SandyS

Got it, I think.

You used Excel to measure the rate of change in sea level for every possible 46 year period since 1700, and then listed them in order of decreasing size. The years listed are the ends of the 46 year periods.

I like it. It's just the sort of thing I enjoy doing myself. I have to revive my Excel skills!

You argue that since the fastest change was between 1804 and 185O and no recent periods match, sea level is not a valid indicator of climate change.

Unfortunately, as I said, I think your pattern reflects the changing amount of noise in the data rather than true differences in the rate of change.

Looking at your graph the data shows three distinct noise levels. Pre-1860 the short term variability is very large, between 1860 and 1940 it is much amaller, and post 1979 it smooths a little more. This probably reflects improving measuring technology , rather than reduced rate variation.

Pulling 46 year periods out of the record, the highest rates will tend to start with a low sea level value and end with a high value. When the noise level is high the lows are lower and the peaks are higher, so the potential difference between them is larger. A lower noise level reduces the potential range.

The result is that a list starting with the largest range and based on the sea level data would tend to include a lot of early periods and fewer late periods, which is what you saw.

If you were examining data with a lot of genuine sea level variation and low noise, this would not be a problem. Unfortunately, in the early years especially, the noise is bigger than the signal. Your list reflects the reduction in noise level with time, not a reduction in ratea of change.

> I have come up with a ready reckoner to compare these two alternatives.
> This graph shows HadCRUt4 from 1970 on.
> I have added a 1970-1998 linear trend in green and a 1998-2013 linear trend
> in blue. I printed out the graph and extended both trends to the year 2040.

EM, as before if you include years before the temperature series properly plateaued you're supressing the start of the trend. Try re-plotting with any start year from 2001.

Or alternatively look at the whole record....

http://www.woodfortrees.org/plot/hadcrut4gl/from:1840/to:2014

...and ask yourself if what's happening now looks so very different from what happened in the 1940s that some "bogeyman" has to be the culprit.

It looks completely natural to me.

" Applying Occam's Razor, there is no point postulating extra unknown processes when the existing ones successfully account for the energy budget." EM

You're assuming the unknown processes are my idea, they're not. I believe Trenberth is the one who suggests that's where the heat is hiding. Others think the increased aerosols from China have bounced the heat back into space. I'm sure some think the oil companies have done a deal with the lizard people to shove the heat down the back of our sofas until the Rapture. or some such.

Look at this from WUWT. Sea level rise is not accelerating. Maybe slowing down?

http://wattsupwiththat.com/2014/11/15/dr-trenberth-redux/

You wrote “Recent temperatures are now back on the 1970-1998 trend line. It will be interesting to see where they go next.” Even if I agreed with you, and by now I have little faith in climate measurement, but even if I agreed, the rate is not accelerating. It’s the acceleration that arose from positive feedbacks. Remember at least one of those feedbacks (Arctic ice cover) is ahead of schedule and yet there is no acceleration.

The point is - the sums aren't adding up. Now I might be rude about climate scientists but the models aren't failing for a wont of trying or brains on their part. With all they’ve got in computing power and smart people, they’re the best answer money can buy. Nothing you can figure out will be better,

I’ve said before, you only get one chance to be taken on trust. Now they’ve got to earn it. That means making their operations more transparent and accountable and issuing predictions with explanations what it means if those predictions don’t come true. Let’s face it, we’d be a lot more forgiving if scientists had explained there might be decade plus long pauses before the current one happened. They could start by issuing a spec for how they measure global temperature and sticking to it. That means no more abusing the historical data.

Nial

Did you try these yourself? The results are not what you expected.

You can start the pause in 2001 quite happily, though I would probably prefer 2002, which is when the GISS 5-year average goes flat.

I drew my second graph using a 1970 start and a pause from 2001.The difference between the two trends is more marked.

The 1970-2001 trend is much like the 1970-1998 trend and still goes straight through the 2014 figures.

The 2001-2013 trend now points down and the 95% confidence limits no longer overlap after 2020. It may be moot. The 2014 figures are already outside the 95% confidence limits for the pause!

Starting the pause later strengthens the case that the pause has already ended. :-)

Starting from 1850, the pause begins with the temperatures already five standard deviations above the long term trend and the pause looks like a regression back towards the trend. If temperatures continue down it would be evidence for a 60 year cycle peaking in 2001, then dropping again until 2030 odd. Unfortunately that hypothesis would then expect rapid warming again after 2030.

Incidentally, my problem with crediting natural variation for the changes is that there is no evidence that natural variation is sufficient . if you use the published measurements, natural variation is insufficient to explain the observed changes. If you follow the denier line and say all the observations are wrong, then you have no evidence at all!

But the hype was built on all the warming being man made and the future warming to be exponential. A moderate amount of warming is not the C in CAGW.

Do your graphs again but plot CO2 and temperature instead of date. If anything the rate of warming is decreasing. The number of years is something of a red herring.

TinyCO2

I am sure Willis Eschenbach is an excellent person, but I doubt his conclusion for two reasons. The first is his rudeness about Trenberth. I automatically doubt anything written in the language of propaganda.

The second is that his graph of sea level stops in 2010 at the end of a year which dropped 500 cubic kilometres of water onto Australia and dropped sea level by 1.4mm. If his graph had continued it would have shown the recovery to the long term trend average.

Which is why I didn't say it was slowing down. I said there was no acceleration, something that is essential to the prophets of doom.

TinyCO2

Have you noticed that CAGW is used by sceptics rather than by accepters? Catastrophe is your meme, not mine.

Regarding rates of warming, that is why I drew the graph. Let's try it for a while and see how the trend goes.

TinyCO2

Depends on your timescale. For the period since 1990 the rate has been 3.2mm/year with confidence limits of +/- 0.4mm. That implies a minimum rate since 1990 larger than 2.8 mm/ year and less than 3.6mm/ year. It is even possible, though not very probable, that the rate has increased by 0.8mm/year over 24 years

Even at 3.2mm/year this rate is higher than any other period since direct tidal measurement began in 1850.

You complain that the data is uncertain, but forget that the uncertainty can be measured and used.

Em, the C might not be your meme but you're not seriously trying to pretend that the majority of messages to the public have not been very catastrophic in their outlook? Do I really have to get you quotes for what you know is true? I don't remember the shock horror of being told the seas might rise a full 32cm in the next hundred years. Flee! Flee now while you have the chance!

You also need to use data prior to 1970. The warming from CO2 started 1950 ish and began with a pause. There were historical periods of warming that match the warming in the 80s and 90s which are hard to match if you start fiddling about with climate driver values. Your mini model has to match the past as well as the present.

If you plot CO2 against temperature back to 1850 the match is scary until you accept that prior to 1950 the temperature was mostly driving the CO2 up rather than the other way round. After 2000 the effect of extra CO2 diminishes.

> The 2014 figures are already outside the 95% confidence limits for the pause!
> Starting the pause later strengthens the case that the pause has already ended. :-)

EM, you're making this up as you go along.

What's your definition of "95% confidence limit"?

> Starting from 1850, the pause begins with the temperatures already five
> standard deviations above the long term trend and the pause looks like
> a regression back towards the trend.

Just as the warming from 1970 - 2000 started significantly _below_ the trend and was merely a regression back to the trend? (Just as happened from 1910 - 1940)....

http://www.woodfortrees.org/plot/hadcrut4gl/from:1840/to:2015/plot/hadcrut4gl/from:1840/to:2015/trend

"If you follow the denier line and say all the observations are wrong, then you have no evidence at all!" I don't know of any deniers who say all the observations are wrong, in fact I believe you'll find that particular stance is more prevalent among the alarmist community. It took over 10 years of no rise in temperature for them to admit that - the temperature hasn't risen. They are undeterred by the fact that the forecast temperature rises are wildly out and they don't believe that the heat has gone out of the system, preferring to postulate that the heat, having warmed the atmosphere from 1880 to 1998 suddenly decided to take itself off into the deep oceans without a shred of evidence.

This whole thing is a farce beyond parody. We have a theory sans conditions of failure - BIG FAIL. We have scientists who can't give us Friday's weather on Wednesday with any certainty telling us they can forecast the exact weather in 2050, 2100, 3142 etc. And these warnings can only be avoided by the dismantling of Western Industrialisation, an agenda that happens to be the objective of the UN, the Club of Rome and the environmentalists.

I gave in a long time ago. I will continue to challenge this rubbish masquerading as science, but the, let's just take the scientists at their word, we don't need them to keep coming up with scare stories, let's for the time being at least pretend that just for once the claim to be able to foretell the future isn't the sole preserve of charlatans and mountebanks, and let's go straight to a viable solution to emitting CO2.

We can accelerate this solution by diverting the money we're pissing away paying for scarier stories from the clisci community and put it into development of safe nuclear energy and the training of engineers to deliver these solutions.

Here's a forecast for you, by the end of the century, regardless of the CAGW scare, we will have developed energy supplies that don't depend on fossil fuels and dispense with the need for national grid. Everyone will power their homes with their own individual power source, probably nuclear and probably cold fusion.

Entropic man
Just three thoughts which have come to mind, before I call it a day.

1. This is the only data I have seen that goes back to 1700, the CU data is only 32 years long and therefore is virtually useless as the is no pre-industrial data.

2. Despite the noise there is no sign of an unprecedented rise in the latest 4 decades.

3. Does the noise mean that every single 46 year period pre-1992 will over estimate the sea level rise.

Therefore sea levels show no sign of a CO2 influenced rise.

Neither of of us can guess what Raff thinks about falsification.

Entropic man

Have you noticed that CAGW is used by sceptics rather than by accepters? Catastrophe is your meme, not mine

If you don't think that there is a C just AGW is there anything to worry about? That seems to me to be what could be considered to be a sceptical position.

Did you see the report by the Universaity of Bonn on the result of drilling in Lake Vann, Turkey.

[URL]http://www.eurekalert.org/pub_releases/2014-11/uob-cco111714.php[/URL]

[Quote]Once they determine the composition of the vegetation present and the requirements of the plants, the scientists can reconstruct with a high degree of accuracy the temperature and amount of rainfall during different epochs. These analyses enable the team of researchers to read the varves of Lake Van like thousands of pages of an archive. With these data, the team was able to demonstrate that fluctuations in climate were due in large part to periodic changes in the Earth's orbit parameters and the commensurate changes in solar insolation levels. However, the influence of North Atlantic currents was also evident. "The analysis of the Lake Van sediments has presented us with an image of how an ecosystem reacts to abrupt changes in climate. This fundamental data will help us to develop potential scenarios of future climate effects," says Doctor Litt.[/QUOTE]

Nial

95% confidence limits. In the context of Hadcrut4, they define their global temperature averages with an uncertainty quoted as +/- 0.1C (+/-2Standard Deviations). 95% of these averages should be within 0.1C of the actual average temperature.

I applied this to the trend lines. If the trend is valid 95% of the quoted averages should be within 0.1C of the trend line. That is why I drew them on the graphs, as a visible aid for judging how the temperatures were behaving. As an added bonus, a quick and dirty way of seeing when a difference between two samples becomes significant. Their 95% confidence limits no longer overlap.

You quote variations which look like a sixty year cycle of amplitude +/- 0.15C superimposed on the long term trend. AMO?

SandyS

1) The current CU data is an ensemble of tide gauge data corrected for isostatic and tidal effects, and satellite data.
Pre satellite data comes from tide gauges alone, without some of the corrections.Pre-1850 data comes from anecdotal records and proxies, hence the greater unc rtainties of the earlier data.

2) Apply a smoothing algorithm to filter out the noise and you get a curve which steepens steadily from start to finish. Thus the current long term rate of 3.2mm/year is higher than the long term trend at any time since 1700. This is unprecedented., for the post 1700 record. Look on a longer timescale and the end of the last glacial period 10,000 years ago, saw rates of 25mm/year.

3) No, the effect is stochastic, not determinist. All else being equal, the earlier data is more likely to produce larger apparent rates, There is no reason to expect that every 46 year period should be larger in the earlier data,

SandyS

A catastrophe would be something like the Chixacub meteorite, which played a big part in the extinction of the dinosaurs.

The problem is that our civilization is hypersensitive to much smaller changes and likely to collapse under stress which is, to the planet, merely a minor adjustment.

Apply a smoothing algorithm to filter out the noise and you get a curve which steepens steadily from start to finish. Thus the current long term rate of 3.2mm/year is higher than the long term trend at any time since 1700.
Nov 19, 2014 at 12:29 AM | Unregistered CommenterEntropic man

It may or may not be harmful, but it is certainly wrong. You must first demonstrate that it has no effect other than visual/presentational. Why else do it?

Michael Mann has carved a career out of carving hockey sticks from time series with such contrivances, but that doesn't make it correct. I've pointed you towards Briggs' comment before. Did you read it?
" Do not smooth times series, you hockey puck!"

Entropic man
In general I agree with Michael hart who has said it better than I could.

Where is the pre-1990s data from Colorado?

For number 3 a simple yes or no would suffice, I still don't really know what your answer is. As looking at the data the noisiest is also the most "stable".

I would say that the current civilisation is no more or less sensitive to changes in climate than any other in history. Especially when one removes the human predisposition to think that we are the most whatever in the history of mankind.

Radical Rodent

If you print out and annotate the first graph I described, the 95% confidence limits no longer overlap in 2031. The warming trend would be expected to produce anomalies then around 0.8C. The pause trend would produce 0.6C.

If you try Nial's suggestion of starting the pause in 2001 it shows a cooling trend. The 95% confidence limits separate in 2021 with the two trends at 07C and 0.5C respectively.

Let's wait and see which way it goes.

Finally got around to typing up the first calculation from observation.

There are three main energy sinks in the climate system, these are the oceans, the ice and the atmosphere. The pause in air temperatures means that we only need to consider ice and oceans.

Ice volume on land measured by GRACE and Cryosat is decreasing by 500 cubic kilometres per year.

The latent heat of fusion of water is 3.34*10^5 J/kg. 500 cu. Km.of ice is 5*10^14Kg.

The melting ice is therefore absorbing 1.67*10^20J/year.

It takes 360 cubic kilometres of extra volume to raise sea level by 1mm. The melting ice is raising sea level by 1.39mm/year.

Sea level is rising at 3.2+/-0.4 mm/year.Take out the ice- related rise and thermal expansion is causing an increase of 1.81mm/year. This is an expansion of 652 cu.km/year.

Total volume of the ocean is 1.37*10/9 cu.km. The thermal expansion coefficient of seawater is ~10^-4/K.

The ocean has expanded by 6.52*10^2 / 1.37*10^9 = 8.9*10^-7. This corresponds to a temperature increase of 8.9*10-7 / 10^-4 = 8.9*10^-3K.

The specific heat of seawater is 3.98*10^3J/Kg/K, 3.98*10^15J/cu. km/K.

To produce the observed rate of sea level rise requires ( 1.37*10^9)*(3.98*10^15)*(8*9*10-3) =4.85*10^22
J.

Total imbalance is 4.85*10^22 + 1.67*10^20, still 4.85*10^22J.

1W = 1J/second, 3.15*10^7J/year.

To produce the imbalance you need an input power of 4.85*10^22 / 3.15*10^7. = 1.54 * 10^15W.

Surface area of the Earth is 5.1*10^14M^2. The imbalance is 1.54*10^15 / 5.1*10^14 = 0.3W/M^2.

The uncertainty in sea level rise is +/- 0.4mm. The uncertainty in energy imbalance is therefore 0.3*( 0.4/3.2) = 0.04.

Stephens et al 2012 gave an imbalance figure at TOA of 0.6+/- 0.4. Nice to see that my own rough amateur estimate is within the uncertainty range of the professional work.