Guest post by John Kehr
Based on the most current data it appears that 2010 is going to show the largest drop in global sea level ever recorded in the modern era. Since many followers of global warming believe that the rate of sea level rise is increasing, a significant drop in the global sea level highlights serious flaws in the IPCC projections. The oceans are truly the best indicator of climate. The oceans drive the world’s weather patterns. A drop in the ocean levels in a year that is being cited as proof that the global warming has arrived shows that there is still much to learned. If the ocean levels dropped in 2010, then there is something very wrong with the IPCC projections.
The best source of sea level data is The University of Colorado. Only government bureaucracy could put the sea level data in one of the places farthest from the ocean, but that is where it is. I use both data sets that includes the seasonal signal. So with and without the inverted barometer applied. This is the source of the data that is used to show that the oceans are rising. Of course the rate of rise is greatly exaggerated and if the rate from 1993-2010 is used there will be a 1m rise in the year 2361.
Of course the rate is not constant. The rate of rise over the past 5 years has been half the overall rate. At the rate of the past 5 years it will be the year 2774 before the oceans rise a single meter. Of course a decrease in the rate is technically an negative acceleration in the rate of rise, so technically the rate of rise is accelerating, but in a negative direction. That statement is misleading though as most people consider acceleration to be a positive effect.
Sea Level Change
Even more interesting is the fact that from 1992-2005 there was an increase each year. 2006 was the first year to show a drop in the global sea level. 2010 will be the 2nd year to show a decrease in sea level. That is correct, 2 of the past 5 years are going to show a decrease in sea level. 2010 could likely show a significant drop global sea level. By significant I mean it is possible that it will likely drop between 2-3 mm from 2009. Since the data has not been updated since August it is difficult to guess more precisely, but the data ends at the time of year that the seasonal drop begins to show up. If the drop does show up as expected it is possible that 2010 will show the largest drop in sea level ever recorded.
2010 could show a significant drop in sea level from 2009.
Of course what will happen won’t be known until the data for the past 5 months is made available. I have been patiently waiting for the data to be updated for several months now, but I got tired of waiting and decided to put the information I have out there.
One fact is certain. A drop in sea level for 2 of the past 5 years is a strong indicator that a changing sea level is not a great concern. In order for the IPCC prediction to be correct of a 1m increase in sea level by 2100, the rate must be almost 11 mm/yr every year for the next 89 years. Since the rate is dropping, it makes the prediction increasingly unlikely. Not even once in the past 20 years has that rate ever been achieved. The average rate of 2.7 mm/yr is only 25% of the rate needed for the IPCC prediction to be correct.
This is yet another serious blow the accuracy of the official IPCC predictions for the coming century. The fact that CO2 levels have been higher in the last 5 years that have the lowest rate of rise than the years with lower CO2 levels is a strong indicator that the claims of CO2 are grossly exaggerated.
=========================================================
John Kehr runs the website The Inconvenient Skeptic – I recommend a visit. – Anthony
[jorkekafkazar says: January 17, 2011 at 3:44 pm
[That assumes the error is random, i.e., zero bias, which is often not the case. Sorry, you’ve still got a sow’s ear, no silk purse.]
Let’s see, now. Up to my previous post, most everyone on this thread was asking the question of how these sea heights were measured. I provided a link to the source agency where they described the technique quite transparently. I explained how statistical processing can be used to tighten up on error spread. This is all perfectly legitimate science.
I have the strong impression you did not bother to read that information. Your comment on statistics (random = zero bias) indicates that you do not understand statistics. “Random” is the usual acceptance of Gaussian error as a description of the random error. Bias is a separate error. They often go together. In the case of the Jason satellite, however, two independent radar altimeters are being used, so a comparison of the means of their measurements would be a way of discovering whether there are biases between the instruments. It is a little difficult to understand where a source of timing bias might arise, for that is the actual measurement process. But if there is “bias,” it represent a permanent offset of the altitude measurement and would be automatically cancelled by any difference comparisions (this year vs. last year, for example).
Finally, have the gumption to write NASA an e-mail asking them how they account for instrumental bias, instead of casting snarky aspersions against people who are performing legitimate science. I’m no fan of NASA in its modern incarnation, but not everyone in it is corrupt. Unfortunately, Jorge, I now have your measure.
Experiment a little… do a WEB search on ‘department of ecology’ + ‘carbon credits’
I did this in my own State of Washington. You will find near countless meeting minutes and conversations about Carbon credit income. It seems as if they’ve already spent the money! This is likely the reason the AGW ‘lie’ is so hard to kill. Bureaucrats have no fear of AGW, instead they are giddy about an unlimited funding source, enough money to fund every dream…. like young children anticipating a trip to Disneyland….. and you want to take it all away from them!!
No wonder they scream “the debate is over!”
Patrick Davis says:
January 17, 2011 at 2:53 am
Argumentum ad ignorantium. Just because you do not know how such measurement works does not mean it does not work. As long as the mechanism (literally bouncing radar off of the surface of the ocean) has sufficient resolution to eliminate aliasing caused from waves, tides, etc., it is possible to extract a mean level of oceans down to any arbitrary scale (limited by many factors, of course.)
That does NOT mean the results are accurate, nor does it mean I endorse the results (I do not know enough about the specifics of their method,) but it does mean you cannot simply say the results are wrong because it is your opinion they are impossible to acheive. Indeed, you would be surprised at what can be done with a little ingenuity (and intense signal processing theory.)
Mark
SteveE posted this link:
http://academics.eckerd.edu/instructor/hastindw/MS1410-001_FA08/handouts/2008SLRSustain.pdf
Also see Fig. 7. Compare model projections with empirical measurements.
I’ve come up with a “AGW save” theory!! Oceans are still rising, but below 2000 meter. 10000 IPCC scientists all agree…. [OK but this is very cryptic]
Mark
I haven’t looked back at all the comments but I may have been the first person to question whether satellites can measure sea level to within 1 mm. I am not saying that they cannot, but rather just throwing the possibility that they may not be able to do so into the melting pot. I am certainly appreciative of any comments explaining how measurements are taken and the specification of the system.
One point that galls me is that frequently data is set out without setting out the limitations of the accuracy of the data set, the relevant margin of error. Why does the satellite data not set out the measurement errors?
As regards errors, the first point is can a satellite maintain its orbital position above the earth to within 1mm? We all know about orbital decay which require satellites from time to time to be repositioned. There are also various gravitational factors which lead to samll orbital changes such as the sun, the moon and the conjunction of the gas giants. Can these cause a 1mm change in orbital course? Personally, I do not know. No doubt someone reading this thread could supply some answers.
Second, if the satellite is orbiting at say 1350 km above the earth, then that is some 1,350,000,000 mm above the target. As I understand matters, the satellite emits a signal which is then timed for rebound. It is measuring a distance of some 2,700,000,000 mm. May be, the system can measure to a wavelength of light but may be it cannot distinguish between 2,700,000,001 mm and 2,700,000,000 mm and 2,699,999,999 mm. Again, I do not know the answer. May be someone else does.
Third, there is the problem that one is measuring an ever changing randomly chaotic system. With waves, swell, wind spray, tides (gravitational bulge of the oceans) etc such that one is measuring a surface that is continuously distorting and on each subsequent pass will never be like the previous pass. I understand that statistical regression techniques are used to improve resolution. However, it is easy to see that this is not like a laser measure a fixed target in a laboratory and some error inevitably must be imparted.
Finally, what are we measuring given plate tectonics? How do we know that as plates ride over and under each other that on average they are not causing a mm or so difference in the depth of ocean floors. Continents are moving relative to one another, the shape of oceans is changing slightly, some land masses are rising some are falling, this causes a displacement effect. So too erosion. Do we know what effect these are having so appropriate adjustment can be made?
On top of these there is the interpretation issue. This is completely separate since it is not a measurement/data issue but rather a question of what are changes in ocean levels telling us, e.g., thermal expansion, ice melt, water extraction from land, reduction in river flows due to damming, irrigation etc?
SteveE says:
January 18, 2011 at 7:05 am
“Have a look at Figure 3a in this paper on sea level.
You’ll notice that it is certainly increasing at an accelerating rate and the time scale that it is taken at will certainly require more than year or two’s data to slow it down to what it was in the late 1800′s.”
An accelerating rate? Look at figure 1.a from the same paper. You’ll see the sea level rising for the past 20,000 years and have a very slow growth for the past 6-7 thousand years. If that graph is any indication then it would appear that our oceans are close to their maximum height – the vast vast majority of that height gained thousands of years before human civilization.
You can’t honestly call the growth in the figure you cite as “accelerating” when it is well known that is the period when the LIA ended and solar activity was increasing sharply. Just look at the graph from the 1930’s to the present and it is clearly a linear trend.
Misleading “trend” lines put in by agenda driven scientists are quite annoying.
“Will the IPCC get around to suggesting positive feedback from solar output, the way it does with greenhouse gases?”
Consider climate is like a pendulum. A forcing will cause it to shift to one side or the other. When a steady wind blows from the side, it will shift the pendulum but it will not increase the amplitude.
Now take the same wind, but only apply it during the phase where the pendulum is swinging away. Over time this will greatly amplify the motion of the pendulum.
What appears to be mostly ignored in climate science is that everything around us has resonant frequencies, and cyclic forcing can have much greater effects than constant forcings, or slowly increasing forcings.
As a result, those forcings that co-incidentally match the harmonic frequencies of natural systems are what likely will drive long term climate cycles.
It is amazing to me that so many commenters are willing to spend time expressing ignorant opinion and speculation, instead of reading up on the subject and learning something: http://sealevel.jpl.nasa.gov/technology/.
Patrick Davis says:
January 17, 2011 at 7:06 am
…You will also find no significant indication in Royal Naval archives of sea level rise in Royal Naval history, nowhere. And nowhere do I see sea level rises relative to changes in land levels, which fluctuate just as much as the tides.
But you can certainly find an interest in the topic, dating back to at least 1965:
http://www.springerlink.com/content/w40hx5073034400r/
I wasn’t about to drop $34 to satisfy my curiosity.
TonyK says:
January 17, 2011 at 11:22 am
“…Yes, boats moored there – no sea level change AT ALL in two thousand years! “
Well the changes under discussion are a matter of a meter or two, not fathoms. And I’d be astonished if subsequent occupants haven’t made substantial alterations, including docking facilities, which may (likely have) disguise the original mooring points.
SteveE says:
January 18, 2011 at 7:05 am
“Have a look at Figure 3a in this paper on sea level.
You’ll notice that it is certainly increasing at an accelerating rate and the time scale that it is taken at will certainly require more than year or two’s data to slow it down to what it was in the late 1800′s.”No Steve, Now look at this chart.
http://upload.wikimedia.org/wikipedia/commons/0/0f/Recent_Sea_Level_Rise.png
The only way to get an increase in the rate of rise is to go from the lowest point in the past 20 years to the highest peak. Of course CAGW propenets do this with the ENSO cycles over thirty years, so why not with sea level. Good idea to promote an agenda, but bad science. The level of rise is below the 20th century for the last decade, matches the 20th century for the last thiry years, and this de-acceleration is accelerating. There is NO indication that CO2 has done anything to increase the 20th century rate of sea level rise.
Michael J. Dunn says:
January 18, 2011 at 1:30 pm
It is amazing to me that so many commenters are willing to spend time expressing ignorant opinion and speculation, instead of reading up on the subject and learning something: http://sealevel.jpl.nasa.gov/technology/.
…………………………………..
Well Michael, it appears from the data sheet that raw measurments are accurate to within 3 cm.
Thereafter, they carry out their statistical regression and suggest that the measurements are, after that process, accurate to within “several” millimetres. It is unfortunate that they do not define “several” but since a couple is two, and a few is three, several is usually regarded as four or more. So I guess all going well, they can state that sea level rise is say 3mm +/- 4mm. So on an annual basis the increase in sea level is within the margins of error.
Hence what this suggest is that say over a period of some years (I would say at least 5), they would give a worthwhile indication of the change, eg 15 mm +/- 4
mm (assuming that each year the sea level rose by 3 mm for each of the 5 years).
However, as I noted this is all going well. It is interesting that they test the accuracy of the orbit via set land data points which do not suffer from the same chaotic measurment problems. There is nothing in the data sheet that suggests that the problems with chaos have been fully addressed (albeit I do accept that the statistical regression does help to reduce this factor). Likewise, there is nothing in that sheet to suggest that they have taken account of possible (and I don’t put it higher than that) changes caused by moving land masses/plate techtonics.
“The best source of sea level data is The University of Colorado.”
There is also CSIRO:
http://www.cmar.csiro.au/sealevel/
And AVISO:
http://www.aviso.oceanobs.com/en/news/ocean-indicators/mean-sea-level/
“The rate of rise over the past 5 years has been half the overall rate. At the rate of the past 5 years it will be the year 2774 before the oceans rise a single meter. Of course a decrease in the rate is technically an negative acceleration in the rate of rise, so technically the rate of rise is accelerating, but in a negative direction”
Only if you use the University of Colorado data, CSIRO and AVISO show a nearly constant rate of sea level rise of approximately 3.2 mm/yr:
http://www.aviso.oceanobs.com/en/news/ocean-indicators/mean-sea-level/products-images/index.html
http://www.cmar.csiro.au/sealevel/sl_hist_last_15.html
“Even more interesting is the fact that from 1992-2005 there was an increase each year. 2006 was the first year to show a drop in the global sea level.”
Not in the AVISO and CSIRO data. There was a drop between 2007 and 2008, that was obviously causated by the strong 2007-2008 La Niña.
“2010 will be the 2nd year to show a decrease in sea level. That is correct, 2 of the past 5 years are going to show a decrease in sea level.”
Sea level goes up and down seasonally. Over it there is a trend of 3.2 mm/yr. Even removing the annual signals, there is a lot of variability in all years. This is totally expected, because weather and climate oscillations (the biggest is ENSO) add noise to the data. There are spikes and drops every year.
” 2010 could likely show a significant drop global sea level. By significant I mean it is possible that it will likely drop between 2-3 mm from 2009.”
If you do not remove the seasonal signal, of course it will, as happens every year. If you remove the seasonal data, there is still the signal of ENSO and all the other oceanic oscillations. A drop in 2-3 mm will not be surprising given the current moderate-to-strong La Niña
“If the drop does show up as expected it is possible that 2010 will show the largest drop in sea level ever recorded.”
Maybe, maybe not. This is just guessing.
“A drop in sea level for 2 of the past 5 years is a strong indicator that a changing sea level is not a great concern.”
No. Its is an indication that there is noise in the data, mainly from ENSO.
“In order for the IPCC prediction to be correct of a 1m increase in sea level by 2100,”
This is quite false. The IPCC does NOT predicted a sea level rise of 1m. It predicted a sea level rise between 20 and 70 cm.
See here:
BRIEFING: a post-IPCC AR4 update on sealevel rise
http://www.cmar.csiro.au/sealevel/downloads/797655_16br01_slr_080911.pdf
“the rate must be almost 11 mm/yr every year for the next 89 years.”
No. For the satellite altimeter period, the predicted mean rate is around 2 mm/yr , not 11 mm/yr. The rates measured are much higher, around 3.2 mm/yr. So the IPCC UNDERestimated the rate of sea level rise.
See here:
Recent Climate Observations Compared to Projections
http://www.pik-potsdam.de/~stefan/Publications/Nature/rahmstorf_etal_science_2007.pdf
“Since the rate is dropping, it makes the prediction increasingly unlikely.”
The rate is not dropping according to AVISO and CSIRO. The University of Colorado shows a pause in sea level rise in the 2007-2008 La Niña, followed by a continuation of trend of the years before (a polynomial fit will give a deceleration, but it is more accurate to say that there was a nearly constant rate of SLR, then a pause, and then again the same rate of SLR)
“Not even once in the past 20 years has that rate ever been achieved. The average rate of 2.7 mm/yr is only 25% of the rate needed for the IPCC prediction to be correct.”
This statement is based on a fantasy “IPCC prediction”. The reality is that the rate of sea level rise (3.3 mm/yr) is 60 % HIGHER than the mean sea level rise rate predicted by the IPCC (2 mm/yr).
From Peru/Mars
For the record, from your aviso link, which give 3.28 mm/yr:
This graph, which for some reason you don’t like, shows 3.0 +/-0.4 mm/yr.
http://sealevel.colorado.edu/current/sl_ib_ns_global.jpg
Using your own numbers for a range of 20 -70 cm rise by 2100, and taking the mid-point, SLR will have to average about 5 mm/yr.
Of course, the low end (20 mm/yr) will merely repeat the slr for the 20th century, so one would hardly give any points for that.
richard verney says: January 17, 2011 at 3:37 pm
I fully concur with the observation “It is a sad fact that many in the scientific community still haven’t resolved the difference between precision and accuracy”.
I recall from geodesy class that precision was half the distance between the finest lines on your ruler, while accuracy was how close your measurement was to the actual truth. At that time, geodetics consisted of setting up a grid of points around the earth and measuring the distances between them, just trying to define the geoid. From our station in Colorado Springs down to the one in Trinidad, we had it down to 30 meters spherical. That was using theodolites, gravimeters, and photos of passing satellites timed from WWV adjusted for propagation error.
Mr Dunn has it right. You can get accuracy far greater than the precision of your instruments by taking a whole bunch of measurements, and the “few hundred thousand” satellite readings is a whole bunch. The reference to an out of date geoid doesn’t matter, as long as it’s the same geoid. It’s like the zero line in the temperature anomaly measurements, just an arbitrary choice.
That doesn’t apply to the temperature readings, though. The old liquid thermometers were marked to 1 °F I believe, but they were taking only one high and one low reading for each separate day, and the all the thermometers were reading different locations. So half a degree precision, no better than half a degree accuracy for the climate network.
The good thing about the sea level readings drop is that it takes some steam out of the mad rush to “do something” about the alleged climate problem. Same cheery thing about the big snowstorms.
I’ll toss in tokyoboy‘s Japan tide gauge chart –
http://i56.tinypic.com/15ewqwp.jpg
The entirety of the rising satellite data is encompassed in just the last rising oscillation of the Japan chart, so we may actually see a significant lowering for years to come. Maybe.
Big oil and the heartland institute are obiously hoarding water to drown kittens in…
Has the amount of snow on continental land masses last winter and in December 2010 been taken into account?
I hypothesise that global warming has caused proliferation of sponges which have soaked up all the water. Where do I get a grant to test this?
A fathom is 6 feet or about 1.8 meters. So 2 meters is “fathoms”.
The updated (late september 2010) AVISO data show a spike, to record levels of the sea level, just as did last year:
http://www.aviso.oceanobs.com/en/news/ocean-indicators/mean-sea-level/products-images/index.html
Will sea level drop in 2010? According to the (probably still preliminary) AVISO data, not.
This despite the current moderate to strong La Niña.
Hmmm, what happens if data indicates a sea-level drop but climate models do not?
Prof. Lindzen knows:
“Inevitably in climate science, when data conflicts with models, a small coterie of scientists can be counted upon to modify the data. Thus, Santer, et al (2008), argue that stretching uncertainties in observations and models might marginally eliminate the inconsistency. That the data should always need correcting to agree with models is totally implausible and indicative of a certain corruption within the climate science community.”
richard verney says:
January 18, 2011 at 10:35 am
Questioning the ability to do so, with the intent of learning more, as you apparently are doing, and flatly stating that it cannot be done with an absence of the understanding of the process are two different things. The former is an attempt to learn, the latter is argumentum ad ignorantium.
The link Michael Dunn provided does not specify the type of radar being used, so I cannot comment on specifics, but I can say this is one instance in which the much touted law of large numbers actually works. In other words, you can average over multiple readings to reduce the noise in the aggregate measurement. By integrating over a long period of time, you constrain your noise bandwidth to a very small range (less noise power) while maintaining the same signal power, i.e., your SNR gets larger with increased integration. The same principle is used to get very accurate GPS measurements, for example. Since the measurement is a distance, an average actually has physical meaning, too, not just mathematical.
Interestingly, the clutter induced by the surface of the water is not typically Gaussian. Depending upon the waveform used in the radar (particularly the frequency,) it could be rather complex. There is significant literature gaining ground in the radar community regarding the treatment of clutter as a compound K distribution, a distribution that does not have a lot of exposition in and of itself.
Mark
The first chart above calls out for cumulative sum analysis.
I had industrial experience of the power of the technique over many years when then new computing power became available to convert from the difficult-to-interpret “mask” presentation to a bar chart.
The management value came because:
1.Cusum “Change points” became time indicators of changes in process.
2. The statistical significance between periods of production could be clearly illustrated
The technique has been advanced beyond my very limited industrial experience, see:
http://www.variation.com/cpa/tech/changepoint.html#Change-Point Analysis
Maybe someone more qualified than I would like to have a look at this?
Don
D. J. Hawkins says:
January 18, 2011 at 2:22 pm
Well the changes under discussion are a matter of a meter or two, not fathoms.
Err, no, 2000 years times 3mm = 6 metres – roughly twenty feet! I can absolutely guarantee the sea level here has not changed that much. In fact, a little way up the coast is Fishbourne Roman Palace which used to be right on the coast – or a least next to an inlet where ships used to call – and now it’s half a mile or more away from the sea. My point is that there must be many places where the rise in sea level is insignificant compared to the change in the elevation of the land for all manner of reasons.