I’ve moved to WordPress: http://bobtisdale.wordpress.com/

Monday, April 27, 2009

Misunderstandings about the PDO – REVISED

I’ve moved to WordPress.  This post can now be found at Misunderstandings about the PDO – REVISED
UPDATE (September 14, 2010): The discussion of Figure 5 has been corrected.


The first version of this post (The Common Misunderstanding About The PDO dated June 26, 2008) incorrectly described the method for calculating the Atlantic Multidecadal Oscillation. I originally intended to do a quick correction in agreement with my post The Atlantic Multidecadal Oscillation - Correcting My Mistake, but then I decided to expand this post.


Many climate change bloggers often note that global temperatures rise when the Pacific Decadal Oscillation (PDO) is positive and drop when the PDO is negative. They then make the assumption that it’s the PDO that causes global temperature to vary. To dispel this, let’s first examine what the PDO is.


The Pacific Decadal Oscillation (PDO), Figure 1, is “derived as the leading PC of monthly SST anomalies in the North Pacific Ocean, poleward of 20N. The monthly mean global average SST anomalies are removed to separate this pattern of variability from any ‘global warming’ signal that may be present in the data.” The quote is from the JISAO website: http://jisao.washington.edu/pdo/PDO.latest
The main JISAO PDO webpage is here:

Figure 1

The semi-periodic variation in the PDO can be better seen when the data is smoothed with a 121-month running-average filter, Figure 2.
Figure 2


Nathan Mantua of the University of Washington and JISAO, in an email, described the process used to calculate the PDO. And it is a process:

“The full method for computing the PDO index came from Zhang, Y., J.M. Wallace, D.S. Battisti, 1997: ENSO-like interdecadal variability: 1900-93. J. Climate, 10, 1004-1020.
“They labeled this same time series "the NP index" (see their figs 5 and 6). The steps are listed below, and files described below can be found at: ftp://ftp.atmos.washington.edu/mantua/pdofiles/
“Data used:
* monthly 5x5 Hadley Center SST 1900-93
1. create monthly anomaly fields for all grid points
2. create a monthly mean global SST anomaly time series for all months, 1900-93, using gridpoints specified in file grid.temp.glob_ocean.977
3. create a "residual SST anomaly" field for the North Pacific by subtracting out the global mean anomaly from each North Pacific grid point in file grid.N_Pac_SST.resi.172 (20N-65N, only in Pacific Basin) for all months and locations

np_resi(mo,loc)= np_ssta(mo,loc) - global_mean(mo)
4. compute the EOFs of the North Pacific residual SST anomaly fields, and ignore all missing data point (set them to zeros)
5. the PDO index is the leading PC from the above analysis
6. for PDO index values post 1993, project observed ‘North Pacific residual SST anomalies’ onto the leading eigenvector (what we call the ‘PDO pattern’ of ssts) from the EOF analysis done in step 4. We now do this with the Reynold's and Smith Optimally Interpolated SST (version 2) data.”
A link to the referenced Zhang et al (1997) paper is here:

The point of listing that multistep process was to show that the PDO is a statistically created dataset. Let’s look at what the PDO does not represent.


SST anomalies for the North Pacific Ocean (20N-65N) and scaled PDO data are illustrated in Figure 3. The PDO does not represent SST anomalies for the North Pacific.
Figure 3


The PDO is not calculated in the same fashion as the Atlantic Multidecadal Oscillation (AMO). NOAA ESRL calculates the AMO by detrending SST anomalies for the North Atlantic. Refer to The ESRL AMO webpage:

In Figure 4, the PDO (scaled) is compared to detrended North Pacific (North of 20N) SST anomalies (calculated the same as the AMO). While there are semi-periodic variations in detrended North Pacific SST anomalies, the PDO does not represent them.
Figure 4


Let’s subtract Global temperature anomalies (LST & SST) from North Pacific SST anomalies to see what that curve looks like. Refer to Figure 5. The PDO does not represent the difference between global temperature anomalies and North Pacific SST anomalies.
UPDATE (September 14, 2010): In a more recent post An Introduction To ENSO, AMO, and PDO -- Part 3, it was pointed out that the two curves in Figure 5 appear to be negatively correlated. I confirmed this and presented that inverse relationship in the post An Inverse Relationship Between The PDO And North Pacific SST Anomaly Residuals.
Figure 5


The PDO represents a pattern of SST anomalies in the North Pacific. The operative word in that sentence is PATTERN. Figure 6 (from the JISAO PDO webpage) illustrates the warm and cool phases of the PDO. When the PDO is positive, SSTs in the eastern North Pacific are warmer than in the central and western North Pacific, and when the PDO is negative, the reverse is true.
Figure 6

Keep in mind, though, that the PDO data itself represents only the North Pacific, north of 20N, which I’ve blocked off in Figure 7. Figure 7 is a map of SST anomalies from April 14–21, 2008 that shows a negative PDO pattern. It’s from the NASA Earth Observatory webpage here:
Specifically, this linked page:
Figure 7


There is also a popular belief that the sign of the PDO dictates whether El Nino or La Nina events dominate. There is, however, an analysis that contradicts that belief. Refer to:
And for those who enjoy PowerPoint presentations for the visuals:

In “ENSO-Forced Variability of the Pacific Decadal Oscillation”, Newman et al state in the conclusions, "The PDO is dependent upon ENSO on all timescales. To first order, the PDO can be considered the reddened response to both atmospheric noise and ENSO, resulting in more decadal variability than either. This null hypothesis needs to be considered when diagnosing and modeling ‘internal’ decadal variability in the North Pacific. For example, the observed spatial pattern of Pacific SST decadal variability, with relatively higher amplitude in the extratropics than in the Tropics, should be at least partly a consequence of a reddened ENSO response."

In the introduction, Newman et al explain, “Anomalous tropical convection induced by ENSO influences global atmospheric circulation and hence alters surface fluxes over the North Pacific, forcing SST anomalies that peak a few months after the ENSO maximum in tropical east Pacific SSTs (Trenberth and Hurrell 1994; Alexander et al. 2002). This ‘atmospheric bridge’ explains as much as half of the variance of January–March seasonal mean anomalies of SST in the central North Pacific (Alexander et al. 2002). Furthermore, North Pacific SSTs have a multiyear memory during the cold season. Deep oceanic mixed layer temperature anomalies from one winter become decoupled from the surface during summer and then ‘reemerge’ through entrainment into the mixed layer as it deepens the following winter (Alexander et al. 1999). Thus, over the course of years, at least during winter and spring, the North Pacific integrates the effects of ENSO." [Emphasis added]

They continue, “The prevailing null hypothesis of mid latitude SST variability posits that the ocean integrates forcing by unpredictable and unrelated weather, approximated as white noise, resulting in ‘reddened’ noise with increased power at low frequencies and decreased power at high frequencies (e.g., Frankignoul and Hasselmann 1977). In this paper, we propose an expanded null hypothesis for the PDO: variability in North Pacific SST on seasonal to decadal timescales results not only from red noise but also from reddening of the ENSO signal.”

Figures 8 and 9 are comparative graphs of the PDO and NINO3.4 SST anomalies, smoothed with 12-month and 121-month filters.
Figure 8
Figure 9


As discussed and illustrated, the PDO cannot directly explain global temperature variations because it represents a pattern of SST variability, not SST. And the Newman et al paper explains why the low frequency variations of the PDO are greater than ENSO. They write in their abstract, “Variability of the Pacific decadal oscillation (PDO), on both interannual and decadal timescales, is well modeled as the sum of direct forcing by El Nino–Southern Oscillation (ENSO), the ‘reemergence’ of North Pacific sea surface temperature anomalies in subsequent winters, and white noise atmospheric forcing.” [Emphasis added]

Do other areas of the Global oceans integrate the effects of ENSO like the North Pacific?


The links for the PDO data are included in the text of the post. HADISST NINO 3.4 SST anomaly data, HADISST North Pacific SST anomaly data, and the combined CRUTEM3+HadSST2 global temperature anomaly data are available through the KNMI Climate Explorer website:

Thursday, April 23, 2009

The Atlantic Multidecadal Oscillation - Correcting My Mistake

I’ve moved to WordPress.  This post can now be found at The Atlantic Multidecadal Oscillation – Correcting My Mistake
Figure 1 illustrates the NOAA Earth System Research Laboratory (ESRL) Atlantic Multidecadal Oscillation (AMO) dataset from 1856 to present, smoothed with a 37-month running-average filter. The ESRL AMO webpage is here:
The unsmoothed AMO data is here:
Figure 1

I’ve recently been providing monthly AMO updates (Mid-April 2009 NINO3.4 SST Anomaly and AMO Update and March 2009 SST Anomaly Update), after noting A Recent Drop in the AMO in yet another post. Those posts presented a short-term (November 1981 to present) view of the data, so the long-term oscillation was not visible.

I did, however, post a long-term “AMO” graph recently in my post Individual Ocean SST Anomalies In Perspective. Refer to Figure 2. Note the major differences between that curve and the ESRL AMO curve in Figure 1. The data in Figure 2 is missing the rise in the late 1800s, and the timing of the decrease then increase in the mid to late 20th century is incorrect.
Figure 2


My error: I had calculated the AMO as the difference between the North Atlantic SST anomalies and Global SST anomalies (North Atlantic SST anomalies MINUS Global SST anomalies).


The ESRL describes the method for creating their AMO data as follows:

“Use the Kaplan SST dataset (5x5).
“Compute the area weighted average over the N Atlantic, basically 0 to 70N.
“Detrend that time series”

Figure 3 illustrates raw North Atlantic SST anomaly data and the corresponding linear trend. As noted above, ESRL uses Kaplan SST anomaly data for the AMO, so I’ve used it in Figure 3. To detrend the North Atlantic SST anomaly data and create the AMO, the monthly values of the trend are subtracted from the monthly SST anomaly data. (I have no idea how I mistakenly got it stuck in my head that the AMO was a residual.)
Figure 3

In Figure 4, I’ve detrended the North Atlantic SST anomaly data, using the simple method described above. I’ve also smoothed it with a 37-month filter. The curve now reproduces the cycles of the AMO.
Figure 4

To confirm that my method was correct, I created a comparative graph of the ESRL AMO data and the detrended Kaplan North Atlantic SST anomalies, Figure 5. Note that the Kaplan SST data through the KNMI Climate Explorer has not been updated since March 2003, but that’s fine for this examination.

There are some minor differences that likely result from differences in the coordinates, climatologies, and base years. KNMI and ESRL may also infill missing data differently. Considering all those possible differences, the Detrended Kaplan North Atlantic SST Anomalies do correlate well with the ESRL AMO data, and the curve is a much better presentation of the AMO than my erroneous residual version.
Figure 5


For information purposes, in Figure 6, I’ve provided a comparison of the ESRL AMO data and Detrended ERSST.v3b North Atlantic SST Anomalies. The coordinates for the ERSST.v3b data are 0 to 70N, 78W to 10E. The ERSST.v3b data is also up-to-date, which allows for a comparison through March 2009.
Figure 6

The short-term comparison, Figure 7, confirms that the AMO has indeed dropped significantly since June 2008. You have to go back to late 1996, before the 1997/98 El Nino, to find AMO values as low. Keep in mind, though, that the AMO is a noisy dataset and the AMO has a long way to go before it reaches its minimum. Referring back to Figure 6, it took approximately 30 years for the AMO to rise from minimum to maximum (assuming that 2004 was the maximum), and it should take about 25 more years for the AMO to reach minimum again.
Figure 7


There will be those who will wonder why a map of the North Atlantic SST anomalies will show predominantly positive anomalies, while the AMO is presently in negative values. Figure 8 illustrates raw ERSST.v3b North Atlantic SST anomaly data and the corresponding linear trend. Present SST anomalies are positive, but since they are below the trend line, the AMO value is negative.
Figure 8


Over the next few weeks, I will be correcting the graphs and text in earlier posts. I will indicate that the corrections have been made with a note at the beginning of the post. For some posts, I will rewrite and update them in their entirety.

Sorry for the misinformation.


The Kaplan and ERSST.v3b SST anomaly data is available through the KNMI Climate Explorer webpage:

Monday, April 20, 2009

Mid-April 2009 NINO3.4 SST Anomaly and AMO Update

I’ve moved to WordPress.  This post can now be found at Mid-April 2009 NINO3.4 SST Anomaly and AMO Update
CORRECTION: The AMO discussion in this post has been corrected to reflect the discussion in The Atlantic Multidecadal Oscillation - Correcting My Mistake.


NINO3.4 SST anomalies have risen over the past few weeks. They’re no longer near the threshold of a La Nina. The NINO3.4 SST anomaly for the week centered on Wednesday April 15, 2009 was -0.132 deg C.
Figure 1 - NINO3.4 SST Anomalies

Figure 2 is the map of weekly SST anomalies, centered on Wednesday April 15th. The Global SST anomalies for the week are approximately 0.17 deg C.

I’ve highlighted the location of the NINO3.4 region. I noted it’s the “approximate” location because I had to eyeball the 5S, 5N, and 170W coordinates. The 120W longitude was provided on the map.
Figure 2 - Global SST Anomaly Map

The Climate Prediction Center’s Pacific Subsurface Temperature Anomaly and Temperature animations are being updated again. They hadn’t been for a few weeks.

Figure 3 - Pacific Subsurface Temperature Anomaly


As noted in The Atlantic Multidecadal Oscillation - Correcting My Mistake, the AMO is detrended North Atlantic SST anomalies. The updated weekly OI.v2 data begins on January 3, 1990, so it does not seem appropriate to detrend it with only 19 years of data. Hence, this update illustrates the weekly North Atlantic SST anomalies, not the AMO.

North Atlantic SST anomalies (and, therefore, the AMO) appear to have ended their precipitous decline. Refer to Figure 4. But that recent drop was sizable. It will be interesting to see what unfolds in the months to come.
Figure 4 – North Atlantic SST Anomalies


The OI.v2 SST anomaly data is available through NOAA’s NOMADS System:

Friday, April 17, 2009

A Closer Look At The ERSST.v3b Southern Ocean Data

I’ve moved to WordPress.  This post can now be found at A Closer Look At The ERSST.v3b Southern Ocean Data
The time-series graph of the ERSST.v3b SST anomaly data for the Southern Ocean, Figure 1, is unique. It clearly shows that Southern Ocean SST anomalies were higher in the late 1800s than they were at their late 20th century peak. Note also how Southern Ocean SST anomalies have been dropping since the early 1990s.
Figure 1

But where does the unique shape come from?


Figure 2 is a comparative graph of the Southern Ocean SST anomalies, where the data has been divided by the approximate southernmost longitudes of the three major oceans: Atlantic (90S-60S, 70W-20E), Pacific (90S-60S, 145E-70W), and Indian (90S-60S, 20E-145E). It’s clear that the major variations originate south of the South Pacific.
Figure 2

Figure 3 illustrates the Southern Ocean SST anomalies south of the South Pacific without the distraction of the other datasets.
Figure 3

Dividing the data of the Southern Ocean south of the Pacific, Figure 4, illustrates that the majority of the variability lies in the East Central (90S-60S, 145W-110W) and East (90S-60S, 110W-70W) segments.
Figure 4

If those two sections of the Southern Ocean south of the Southeast Pacific are combined (90S-60S, 145W-70W) and compared to remainder of the data for the Southern Ocean (90S-60S, 70W-145W), Figure 5, two things stand out. First, the remainder of the Southern Ocean made a slow dip (from the late 1870s to the early 1930s) and rebound (from the early 1930s to the late 1970s). And since the 1970s, the SST anomalies for that major portion of the Southern Ocean have been dropping. Is the dip and rebound and recent decline part of a ~100-year oscillation? Second, something adds to the apparent natural oscillation in the Southern Ocean south of the Southeast Pacific, which represents about 21% (75 deg longitude/360 deg longitude) of the Southern Ocean, to create the additional variability.
Figure 5

And the logical contributor to the variability of the Southern Ocean south of the Southeast Pacific would be ENSO. Figure 6 compares scaled NINO3.4 SST anomalies to those of the Southern Ocean south of the Southeast Pacific. The timing of the perturbations agree for the most part. At other times, that portion of the Southern Ocean appears to respond to some other forcing.
Figure 6

Figure 7 illustrates the SST anomalies of the Southern Ocean south of the South Atlantic and Indian Oceans. Note how the two datasets appear to modulate out of sync at times. Is this evidence of Antarctic Circumpolar Waves? Refer to:
Figure 7

Refer back to Figure 2. After 1910, note how two of the three datasets appear to vary in unison, while the third opposes them. There are occasions when all three vary in sync, but they occur less often.

As an additional reference, Figures 8 and 9 are the individual SST anomaly graphs for the Southern Ocean south of the South Atlantic and Indian Oceans.
Figure 8
Figure 9


We often hear that global warming is causing the Wilkins Ice Shelf to break free of Antarctica. Refer to Figure 10. Many times the article will note that the local Southern Ocean SST anomalies have risen for the past 50 years. What they fail to mention are:
-SST anomalies for that location show a negative trend over the past 150+ years,
-SST anomalies in that area were higher in the late 1800s than they were in the late 1900s, and
-SST anomalies have dropped significantly since the late 1990s.
Figure 10


ERSST.v3b SST anomaly data is available through the KNMI Climate Explorer website:http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere

Friday, April 10, 2009

Revisiting Bratcher and Giese (2002)

I’ve moved to WordPress.  This post can now be found at Revisiting Bratcher and Giese (2002)

In a comment in the March 2009 SST Anomaly Update thread, Blogger DB reminded me of the Bratcher and Giese (2002) paper “Tropical Pacific Decadal Variability and Global Warming” [GEOPHYSICAL RESEARCH LETTERS, VOL. 29, NO. 19, 1918, doi:10.1029/2002GL015191, 2002].

“An analysis of ocean surface temperature records show that low frequency changes of tropical Pacific temperature lead global surface air temperature changes by about 4 years. Anomalies of tropical Pacific surface temperature are in turn preceded by subsurface temperature anomalies in the southern tropical Pacific by approximately 7 years. The results suggest that much of the decade to decade variations in global air temperature may be attributed to tropical Pacific decadal variability. The results also suggest that subsurface temperature anomalies in the southern tropical Pacific can be used as a predictor for decadal variations of global surface air temperature. Since the southern tropical Pacific temperature shows a distinct cooling over the last 8 years, the possibility exists that the warming trend in global surface air temperature observed since the late 1970’s may soon weaken.”

Link to GRL Abstract:

Also refer to the copy of the Bratcher and Giese slide presentation:http://www.decvar.org/documents/CCR_workshop/bratcher.htm?PHPSESSID=df81bed52419c895efe9135099fb26e9

And theWorldClimateReport post on the study here:http://www.worldclimatereport.com/archive/previous_issues/vol8/v8n04/feature1.htm

And CO2Science did a write up here:http://www.co2science.org/articles/V6/N20/C1.php

As illustrated in Figure 1 (Figure 1 of Bratcher and Giese 2002), their comparison of Global Temperature Anomaly (GISTEMP) and NINO3 SST Anomaly (Simple Ocean Data Assimilation-SODA) ran from 1948 to 2000. The graph appears to be of annual (not monthly) data, with 5-year smoothing as discussed in the paper. The two questions that struck me were: How far back in time do the two datasets coincide and what would the updated graph look like? (The latter question was also part of DB’s comments.)

Figure 1

The problem: the easily available SODA data through the KNMI Climate Explorer website only includes the years of 1958 to 2004. So I’ve substituted ERSST.v3b data in place of the SODA data. I’ve also used monthly instead of annual data.


Figure 2 illustrates a reasonable facsimile of Cell A of the Bratcher and Giese comparison of Global Surface Temperature and NINO3 SST anomalies. It has been extended in time through February 2009. The global surface temperature data is GISTEMP and the NINO3 SST anomaly data is ERSST.v3b. Bratcher and Giese appear to use different base years than the ones used here (1971-2000), and, of course, the variability appears greater with the monthly data.
Figure 2

Figure 3 updates Cell B of the Bratcher and Giese Figure 1. Without accounting for volcanic aerosols, any attempt to determine the 4-year lag between NINO3 and Global Surface Temperatures as claimed by Bratcher and Giese would be difficult with these time-series graphs, including Figure 1. However, Global Temperatures do appear to respond gradually over time to the shift in NINO3 SST anomaly. Note the impact of the Pacific Climate Shift of 1976 on the smoothed NINO3 SST anomalies. It stands out in Figures 1, 2, and 3.
Figure 3

Based on the averages of the smoothed NINO3 SST anomalies for the periods of 1950 through 1975 and of 1978 to present, Figure 4, the magnitude of the 1976 shift in NINO3 SST anomalies is more than 0.4 Deg C. If one considers NINO3 SST anomalies as a forcing, then the rise of global temperatures from 1976 to the early 2000s would then appear to be a natural response to a natural variation.
Figure 4


Figure 5 illustrates the NINO3 SST and global surface temperature anomalies from 1880 to present. Both datasets are smoothed with 61-month (5-year), running-average filters. Two things to consider when looking at the data before 1948: the discontinuity in the SST anomalies at 1945 would have impacted both datasets, and prior to 1914 and the opening of the Panama Canal, there were very few SST samples in the NINO regions. With those in mind, the fact the two datasets do seem to “track” is quite remarkable.
Figure 5

The NINO3 data was still noisy with the 61-month smoothing, so I changed to a 121-month filter in Figure 6. This seems to aid in illustrating the influence of NINO3 SST anomalies on global surface temperature. ENSO appears to dictate whether global surface temperatures rise or fall over decadal periods. It also illustrates a gradual “ramp up” required to overcome global thermal inertia.
Figure 6

There’s a mismatch with those two datasets. The NINO3 SST anomalies are based on ERSST.v3b data, while GISS uses HADSST data prior to November 1981 for their global surface temperature product. So let’s look at matching data.


In Figure 7, the GISS Global Surface Temperature data has been replaced by ERSST.v3b Global SST Anomaly data. The most significant difference between the GISTEMP Global Surface Temperature and the ERSST.v3b SST data can be found between ~1880 and 1900. Note how the drop in the temperature from 1880 to 1900 is exaggerated in the ERSST.v3b Global SST anomaly data.

Note also how the effect of NINO3 SST anomalies on Global SST anomalies is still clear after 1914. This is especially true following the shift in NINO3 SST anomalies in 1976.
Figure 7


As noted earlier, Bratcher and Giese wrote in the Abstract, “Anomalies of tropical Pacific surface temperature are in turn preceded by subsurface temperature anomalies in the southern tropical Pacific by approximately 7 years.” They illustrated this lag in their Figure 3, my Figure 8.
Figure 8

In the body of the paper, Bratcher and Giese discuss and illustrate the lag between the subsurface Tropical Pacific temperature and NINO3 SST anomalies. I am not trying to undermine that in any way. But the second problem I encountered while trying to update the Bratcher and Giese 2002 paper was the availability of Subsurface Temperature data for the Tropical Pacific Ocean. Simple Ocean Data Assimilation (SODA) data through KNMI does not include the subsurface temperature data.

It was my original intent to end the comparisons here, which is why the graphs are titled “Revisiting Part of Bratcher & Giese 2002”, but then it struck me that this would be a good time to illustrate a possible influence of the Southern Ocean on ENSO.


Figure 9 shows the locations of NINO3 region and the portion of the Antarctic Circumpolar Current (ACC) in the extreme Southeast Pacific used in the following comparison. Note how the Humboldt Current carries waters from the ACC along the coasts of Chile and Peru and up to the eastern equatorial Pacific. The SST anomalies of the Southeast Pacific ACC should have an influence on NINO3 SST anomalies.
Figure 9

Figure 10 is a comparison of NINO3 and Southeast Pacific ACC SST anomalies. Again, both datasets have been smoothed with 121-month filters. From 1940 to present, there is a reasonable agreement between the two datasets, indicating that the underlying SST for the equatorial Pacific is impacted by the ACC and Southern Ocean SST anomalies. There does not appear to be the 7-year lag suggested by Bratcher and Giese, though.

I found the correlation between the Southeast Pacific ACC and NINO3 SST anomalies interesting, but not conclusive. And I have no explanation for the divergence between the two datasets from ~1915 to 1945. Did the North Pacific have a greater influence during those times? I can’t say. I’ll have to investigate that and the SST anomalies along the Humboldt Current in a future post to try to determine the reason for the disagreement during that period.
Figure 10


In my series of posts “Can El Nino Events Explain All of the Warming Since 1976?” I illustrated the processes that cause step changes in the East Indian and West Pacific SST anomalies, which in turn result in increased global SST anomalies. Refer to:
-Can El Nino Events Explain All of the Global Warming Since 1976? – Part 1
-Can El Nino Events Explain All of the Global Warming Since 1976? – Part 2
-Supplement To “Can El Nino Events Explain All Of The Warming Since 1976?”
-Supplement 2 To “Can El Nino Events Explain All Of The Warming Since 1976?”

I illustrated the similar impacts of significant ENSO events on the North Atlantic Ocean in There Are Also El Nino-Induced Step Changes In The North Atlantic.

I’ve shown how Global Surface Temperature time-series data can be replicated using natural variables in Reproducing Global Temperature Anomalies With Natural Forcings. In that post, a running total of NINO3.4 SST anomalies establishes the underlying curve.

Based on the findings of Bratcher and Giese 2002, this post presents yet another way to illustrate that ENSO dictates long-term Global Surface Temperature Anomalies.


-IF the 5-year smoothing used by Bratcher and Giese (or the 61- and 121-month smoothing that I used) reflects the underlying NINO SST anomalies, and

-IF global temperatures do respond as implied by the correlation of the NINO3 SST anomalies and Global Surface Temperature anomalies as shown in the preceding and as discussed in Bratcher and Giese 2002, and

-IF NINO3 SST anomalies continue to follow the Southern Ocean and ACC SST anomalies,

-THEN Global Surface Temperatures should continue to decrease in response.

Will these natural variations overwhelm any anthropogenic sources of warming and drive global temperatures down, as opposed to only flattening the curve as it has recently and as it had from the 1940s to the late 1970s? Only time will tell.


The GISS Global Surface Temperature anomaly data and the ERSST.v3b SST anomaly data are available through the KNMI Climate Explorer website:

Wednesday, April 8, 2009

Individual Ocean SST Anomalies In Perspective

I’ve moved to WordPress.  This post can now be found at Individual Ocean SST Anomalies In Perspective
CORRECTION: Figure 2 and its discussion has been corrected to reflect the discussion in The Atlantic Multidecadal Oscillation - Correcting My Mistake. The North Atlantic Residual is not the AMO.


I first posted the ERSST.v3b version of global, hemispheric and individual ocean (north and south) SST anomalies in KNMI Added ERSST.v3b Data To Climate Explorer. This is follow-up post. In it, I compare the individual ocean SST anomalies to global SST anomalies, which should help put the variability of the individual ocean datasets into perspective. I also present the residuals (individual ocean SST anomalies minus global SST anomalies) and compare the residuals where appropriate.


The SST anomaly graphs are based on ERSST.v3b data, from January 1854 to February 2009. The ERSST.v3b data is available through the KNMI Climate Explorer website. The data have been smoothed with a 37-month running-average filter. The linear trends are calculated by EXCEL for the entire time span of the data. The coordinates used for the individual oceans are listed on the graphs.


Figure 1 is a comparative graph of North Atlantic and Global SST anomalies. The deviations of the North Atlantic SST anomalies from the global dataset are the results of Atlantic Meridional Overturning Circulation and the impact of ENSO events on North Atlantic SST anomalies. Refer to my post There Are Also El Nino-Induced Step Changes In The North Atlantic for further information. Note the exaggerated rise in the North Atlantic SST anomalies from the early 1990s to the early 2000s, but in looking back to the warming period in late 1920s-early 1930s, the recent rate of rise is not unusual. Also note that the linear trend for the North Atlantic SST anomalies (0.025 deg C/decade) is less than the global trend (0.031 deg C/decade).
Figure 1

Figure 2 shows the North Atlantic residual. (It is not the same as the Atlantic Multidecadal Oscillation (AMO), which is detrended North Atlantic SST anomaly.) The lack of a sizable multidecadal oscillation in the early years may result from sparseness of the data, or it is possible that the two variables, North Atlantic and Global SST anomalies, were in synch at the time. At least one longer-term reconstruction shows that North Atlantic SST anomalies do vary on a semi-periodic basis. Refer to Figures 6 and 7 in my post SST Reconstructions.
Figure 2


The South Atlantic and Global SST anomalies are compared in Figure 3. The drop in South Atlantic SST anomalies from the late 1880s to the about 1905 is severe and contributes to the higher linear trend of that dataset, as does the anomalous spike in the early 1970s. Note that the linear trend for the South Atlantic (0.052 deg C/decade) is more than double that of the North Atlantic (0.025 deg C/decade).
Figure 3

The late-19th/early-20th century dip and the unusual early 1970s rise are clearly visible in the South Atlantic residual data, Figure 4.
Figure 4


Considering the contribution of the Pacific Ocean to the global SST anomaly dataset, it makes sense that the North Pacific and Global SST anomalies would be similar. Refer to Figure 5. The two linear trends are fundamentally the same.
Figure 5

The North Pacific Residual is illustrated in Figure 6. It bears no likeness to the Pacific Decadal Oscillation (PDO). Refer to the post The Common Misunderstanding about the PDO for additional clarification on what the PDO is and what it is not.
Figure 6

Note: I’ll show an interesting correlation with the North Pacific Residual later in the post.


Like the North Pacific, the South Pacific SST anomalies, Figure 7, are very similar to the global dataset. The linear trend for the South Pacific (0.027 deg C/decade) is slightly less than the global linear trend (0.031 deg C/decade).
Figure 7

The South Pacific Residual is shown in Figure 8.
Figure 8


Where the SST anomaly linear trend for the South Pacific was slightly less than the global trend, the Indian Ocean linear trend, Figure 9, is slightly higher (0.036 deg C/decade). Note how the Indian Ocean SST anomalies are also very similar to the global SST anomalies, but that the Indian Ocean SST anomalies have been increasing faster that the global SST anomalies since 1970.
Figure 9

This is very visible as a shift in the Indian Ocean Residual data in 1970, Figure 10.
Figure 10

Note that the shift in the Indian Ocean Residual corresponds with the opposing shift in the North Pacific Residuals. See Figure 11. The two datasets oppose one another as far back as the 1870s. What causes the relationship? We’ll have to examine that further in a future post.
Figure 11

And how much did the rise (and fall) in the Southern Ocean SST anomalies that began in the mid-1960s influence the shift in the Indian Ocean Residuals? Refer to Figure 12, which is a comparative graph of Southern Ocean SST anomalies and Indian Ocean Residuals.
Figure 12


Figure 13 compares Arctic Ocean and Global SST anomalies. As you’ll note, the linear trend in the Arctic Ocean SST anomalies (0.014 deg C/decade) is significantly less than the Global linear trend (0.031 deg C/decade). It’s less than half, actually. This should be due in part to the low density of readings in early years. Part of it should also result from there being greater ice cover in earlier years. Visually note the rate of Arctic SST anomaly rise from the late 1960s to approximately 1998. In that period, about 30 years, Arctic Ocean SST anomalies rose ~0.17 deg C. Then from ~1998 to ~2006, after the 1997/98 El Nino, Arctic Ocean SST anomalies accelerated and rose 0.24 deg C in less than 10 years.
Figure 13

The Arctic Ocean Residuals, Figure 14, are misleading, because the Global SST anomalies are the dataset with the greater variability.
Figure 14


The Southern Ocean and Global SST anomalies are compared in Figure 15. Note that the Southern Ocean linear trend is negative. The two late 19th century peaks in Southern Ocean SST anomalies were higher than they were in the late 20th century. Note also that the Southern Ocean SST anomalies have been declining since the 1980s, or the early 1990s, depending on your perspective, while the global SST anomalies rose until recently. This is not unusual. Southern Ocean and Global SST anomalies were also out of synch from ~1910 to the early 1930s.
Figure 15

The Southern Ocean Residual, Figure 16, appears similar to an inverted Global SST anomaly curve. This makes sense, since the Southern Ocean SST anomalies are comparatively flat.
Figure 16


The ERSST.v3b Extended Reconstructed Sea Surface Temperature anomaly data is available through the KNMI Climate Explorer website:



Tips are now being accepted.

Comment Policy, SST Posts, and Notes

Comments that are political in nature or that have nothing to do with the post will be deleted.
The Smith and Reynolds SST Posts DOES NOT LIST ALL SST POSTS. I stopped using ERSST.v2 data for SST when NOAA deleted it from NOMADS early in 2009.

Please use the search feature in the upper left-hand corner of the page for posts on specific subjects.
NOTE: I’ve discovered that some of the links to older posts provide blank pages. While it’s possible to access that post by scrolling through the history, that’s time consuming. There’s a quick fix for the problem, so if you run into an absent post, please advise me. Thanks.
If you use the graphs, please cite or link to the address of the blog post or this website.