Global warming temperature predictions

[duwayne]

Mar 31, 2012 14:45:49 GMT socold said:

Mar 26, 2012 14:00:09 GMT duwayne said:

Socold, you continue to point out that the ocean currents have reduced the amount of global warming over the recent past and rightly so. But you seem to ignore the fact that ocean currents contributed significantly to the warming during the 1977-2007 period when the ocean currents were in their warm phase. Why does this only work in one direction?

It doesn't work in only one direction. From 1977-2007 though ENSO hasn't trended upwards enough to explain much of the warming at all.

Since 2002 it has trended downwards enough to explain significant cooling

Socold, you indicate ENSO significantly impacted temperatures during the 2007-2011 period but had very little impact for the 1977-2006 period. If my calculations are correct, the average annual MEI for 1977-2006 was +0.40 and the average for 2007-2011 was -0.39. Why would -0.39 have far greater impact than +0.40?

[sigurdur]

dwayne:
Thank you for the confirmation about ENSO. Your calculations are correct.

[socold]

Mar 31, 2012 22:38:31 GMT duwayne said:

Mar 31, 2012 14:45:49 GMT socold said:

It doesn't work in only one direction. From 1977-2007 though ENSO hasn't trended upwards enough to explain much of the warming at all.

Since 2002 it has trended downwards enough to explain significant cooling

Socold, you indicate ENSO significantly impacted temperatures during the 2007-2011 period but had very little impact for the 1977-2006 period. If my calculations are correct, the average annual MEI for 1977-2006 was +0.40 and the average for 2007-2011 was -0.39. Why would -0.39 have far greater impact than +0.40?

MEI falls over 1977-2006, it's a cooling trend. It falls at a rate of about -0.1 MEI index/decade.

From 2007 to 2012 the negative slope is 10 times greater, -1.1 MEI index/decade.

In terms of temperature the relationship I use is that the ENSO temperature influence in hadcrut is about 12% of MEI index.

So over the period 1977-2006 ENSO caused about -0.01C/decade cooling (effectively no contribution at all), but from 2007 it's 0.13C/decade cooling, a significant amount compared to the longterm warming rate, enough to mask almost all of it.

[duwayne]

Apr 1, 2012 20:55:24 GMT socold said:

Mar 31, 2012 22:38:31 GMT duwayne said:

Socold, you indicate ENSO significantly impacted temperatures during the 2007-2011 period but had very little impact for the 1977-2006 period. If my calculations are correct, the average annual MEI for 1977-2006 was +0.40 and the average for 2007-2011 was -0.39. Why would -0.39 have far greater impact than +0.40?

MEI falls over 1977-2006, it's a cooling trend. It falls at a rate of about -0.1 MEI index/decade.

From 2007 to 2012 the negative slope is 10 times greater, -1.1 MEI index/decade.

In terms of temperature the relationship I use is that the ENSO temperature influence in hadcrut is about 12% of MEI index.

So over the period 1977-2006 ENSO caused about -0.01C/decade cooling (effectively no contribution at all), but from 2007 it's 0.13C/decade cooling, a significant amount compared to the longterm warming rate, enough to mask almost all of it.

Let's try this another way. Would you agree that the average temperature over the 1977-2006 period was higher because of ENSO? Would you agree that if ENSO is significantly negative over the 2007-2037 period than the average temperature will be reduced accordingly? Then would you agree that if you extrapolate the 1977-2006 trend, it will overstate the 2007-2037 average temperature if ENSO is significantly negative?

[socold]

"Would you agree that the average temperature over the 1977-2006 period was higher because of ENSO?Would you agree that if ENSO is significantly negative over the 2007-2037 period than the average temperature will be reduced accordingly?"

Yes and yes

"Then would you agree that if you extrapolate the 1977-2006 trend, it will overstate the 2007-2037 average temperature if ENSO is significantly negative?"

No, because average does not tell us about trend. Eg over the period 1977-2006 average ENSO was positive, but the trend in ENSO fell, which I interpret as a cooling influence on the global temperature trend. The trend can a falling (cooling) one whether ENSO is average positive or negative over the period.

[socold]

Tamino has a post in which he explains what I was saying about the last 10 year period since 2002 better than me.
tamino.wordpress.com/2012/04/05/decadal-trend-in-temperature/

Solar impact on UAH temperatures:


ENSO impact on UAH temperatures:


Combined impact on UAH temperatures according to his calculations is about -0.2 deg.C/decade since 2002:


Then it's just a process of elimination:
Solar + ENSO + OtherStuff = global temperature change

UAH currently reads flat since 2002, so:

-0.2C + ? = 0C

? = 0.2C, ie "other stuff" has caused 0.2C warming since 2002.

I think Dr Spencer was planning to fix something with UAH which will probably drop the trend and reduce the "other stuffs" contribution. RSS shows -0.1C since 2002, but I don't know if the solar/enso contributions for that to be able to use it.

[duwayne]

Apr 6, 2012 11:42:57 GMT socold said:

"Would you agree that the average temperature over the 1977-2006 period was higher because of ENSO?Would you agree that if ENSO is significantly negative over the 2007-2037 period than the average temperature will be reduced accordingly?"

Yes and yes

"Then would you agree that if you extrapolate the 1977-2006 trend, it will overstate the 2007-2037 average temperature if ENSO is significantly negative?"

No, because average does not tell us about trend. Eg over the period 1977-2006 average ENSO was positive, but the trend in ENSO fell, which I interpret as a cooling influence on the global temperature trend. The trend can a falling (cooling) one whether ENSO is average positive or negative over the period.

The slope of the Hadcrut3 anomalies for the 1977-2006 period is 0.017. I calculate that an extrapolation of the 1977-2006 anomalies would give an average anomaly for the 2007-2037 period of 0.71C. Using your ENSO adjustment factors, can you give me an ENSO scenario for the 2007-2037 period that is significantly negative on average which results in a projected average anomaly of 0.71C or higher ?

[duwayne]

Apr 6, 2012 11:44:53 GMT socold said:

Tamino has a post in which he explains what I was saying about the last 10 year period since 2002 better than me.
tamino.wordpress.com/2012/04/05/decadal-trend-in-temperature/


ENSO impact on UAH temperatures:




Then it's just a process of elimination:
Solar + ENSO + OtherStuff = global temperature change

UAH currently reads flat since 2002, so:

-0.2C + ? = 0C

? = 0.2C, ie "other stuff" has caused 0.2C warming since 2002.

I think Dr Spencer was planning to fix something with UAH which will probably drop the trend and reduce the "other stuffs" contribution. RSS shows -0.1C since 2002, but I don't know if the solar/enso contributions for that to be able to use it.

What result do you get for Hadcrut3?

[sigurdur]

Socold:
I will look at Fosters link, but do so critically. I have found his knowledge of proper use of stats to be lacking. I am sure he is a fine guy, but he isn't the brightest and then refuses to acknowledge his errors from what I have seen.

[hrizzo]

Socold:

So Tamino calculates what temperatures would have been "if" some forcings would have not happened...

A few caveats:

1) In order to do so, he should know every extra forcing and then calculate; if he doesn´t know every one of them, then he is just guessing.

2) Even if he knows them, he shouldn´t do comparisons with other periods without applying to them the same adjustments, that is, he should adjust all registered older data to be true. If not, he is just trying to mislead us.

3) But truth is that we don´t know all of the possible "extra forcings", so in order to have an idea about what reality is, the only sensible thing to do is accept and use raw data at face value, without "massaging", and then make the comparisons.

That is, if you really want to do science... but I don´t think he wants to do that.

[duwayne]

This WUWT post wattsupwiththat.com/2012/04/05/preview-of-cmip5ipcc-ar5-global-surface-temperature-simulations-and-the-hadcrut4-dataset/#comment-948248 contains this chart

which shows a preview of how the new IPCC models might look versus actual observations.

The post says......

But based on the model mean, the CMIP5-based hindcasts of the 20thCentury are:

1. not able to simulate the rate at which global surface temperatures cooled from 1944 to 1976 (Figure 11),

2. incapable of simulating how quickly global surface temperatures warmed from 1917 to 1944 (Figure 12), the observations warmed at a rate that’s more than 3 times faster than simulated by the models, and,

3. not capable of simulating the low rate at which global surface temperatures warmed from 1901 to 1917 (Figure 13).

The article does not mention the PDO but notice how closely the problem areas relate to the PDO half cycles I have mentioned here several times. A down leg prior to 1917, an up leg in the 1917-1947 period and a down leg in the 1947-1977 period and an up leg in the 1977-2007 period. This is an obvious explanation for the hindcast errors of the models.

And it's also obvious that ignoring the Ocean Cycle effect will mean the models will over forecast the warming for the 2007-2037 period and this should soon apparent as more observations are added to the chart.

[socold]

Apr 6, 2012 16:14:27 GMT hrizzo said:

Socold:

So Tamino calculates what temperatures would have been "if" some forcings would have not happened...

A few caveats:

1) In order to do so, he should know every extra forcing and then calculate; if he doesn´t know every one of them, then he is just guessing.

What Tamino has estimated is what temperatures would have been without ENSO and solar cycle forcings. It's not temperature without *every* forcing. Removing just two is useful to see what is left.

2) Even if he knows them, he shouldn´t do comparisons with other periods without applying to them the same adjustments, that is, he should adjust all registered older data to be true. If not, he is just trying to mislead us.

He has applied the same adjustments back to 1979:

3) But truth is that we don´t know all of the possible "extra forcings", so in order to have an idea about what reality is, the only sensible thing to do is accept and use raw data at face value, without "massaging", and then make the comparisons.

If you use raw data at face value you should really only look at 30 year trends, not 10 year trends to determine if global warming has stopped. 10 year trends are prone to being flat even if global warming hasn't stopped.

Eg a flat temperature trend since 2002 can mean two things:

1) Global warming of ~0.2C/decade has stopped
2) Global warming of ~0.2C/decade hasn't stopped but solar activity and ENSO have induced 0.2C cooling since 2002.

If you take raw temperature at face value then you can't tell. Only by trying to remove ENSO and the solar cycle can you rule out #2.

[socold]

Apr 7, 2012 15:05:51 GMT duwayne said:

This WUWT post wattsupwiththat.com/2012/04/05/preview-of-cmip5ipcc-ar5-global-surface-temperature-simulations-and-the-hadcrut4-dataset/#comment-948248 contains this chart

which shows a preview of how the new IPCC models might look versus actual observations.

The post says......

But based on the model mean, the CMIP5-based hindcasts of the 20thCentury are:

1. not able to simulate the rate at which global surface temperatures cooled from 1944 to 1976 (Figure 11),

2. incapable of simulating how quickly global surface temperatures warmed from 1917 to 1944 (Figure 12), the observations warmed at a rate that’s more than 3 times faster than simulated by the models, and,

3. not capable of simulating the low rate at which global surface temperatures warmed from 1901 to 1917 (Figure 13).

Of course anyone who was skeptical of the usable accuracy of the surface temperature record wouldn't be able to conclude either #1, #2, or #3. They wouldn't know how fast it warmed from 1917 to 1944 for example so for all they know the models are right and HadCRUT is wrong.

[trbixler]

wattsupwiththat.com/2012/04/09/march-2012-sea-surface-temperature-sst-anomaly-update-a-new-look/

[icefisher]

Apr 10, 2012 0:45:09 GMT trbixler said:

wattsupwiththat.com/2012/04/09/march-2012-sea-surface-temperature-sst-anomaly-update-a-new-look/

Yep take out the eruptions of El Chichon and Pinatubo in the first 10 years and it would be a lot worse!