Part 2 – Comments on the UKMO Report “The Recent Pause in Global Warming”

This is part 2 of the 3-part series. In it, we’ll discuss the recently released UK Met Office paper The recent pause in global warming (2): What are the potential causes? See the UKMO webpage here.

The first post on the UKMO report dealt with their Part 1. In Part 2, they’ve restated a number of points they tried and failed to make in Part 1, so I’m not going to bother with them again in this post.

Let’s start with their first illustration in Part 2.


My Figure 1 is Figure 1 from Part 2 of the UKMO report on the recent pause in global warming. It’s curious for a number of reasons. First, the UKMO has presented yet another ocean heat content dataset. In Part 1 of their report, the UKMO included ocean heat content for the depths of 0-700 meters in two illustrations: in Cell W of Figure 1 and in their Figure 18. See the comparison of the two graphs from Part 1 here. Now, in Part 2 of their report, the UKMO has presented an ocean heat content dataset for 0-800 meters. It, too, shows a flattening after the year 2000, but it also includes a curious uptick in 2012. The UKMO cites Smith and Murphy (2007) as the source of the ocean heat content data in this latest graph, but that paper is titled “An objective ocean temperature and salinity analysis using covariances from a global climate model” [paywalled]. The UKMO does not cite Smith and Murphy (2007) on their EN-3 Ocean Heat Content data webpage. So it appears that the ocean heat content “Data” the UKMO presented in their Figure 1 from Part 2 may be a reanalysis and not data. Regardless, it appears the UKMO is picking and choosing ocean heat content datasets in each graph to satisfy a particular point they’re trying to make.

Figure 1

Figure 1

Second, sea level: The third plot in the UKMO’s Figure 1 shows sea level rising at an almost monotonous monotonic rate. The obvious reason they included sea level in Figure 1 was to show that sea level continues to rise even though global temperatures flattened since 1997 (the year the UKMO selected as the onset of the flattening). In fact, later in the paper the UKMO states:

As shown in Figure 1 and discussed in the first report in this series, sea level has continued to rise throughout the period of the recent pause.

What they fail to acknowledge is that sea level is a terrible proxy for global temperatures. In looking at the global temperature plots in the top panel of their Figure 1, global surface temperatures cooled from 1880 to about 1917 and then warmed from 1917 to 1944. As if repeating a cycle, global surface temperatures then cooled from about 1944 to 1976 and warmed once again from 1976 to 1998/2000. But the rise in sea level did not slow during the two cooling periods.

My Figure 2 presents the 30-year linear trends in surface temperature and sea level from 1880 to 2012. Each data point in the graph represents the 30-year trends in mm/decade for sea level or scaled Deg C/decade for global surface temperature, for the period ending in that year. The surface temperature data is the annual UKMO HADCRUT4 data available here. Note that it has been arbitrarily scaled in the graph by multiplying it by a factor of 100. And the reconstructed global mean sea level for 1880 to 2009 as described in Church and White (2011) is available from through the CSIRO webpage here. (CSIRO is Australia’s national science agency.) So we have the global surface temperature and sea level data presented by the UKMO. As shown in Figure 2, there is no agreement in the 30-year linear trends. The 30-year trends in global sea levels increased for the 30-year periods ending in 1940 through 1960, but the 30-trend in global surface temperatures decreased from 1945 to 1960.

Figure 2

Figure 2

In short, based on the 30-year trends, there is no agreement between the rates as which global surface temperatures warm or cool and the rates at which global sea level rises. So there is no reason based on past cooling periods since 1880 to believe that the rise in sea levels would slow during the recent hiatus in global warming. This might lead one to believe the UKMO only included sea level in the graph as a distraction. Global sea level has risen since the end of the last ice age, and based on data presented by the UKMO, the rate at which it has risen during the instrument record has no relationship with the rate at which global temperatures have warmed or cooled.

The third point with respect to the UKMO’s Figure 1 in Part 2 of their report on the pause in global warming: In the bottom panel, they present the 9-year trends in the 3 global surface temperature datasets and in the ocean heat content (reanalysis?). Where did 9 years come from? The UKMO is very specific in the discussion of when the hiatus period started. They wrote:

The start of the current pause is difficult to determine precisely. Although 1998 is often quoted as the start of the current pause, this was an exceptionally warm year because of the largest El Niño in the instrumental record. This was followed by a strong La Niña event and a fall in global surface temperature of around 0.2 deg C (Figure 1), equivalent in magnitude to the average decadal warming trend in recent decades. It is only really since 2000 that the rise in global surface temperatures has paused.

The annual data from 2000 to 2012 cover a period of 13 years, not 9 years. And 1998 to 2012 covers 15 years, not 9 years. It appears the UKMO chose 9 years so that they could claim that there is nothing unusual about a 9-year period without global warming. But 9 years is not the time period in question. They’re 13 years and 15 years. It’s blatantly obvious why the UKMO selected 9 years instead of 13 or 15 years. Refer to my Figure 3. If they had chosen 13 or 15 years, the recent warming rates would look unusual; that is, we haven’t seen global warming trends that low since the late 1970s.

Figure 3

Figure 3

It appears that most of the UKMO’s Figure 1 was intended to misdirect the public, to steer them away from the fact that the recent pause in global warming is unusual in a period when climate models can only simulate warming if they’re forced with manmade greenhouse gases.


The Atlantic Multidecadal Oscillation (AMO) is the subject of the next portion. For further information about the Atlantic Multidecadal Oscillation, refer the NOAA Atlantic Oceanographic and Meteorological Laboratory (AOML) Frequently Asked Questions webpage here, and to my blog post here and my introduction to the Atlantic Multidecadal Oscillation here.

In their discussion of their Figure 1, the UKMO refers to the multidecadal changes in global surface temperature. They state (my boldface):

It is clear that there have been other periods with little or no surface warming in the relatively recent past, a good example being the period between the 1940s and the 1970s (Figure 1). The trend in warming over that period is well understood, and linked to a substantial increase in the amount of aerosol in the atmosphere.

The UKMO resurrected the aerosols conjecture once again, and claim it’s “well understood”. Then they contradict themselves later in the report. Under the heading of “2.2 Redistribution of energy within the climate system – Storing the heat below the surface”, and as a preface to their discussion of the impact of the Atlantic Multidecadal Oscillation on global surface temperatures, the UKMO writes:

What drives these multi-decadal variations in the North Atlantic temperatures is still unclear and there is a debate about the role of anthropogenic aerosols in forcing the AMO (Booth et al 2012, Zhang et al 2013).

If the UKMO states there’s debate about to the impacts of anthropogenic aerosols on the AMO, and if the AMO is the most dominant mode of multidecadal variations in surface temperatures, then how can UKMO claim it’s “well understood” and that the 1940s-1970s pause in global warming is “linked to a substantial increase in the amount of aerosol in the atmosphere”?

We discussed Booth et al (2012) and Zhang et al (2013) in the March 2013 blog post Blog Memo to Lead Authors of NCADAC Climate Assessment Report. In it, I noted:

Booth et al (2012)—a climate model-based study—claimed in their abstract that a “…state-of-the-art Earth system climate model…” was used “…to show that aerosol emissions and periods of volcanic activity explain 76 per cent of the simulated multidecadal variance in detrended 1860–2005 North Atlantic sea surface temperatures.” Zhang et al (2013), on the other hand, utilized the limited data that is available to dispute those claims. Apparently a “state-of-the-art Earth system climate model” is still inadequate for attribution studies. Do we know what causes the multidecadal variations in North Atlantic sea surface temperatures based on computers models and on very limited data? The realistic answer is no.

Then there’s the recent paper that discusses the many faults with climate model simulations of the Atlantic Multidecadal Oscillation. It’s Ruiz-Barradas et al (2013) The Atlantic Multidecadal Oscillation in twentieth century climate simulations: uneven progress from CMIP3 to CMIP5. The full paper is here. The beginning of the “Concluding remarks” provides an explanation of why it’s important for climate models to be able to properly simulate the Atlantic Multidecadal Oscillation. They write (my boldface):

Decadal variability in the climate system from the AMO is one of the major sources of variability at this temporal scale that climate models must aim to properly incorporate because its surface climate impact on the neighboring continents. This issue has particular relevance for the current effort on decadal climate prediction experiments been analyzed for the IPCC in preparation for the fifth assessment report. The current analysis does not pretend to investigate into the mechanisms behind the generation of the AMO in model simulations, but to provide evidence of improvements, or lack of them, in the portrayal of spatiotemporal features of the AMO from the previous to the currents models participating in the IPCC. If climate models do not incorporate the mechanisms associated to the generation of the AMO (or any other source of decadal variability like the PDO) and in turn incorporate or enhance variability at other frequencies, then the models ability to simulate and predict at decadal time scales will be compromised and so the way they transmit this variability to the surface climate affecting human societies.

The only way Ruiz-Barradas et al could have been clearer would have been to state point blank that climate climate models will only have value if and when they can ever simulate the decadal and multidecadal characteristics of natural ocean processes. Ruiz-Barradas et al (2013) then describe the many problems with climate model simulations of the Atlantic Multidecadal Oscillation. And the paper ends with (my boldface):

The current analysis does not provide evidence on why the models perform in the way they do but suggests that that the spurious increase in high 10–20 year variability from CMIP3 to CMIP5 models may be behind the unsatisfying progress in depicting the spatiotemporal features of the AMO. This problem, coupled with the inability of the models to perturb the regional low-level circulation, the driver of moisture fluxes, seem to be at the center of the poor representation of the hydroclimate impact of the AMO.

If climate modelers cannot simulate the AMO properly, then it’s very obvious that the pause in global surface temperatures from the 1940s to the 1970s is not “well understood”…as the UKMO claims.

NOTE:  Figures 2 and 3 above presented variations in trends.  In the following, Figures 4 through 7 present variations in temperature anomalies.


Moving on the Pacific Decadal Oscillation: The UKMO starts out well by noting that there are multidecadal variations in the sea surface temperatures of the North Pacific. But then they go off on a tangent by discussing the Pacific Decadal Oscillation (PDO) and other abstract forms of the sea surface temperatures of the Pacific, as if those abstract forms represent the multidecadal variations in the sea surface temperatures of the North Pacific north of 20N or the Pacific Ocean:

The Pacific Ocean also displays longer timescale variations (Figure 5), variously referred to as the Pacific Decadal Oscillation (PDO; Mantua et al. 1997), Pacific Decadal Variability (PDV; Deser et al. 2012) and the Interdecadal Pacific Oscillation (IPO; Power et al. 1999).

Then the UKMO makes a rather surprising admission (my boldface):

The characteristics and mechanisms for the PDO are less well understood than for the AMO and there is some debate whether it just represents low frequency behaviour of ENSO and its influence on mid-latitude atmospheric circulations and surface fluxes (Deser et al. 2012), or whether ocean dynamics are involved too.

Let’s drop back for a moment. The current generation of climate models can’t simulate the Atlantic Multidecadal Oscillation (AMO), there’s also a debate as to whether or not manmade aerosols play a role in the AMO, and now the UKMO is stating that the multidecadal variations of the North Pacific (and the Pacific Ocean as a whole) are even less understood! And the UKMO wants us to believe, “The trend in warming over that period is well understood, and linked to a substantial increase in the amount of aerosol in the atmosphere.” That’s actually laughable. (And I’m not laughing with them; I’m laughing at them.) What gall!

Back to the PDO: The UKMO then goes on to discuss the PDO index and illustrate it with their Figure 4. But they’re ignoring (Willingly? Unknowingly?) the fact that the PDO index does not represent sea surface temperatures of the North Pacific—that the PDO index represents an abstract form of the sea surface temperatures of the North Pacific.

Brief description of the PDO index: El Niño and La Niña events create a specific spatial pattern in the sea surface temperatures of the North Pacific, north of 20N. For example, during a strong El Niño, sea surface temperature anomalies are warm in the east portion of the North Pacific and cool in the west and central portions. The PDO index basically represents how closely the spatial pattern in the North Pacific, and its intensity, match the spatial pattern created by El Niño and La Niña events. The PDO index does not relate to the surface temperature of the North Pacific (1) because that spatial pattern could exist at any temperature and (2) because the sea level pressure (and interrelated wind patterns) also impact the spatial pattern and, in turn, the PDO index.

And the UKMO is ignoring the fact that the sea surface temperatures of the North Pacific have multidecadal variations that can be comparable in strength to those in the North Atlantic. This is easy to see by detrending the sea surface temperature anomalies of the North Pacific (north of 20N) in the same fashion as the Atlantic Multidecadal Oscillation Index prepared by the NOAA ESRL. My Figure 4 compares detrended North Atlantic (0-70N, 80W-0) and detrended North Pacific (0-65N, 100E-90W) sea surface temperature anomalies. In this graph, we’re using the UKMO HADSST3 dataset. Both subsets have also been smoothed with 61-month filters to reduce the monthly and annual fluctuations. [Note: I typically present the detrended extratropical North Pacific sea surface temperature anomalies (20N-65N, 100E-100W) in these comparisons because the PDO is derived from it, but in this case I’ve used the entire North Pacific (0-65N, 100E-90W).]

Figure 4

Figure 4

It’s really difficult to miss the multidecadal variations in sea surface temperature in the North Pacific. They can vary in and out of synch with the North Atlantic. But let’s look at the sea surface temperature anomalies of the North Atlantic and North Pacific without the detrending, Figure 5. The sea surface temperatures in the North Atlantic may have peaked, but the plateau is very short. On the other hand, the North Pacific has cooled quite drastically since about 2003.

Figure 5

Figure 5

Of the two ocean basins, the North Pacific is the one that is contributing most to the hiatus in warming, but according to the UKMO, the variations in surface temperatures are even less understood in the North Pacific than they are in the North Atlantic, which aren’t understood either. But wait, there’s even more.

The cooling in the North Pacific extends to the entire Pacific basin, but the flattening of the sea surface temperature anomalies in the North Atlantic does not extend to the entire Atlantic Ocean. Refer to Figure 6, which presents the sea surface temperature anomalies for the Pacific Ocean from pole to pole (90S-90N, 120E-80W) and for the Atlantic Ocean from pole to pole (90S-90N, 70W-20E).

Figure 6

Figure 6

Those two basin-wide subsets have been detrended in Figure 7 to show that the multidecadal variability carries over to the entire basins.

Figure 7

Figure 7

Based on the UKMO’s own statements, they and the climate science community in general do not understand the causes of the multidecadal variations in the sea surface temperatures of the Atlantic and Pacific Oceans. This indicates that they also do not understand what caused the global warming from the mid-1970s to about 2000.  If they can’t explain the pause, then they surely can’t explain the warming.

It’s very important for the climate science community to be able to model the natural processes that contribute to the warming and cooling of the global oceans. Land surface temperatures mimic and exaggerate the annual, decadal and multidecadal variations in sea surface temperatures. This was presented by Compo and Sardeshmukh (2009) in Oceanic influences on recent continental warming. The abstract begins:

Evidence is presented that the recent worldwide land warming has occurred largely in response to a worldwide warming of the oceans rather than as a direct response to increasing greenhouse gases (GHGs) over land.

We can also use correlation maps to illustrate the influence of sea surface temperatures of the North Atlantic and North Pacific on Northern Hemisphere land surface air temperatures. I’ve used Berkeley Earth Surface Temperature (BEST) data in Figure 8 because CRUTEM4 is not spatially complete, and I’ve used Reynolds OI.v2 data in place of HADSST3 data for the same reason. The data extends from 1982 (the start of the satellite-based Reynolds OI.v2 data) to 2011 (the last full year of the BEST data). The top map shows the correlations of land surface temperature anomalies with North Atlantic sea surface temperature anomalies and the bottom map presents the correlations with North Pacific sea surface temperature anomalies. It’s easy to see why researchers believe the North Atlantic has a strong influence on the surface temperatures of the Northern Hemisphere, but the North Pacific can’t be ignored because it has an impact as well.

Figure 8

Figure 8

The UKMO briefly discussed the possible causes of the multidecadal variations in the sea surface temperatures of the Kuroshio Extension; also known as the Kuroshio-Oyashio Extension or KOE.

There is some evidence that changes in the ocean heat transports associated with the Kuroshio Extension region (an eastward flowing ocean current in the North Pacific Ocean), and the generation of internal ocean circulations called Rossby waves by the surface winds may be contributory factors (e.g. Alexander 2010).

We have discussed the processes through which leftover warm water from strong El Niño events is carried to the Kuroshio-Oyashio Extension during the trailing La Niñas and that this warming following strong El Niños counteracts the impacts of the trailing La Niña events on land surface and sea surface temperatures in the Northern Hemisphere. Refer to the post here.

As noted earlier, the PDO index does not represent the sea surface temperatures of the North Pacific. We’ve discussed this in numerous posts in recent years. See here, here and here. As presented in those posts, the PDO index is inversely related to the sea surface temperatures of the North Pacific. And as shown in Figure 9, the PDO index is also anti-correlated with variations in land surface temperatures for many parts of the Northern Hemisphere. That is, when the PDO index rises, land surface temperatures cool for many parts of the globe and vice versa when the PDO index falls. On the other hand, the sea surface temperature anomalies of the Kuroshio-Oyashio Extension are a measure of sea surface temperature and they are positively correlated the land surface temperatures of similar areas of the Northern Hemisphere.

Figure 9

Figure 9

To sum up this section, the PDO is an abstract representation of North Pacific sea surface temperatures, and as such, it is the wrong dataset to use in discussions of global warming—or a lack thereof.


The UKMO then included a discussion of their simulations of sea surface temperatures of the Pacific Ocean with different versions of their climate models, and they presented some maps of the simulations to show the prevalence of a La Niña pattern during the modeled decade-long hiatus periods. The UKMO wrote (my boldface):

The importance of decadal variability in the Pacific Ocean for driving extended periods of surface cooling has also been identified in multi-century simulations with fixed radiative forcing. Figure 6 shows the spatial distribution of the temperature trend averaged over extreme cooling periods from a number of climate model simulations from the Met Office Hadley Centre, including the most recent model (HadGEM2) that has contributed to the IPCC 5th Assessment Report. They all look remarkably like the observed pattern over the most recent decade (Figure 6, lower right panel), and provide fairly robust evidence of the importance of understanding the role of the Pacific Ocean in the current pause in global mean surface warming.

“Fixed radiative forcing” means infrared radiation from manmade greenhouse gases aren’t increasing in the models, there are no temporary decreases in sunlight caused by volcanic eruptions, there are no variations in radiation from the sun, etc. In other words, that’s not the real world. It’s irrelevant.

I’m always amazed that modeling groups will highlight the failings of their models. The UKMO presentation of model outputs with “fixed radiative forcing” strongly suggests that the UKMO’s climate models cannot simulate the pause if the models are forced by real world conditions. Or maybe the only way the UKMO’s climate models can simulate decade-long pauses in sea surface temperatures is by removing the effects of greenhouse gases. Ponder that for a while.

The UKMO then refer to a study by Meehl et al (2011) “Model-based evidence of deep-ocean heat uptake during surface-temperature hiatus periods.” It was a study based on the NCAR CCSM4 climate model. Meehl et al basically searched the future projections of the CCSM4 model for decade-long global warming hiatus periods. During those periods, the models were simulating La Niña-like conditions. But we know that climate models do not properly simulate the processes of El Niño and La Niña. I have used Guilyardi et al (2009) as a reference for these failings in past posts. The following paper is a more recent confirmation of how poorly climate model simulate ENSO. It’s Bellenger et al (2013): ENSO representation in climate models: from CMIP3 to CMIP5. Preprint copy is here. The section titled “Discussion and perspectives” begins:

Much development work for modeling group is still needed in order to correctly represent ENSO, its basic characteristics (amplitude, evolution, timescale, seasonal phaselock…) and fundamental processes such as the Bjerknes and surface fluxes feedbacks.

Additionally, in the post here, we presented and discussed how the NCAR CCSM4 climate model was not capable of simulating the observed warming pattern of the Pacific Ocean for the past 31 years. See Figure 10, which presents modeled and observed sea surface temperature trends for the Pacific Ocean since November 1981 on a latitudinal (zonal means) basis. The fact that those models can occasionally produce decade-land hiatus periods is, therefore, immaterial.

Figure 10

Figure 10


In their discussion of Figure 1, the UKMO states (my boldface):

Furthermore, upper ocean heat content continued to increase for a few years after 1998 (Figure 1). Nevertheless, it is clear that the rate of warming both at the surface and in the upper ocean has declined recently (Figure 1, lower panel). This is despite a continuing rise in atmospheric carbon dioxide concentrations, which act to trap radiation within the Earth system.

The data (maybe a reanalysis?) for the upper ocean in their Figure 1 is for the top 800 meters or about 2600 feet. The UKMO then goes on to discuss ocean heat content data under the heading of “2.3 What do we know about changes in the ocean heat budget?” They state (my boldface):

If indeed the ocean has hidden heat below the surface, then the question is whether changes in heat content, especially in the deeper ocean can be detected. There have been important advances in estimates of ocean heat storage over the last decade due to the correction of biases in conventional soundings, and the deployment of Argo floats.

That’s a big IF to start that paragraph, especially when they admit the data have to be modified, and, of course, those modifications were such that the data showed warming. They continue (my boldface):

The Argo measurements indicate a rapid increase in heat content down to 700m between 1999 and 2004 (Guemas et al, 2013), as indicated in Figure 1, but the rate of increase is lower thereafter. This period does, however, coincide with substantial changes in the observing system due to the rapid introduction of Argo floats. Although Argo floats are increasingly providing robust global measurements of the upper ocean, drawing conclusions from periods during which there are large changes in an observing system must be treated with caution.

The UKMO should be congratulated for noting that “…periods during which there are large changes in an observing system must be treated with caution.” But that was a discussion of the upper ocean. They continued (my boldface):

Measurements of heat content in the deeper ocean are much sparser and hence less certain. Using a combination of satellite and ocean measurements down to 1800m, Loeb et al (2012) estimate the Earth system has been accumulating energy at a rate of 0.5±0.4 Wm-2 from 2001 to 2010, similar to the 0.4Wm-2 from 2005 to 2010 down to around 1500m estimated from Argo floats alone (von Schuckman and Le Traon 2011). Reanalyses3 of ocean data give an average rate of warming from 2000 to 2010 of about 0.9Wm-2 averaged over the globe, with 30% of the increase occurring below 700m (Balmaseda et al. 2013). We conclude that the Earth system has continued to absorb a substantial amount of heat during the last 15 years, despite the pause in surface warming.

We discussed the flaws with Balmaseda et al (2013) in numerous posts over the past few months. See here, here, here, here, and, lastly, here under the heading of Balmaseda et al (2013).

The UKMO footnote 3 reads:

Ocean reanalysis uses a global ocean model, driven by driven by historical estimates of surface winds, heat, and freshwater fluxes, to synthesise historical observations of the ocean from a range of platforms using the process of data assimilation, in order to reconstruct historical changes in the state of the global oceans.

And the Reanalysis used by Balmaseda et al (2013) was the ORAS4 by the European Centre for Medium-Range Weather Forecasts (ECMWF). See the ECMWF webpage here for more information about ORAS4. It includes the following disclaimer (my boldface):

There is large uncertainty in the ocean reanalysis products (especially in the transports), difficult to quantify. These web pages are aimed at the research community. Any outstanding climate feature should be investigated futher [sic] and not taken as truth.

That simple fact does not come across in the UKMO paper, which reads as if Balmaseda et al (2013) are presenting gospel.

Above I congratulated the UKMO for admitting to problems during periods when there are large changes in the observing system, when discussing the ocean heat content data for the upper oceans. They seem now to have forgotten that by agreeing with the data at depths to 2000 meters. In the post here, we recently presented how the subsurface temperatures for the depths of 700 to 2000 meters showed little long-term trend from 1955 to 2002. See Figure 11. Then, as soon as the ARGO floats are introduced, the subsurface temperatures skyrocket. Also keep in mind that ARGO data have to be adjusted to show warming.

Figure 11

Figure 11

The change in the observing system appears to have had a very strong impact on the measurement of ocean temperatures to depths of 700-2000 meters, but it appears the UKMO is willing to overlook that impact because the data supports their conjectures.

The UKMO then introduces radiative imbalance to the discussion:

The sea level budget can also be used to constrain the ocean heat budget. As discussed earlier, the pause in global mean surface warming requires either a reduction in the energy input to the planet of ~0.6Wm-2, or the storage of that amount of energy within the ocean. If an additional heating of 0.6Wm-2 was indeed stored in the ocean, then this would lead to a thermosteric (thermal expansion) rise in sea level of about 1.2mm/year, assuming all the increase in heat is confined to the top 700m. This number would be smaller if some of the heat was mixed below 700m where the coefficient of expansion is smaller.

The assumed radiative imbalance at the surface was also recently presented in Stephens et al (2013) An update on Earth’s energy balance in light of the latest global observations. See their Figure 1. They illustrate an imbalance of 0.6 watts/meter^2 at the surface with an estimated uncertainty of +/- 17 watts/meter^2. We can see why the UKMO failed to discuss that uncertainty.

The UKMO comes to an amazing conclusion for that section. They state with unbelievable certainty:

In summary, observations of ocean heat content and of sea-level rise suggest that the Earth system has continued to absorb heat energy over the past 15 years, and that this additional heat has been absorbed in the ocean.

But they’ve presented nothing to support their certainty.


All in all, the UKMO continues to present the myth that only an assumed energy imbalance caused by manmade greenhouse gases can explain the warming of the global oceans, while the data, when it’s divided into logical subsets, indicate that manmade greenhouse gases had no visible impact on the warming of the global oceans. If this subject is new to you, refer to “The Manmade Global Warming Challenge” [42MB].


As a general note, the UKMO has presented numerous datasets in their efforts to show that global warming continues in theory, even though surface temperatures are no longer warming. Does the general public care about the temperature of the global oceans at depths below 2000 meters?  No.  Do they care about radiative imbalance?  No. People care about what the temperatures and precipitation will be like where they live.  And if those people live near shorelines, they care about sea levels. And sea level is probably the biggest concern.  But sea levels will continue to rise no matter what we do about greenhouse gases.  Does the UKMO address this?  No.  Then they’ve failed the people who pay the taxes that support their research.

The final paragraph of Part 2 of the UKMO’s report on the warming hiatus reads (my boldface):

The scientific questions posed by the current pause in global surface warming require us to understand in much greater detail the flows of energy into, out of, and around the Earth system. Current observations are not detailed enough or of long enough duration to provide definitive answers on the causes of the recent pause, and therefore do not enable us to close the Earth’s energy budget. These are major scientific challenges that the research community is actively pursuing, drawing on exploration and experimentation using a combination of theory, models and observations.

In other words, based on the limitations of available data, the UKMO does not understand the reason (or reasons) for the pause in global warming, and that strongly suggests that they do not understand the cause of the warming from the mid-1970s to 2000—a period when data are even less certain.

We’ll take a look at Part 3 of the UKMO report in a few days.

About Bob Tisdale

Research interest: the long-term aftereffects of El Niño and La Nina events on global sea surface temperature and ocean heat content. Author of the ebook Who Turned on the Heat? and regular contributor at WattsUpWithThat.
This entry was posted in CAGW Proponent Arguments, The Pause. Bookmark the permalink.

11 Responses to Part 2 – Comments on the UKMO Report “The Recent Pause in Global Warming”

  1. Dr. Lurtz says:

    Hi Bob,

    I think that there is an explanation for the extreme sea surface temperature anomalous indications from Unisys. I don’t think that they incorporate many actual sea-surface temperature readings. They just report the results of the satellite data. Now, that data is an interpretation of the radar altimeters combined with gravity to give a temperature [I think that this is a correct analysis, hard to get actual measurement techniques from the web site].

    Since they don’t take into account actual temperature measurement in all places, ocean currents causing ocean bulges could be interpreted as an increase in temperature; where none exists. I am amazed at temperature increases of 5C in a matter of 2 to 3 days!

    Just my thoughts. What do you think?



  2. Bob Tisdale says:

    Dr. Lurtz: In a couple of weeks, the sea surface temperature datasets that rely only on in situ measurements will be available. We’ll check your hypothesis then.


  3. tomwys says:

    Well done, Bob! Your piece is littered with GEMS that jump out of the text to delight the reader. I’ll leave it to others to find them too.

    I’m still recovering from the addition of well over 100 Arctic stations in Had4 vs Had3 (and no similar representation for the Antarctic) that serves to translate (mathematically) the data set uniformly upward.

    Thanks for your intrepid work and a valuable post!!!

  4. plazaeme says:

    Hi, Bob.

    Just a question, not too much off topic. Have you seen any papers addressing the question of the difference between the last two gobal warmings, related to the separation of land and sea surface temperatures during the last one? I can’t think of any obvious reason.

    It is shown in your figure 1. More clear here:

    I’ll appreciate a reference if you have any. Or your own thoughts.


  5. Bob Tisdale says:

    plazaeme: I don’t recall any recent papers about the divergence between land surface temperature anomalies and sea surface temperature anomalies, but I do believe it’s discussed in the IPCC’s AR4.

    My understandings of this topic: Land surface air temperatures mimic and exaggerate sea surface temperatures. Land surface air temperatures are more volatile than sea surface temperatures because the oceans have thermal inertia. In addition, land surface air temperatures are also impacted by factors that are not present in the warming of sea surface temperatures. These include heat island effect, black carbon on snow, land-use changes…and all of those modifications to the land surface temperature data that always seem to increase the long-term trend. One might also suspect, because there’s no evidence of human-induced global warming of the oceans, that manmade greenhouse gases may also impact land surface air temperatures.


  6. plazaeme says:

    Thank you, Bob.

    Yes, I understand the thermal inertia idea. But it should have been about the same case in the 1915-45 warming as in the 1975-2000. And, they weren’t. Unless it has something to do with CO2, but I don’t remember reading about this particularity of CO2 forcing. What I remember is they had to get rid off the mid century blip of sea surface temps.

    Forget it. Just another climate mystery. Or my ignorance.

  7. Bob Tisdale says:

    plazaeme, you apparently overlooked most of my reply to you:

    In addition, land surface air temperatures are also impacted by factors that are not present in the warming of sea surface temperatures. These include heat island effect, black carbon on snow, land-use changes…and all of those modifications to the land surface temperature data that always seem to increase the long-term trend.

  8. plazaeme says:

    Lets try an image.

  9. Bob Tisdale says:

    plazaeme: I’m not sure what additional point you’re trying to make with your annotated graph. I provided explanations for the divergence. Do you agree with those explanations?

  10. plazaeme says:

    I agree with the explanations, Bob. Thank you. I agreed before you provided them.

    In the right hand square, you have a warming, and a divergence. In the left hand square, you have about the same (global) warming with no divergence. Why? Why the difference divergence / no divergence? In case you have heard or thought of an explanation

  11. Diary Vazha says:

    Oh No Its Not Good for World 😦

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s