climate records. Jim has also done another essay I’ve posted which relates The Pacific Decadal Oscillation, Sunspots, and Length of Day variances to rainfall patterns in California which you can read here. […]

January 25, 2008 9:45 am

[…] PDO and TSI to surface temperature connection has also been pointed out in previous post I made here, for former California State Climatologist, Jim Goodridge. PDO affects the USA more than the […]

February 1, 2008 10:12 am

In fact, what you’ve proposed is something everyone working in the field has noted and never published mainly because it is not possible to sustain that scientifically. There’s even people who already noted that the LOD is connected to PDO and then with the salmon variability… However what you have to know if even if there’s a relationship between LOD and AAM, this true relationship is replaced by the mantle core exchange of angular momentum at lower frecquencies, therefore at longer periods i.e., more than 20 years. Therefore the relationship between the LOD and the surface temperature is only a mere coincidence or you have then to imagine that the mantle core exchange of angular momentum is influencing climate. Instead others has proposed that the solar dynamo is influencing the earth dynamo, and that solar activity is influencing climate and therefore the fact that LOD and PDO are so close together means that both are a confirmation of the two influences…

December 2, 2008 1:21 am

Did it snow in Sacramento winter 1976?

Verified by MonsterInsights