By Andy May
We’ve all heard the question. We point out there is no evidence that current climate changes, whether man-made or natural, are dangerous or unusual. Then we are asked, but “What about sea level rise? Isn’t that dangerous?” There are many very good technical arguments why the current rate of sea level rise will not threaten humans, New York City, Miami, or Tuvalu. These are urban legends that spawn from silly IPCC models as explained by Ole Humlum in Chapter 10 of our latest book (Crok & May, 2023). How are these myths dismissed quickly in clear language? This is my best attempt in ~600 words. Let me know how I did.
The current rate of global sea level rise is below the accuracy of our current ability to measure it as discussed in Kip Hansen’s Chapter 5 in Crok & May. Figure 1 shows three respected estimated rates. The sea levels are shown as reported and they have different zero points. The Jevrejeva, et al. estimate in blue is 2 mm/year (± ~0.3), the one below that is the Church and White estimate of 1.7 (± ~0.3) mm/year. These estimates are both from tide gauges, although the Jevrejeva estimate does try and include the satellite data from 1993 to 2009. Due to the overlap of the author’s estimates of uncertainty, the two estimates are statistically equivalent.
The lower estimate, shown in gray, uses all the NASA satellite data since 1993. It shows a rate of 3.3 mm/year (Beckley, Callahan, Hancock, Mitchum, & Ray, 2017). The satellite record is too short to be meaningful, we need at least 60 more years of data before we can derive a meaningful rate from satellites. The satellite data only covers the upward part of a ~60-year cycle or oscillation that began in 1991.
Global mean sea level has been rising for the past 170 years, but the rate is cyclical. The cycle shows up prominently in Figure 1 from about 1930 to 1991 in the Jevrejeva et al. reconstruction. It is also seen, albeit in a more subdued fashion, in the Church and White reconstruction. Thus, any estimate of the rate of rise based on a reconstruction (satellite or tide gauge or combined) that is shorter than 90-120 years is erroneous. Sea level fell during the Little Ice Age until about 1861 when it began to rise.
All the rates (since 1900) are projected to 2100 in the upper left box and show sea level rises between 5 and 10 inches by 2100. These increases are much less than the average daily tidal range of over one meter. Such a small rise in 76 years is unlikely to be noticed. It should be noted here that sea level is not the same in all oceans as one might expect. In Panama, sea level is 20 cm (~8 inches) higher on the Pacific side of the country than on the Atlantic/Caribbean side, this is about the global sea level rise expected between now and 2100. Even more important, the tidal range on the Pacific side is much larger than on the Caribbean side.
Measuring global mean sea level is made extraordinarily difficult by the changes in mean sea level from ocean to ocean and the related changes in the daily tidal range. Jevrejeva points out that tide gauges are confined to continental and island margins and most of them are in the Northern Hemisphere, the tide gauge records don’t all cover the same time period, and they are attached to land that sometimes rises and falls itself. Jevrejeva points out that it is not easy to combine the various records into a single global sea level curve.
In a like fashion, any estimate of acceleration in the rate with any of this data, or any combination of it, is little more than a guess. Polynomial fits to all these series can show some acceleration, but the fit to the polynomials is statistically no different than a linear fit to the same data.
As many of my regular readers know, I often throw ideas out there for comment. I want to know what the best arguments against my ideas are. Or did I miss a better argument? So positive or negative, give me your best shot. Am I right, or wrong? Is there a better argument than what I’ve offered?
Works Cited
Beckley, B. D., Callahan, P. S., Hancock, D. W., Mitchum, G. T., & Ray, R. D. (2017). On the “cal-mode” correction to TOPEX satellite altimetry and its effect on the global mean sea level time series. Journal of Geophysical Research: Oceans, 122, 8371–8384. doi:10.1002/2017JC013090
Church, J. A., & White, N. J. (2006). A 20th century acceleration in global sea-level rise. Geophys. Res. Lett., 33. doi:10.1029/2005GL024826
Crok, M., & May, A. (2023). The Frozen Climate Views of the IPCC, An Analysis of AR6. Andy May Petrophysicist LLC.
Jevrejeva, Moore, J., Grinsted, A., Matthews, A., & Spada, G. (2014). Trends and acceleration in global and regional sea levels since 1807. Global and Planetary Change, 113, 11-22. doi:10.1016/j.gloplacha.2013.12.004
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
The satellite data appears to be an artifact, and not comparable to tide gauge records.
Correct. See long comment below referencing previous quest posts here.
Sea level rise projections are not derived from simple linear extrapolations of tide gauge data, nor are they derived from polynomial extrapolations. They are derived from sophisticated models such as process based models or “semi-empirical” models. It may (or may not) be true that we lack sufficient observational data to detect an acceleration in ongoing sea level rise, but that is separate from the question of what we should likely expect to see over the coming years. And we should expect to see near-term acceleration, since sea level lags behind global temperature change by several years.
Projections are not facts. Models do not output data. The best measurements available do not show an acceleration in sea level rise.
Humanity has always adapted to changes in sea level. Adaptation remains the best choice for dealing with climate change, regardless of the cause.
And of course there is no evidence that humans have caused any of the tiny sea level rise over the past century or so.
As AM showed in his post, the steady slow SLR we are seeing is a simple consequence of coming out of the qualitatively indisputable LIA. Last Thames ice fair was 1818.
“I got taller because I grew up” isn’t an auxological model.
Getting taller is at least an observable event.
Future sea level rise = models upon models upon models upon models . . .
The climate modelers would be panicking that their 12 year old child would be 3m tall by the time they are in their mid 20s. Maybe even more, as their growth is “accelerating” during the early stages of puberty.
They would lobby for tax payer funding to increase their ceiling height.
epimodels!
When you have to lie about what others have said, you are admitting that you know you can’t win on the facts.
Nobody said that the seas aren’t rising, what they are denying is that humans are responsible for more than a tiny fraction of it.
…and that it is rising at a catastrophic rate, or even accelerating. Which it is not.
Thinking that the rise of the sea is an emergency- is for people who don’t have any real problems.
What a moronically irrelevant comment !
Then your height levelled off and will soon start decreasing.
“… will soon start decreasing“
I passed that stage about 40 years ago.
What is interesting is that from birth to age 18,
I grew about 54 inches. In the last 60 years,
I have shrunk only 2 inches — a much slower process.
I think the rate of shrinkage is decreasing, but I await
another 10 years of data. 🙂
“I have shrunk only 2 inches”
Height or length ???
I’ll concede that the point is irrelevant if you can convince me that you actually grasp it and demonstrate why it is irrelevant.
Another moronic self-serving piece of gibberish.
You even KNOW it was irrelevant. so dumb !!
Oof, sorry, gotta play to win. Try again.
Poor Little AJ… still an irrelevant self-serving LOSER. !
Can’t do it? Understandable. Better luck next time.
You continue to FAIL to produce anything of any relevance what-so-ever.
You know what you posted is totally irrelevant, and try to “blah-blah” your way around that fact.
You might have groan in the body… but your mind is still stuck at the lowest level possible for human functionality..
Only FAIL is YOU !!
Strawman argument. You made the original assertion, therefore, it is up to you to show WHY criticisms are not based on facts.
It is not at all, I made bnice a fair offer, and he was unable to rise to the challenge. The reason he failed is simply because the point I made was germane and sensible. Claiming that sea level rise is simply explained as a consequence of the end of the LIA is circular reasoning, and does not explain the cause of sea level rise.
“Why did sea level rise at the end of the LIA?”
“Because the climate warmed.”
“Why did the climate warm?”
“Because the LIA ended.”
“Why did the LIA end?”
“Because the climate warmed.”
That isn’t science, although I see that most here find the logic quite compelling.
Actually, what’s ‘compelling’ is that there was a LIA and many other natural cycles that are demonstrably visible in the ice core and paleo records, none of which had anything to do with human activity or CO2.
Yet the Left, including their useful idiots within the climate alarmist community, insist that we trash our economies and surrender our freedoms to serve their nefarious ends.
Yeah true, Since OJ Simpson got cremated, he now fits in his glove …….
What’s this AlanJ dude on about? Is he not keeping up ?
Somehow thinks that human growth models are relevant to sea level rise.
You can’t get any more GORMLESS and STUPID than that !!
I am getting taller, than my hair.
True, they are not facts, and the projections carry uncertainty. But they are not linear extrapolations of linear best fit lines. Andy is conflating the uncertainty in the tide gauge observations with the uncertainty in the projections themselves, which are distinct.
They are mindless conjecture-driven junk… totally meaningless.
That is what you are saying.
Another trendologist who doesn’t understand measurement uncertainty…but can sling the word around.
No, they’re not linear extrapolations, they find polynomials to support the idea of acceleration.
The rate of sea level rise is directly proportional to the amount of energy used to melt ice and warm the ocean and that is a measure of the TOA radiative imbalance. The TOA radiative imbalance needs to accelerate for SLR to accelerate and there isn’t even an expectation that will happen. Alarmists expect it to linearly increase in line with CO2 levels.
Too bad the alarmists are too stupid to understand this shortcoming in their non physical, model based, arguments.
hmmm… seems reasonable to spend hundreds of trillions of dollars to slow the rise rather than add a few inches to your sea wall- after all, masons are hard to come by nowadays /s
sea level is like global temperatures…
…without the constant yapping of the climate scammers.. NO-ONE WOULD NOTICE.
There is no evidence of any past El Nino effect in, say, Sydney sea level data, so why should there be from the now subsiding 2023 El Nino.
The sea level rise along the east coast of NSW over 5 decades would be about 4cm, according to the Bondi or Fort Denison high tide data.
The standard deviation of the high tides at Bondi over that period, is about 12cm.
No-one, not even the old codgers that sit ogling the beach girls, would have noticed any sea level change over that time.
They would have noticed the change in beach wear though 😉
Get back with us when coastal car dealers start selling boats instead of EVs.
Great graph of SLR..
…. but the inclusion of massively tainted and fabricated urban surface data is basically meaningless.
That’s funny.
Models are computerized opinions with about the same value.
Models all the way down.
AlanJ, your presumably sincere comment has two objective easily established problems. Research them for your own edification.
First, Hansen has been predicting accelerating sea level rise since 1988–his most recent post retirement paper said maybe about 0.5 meters by 2050– and it just hasn’t happened. His math works out to significantly more than 10mm/year on average—ridiculous.
Second, the climate model projections are all (save one in CMIP6, INM CM5) provably wrong in several distinct ways. Relying on them for the future is a fools errand.
See also my long comment just posted below. You came to the wrong place to spout your faith based alarmist ‘didn’t happen yet, but gonna’ nonsense, versus folks who have spent many years researching and written copiously footnoted books debunking the global warming alarm.
According to the IPCC AR6: “Global mean sea level (GMSL) rose faster in the 20th century than in any prior century over the last three millennia (high confidence), with a 0.20 [0.15 to 0.25] m rise over the period 1901–2018 (high confidence). GMSL rise has accelerated since the late 1960s, with an average rate of 2.3 [1.6 to 3.1] mm yr –1 over the period 1971–2018 increasing to 3.7 [3.2 to 4.2] mm yr –1 over the period 2006–2018 (high confidence).”
Prove it, then, please and thanks.
They tacked on their tainted and mal-adjusted satellite data onto actual tide measurement.
I’m sure you are totally aware of that.
So you are deliberately posting FAKED information.. again. !
Show us 5 tide data series with that faked acceleration..
I bet you can’t.
Here is a very stable tide site. trend about 1mm/year…
.. which is about 10x less than the standard deviation of the high tide data.
You haven’t run it through a model. Do that, and you’ll get the “correct” result. Because we know what the model says is right.
All of those numbers are less than our ability to measure.
As noted above and in the post, this ignores the 60-year cycle very apparent in the data. See this plot from Frederikse (2020) of the rate of change in sea level of the various components of change, observed is blue, black is the sum, red is barystatic (volume of water) change, orange is thermosteric (warming and expansion of the water).
From:
The causes of sea-level rise since 1900 | Nature
What you are seeing in the data is not a mode of cyclic variability in sea level, rather it is the sum result of the various contributors to sea level over the 20th century. Frederikse et al. deconvolve these various contributions in their paper:
Saying, importantly:
“The acceleration in sea-level rise since the 1970s is caused by the combination of thermal expansion of the ocean and increased ice-mass loss from Greenland.”
It’s important to exercise caution in interpreting what might naively appear to be repeating patterns in time series data as true cyclic oscillations.
More word salad.
The plot I showed has the sum of barystatic and thermosteric as a black line. Observations are in blue, they do not match. In any case, the ~60-year cycle is apparent in both.
I must be going blind, because the curves overlap perfectly except for a brief period in the 1920s, which Frederikse et al. states is possibly due to deficiency in the early part of the reconstruction. In any case, the point is that the apparent cycle you are stating is present in the data is explained by non-cyclic, primarily barystatic processes (dam projects), as shown by Frederikse et al. It isn’t a mode of cyclic variability in sea level, it’s just a quasi-cycle arising by coincidence.
That is your opinion and Frederikse’s, but it is just an opinion. In my opinion, the overlay of the AMO and the sea level trends suggests that the reduction in rate the 1960s and 1970s is due to the reduction in temperature at that time, especially SST.
They are both opinions and we won’t really know until sometime after 2050.
It is the conclusion of their research, yes. Calling that an “opinion” strikes me as a bit glib, but you are free to disagree with it. We are talking about historical rates of change, though, so we aren’t going to have more clarity by 2050 unless some new discovery is made about the historical record that helps us understand differently.
It is an opinion, since alternative opinions, like mine, are equally supported by their research and the research of others. However, we agree that a definitive answer will have to wait until 2050.
Greenland ice area is only slightly the maximum of the last 10,000 years.
Stop being an ignorant twit.
And you still haven’t produced any actual tide data that shows this mythical acceleration.
Except: there is no ice-mass loss from Greenland, let alone an ‘increase’.
Greenland Ice mass since 1900
Someone always has to throw the abstract so called science of modelling. Also known as. “What If” science.
Thus demonstrating the scientific principle known as “making crap up”.
In other words, the data is cooked until it shows the results we are looking for.
Comment says:”… we should expect to see near-term acceleration…”
This says that we haven’t seen it yet. So all the caterwauling about sea level is not demonstrable today. In fact you say that all the headlines we have seen are a lie. Nothing has been harmed as of now by sea level rise.
“ we should expect to see near-term acceleration…”
We also expect the tooth-fairy to visit us every night.
And goldilocks !
It does not say that.
Yes, the bad stuff isn’t projected to happen for a while yet.
So it is a total fairy-tale…. a fantasy…
That is your world AJ… change your meds….
We live in the real one.
Well, that is true. It is always 5-10 years away. And keeps following the same pattern. So, in 10 years time another set of reasons (or modeled datapoints) will be put forward to state that in 10 years time (model projections) bad things will happen. No need to put up the list of failed predictions. We all know what really happened..
Wake me up when the Maldives are under water so i will check again..
Yes, sea level projections are derived from modeled data rather than empirical data. They have no basis in reality other than the virtual reality of modeled space
if sufficient observational data is lacking to detect any sea level rise acceleration how can a 95% confidence level be claimed otherwise, as touted in most academic papers? Are the associated papers then even worth the paper they’re written on?
Extraordinary claims require extraordinary proof.
HAHAHAHAHAHAHAHAHAHAHA
Dork
AJ,
You believe model based projections that don’t fit well with observed SL rise data. This is why you get so many downvotes.
The projections have historically under-predicted the observed rate of SLR (the semi-empirical method was developed because the process-based approach was consistently lower than observations). But I doubt that fits with the narrative you want to paint for yourself.
I also don’t think that’s the reason for the downvotes – the downvotes I receive are in exact direct proportion to how dearly held the misconception I’m challenging is.
Please stop with the irony.
We must learn to not feed the trolls.
AJ – you obviously also believe the satellite measurements are precise enough and accurate enough to determine sea-level height to a fraction of a millimetre. The base resolution is around 3.5cm, and you also need to decide where on the surface waves you say the actual level is. There’s a belief that if you average enough measurements, then you can increase resolution, so average ten times the number of measurements to get an extra decimal point of resolution. In truth it really isn’t that certain or precise, whether you’re measuring temperature or a difference in the mm range at a distance of 180km or more. There are systematic errors involved too.
Though satellite measurements do give information on the surface height of the ocean away from the coast, you’ll get far better precision and accuracy by a measurement from a tide gauge, though even there there are systematic errors that are unavoidable, though the measurements can reasonably be compared over centuries (wave shape and thus weather will affect the damping imposed on the water in the tide gauge).
Net though is that during the time (couple of centuries) we’ve been actually measuring sea-level, it’s been rising slowly. To counter it, we need to add an extra brick to the height of the sea-walls every half-century or so, unless of course your local coast is subsiding, like it is in Miami.
I suggest you look at the predictions of dramatic sea-level rise over the years and see if they actually happened. Did New York get flooded?
Now figure out just how hard it is to get an accurate measurement of sea-level when you have waves several metres high and a tidal range maybe larger. Go put a ruler in the sea near you and see how high the wet gets. You need a really long series of measurements to get an idea of maximum height, and it will vary a lot day to day.
The downvotes are, I think, mainly because you think that these are things we can control and thus should control. As King Canute (now called Cnut) proved a long time back we don’t have control of the tides, and we also don’t control sea-level, global temperatures, or amount of CO2 in the atmosphere. Useful to measure such things, and to try to predict the future, but given the scale of the natural processes we can’t affect things that much.
Curious. How do satellites account for waves?
Sparta Nova 4 – yep, I wondered about that too, but can’t find relevant data. Do we only get reflection from the water surface normal to the radar direction, in which case we’d only see a bit of the trough? In that case, the higher the wave, the lower the level the radar would say it was. In any case you’d have a return over an interval as the reflections from patches of sea further from vertical return. Looks like there may be a lot of corrections needed to be applied to the data to calculate what the actual water surface level is. It certainly isn’t a simple problem.
Some clues on this at https://nl.mathworks.com/help/radar/ref/seareflectivity.html but looks a lot of effort needed to get an answer to the question.
Averaging, regardless of uncertainty. After researching this, I can’t find a NASA site that lists the standard deviation of all these measurements, let alone any kind of uncertainty budget/calculation.
“Did New York get flooded?”
Frequently!
?crop=16:9&width=980&format=pjpg&auto=webp&quality=60
https://s7d2.scene7.com/is/image/TWCNews/nyc_rain_storm_92923_AP23272566230719(1)?wid=1250&hei=703&$wide-bg$
Phil – yep, I should have stated that as “permanently flooded” rather than just occasionally. Still, that was just one of the predicted climate problems that just didn’t happen. There were so many.
And completely ignoring tide Guage data, especially when adjusted for geo-static land movements. (!.4 mm pre year)
Modelling is What If science. Once that’s your argument, you’ve lost the argument.
“sophisticated models”
They may be sophisticated but that doesn’t mean they’re correct.
Note that the word sophisticated is derived from the word sophistry:
Noun: sophistry|só‑fis‑tree|
A deliberately invalid argument displayingingenuity in reasoning in the hope of deceivingsomeone
=sophism, sophistication
People can be sophisticated- things cannot.
Please do not equate sophism (a philosophy from 450 B.C.) with sophistry, correctly defined in your post.
WRT:
“Sea level rise projections are not derived from simple linear extrapolations of tide gauge data”
From AR6, Ch 9, page 1216:
“Global mean sea level (GMSL) rose faster in the 20th century than in any prior century over the last three millennia (high confidence), with a 0.20 [0.15–0.25] m rise over the period 1901 to 2018 (high confidence). GMSL rise has accelerated since the late 1960s, with an average rate of 2.3 [1.6–3.1] mm yr-1 over the period 1971–2018 increasing to 3.7 [3.2–4.2] mm yr-1 over the period 2006–2018 (high confidence).”
Sea level has a clear ~60-year cycle, just like our climate and the AMO. So AR6 compares a “simple linear” fit to 117 years of data, a 47-year subset and a 12-year subset and proclaims acceleration. I don’t think so.
More here:
https://andymaypetrophysicist.com/2022/03/19/ar6-and-sea-level-rise-part-1/
You misunderstand, I am referring to your own projections of future sea level rise, not merely an assessment of the rise that has occurred, which can suitably be represented with a linear function. Your projections simply take the historic trend and extrapolate it into the future, and, you say, don’t look concerning. But that is not how future sea level rise should be projected.
You are saying I should make up some elaborate model that has no basis, then I would be no better than the IPCC or Frederikse. Instead, I found that a linear fit over 124 years of data appears to be significant. Then I checked for a polynomial fit to the same data and the correlation coefficient barely changed from the linear fit. Hmmm … maybe the linear fit is correct? If so, what does that mean?
So, as I say in the post, prove me wrong. I’ve seen nothing better in Frederikse, Jevrejeva, or in AR6.
I’m saying that finding that a linear model describes the series adequately does not mean you can assume that the linear model can be used to extrapolate future changes. How could it? It is not based on any knowledge, theoretical or otherwise, of any future changes in the processes driving SLR. This is curve-fitting 101.
You don’t have to adopt the models widely used by the scientists who are experts in this subject area, but you need to propose some model based on the underlying processes. Ideally you’ll publish that model in a reputable journal, so that it can be scrutinized by experts.
First you say I cannot use a model to extrapolate into the future, then you say I must propose a model and publish it, which is it?
Whether the model is a LSR (linear least squares regression) or based on assumptions about natural processes doesn’t matter, an extrapolation using either model is risky. You are making no sense.
In this case the test for linearity from 1900 to 2000 is good at the 5% level. Simple models that meet basic statistical tests are superior to complex models based on numerous complex assumptions IMHO.
The famous George Box once said:
That is the best criticism of CMIP6 and Frederikse I know of. Sometimes complex models are created to baffle the reader with BS.
I am saying that you cannot assume that a simple curve-fitting exercise will yield predictive power. It has little to do with the simplicity or complexity of the model, it’s just the basic nature of curve fitting. You can devise a very simple model of SLR by doing some basic estimates of thermal expansion and ice sheet runoff under various warming/cooling scenarios, and these would at least be based on the underlying processes. Claiming that all models carry uncertainty and therefore all models are equally valid is simply fallacious reasoning.
Curve fitting is best used to describe the characteristics of the existing series, not to make predictions about how it will behave in the future.
Surely you can see this just shifts the assumption and is no better than a curve fit.
But in your case you’re so bought into the idea you don’t understand the energy implications of that acceleration and that they’re not expected or valid.
I swear most of the folks relying on any kind of “modeling” just don’t understand that multivariable curve fitting is DANGEROUS when making projections. They have not studied Quality Control or business and are ignorant about making predictions of the future.
Quality control monitoring of output in a multistep manufacturing process can only tell you what has occurred. It won’t tell you what step in the process is failing nor tell you what will occur in the future. Only detailed knowledge of each step will provide a solution.
Likewise in business. Only detailed knowledge of each piece of revenue will allow one to decipher what product is causing a change in revenue. Making predictions for next year’s revenue based on a simple extrapolation of what HAS ALREADY OCCURED in the total is not recommended.
I’m happy to see us in agreement on something. A rare and welcome occurrence.
KISS is the best method in all design.
Most natural phenomena are cyclical, aka sine wave functions.
Did you get a gaussian error distribution model to data?
That does not mean that every time you think you see a cyclic pattern in a natural dataset that you have discovered a new natural phenomenon.
I did not say that.
Nor does it rule it out. Give the reasons why you declare a cyclic pattern is not a natural phenomenon. Sea level is certainly cyclical at its most fundamental basis. Why would you not start theere?
Frederikse show that the apparent cycle can be explained via barystatic processes, consisting of an increase due to freshwater runoff (glacier and ice sheet melting) in the early part of the century, and a decrease due to global dam projects in the middle of the century:
So, no, nothing can ever be 100% ruled out in science, but to posit the presence of some natural cyclic process, you have to
a. show that the estimates made by Frederikse et al. are completely wrong (or you have to show an exact counterbalancing change in sea level that undoes the contributions from the processes they describe).
b. Provide a physical model explaining the mechanism by which your proposed natural cycle is driving the change in sea level (it isn’t enough to say, “I think it might be SSTs” – that is barely a hypothesis).
Alàn, we would see sea level acceleration if official temperatures were actually correct and not jiggered to fit an alarmist meme. Models used to to forecast the temperature anomaly in 25 years from 1990 proved to be 300% too hot on average compared to two independent empirical measurements. The forecast averaged a bundle of ~30 models. Ths means that half the models ranged from 300% to ~ 500-600% too hot.
Modeler Gavin Schmidt, GISS (NASA), remarked: “Models are running a way too hot and we don’t know why.” This was just before the IPCC AR6 report release and GISS wanted the AR6 delayed to discuss the problem. Here is a recent article that discusses this “Too hot” problem:
https://www.nature.com/articles/d41586-022-01192-2
A geological perspective on global sea level change…
Perfect!
Thanks
When a geologist takes a picture with people in it, they’re only for scale… 🙂
My wife has always known she is my 1.5m scale bar, ever since student days. Add in a geopick and a coin, what more do you need for geological photography.
I dunno why, but I was expecting to see Tom Cruise in there.
Cruise is to the right of Peter Dinklage 🤣
I forget which state, but I saw a story recently where someone found a 6.5 inch Megalodon tooth several miles inland.
Megalodons must have been VERY shallow water feeders.
Bricen Columbia, 19, found a 6.5-inch megalodon tooth in South Carolina while fossil hunting with his family and Palmetto Fossil Excursions. On June 10, the tour group posted a photo of the tooth to Facebook.
The Lightning Site – Palmetto Fossil Excursions (palmetto-fossil-excursions.com)
Maybe someone carried it there.
Maybe a megalodon a long, long time ago?
The oceans have, in paleo geological times been 200 meters higher and 200 meters lower than today.
The ocean once went up as far as Montana and Florida was less than a sand bar.
So things “changed” even before Man was around?
(Must have been the dino CO2 and methane emissions.)
That’s better… not ruined by the HadCRUD non-data.
Despite its flaws. HadCRUT is the least fraudulent of the surface temperature reconstructions. Even with its flaws, it doesn’t support the notion that an SLR acceleration should be expected. Modern SLR preceeded the 20th century temperature rises.
Still based on massively tainted unfit-for-purpose urban surface sites. !
Any time you’re showing a “global temperature”, it’s fraudulent.
made me go out in to the garage and find my Estwing B3 “mini-sledge”. hit a few rocks in the Wyoming overthrust belt.
Thank you.
Got one of those too!
As I understand the situation, northern latitudes that have experienced glaciation are experiencing isostatic rebound and, consequently, a lowering of sea level with respect to the rising land. The shallowing of the basin is reducing the ocean basin volume and the water is being displaced by land. That water has to go somewhere. So, the rise in sea level isn’t just from melting glaciers and thermosteric expansion, but is also the result of displacement of water by the rising land. I haven’t been following this closely. Can someone tell me if this displaced volume of water has been adequately addressed, in a quantitative manner, as a contribution to rising sea levels?
Great question! Both Jevrejeva and Church and White try to take VLM (vertical land movement) into account. With modern tide gauges this is not a problem because they have GPS’s and can measure the land movement, but in 1930? No, that data requires a lot of assumptions about VLM.
In reality, we know very little about sea level change over a long enough period (90 years or more) to make a difference.
Andy:
Nice post.
What is the uncertainty in a GPS satellites’ position?
I recall an article about a California earthquake where they measured the land’s movement
using GPS but there was a ~2-4cm uncertainty.
NASA’s satellites measuring SLR have similar uncertainties. So how do they justfy quoting
sea levels to a tenth of a millimeter?
Measuring sea level with a 30ft ocean swell ..yeah, right !!
Satellite altimeters can make a single point-measurement with an accuracy of about 1cm. Averaged over the several-day period it takes a satellite to cover the ocean, satellites can provide a global mean sea level estimate with an accuracy better than a couple millimeters. For annual averages this is reduced to fractions of an mm.
More pretend maths and total BS from the AGW moron !
Bingo.
Even the Grate Josh Willis says only 1″ accuracy…. which is also a load of anti-science nonsense.
To justify dividing by the square root of the number of samples, to improve the precision, a data set must have the property of stationarity (https://blog.quantinsti.com/stationarity/ ). Few natural time-series meet the requirement of stationarity, having a constant mean and variance. Also, implicit in stationarity is that the same thing must be measured multiple times by the same device. Just as one cannot step into the same river twice, sea level varies with wind direction and speed, atmospheric pressure, temperature, melting rates, evaporation rates, precipitation rates, and changes in the shape and volume of the ocean basins, some of which can change in a matter of seconds, such as when there is a major earthquake. Therefore, strictly speaking, one is not measuring the same sea level multiple times. Rather, one is measuring many different sea levels one time. Consequently, one has to propagate the uncertainty in quadrature — meaning summing the squares of all the uncertainties and taking the square root. It is a fundamental problem that most climatologists pretend doesn’t exist.
The approach of dividing by the square root of the number of observations has long been used by surveyors when turning the exact same angle, which doesn’t change with time, or determining the diameter of something like a ball bearing, in a climate-controlled environment. Those are sequential readings, thus a form of a time series. However, any changes in the mean or variance can be expected to be random and have predictable probabilities. That is very different from natural time-series with trends, seasonal behavior, or long-term cyclical behavior, such as tides and waves.
Well-stated, Clyde.
Time-varying effects such as ocean tides, atmospheric pressure, glacial isostatic adjustment, etc. are subtracted from the geocentric sea surface height prior to averaging, so this comment does not apply, but you are correct that there are subtleties to consider. The mistake is in assuming that scientists have failed to consider them.
And you continue to demonstrate your abject ignorance.
But nice word salad…
Estimates and averages and bingo, 100 decimal places.
We’re off to see the Wizard…..
Bull puckey! If you begin not knowing the height any closer than 4-5 cm, you simply can’t adjust that away. Every adjustment will contain that uncertainty regardless.
If I measure 10 ±5, The interval is {5, 15}.
I can subtract 5 but the uncertainty remains. The interval is {0, 10}
I can subtract 5.12345 but the uncertainty remains. The interval is {-0.12345, 9.88655}
This is not a question of reducing the uncertainty in individual measurements, but about the stationarity of the series used to calculate the GMSL for each full ground track, and how that relates to the uncertainty in the estimate of the global mean.
One can de-trend a time-series to obtain a quasi-stationary data set, to have a better idea of just what the uncertainty is for the individual measurements. However, the trend itself has an uncertainty, which is typically greatest at the beginning and end of the time-series. Thus, de-trending adds a variable uncertainty to the time-series. You can’t make it magically disappear.
Another word salad. Stationarity has nothing to do with uncertainty. If the measurements are uncertain, you cannot remove that uncertainty by data processing. The uncertainty must be propagated throughout the calculations.
The claim Clyde is making is that the measurements taken across a full satellite track over the ten or so days required to cover the ocean produce a nonstationary time series. Because the mean and variance of a nonstationary series can change over time, Clyde argues that the standard error of the mean is an inappropriate characterization of the uncertainty in the estimate of the global mean, and thus the reported uncertainty of the global mean sea level to a couple of mm must be too low.
I have simply pointed out that nonstationarity is removed prior to averaging via differencing.
You misunderstand. De-trending the time-series provides insight on the variance of the measurement process. However, to deal with the rise in sea level, you can’t throw the trend away, and it has its own associated uncertainty.
I do not misunderstand, but I think we are talking about different things. I am talking about the individual estimates of global mean sea level produced by averaging gridded means of the full global satellite track every ten or so days. These hundreds of thousands of measurements do not exhibit long term trends, but are affected by processes that make the series nonstationary. These time-varying processes are removed prior to averaging.
For the annual mean, scientists are careful to account for seasonality or other time-varying effects like ENSO.
One would be unlikely to know because those in climatology and associated fields so rarely explicitly state the confidence interval and discuss those “subtleties.” Most disciplines use a default 2-sigma uncertainty envelope, providing a 95% confidence. On the rare occasion that a climatologist does state a confidence interval, it is usually only 1-sigma, albeit even more rarely explicitly stated, giving the impression that the measurement is more precise than it actually is when read by those in other disciplines.
Please explain just how those adjustments are made when even tide tables are not correct and the effects are generally semi-quantitative at best. What do you suppose the error bars are for isostatic adjustment? Do you suppose it might vary laterally with the type of rock or over time? How does one correct for wind piling up water when the direction and speed of the wind varies continuously, and moving air masses with different pressures can either add or subtract (or both) to the water height?
You need to turn to the relevant literature. There is a vast body of work answering these very types of questions that you seem to be entirely unaware of, yet you are here accusing scientists of not doing the work. It is rather crass. Let’s call it “the conceit of the armchair expert.”
No, let us call it the arrogance and ignorance of experts, when more accurate long term data shows it is incorrect. You are not an expert either, however you appear to have clear bias.
Entirely unaware? I’m no stranger to processing data and cleaning it. You made an assertion that seems unreasonable to me. I’m asking that you explain the basis of your claimed corrections or adjustments. I’m challenging you to support your claim, not the anonymous authors that you assure everyone have dealt with the topic. Not so much as a “Tweedle, et al., (1942)” from you, just an assertion that someone has it under control. You attempt to make it look like I have no knowledge of the issues. Yet, the fact that I asked questions you can’t answer readily suggests otherwise.
The real question of the day is, “Who is the armchair expert? You or me?” I have had 15 articles published here, complete with original graphs and citations for my claims. How many have you had published? Suffice it to say that I asked you some direct questions and rather than answer them, you insult me. That is what I call deflection through ad hominem.
Here are some references to get you started:
Nerem et al., 2010, Satellite Altimetry-Based Sea Level at Global and Regional Scales
Nerem, et al., 2010, Estimating Mean Sea Level Change from theTOPEX and Jason Altimeter Missions
Mitchum, 2010, An Improved Calibration of Satellite Altimetric Heights Using Tide Gauge Sea Levels with Adjustment for Land Motion
Beckley, et al., 2010, Assessment of the Jason-2 Extension to the TOPEX/Poseidon, Jason-1 Sea-Surface Height Time Series for Global Mean Sea Level Monitoring
Dieng, et al., 2017, New estimate of the current rate of sea level rise from a sea level budget approach
Labroue, et al., 2004, Nonparametric Estimates of the Sea State Bias for the Jason-1 Radar Altimeter
Chambers, et al., 2003, New TOPEX sea state bias models and their effect on global mean sea level
Desai & Haines, 2004, Monitoring Measurements from the Jason-1 Microwave Radiometer and Independent Validation with GPS
Brown et al., 2003, Jason Microwave Radiometer Performance and On-Orbit Calibration
And this is barely scratching the surface of this subfield of a subfield. The issue isn’t that you are asking questions, the issue is that you are using your own ignorance as a basis for rejecting expert input on this issue from the scientists who understand it far better than you.
It is certainly not I, since I’ve done nothing but defer to the scientists who actually know what they are talking about.
I suspect that you just did an online search for applicable research, without actually reading, and understanding what was said. For example, in your first link (Nerem, et al. 2010) they say:
“Yet uncertainties of components of the sea level budget equation (including sea level) are still large, in the order of 1 mm/year (2-sigma) (Church et al. 2013). The challenge is thus to reduce the components’ errors, in order to check the statistical significance of the difference between observed sea level and sum of
contributions. The satellite altimetry-based sea level record is affected by errors due to the imperfect altimeter corrections applied to the data (with the orbit solution and the wet tropospheric correction displaying the largest uncertainties), geographical averaging process and imperfect linkage between successive altimetry missions. In terms of long-term (decadal) trends, such factors contribute to the 0.4 mm/year difference quoted above (Ablain et al. 2009, 2015). At the interannual time scale, errors in the GMSL record are also significant and amount to 2–4 mm”
And:
“Similarly at the interannual time scale, the global mean sea level still displays 2–4 mm errors that are not yet fully understood.
As I understand it, those are closure errors in the sea level budget equation, which is a different issue from the uncertainty interval.
They don’t address most of the points I raised, and they most specifically don’t discuss the issues involved in whether they are justified in processing non-stationary data as though it were stationary.
I’ve provided a sample of available research on the topic of compiling global mean sea level estimates via satellite altimetry, with a historic context. These papers include discussions of closure and bias corrections.
They are not doing this, as I’ve demonstrated in earlier comments, and which you can easily see by examining the treatment of time-varying effects in the papers I’ve provided.
The mistake is assuming the WAGs are close to correct. (It is irrelevant that the math works) We have extensive and long historic tide Guage history that shows the assumptions are simply wrong.
Seems incredible that a sensor with a 1 cm accuracy can gain accuracy to a couple of mm by taking hundreds of measurements and fractions of a mm by thousands.
The accuracy of a sensor does not change. Statistical RSS can give you a statistically valid range, but the individual accuracies do not improve.
Averaging does not improve accuracy.
Allow me to add that RSS involved independence and nothing about sea level measurements is independent.
It improves the precision in the estimate of the global mean. That is all that is being said.
Averaging can only improve the precision of measurements of the SAME thing, which has a fixed value, by removing random variations of the measurement system. When the thing being measured does not have a fixed, unchanging value (stationarity), the precision is likely to degrade because the variance grows over time. If one defines the ‘average’ being calculated as the mean weight of all the particles in the universe, which is effectively infinite, then what is obtained is a mean between zero and some very large number and a standard deviation that is so large as to make the mean of no practical value. I think that I would be on safe ground to say that for averages of non-stationary measurements, one should expect that the standard deviation will be proportional to the length of time of the time-series. That is, the longer measurements are taken, the larger the variance will be. All those “subtleties” have to be accounted for and defined very carefully for the statement of what the average is to have any meaning.
The length of time over which the measurements are taken is about 10 days, which is how long it takes a satellite to cover the global ocean. The time varying effects that might make the series non-stationary over those ten days are removed, as stated above.
The standard deviation of the mean is only applicable when you are measuring the SAME THING, multiple times under repeatable conditions. Clyde has tried to educate you but you just keep quoting standard statistical sampling theory. These measurements are NOT samples of a non-varying population. The nearest thing would be sampling a population where the populations keeps changing while you are sampling. You can keep on sampling and proceed to calculating a standard deviation of the mean. It will be a worthless indicator of how accurately you know the mean.
Taking measurements, even independent measurements, of different things is just like what i just described. But, they are different things. each with their own distribution of uncertainty. They ARE NOT the same thing that is measured multiple times.
Read this instruction from Eurachem:
Microsoft Word – SQ root n_29a.docx (eurachem.org)
uₓ̅ = s / √n
Read that last sentence carefully. “stable test sample”, i.e., the same thing. “same measurement conditions”, i.e., repeatability conditions. You are never going to convince anyone here that is familiar with making measurements that you can take measurements from all over the earth over ten days and claim that the measurements are done under repeatable conditions.
If you want to make a good faith argument, then show sections of the documents you listed that show both the standard deviation and the standard deviation of the mean.
All of the papers I looked at were paywalled and I am unwilling to pay those amounts for information that you should provide.
The concerns you raise about the requirements for ‘sameness’ and ‘repeatability’ in measurements are indeed important considerations in any scientific analysis. However, they do not preclude the use of standard statistical methodologies for analyzing satellite altimeter data for global sea level measurements. As has been discussed extensively, time-varying effects such as atmospheric conditions, sea state biases, and tidal influences are systematically removed through differencing. This process effectively stabilizes the test sample and mitigates the issues of nonstationarity in the data series.
Moreover, the satellite altimetry method involves a high degree of repeatability due to the satellite’s precise orbital path and the consistent application of measurement techniques across all global points along this path, all under similarly corrected conditions.
Obviously environmental measurements do not have the same level of control as laboratory experiments, but this hardly invalidates them. Your stance implies that it is impossible for scientists to measure anything outside the confines of a lab, such as stream stage, or the characteristics of wild animal populations, or the motion of the continents. In reality we have robust measurement and analytical techniques that yield reliable and useful data from dynamic environments. You claim that the lack of control on the measurement environment means the mm-level uncertainty of GMSL measurements is impossible, when in reality, a more controlled environment would simply yield measurements with sub-mm uncertainty (i.e. even better than what we have now). Thankfully, mm-level certainty is adequate for the needs of monitoring long term global sea level rise.
Yet this is Mantra Number One in climate science, the air temperature jockeys repeat it ad nauseum.
Maybe you should read this page.
Technology | Missions – Ocean Surface Topography from Space (nasa.gov)
4 – 5 cm is the basic accuracy. What you don’t know unless you dig further is that height can vary by feet, say 2m ± (4-5) cm. The satellite moves up and down a large distance due to gravity and topographical variations of the earth.
There are numerous other measurements taken to ADJUST these variations that NASA would like you to believe can reduce the uncertainty to 1 cm. The problem with this is the fundamental uncertainty of 4-5 cm. That doesn’t go away although in typical NASA/NOAA fashion it gets tossed in the waste bin just like with temperature measurement uncertainty.
The article goes on to say that the altitude corrections result in a ±3 cm for individual altitude measurements.
If they start with an uncertainty of ±4-5 cm, there is no way to reduce that uncertainty. I don’t care how many adjustments are made.
Everyone should read the page you linked, it’s excellent information. I’m honestly surprised to see you endorsing an official NASA webpage, that’s nice to see.
An improvement to the accuracy after systematic artifacts are removed is precisely what we should expect. They’re essentially calibrating the measurement against satellite tracking data, atmospheric conditions, and sea state bias.
Nonsense. Calibration requires comparison against a known and published standard that is more accurate than the item being compared.
You can’t do that with a satellite. So whatever statistical method used that you are calling “calibration”, isn’t. It is just that simple.
That ±4-5 cm is a fixed, maximum precision, before systematic errors are removed. Prior to their removal, the precision was worse.
It’s not a precision, it’s the measurement accuracy.
NO.. You are proving you are a mathematical idiot.. yet again.
You previously said above, “It improves the precision in the estimate of the global mean. That is all that is being said.” Which is it? Or does it depend on the day of the week?
How can you assert a precision of ±0.1 mm if the measurement can have a 95% probability of ranging between ±50 mm? The true value of any measurement can have a range of 100 mm. Claiming to know the value to ±0.1 mm is a contradiction.
Instrument calibration improves measurement accuracy, taking a large sample improves the precision of the estimate of the mean by reducing random error. Is this distinction clearer for you?
Because we are talking about the uncertainty of the mean versus the accuracy of individual readings.
That is the crux of the problem. One is only justified in doing that for the exact same thing measured multiple times, with all other variables held constant except those subject to random variations in the measuring system. The article by Nerem (2016) was principally addressing corrections for systematic biases to correct the accuracy, not the precision.
Assuming a perfect measurement system, with perfect accuracy — to as many significant figures as desired — a varying measurand is going to have a mean value with a probability distribution that reflects the variance of the the thing being measured. If the surface of the water regularly varies by about 2 meters daily, with an occasional large wave adding an additional couple of meters to the range, the Empirical Rule suggests that the standard deviation will be about 1/4 of the range, or about 1 meter. That means the mean water level has about a 95% probability of being between ±2 m of the calculated mean. You want us to believe that taking a lot more measurements during the day will provide a meaningful improvement to the mean precision? Over a 10-day period, there is enough time for a couple of pressure domes to pass over the area, changing the sea level, and increasing the standard deviation. Yet, you claim that we will know the mean sea level more precisely with a larger standard deviation.
In summary, when measuring something with a fixed value, multiple measurements reduce random error associated with the measuring system. Measuring something with a varying value provides one with information about the variance of the measurand, but no improvement in the precision above what the measuring system provides.
Far more accurate tide guage adjusted to geo-static land level is far more accurate. You above statement incredibly simplifies ocean changes, 18 year lunar cycles, disparate storm patternes, etc…
Somebody will complain about this, but the idea is that if you record a number of GPS elevations a day and then average them and create a trend, the trend will be accurate. I suppose that is true as long as the GPS doesn’t move or change.
Having some conventional elevations measured as a check every once in a while will help.
“long as the GPS doesn’t move or change”
And as long as the thing you are measuring doesn’t change…
… you know.. like by wave action.
The mathematical idiocy of thinking you can take measurements of random points on random wave action and create something even remotely “sensible” as a sea level measurement ….is beyond a JOKE!
Besides, isn’t NASA’s claim that individual radar measurements have a resolution of about 5 cm and an accuracy of +/- somewhere around 4 cm?
You’re as likely to get a crest as to get a trough, so indeed they do balance out over a few hundred thousand readings across the global ocean.
Again , total BS and mathematical absurdity. !
You have ZERO clue where you have measured, or what has happened between each measurement.
You even admit that you are measuring at different places in the ocean, so all the “large sample” maths cannot be applied.
Even the urban land temperature sites don’t move from minute to minute !!
The sample is all the places you have measured, the “large sample math” applies perfectly. In general when estimating global mean sea level from satellite altimetry (apart from, e.g., sea state bias), the concern about whether the satellite is measuring a wave crest or trough is generally minimized due to the law of large numbers and the averaging processes involved. The metric of interest is the global mean trend over time, not the instantaneous reading at any one location.
Again, you have shown your complete lack of basic mathematical understanding
The law of large numbers DOES NOT APPLY when making measurements on a changing object, especially when it from a moving object and not even in the same place.
You really are displaying your incredible ignorance this time.. !
Read Clyde’s comment above
https://wattsupwiththat.com/2024/06/24/what-about-sea-level/#comment-3931228
..and try to learn something beyond the basic junior high school mathematical understanding you are stuck at.
GPS and other satellite orbits are not perfect and stable. That is a source of error.
Which GPS makes the measurement varies. Each has its own unique error budget. That is a source of error.
Unless the errors are properly included, the answer is bogus. No better than sticking a yard sick in a creek and making a swag.
Only if identically sized crests and troughs.
And identical periods (aka wavelengths).
Modern tide gauges are firmly locked to the sea floor. The only vertical movement of the GPS is due to changes in the bedrock under the sea floor, which is what they are there to measure. A separate instrument measures the sea level, not the GPS.
The conventional wisdom quoted numerous times is that 3 to 5 years of GPS data is necessary for any meaningful insight.
And that is on a basically stationary point, to determine any movement.
You can’t average different things, and then divide by the √n to reduce the uncertainty. In metrology, that is only allowed when you measure the SAME THING, MULTIPLE times, with the same device, and the same operator over a short period of time, i.e., repeatability uncertainty When you do this, AND have a Gaussian distribution of measurements, you can assume the there are random errors in measurement that cancel. This is only due to the Gaussian distribution surrounding the mean value.
when measuring non-repeatable measurands, this does not apply. Doing ten pipettes into a beaker provides ten non-repeatable experiments. Timing a 100 yard dash is a non-repeatable experiment. These are called reproducibility uncertainty cases and temperature is non-repeatable, especially between stations or between days. The standard deviation is used in these cases.
Dr. Taylor in his book in Chapter 5.7 explains this. You need to do multiple experiments, each with multiple measurements. The experiments must have the same mean “μ” and the same σ in order to justify using the standard deviation of the mean.
I remember when satellites came up with their number for the “Palmdale Bulge” ( a desert area in California that is buckled due to the San Andreas fault.) Those measurements, verses ground surveys, came up with very very large differences, and nobody claimed mm percision. Late the satellite numbers themselves were all over the field.
It seems to me that there are other factors that probably have some impact on sea level changes, such as tectonic plate movements, sea floor spreading, river silt discharges, and large rift valley developments.
One of the things about VLM is that there are places that are moving up, and there are other places that are moving down.
I’m not saying that the two actions cancel, but rather that these opposing actions mean that any change resulting from VLM is going to be less it might appear if you just looked at the upward or downward motions.
It can be quite meaningful for any given location, assuming the measurements have any useful accuracy.
Have we asked HOW we measure the current sea level?
In an ideal measurement, the water level would be measured when there are no waves. Does that exclude some measurement days or is all data included.
Is the water level, like the temperature average, just the average between the peak of the day and the lowest point of the day or is it a an average based on the instantaneous measurements, say every 10 seconds, all day, divided by the total sample size? Is the data collected culled if several readings are missed, say at low tide or is the average, (now weighted higher by omission of a few low scores), still kept? How many points can be missed and the data still used?
If using a pressure gauge for the depth of water, is the gauge calibrated once or often? Is the data either side of the calibration checked for compliance and the how can you remove any data from before the calibration if a step change was noted?
Are the measurements tied to a fixed height datum that is relative to a static elevation? How accurate is that? Tides obviously affect water, ice ages lead to rebound of land, sedimentation, volcanism, etc, etc affect mass, are all of these points corrected and to what accuracy?
I think we need a full answer of the above. And in particular, if the data now uses continuous depth measurements throughout the day, how do you link that to the historical data where surely only the low and high points were recorded, (at best), daily, more likely monthly.
Great points. Some of the problems are solved for now, but the data prior to ~1980 is crap.
Don’t forget that tides vary over long periods of time, not just daily, that needs to be accounted for as well.
Tides have not just an annual neap/king tide cycle, they have a ‘lunar nodular’ 18.6 year orbital eccentricity cycle. Which is why you need about 60 years (~18.6 * 3) to say anything meaningful at all about SLR.
‘lunar nodular’ 18.6 year orbital eccentricity cycle.
Presented and discussed on E. M. Smith – Chiefio’s site about 7 or 8 years ago.
Wikipedia has a page titled: Lunar standstill
“but the data prior to ~1980 is crap.”
Fort Denison still uses the same basic system, in the same place, as it did in 1910.
VLM (vertical land movement) at Fort Denison is measured according to Phil Watson:
https://meridian.allenpress.com/jcr/article/36/5/915/441297/Updated-Mean-Sea-Level-Analysis-Australia
Sea level rise at Fort Denison, corrected for VLM is 1.9 mm/yr. Attached is a conceptual diagram of the Fort Denison set up today. The instruments have changed since inception.
Another more detailed article on Fort Denison contains the following:
“Leuliette et al (2004) found that the errors from the vertical movement of the land at the tide gauge locations are apparently larger than the altimeter drift in estimating global sea level change from the T/P altimetry data.”
https://search.informit.org/doi/abs/10.3316/INFORMIT.863348332016788
Don’t forget there are solar tides in addition to lunar tides. In fact, all of the celestial bodies in the solar system have some level of gravitational effect, although most are sufficiently trivial they can be zeroed out without incurring measurable error.
The few mechanical tide measures I’ve looked at consisted of a pipe that is closed on the bottom and has small holes in it. Because of this when the peak of a swell passes by, only a small amount of water enters the pipe. When a trough passes by, only a small amount of water leaves the pipe.
You can think of this like a mechanical low-pass filter. Combine this with the mass of the sensor itself, which further dampens any rapid movements of the water.
Bondi Baths stilling basin is a good example.
Has a pipe that runs a fair way out to the sea bed, so is basically unaffected by wave action at shore (which can be quite large at times.)
The data is a very close match to the Fort Denison data I posted further up.
For those that don’t know: Fort Denison is inside Sydney harbour, and Bondi is an ocean beach about 7km away as the crow flies.
Neither show any acceleration at all.
Look of some of Kip Hansen’s excellent articles on the instruments and mechanics, published here at WUWT.
Ohhhhhhhh mommmaaaaa !!!! Santa Barbara is totally FLOODED!!! I’ve seen the map !!!!
https://www.independent.com/2019/01/17/santa-barbara-struggles-adapt-sea-level-rise/
Oh, good grief! The nut cases are out in force.
All the homeless people will drown!
Who wouldn’t want to be homeless in Santa Barbara!? Cruising State Street panhandling and getting free handouts. It’s ALWAYS a balmy 82deg.F … the beaches … the hot coeds … if it’s good enough for Al Gore and Prince Harry … ?
You DO know I was mocking the city Mothers of Santa Barbara … ?
“… models show how far inland the ocean would wash during annual storms if the overall sea level rose by 6.6 feet.
“… based on the best available science measuring how quickly carbon in the atmosphere causes Arctic ice sheets to melt”
This sort of schist is enough to ruin the taste of my wine. Horrors!
Arctic Sea ice has been stable for 15+ years..
… and is still in the top 5% or so of the last 10,000 years.
Regarding acceleration, below is a distribution histogram of acceleration for 67 tide gauges with 100 years of reasonable data, it shows a very tight distribution around 0.01mm/yr².
Great histogram. I’ve seen that before, very helpful.
Won’t be the last time.
I love it when the scale of graphs are blown-up to GIGANTIC proportions such that a 0.01mm/yr. looks really scary … ohhhhhhhhh mommmaaaaaa!!!
Willis demonstrated in an article published here that with various locations of the world’s tidal gauge records selected by different criteria, unrelated to amount of variation, but all having a fairly long period of continuous measurements with very little missing data, show that acceleration, deceleration, or no change in rate can be demonstrated. Using all good data shows no acceleration. Of course all the claims are based on statistical nonsense where the uncertainty is many, many times the claimed results.
There are two reasons the NASA satellite record should be rejected in addition to the record being too short as noted here by AM.
First, NASA very incorrectly claims it to be sufficiently precise. See old post here ‘Jason 3 fit for purpose?’ for details. And the newest bird, Sentinal 6, is no better. See two then current after launch Sentinel 6 explanatory posts. Intrinsic accuracy in both birds is worse than 3cm for several distinct and inescapable reasons. (Earth geoid is not uniform, low orbital altitude degradation, varying atmospheric water vapor altimeter signal impedance)—when SLR is known measured in low single mm/yr. NASA satalt SLR is unavoidably off by over 30x. Yet they keep pretending it isn’t.
Second, the NASA SLR estimates do not close by a factor of ~2x. See my old post ‘SLR, Acceleration, and Closure’ for details. (Closure is the first order math that SLR must approximately sum to thermosteric rise plus ice sheet loss, all else being de minimus on multidecadal time scales (where ground water depletion, riverine erosion deposition, and ocean basin depth shifts from plate tectonics do NOT count). In fact, on the NASA SLR website, they have the whole closure thing completely bollixed given their own estimates of that stuff given there. Whoever created the site did not understand closure.
Separate minor critique of Jerejeva and Church/White. They both use large datasets of tide gauges, assuming vertical land motion cancels. That assumption is suspect because of the NH geographic concentration of long record tide gauges given the last mainly NH glaciation.
My own preferred estimate is the most recent from Nils Axel Moerner, who used only long record tide gauges (>60 years) corrected by reasonably proximate dGPS for vertical land motion. Easily replicated. His estimate was 2.2mm/year with NO acceleration. Makes intuitive sense, since the NH land rebound from last glaciation would artificially lower the historically NH weighted SLR. His estimate also closes precisely for about the last two decades with ARGO for thermosteric rise plus NASA ice sheet loss estimates for Greenland and Antarctica.
All above comments are just data driven quibbles. I agree with AM there is no SLR issue, let alone an SLR acceleration issue. Comment just shows how the deeper you dig into the ‘science’, the more there is no SLR ‘science’ issue at all.
The level of precision (one inch) claimed by NASA is so obviously wrong I didn’t want to put it in the post.
It is absurd on its face:
New Climate Satellite Measures Sea Level Rise : NPR
I like this sentence…
Yes Josh, we are well aware you keep “adjusting” everything as part of the AGW scam. !
I have not seen it stated so succinctly as you just have, though..
Agree. The bigger, more important question is, if absurd on its face, why do they keep repeating it when so easily disproven by their own stuff?
Of course, there is an easy answer to that question. Money+careers.
See where AlanJ claimed an “accuracy better than a couple millimeters.” above…
…. roflmao
The fool is totally delusional !!
And of course still totally avoiding the fact that there is no evidence of nay human causation.
The “adjustments” to satellite data started around 2002 when they “created™” a trend of about 2.5 times the Topex/Poseiden data when they changed to Jason.
Red is the original T/P data graphically overlaid on the 2003 Jason data.
For a single measurement, yes. We are looking at annually average global mean sea level estimates, which are more precise.
The NASA page you cite in your older post doesn’t seem to be working, but NASA states the sea level budget can be closed quite precisely:
“Our results demonstrate that the global mean sea level can be closed to within 0.3 mm yr−1 (1σ). “
There is a great problem with credulity on WUWT. Someone proposes a hypothetical they think sounds compelling, decides without evidence that because it is compelling it is true, restates their hypothesis as fact in an authoritative front-page post, and the hypothetical is then taken as gospel by all other readers for all time ever afterward.
The usual trendology bullshit.
ONLY credibility problem is NASA’s satellite sea level fakery !!
And the morons that “believe” it.
‘Someone proposes a hypothetical they think sounds compelling, decides without evidence that because it is compelling it is true, restates their hypothesis as fact in an authoritative front-page post, and the hypothetical is then taken as gospel by all other readers for all time ever afterward.’
Alan,
Well done – you’ve just described the media’s handling of climate alarmism to a ‘T’.
Mr.J: “There is a great problem with credulity on WUWT” What’s great about it? Sometimes, your word choices tell us that you are mentally limited. Oh, and the fact that you have fallen head-over-heels for an evidence-free hypothetical, which you take for gospel, is a give-away as well.
Estimates are more precise than measurements?
Something that is rarely mentioned is that the accuracy and precision of the sea surface measurement is absolutely dependent on knowing the altitude of the satellite. That can be determined with high accuracy over land. However, over the 71% of the surface covered by water, a model is used to estimate the satellite’s altitude and speed based on the assumed gravity. It leaves some room for error.
I’ll just throw in “Who’s afraid of Sea-Level Acceleration.”
Geologists know that the rock basins that contain the ocean waters are moving. For example, “Today, the Indian plate continues to move to the northeast at roughly 5 cm/year, pushing the Himalayas higher as the Indian plate underthrusts the Eurasian plate”.
https://www.ncess.gov.in/hindi/images/cess_images/14-Vineet-Gahalaut.pdf
It is entirely plausible that the ocean basin volume is chaning as we measure it, not by ways involving climate change but simply from the natural movements of the Earth. Also, now and then we read of a new island being formed as a sea mount grows bigger and punctures the ocean surface, in the way that the Hawaiian islands grew. These movements make it impossible to be certain that what we measure is anything more than plain old natural variation.
While we know that the basins change volume, we do not know by how much over what time. The change might be so gradual that we can ignore it – but we do not know that yet.We might be seeing some evidence from the 1906 San Francisco earthquake where the local SF tide gauge showed a jump of some 60 mm to confirm that movement had indeed affected sea level there.
https://www.usgs.gov/media/images/presidio-tide-gauge-graph-april-18-1906
We should be highly suspicious of climate change publications that ignore this known type of error, pretending that it does not exist when faced with the enormous power of the modern computer, with advanced computer models and the vast flow of data from altimeters on highly sophisticated satellites.
Forget all that high tech drama. In reality, you are watching a global warming con job by peole with a mission.
Geoff S
I always wondered about that. Thank you.
I think that it is worth reinforcing the point that tectonic plates are moving at an average rate the is an order of magnitude greater than the estimate for sea level. Yet, plate tectonics gets very little visibility in the discussions.
Per USGS, the Pacific plate is moving 4.6 cm per year (~ 1.8 inches) relative to the North American plate.
Per the Pacific Northwest Seismic Network, the movement is between 7 cm and 11 cm per year.
Story tip – Duggan announces Detroit’s first three solar neighborhoods (detroitnews.com)
Detroit — Mayor Mike Duggan on Monday announced the first three neighborhoods chosen to take part in a project to convert blighted lots into solar farms to help power municipal buildings.
Last year, Duggan unveiled his plan to convert 127 municipal buildings from 33 megawatts of fossil-based energy into solar by installing six neighborhood solar fields totaling 200 acres. The city hopes the fields will reduce its $8 million annual power bill by $4.4 million.
I bet they lose at least a million in copper.
Wait, there are still lights on in Detroit?
Suggest they keep good records and do a yearly audit.
With good luck they might not lose money.
200 acres is a lot of urban land. Perhaps, put high-pole
panels shading the “new parks & playgrounds” below.
Such neighborhood amenities do have value.
Such construction raises the cost tremendously, greatly dropping the possibility of any intrinsic value gain from the power generated.
How much does the city lose on land taxes, business taxes, rents, etc., it would have gained had the blighted lots been improved by investing the solar funds into something that generates economy.
I don’t imagine that it will take long before the solar panels are vandalized.
Another good source on the topic is the “Sea-Level Scares” chapter of Steven Koonin’s Unsettled. The discussion accompanying Fig. 4 of a Naptown Numbers post of mine describes how Raymond Pierrehumbert falsely charged Dr. Koonin of cherry-picking.
Except for decades-long droughts or massive volcanic explosions that ended some civilizations, humanity generally adjusted successfully – through a Pleistocene Ice Age, a Little Ice Age, a Dust Bowl, and other natural crises.
That last glacial period was more like 115,000 to 11,700 years ago. 11,700 – about 6,000 years ago was the Holocene Optimum which was the warmest period since the end of the previous glacial period.
the 7000 years was wrong should 11,000 years ago otherwise fine
Two millimeters a year is lost in the day to day noise of up to 10 meters or more in a single day of tidal inflow and outflow, as well as isostatic land changes (Nova Scotia land, tilts a few millimeters every day because of tides. And then there are gyres, ocean currents. El Nino and la Nina, winds, temperature variations etc.
Not worth bothering with.
The Dutch solved their sea problems hundreds of years ago.
Cuddy:
Yep! We did a river cruise in the Netherlands this past April [very nice!].
The highlight to me [my wife liked the Keukenhoff Gardens better – tulips!] was the day trip tour of the “Deltawerken” system of dikes, dams & levees. One of my questions to the facility guide was
“Aren’t you worried about the sea level rise the climate scientists are warning us about?”
He actually laughed, and said something like “We’ve been managing sea level for 100’s of years with only picks, shovels and horse-drawn wagons. Do you think with today’s equipment we are not up to the task?” Then he said there was a delegation from Houston here learning from the Dutch.
Note that ~ 25% of the Netherlands is below sea level.
RCP 8.5 would be a problem for the dutch so what he is saying is that its BS so there is no “climate emergency”. QED.
Venice even longer ago but recently they markedly changed the dynamics by providing for very large cruse ships to approach and dock in order to increase tourism still more. I also seem to recall some man made changes further west from Venice that resulted in large storm swell. Then there was the historic landmark movement that killed Venice’s greatest adaption scheme.
Very nice Andy. Why are the oceans at different levels. You would think they are all the same.
For the most part it’s because wind pushes water around.
Water is fluid and the basins are restricted.
Mostly the different basin shapes and geometry and the influence of the prevailing winds and currents.
And the gravity based action of the Moon and planets.
and the uneven masses within the Earth itself, effecting local gravity.
Yep.. like moving magma sacs of different densities…
Oh look , some red thumb moron thinks that gravity of the moon and planets doesn’t effect sea levels.
No accounting for total ignorance…… Is that you AJ !!
Think of it in terms of a state-wide election in a US ‘swing state’, e.g., Michigan. In order for any Republican to win, he/she first needs to overcome the considerable level of vote fraud that takes place in large Democrat-machine enclaves like Detroit. You should consider yourself lucky to reside in a country where elections aren’t rigged.
salinity, air pressure and wind.
Re satellite sea level measurement, as the satellite is moving in orbit around the earth at nearly 18000mph, and the earth is rotating once per day, when the satellite takes a measurement, it takes around 14 DAYS or so to take a repeat meaurent. All this talk of averaging measurements!!
The wind, currents, swells, troughs and peaks will be totally different after 14 days or so.
+/- 1mm my a*se.
If sea level was in any way a concern to anybody, then the obvious answer is to begin back filling the low points on the earth that could store the excess. The Dead sea, the deserts of north Africa maybe even fill up Death Valley.
As none of those areas are being looked at we can safely conclude, sea level rise is not an actual problem worth worrying about.
The volcanoe at Mount Pinatubo in the Philippines had a massive explosion in 1991 and again in 1992. This suppressed temperatures and sea levels around the world.
So when the satellites went up in 1993 sea levels were returning to normal which looked like accleration.
A bit later there was a large El Nino which also contributed.
Which was nice.
“What about sea level rise? Isn’t that dangerous?”
Great Britain on.y became an island about 7 000 years ago thanks to global warming, climate change, sea level rise.
Thanks to that we band of brothers, our sceptred isle were protected from invasion by the forces of Imperial Spain, Napoleonic France, Imperial Germany and National Socialist Germany.
So, dangerous? Depends.
Not so much Vikings, Romans, William of Normandy…. But, hey, you are still with us, the island did not sink!
Doggerland did! So did Cantre’r Gwaelod off the Welsh coast.
The way media portrays this is, if we do not stop all CO2 by July 1, 2024, we will suffer 1000 foot tsunamis.
Bogus? Of course.
Even 20 meters over 1 century is manageable. Just look at Netherlands dike system.
Factors not accounted for:
Shipping. The displacement of the ships raises the ocean levels. Not much, but still not accounted for.
Also, the Chinese are building islands (for military purposes) using coral reefs as the foundation. They dump stone and dirt and concrete to make these islands and it is more than just a couple.
How much ocean rise is due to the water displacement caused by these unnatural constructions?
The biology of the oceans is not accounted for. We harvest fish, etc. and their displacement effect on the ocean would temporarily lower the ocean level. Keep in mind, the devil’s in the details and unless things like this are addressed, how accurate can any claim be?
We also dump sand to create beaches or to replace erosion. Same point.
Tide gauges in the region around Victoria BC are interesting, there are indications of tipping of relatively small tectonic plates between New Westminster and La Push.
There may be some land rise from removal of water in Richmond BC which is now behind dikes.
The significant difference between tide gauges and satellite measurements are due at least in part to complications that satellite’s have near the coast. This is a 2019 study so perhaps there have been improvements. However, the linearity of the tide gauge rise is uncorrelated with the exponential rise in CO2 which rules out CO2 as a significant factor. For example NOAA’s New York data:
“ However, the linearity of the tide gauge rise is uncorrelated with the exponential rise in CO2 which rules out CO2 as a significant factor.” Not if the response to CO2 is logarithmic as it is for temperature.
That’s not the case. Both scales are linear.
Yes both scales are linear, but the CO2 one should be logarithmic, that’s my point.
“Radiative forcing of climate by CO2 changes is logarithmic in nature, and the relationship between ln(CO2/C0) (where C0 = 278 ppm = preindustrial CO2) and sea level over the past 550,000 y can be well approximated by a linear fit (Fig. 1B).”
The chart is for a long non-instrumental period prior to when according to Figure 1B CO2 was between 180-280 ppm. Based on the NOAA chart since 1855 sea level rise has been linear while atmospheric CO2 increase has been exponential. The author cautions that “this linear relationship cannot be simply extended beyond the data”. And given the exponential rise in CO2 while sea level rise is linear the linear relationship will not hold.
Bottom line – exponentially increasing atmospheric CO2 does not change the rate of sea level rise. This lack of correlation eliminates a causality argument.