Guest Post by Willis Eschenbach [see Update at the end]
I thought I might write about how I research a subject. Over at Dr. Judith’s excellent website, she periodically puts out a list of interesting papers that she has come across. This time it was “Week In Review: Water Edition”. She gave a link to an article from a Tampa Bay news station headlined Study: Sea level rise may severely impact Tampa by 2040.
Why did I pick this article? To me it’s obviously bogus. Sea level is rising around the world at something like 8-12 inches (200 – 300 mm) per century. It’s only twenty-four years until 2040, call it a quarter century. So by then Tampa will likely see on the order of 2 – 3 inches of sea level rise. That will not have a “severe impact” anywhere. So I went off to read the article.
Reading the article, I said “Well, there’s yer problem”, viz:
The study was based on sea level rise predictions from the National Oceanic and Atmospheric Administration and the findings of the Tampa Bay Climate Science Advisory Board.
The advisory board concluded that the region could experience sea level rise between a half-foot to 2.5 feet of sea level rise by 2050.
Two and a half feet of sea level rise over the next thirty-four years, a third of a century?? That doesn’t pass the laugh test. Consider that to do that we’d have to AVERAGE a sea level rise of seven and a half feet per century. So I set off to unearth the “findings of the Tampa Bay Climate Science Advisory Board”.
From there the trail got murkier. But after several false starts, I came across a Tampa Bay City Council Agenda that referred in passing to the report by the Advisory Board. With that I was able to track it down. It is called “Recommended Projection of Sea Level Rise in the Tampa Bay Region“.
The report starts out reasonably, showing the longest local historical sea level record, from the adjacent city of St. Petersburg.
Regarding this chart they say:
Data measured at the St. Petersburg tide station shows that water levels in Tampa Bay have increased approximately 6.6 inches or approximately 1 inch/decade (see Figure 2).
That’s all quite reasonable, and at ten inches (250 mm) per century the rise is right in the general world range I gave above. So I continue reading and find this:
The final parameter, projections of how much sea level will change globally over the next 100 years, is derived from experts engaged in climate science.
Uh-oh, sez I. We are now in the hands of “experts engaged in climate science”, so hide the good silver and watch your wallets … I continue reading …
The 2012 National Oceanic and Atmospheric Administration (NOAA) Technical Report, Global Sea Level Rise Scenarios for the United States National Climate Assessment, was produced as a coordinated, interagency effort to identify nationally agreed upon estimates for global SLR. The report synthesized the scientific literature on global SLR, included input from national experts in climate science, physical coastal processes and coastal management, and produced a set of four plausible SLR scenarios that can easily be adjusted for regional conditions throughout the United States.
Plausible? Er … um … I continue reading …
Future SLR estimates can be calculated for the Tampa Bay region, integrating data from the local St. Petersburg tide gauge, using a flexible, well-supported tool developed by the United States Army Corps of Engineers (USACE). The tool takes the three parameters discussed above (datum, rate of change, projection) and produces the plots or tables that describe how sea level will change in the future, such as those included as Figure 3 and Table 1.
Here is their Figure 3, in all its glory:
Now, this is good, because it starts in 1992. This means we are already almost a quarter century into this graph, so we can see how well the various projections are doing in 2016. Now, the “NOAA Low” estimate (bottom line, dotted blue) is simply an extension of the historical rate.
To look at the others, I took their Table 1 showing their projections, hand-typed it into Excel (because it’s a graphic and not text and numbers, bad scientists, no cookies), and used the Excel Solver tool to determine the equations of those lines. This lets me calculate the annual sea level rise for each of the four scenarios. Remember, the lowest one called “NOAA Low” is the observed rate extended out to 2100. In the figure below we can see the observed rate of sea level rise (2.6 mm/year) compared to the three NOAA projections for the year 2016.
How are they doing? Let’s start from the “NOAA High” projection. NOAA was definitely high for this one. It’s more than four times the current rate. I’d throw that model in the trash. After a quarter century it’s rising at more than four times the observed rate. Sorry, not valid.
Same thing for the “NOAA Int High”. After a quarter century that sucker is almost three times actual observations. Into the trash with that one as well.
Finally, the NOAA Int Low … well, it is about fifty percent higher than observed. I suppose that is in a feasible range … or so it would seem until we look at where the rates of sea level rise are projected to be by the year 2100.
The fastest sea level rise in the paleo record is from about 16000 to 8000 years before present, at the end of the last ice age glacial period. The rapid rise was from the melting of the ice that was a mile thick where Chicago sits today. As that incredible mass of ice in the temperate zones melted, sea level rose about 110 metres. That is a sustained rate of rise of about 14 mm/year. That rate has not been seen for the last 8000 years, and for a good reason—there is no mile of ice over Chicago to melt. Despite that, NOAA predicts that we may well see twice that rate of rise by the end of the century?? Like I said … that doesn’t pass the laugh test. Even their lowest estimate (NOAA Int Low) is more than half the historical record post-glacial rate … not believable.
Finally, look at the St. Petersburg sea level dataset, or any Florida sea level dataset. None of them show any significant acceleration, despite covering the period of recent warming. Warming but no acceleration of sea level rise … oops.
These alarmist claims of accelerating sea level rise have been being made since about 1988, when Jim Hansen conned the US Senate into buying into his hysteria. Despite the claims, there is no sign of said acceleration in the Florida sea level data.
Now, could it happen? Could sea level rise start to accelerate?
Sure, it’s possible … but until you see it, it is just a James Hansen fantasy.
These NOAA claims of wildly accelerating rates of sea level rise are not science in any form. They are government sponsored hysteria, and whoever did this at NOAA should resign.
Unfortunately, the costs of this rampant alarmism will be huge, as coastal communities will struggle to comply with a meaninglessly exaggerated risk.
The only good news is that sea level pays no attention to what NOAA and the rest of the activist-ridden government and non-government organizations say … and sooner or later, this will become too evident for even the most ardent climate activist to ignore.
My best wishes to everyone,
w.
PS—If you are commenting please QUOTE THE EXACT WORDS YOU ARE REFERRING TO, so that everyone can understand your subject.
UPDATE: As usual, commenters provide pure gold.
Larry Hamlin January 15, 2017 at 11:18 am says:
The Tampa Bay study is question relies upon a “tool” developed by the USACE as noted in the quote in the subject article provided below.
“Future SLR estimates can be calculated for the Tampa Bay region, integrating data from the local St. Petersburg tide gauge, using a flexible, well-supported tool developed by the United States Army Corps of Engineers (USACE). The tool takes the three parameters discussed above (datum, rate of change, projection) and produces the plots or tables that describe how sea level will change in the future, such as those included as Figure 3 and Table 1.”
The USACE model methodology can be found at: http://www.corpsclimate.us/ccaceslcurves.cfm.
This USACE tool is in fact a semi-emprical methodology model of the type whose significant shortcomings was addressed in the UN IPCC AR5 report which specifically noted:
“Many semi-empirical model projections of global mean sea level rise are higher than process-based model projections (up to about twice as large), but there is no consensus in the scientific community about their reliability and there is thus low confidence in their projections. {13.5}”
“Despite the successful calibration and evaluation of semi-empirical models against the observed 20th century sea level record, there is no consensus in the scientific community about their reliability, and consequently low confidence in projections based on them. {13.5.2, 13.5.3, Figure 13.12}”
“Because of the limited or medium evidence supporting SEMs, and the low agreement about their reliability, we have low confidence in their projections.” (Section 13.5.3)
Thus the Tampa Bay study addressed in this post relies upon highly speculative methodology which was specifically cautioned against in the UN IPCC AR5 report as providing results which the IPCC have “low confidence in their projections.”
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Willis
Really enjoyed the article particularly since I grew up in St. Petersburg, Florida. I lived on a landfill in Boca Ciega Bay. I did not live on the water but my neighbor across the street did. He let me use his boat dock and many days after school I would get some frozen shrimp out of the freezer and fish. Sometimes I would even catch dinner.
The headline does not go far enough. This is more than jumping the shark. This is the worst kind of deceit. If there were even a grain of truth to this your only option if you lived on one of those land fill would be to find a sucker to buy your property. You know that in that short of a period of time there is nothing you can do. Another futile gesture (Animal House) just won’t do.
I did try to find the raw data you were using and could not get to it. In a comment by Steve Case I did find a link where I could find some raw data. I have analyzed the satellite record of sea level change and I wanted to apply the technique to the gauge he referenced.
His reference was here.
http://www.psmsl.org/data/obtaining/stations/520.php
The location is shown here.
https://1drv.ms/i/s!AkPliAI0REKhgYw44IsxoAx3Durr_w
I lived on the bay between the barrier islands and the peninsula and is just to the left of the location shown.
The figure below shows how well the fit worked to the measured data. The figure also reveals the modeling equation. For this I was not interested in the sinusoidal behavior but only the underlying quadratic fit.
https://1drv.ms/i/s!AkPliAI0REKhgYw3uQlrTnJvrN1s_Q
https://1drv.ms/i/s!AkPliAI0REKhgYw2Uzw17_n1A3FRJA
Here are the values for b determined by the analysis.
-92.62998619387541
1.000011713071845
19.39853550453368
.01212077987873379
-45.34907824847151
49362.74622255816
-8.354019154726986
.09087440197605502
217.1968803590611
Here is the rate of change.
https://1drv.ms/i/s!AkPliAI0REKhgYw5-1vTcszLpuQSJA
Here is what might be expected until 2100.
https://1drv.ms/i/s!AkPliAI0REKhgYw1PkFL1rQPhS19cA
I think all of this supports what you are saying.
The Tampa Bay study is question relies upon a “tool” developed by the USACE as noted in the quote in the subject article provided below.
“Future SLR estimates can be calculated for the Tampa Bay region, integrating data from the local St. Petersburg tide gauge, using a flexible, well-supported tool developed by the United States Army Corps of Engineers (USACE). The tool takes the three parameters discussed above (datum, rate of change, projection) and produces the plots or tables that describe how sea level will change in the future, such as those included as Figure 3 and Table 1.”
The USACE model methodology can be found at: http://www.corpsclimate.us/ccaceslcurves.cfm.
This USACE tool is in fact a semi-emprical methodology model of the type whose significant shortcomings was addressed in the UN IPCC AR5 report which specifically noted:
“Many semi-empirical model projections of global mean sea level rise are higher than process-based model projections (up to about twice as large), but there is no consensus in the scientific community about their reliability and there is thus low confidence in their projections. {13.5}”
“Despite the successful calibration and evaluation of semi-empirical models against the observed 20th century sea level record, there is no consensus in the scientific community about their reliability, and consequently low confidence in projections based on them. {13.5.2, 13.5.3, Figure 13.12}”
“Because of the limited or medium evidence supporting SEMs, and the low agreement about their reliability, we have low confidence in their projections.” (Section 13.5.3)
Thus the Tampa Bay study addressed in this post relies upon highly speculative methodology which was specifically cautioned against in the UN IPCC AR5 report as providing results which the IPCC have “low confidence in their projections.”
There has been tremendous amounts of groundwater extraction along the eastern US coast. This surely has caused considerable subsidence, which probably is the main contributor to relative sea level rise along the east coast (and in places where the hysterics are waving their arms telling us they are going to be under 5′ of water in a year or two). Love those scare stories of how the entire state of Florida will be under several feet of water ‘by 2100.’ I have to suppress my laughter at those people who actually take such things seriously.
Congratulations on another excellent article, Willis.
We here in Gustavus Alaska have proof, I tell you, proof the sea level is falling! Sea level has fallen so much that we had to build a new boat dock because the old one was now out to the water.
At the rate we are going the oceans be much smaller, we must start burning more fossil fuels to make it warmer so we can stop this falling sea level.
You are probably experiencing uplift. This is very normal in a geological setting like Alaska – the point being that without measuring uplift or subsidence, tide gauges are useless for the establishment of GSL change
Their recommendations are reasonable
“CSAP encourages local governments and other agencies to use multiple scenarios in order to
allow experts and decision makers the flexibility to consider a variety of contextual factors,
including the expected lifespan of the project, project cost, and criticality of function, when
developing adaptation strategies. Scenario planning offers opportunities to initiate actions now
by balancing the costs of inaction against reasonable returns on investments made to
infrastructure that may reduce future impacts and vulnerabilities. For example, decision makers
may decide to plan for the NOAA Intermediate Low or Intermediate High scenarios when faced
with projects with low to moderate risk tolerance such as temporary projects or infrastructure
projects with a relatively short life cycle, while they may choose to plan for the NOAA High
scenario in situations where there is little tolerance for risk (e.g. new infrastructure with a long
anticipated life cycle such as a power plant, waste water treatment facility, or bridge) (NOAA,
2012). The level of adaptation planning necessary will be up to the planning entity and based on
the acceptable level of risk and vulnerability.”
When it comes to climate related planning you are really on the edge of science.. its more like OA where all you have is sketchy data and various models.
Like war planning.. historical data is small, incomplete, not collected with science in mind. And where models of how war happens and evolves are basically narratives with numbers attached. Still, it’s precisely
narratives and models like this that led Reagan to do Star Wars and the huge build up our military to counter a threat that ONLY EXISTED in the minds of modelers. You read that right. To plan for the future war we had to project, imagine, “predict” what we thought the soviets would look like 20 years into the future. Turns out we were wrong. Opps
But sometimes planning for the worst is vindicated
http://www.dailymail.co.uk/news/article-1386978/The-Japanese-mayor-laughed-building-huge-sea-wall–village-left-untouched-tsunami.html
It is one thing to plan for known risks, like hurricanes, which are relatively frequent; and for others less frequent but still known to occur, like tsunamis; and even for speculative but possible dangers, like an enemy with an existing arsenal of ICBMs. But it makes no sense to plan for fantastic events that cannot occur in less than thousands of years, like an ice-age glaciation melting. /Mr Lynn
Most offshore sedimentary basins subside under their own weight. This provides accommodation for further sediment deposition. Without this process there would be no hydrocarbons, as organic material must be buried at depths ranging from 2 – 5.5 km to form hydrocarbons i.e. sufficient heat and pressure.
The US eastern seaboard is a passive margin which no doubt has a very extensive and thick sedimentary basin offshore. It is quite possible for the continental coastal region to be dragged down too. I doubt that this would be as high as 1 mm/yr.
Just some basic geology
IOn 2007 Chao. Yu. and Li determined that if the impoundment of water in artrificial reservoirs is taken account of the sdea level rise foir the last 80 years becomes linear with a slope of 2.47 millimeters per year. This works out to judst under 10 ionches per yuear, nmot the twenmty feet as in Al Gore’s book that was awartded a Nobel Prize. I If something has been linear asw long as that there is no rason to thuinkj that it is abpot to change anytime soon. Lookimng at more recent sea levcel rise curves, including the current one from NOAA verifies thsat the linear trend is still present in currently available vdata. The problem is that people studying it simply want to rewad morer warming into it thjanj there is. I am willing to bet that the centennial sea level value will come out at ten inches, just as the work of Chao, Yu. and Li from 2007 would predict, plus or minus at most one inch for unforeseen circumstances. Thje numerous asttempts to find some other value since then are testimony to the strenth of the belief in pseudo-science that predicts iditioc values in the meters or tens of meters range.
Ten inches perr century is correct. (Sorry!)
The New Zealand Governments Parliamentary Commisioner for the Environment (PCE) Dr Jan Wright went on a similar rant in her first report on Sea level rise. Page 19 had a graphic from the IPCC report indicating rises of 50 to 100 cm by 2100. the current average for NZ and for Auckland is 1.4mm/an , the same as for the Globe. Dr Wrights report rate requires a rise of 9mm/yr every year from now to the end of the century. There is no CO2 signal in the New Zealand data and there is no acceleration evident in the real data. The PCE’s report did not reference New Zealands current expert in sea level (John Hannah) nor the retired head of The Department of Scientific and Industrial Reseach (Dr Kerr) who did his Doctorate in Sea Level.
When I pointed out these flaws(absurdities) in the report a second report appeared a year later but it still blamed sea level rise on CO2.
I decided to use a different different method to detect the possibility of catastrophic sea level rise. Our recently departed Prime Minister has a holiday home on a sand spit (roughly sea level) just north of Auckland, when he sells it I will take another look at sea level.
Enjoy
Willis: The first half of your post deals with the current rate of SLR. The second deals with the key issue acceleration of SLR. Current SLR is about 1 inch/decade. For SLR to reach 1 m by the end of the century, acceleration of SLR needs to be 1 inch/decade/decade. We obviously haven’t seen that much acceleration over the past few decades. If we are experiencing sea level rise of 2 inch/decade in 2025, then we might experience 1 meter of SLR by 2100. So we appear to be safe at the moment.
Or are we? Tide gauge records are too noisy to provide reliable rates of SLR in periods shorter than about 50 years. Clearly tide gauges aren’t going to be of much use telling us that SLR has risen from 1 to 2 inch/decade between now and 2027. They don’t provide an instantaneous measure of SLR, they provide half-century averages. Perhaps combining several hundred tide gauges will help some, but they aren’t up to the job of measuring acceleration.
That leaves us satellite measurements. Unfortunately, they show that global SLR is quite noisy. So we need to detect acceleration of SLR in noisy data. A few years of data around today’s SLR of 1 inch/yr and a hypothetical 2″ inch/yr in 2027 will unfortunately not be statistically significant due to the amount of noise in the data we have been seeing.
I agree with you that a 5-fold increase in SLR by 2040 looks absurd, but detecting and quantifying acceleration in SLR is a challenging problem. Fortunately, the 30-year satellite record doesn’t show evidence for worrisome acceleration.
Some interesting info here emphasising the influence of both geological subsidence and ground water extraction. Until these kinds of studies are done on an extensive scale all tide gauge data should be considered useful only for relative sea level change. Most coastal zones around the globe are moving, either up or down in relation to mean sea level. These are simple principles taught in first year Geology. Why don’t they get it?
http://www.gps.gov/cgsic/meetings/2015/dixon.pdf
Your graph of the St Pete Tide Gauge data seems to be missing some of the most recent data, the most recent version with data up to Nov 2016 is as follows:
https://tidesandcurrents.noaa.gov/sltrends/sltrends_station.shtml?stnid=8726520
The last few years show levels above the 95% CI, a quadratic fit to the data gives a slightly better fit to the data showing a modest acceleration, and appears to be the form of the USACE projection.
Phil
I did post such a fit in an earlier comment. It is virtually the same result I get when looking at the satellite record of sea level change. It is modestly accelerating.
Phil. January 16, 2017 at 7:42 am
First, note that the 95%CI is NOT the 95%CI of the data. It is the 95%C! of the trend. Most of the data is either above or below the CI of the trend … so? Look at the data. It is both above and below the 95% CI everywhere.
Second, while the fit is “slightly better” with a power curve, it is not significantly better. As a result, Occam’s Razor suggests that we should use the straight line, because there are less tunable parameters.
Next, we cannot use the power fit thus determined, because the power fit starts in 1947, and the USACE power fit starts in 1992.
Finally, I strongly object to the idea that a power curve can simply be extended a century into the future. You MIGHT, and I emphasize might get away with extending a straight line for a few years, that’s physically possible.
But as I showed in the head post, if you extend a power curve very soon you are into unphysical, never-seen rates of annual sea level rise.
All the best, thanks for the thoughts,
w.
“You MIGHT, and I emphasize might get away with extending a straight line for a few years”
Not legitimately. No empirical model can be used legitimately to predict the behavior of a system. Unless there’s a sound and well verified theory backing the model, it can’t be used to predict.
This is a feature of statistics regularly ignored by the people who produce these “predictions”. They’re all invalid. Books have been written on the subject. It’s drilled into every trained statistician; you cannot use an empirical model outside the boundaries of observation. It’s the same problem Mike Mann had in his absurd paleo reconstructions based on tree rings and other proxies that were only calibrated over the short time we’ve had instruments. It’s all pure nonsense. Doesn’t matter if it’s a straight line or a power curve; it’s an empirical model. It isn’t predictive.
I wrote: “No empirical model can be used legitimately to predict the behavior of a system”.
That was too emphatic.
An empirical model can be used to predict observations within the calibration range. So if a model was developed to record the behavior of Tampa SLR between 1945 and 2016, but it was based on annual averages and you wanted to make a statement about the sea level in Tampa on January 16th, 1962, you could legitimately use this model to estimate it.
But you can’t use it to predict the SLR in 2035. That’s not cricket.
And I’ll add, “bad scientist. No cookie”. 🙂
Willis Eschenbach January 16, 2017 at 10:36 am
Second, while the fit is “slightly better” with a power curve, it is not significantly better. As a result, Occam’s Razor suggests that we should use the straight line, because there are less tunable parameters.
The version attributed to Einstein is:
“Everything should be kept as simple as possible, but no simpler.”
Since the straight line doesn’t allow for the possibility of acceleration a straight line would appear to be ‘too simple’.
Extrapolation of any fit without an underlying justification is speculative, and should be done in knowledge of the limitations.
Willis,
Did you realize you have been Taminoed?
https://tamino.wordpress.com/2017/01/16/coping-with-sea-level-rise/#more-9231
Of course, he says that using a power curve is the only scientific choice and that anything else is unscientific. But then he always likes power curves as they show rampant acceleration in the out years.
Once again AGW (cough, hack, gag) scientists project a sea level rise caused by a tiny change in energy not even worthy of a B-rated sci-fi movie. At least their premis of enlarged insects was based on nuclear energy. This sea level rise monster is based on… wait for it… organic gas fertilizer increasing by a parts per million tiny fraction.
I dare say the Inquirer, back in the days of Alians, had better research behind their reports than this peer reviewed drivel.
The comparison between sea level rise and tide range, is rather like that between the warming effect of CO2 and seasonal temperature changes. Both involve measurement of a tiny quantity which is totally masked by a far larger cyclic change. Worse, the cyclic change has a random component as well.
When I was at uni, I was told that it is poor scientific practice to derive conclusions from measurements made ‘below the noise floor’ of the system. That salutary piece of wisdom seems to have been forgotten by the climate change movement.
“whoever did this at NOAA should resign.”
Their resignation should be accepted. It’s pretty hard to fire a GS level anything “scientist”. Another argument against public unions.
I don’t notice any mention of subsidence in your report? My understanding is most of Florida (like Holland) is sinking due to depletion of the fresh water aquifer?
The place to look for the cost of this nonsense is the local building and planing department, who very likely will place all sorts of new restrictions on building, along with requiring modifications of existing property to meet new codes based on these findings. The real cost will be felt by Florida property owners and the beneficiaries will be local contractors and, of course, local government.
This issue is already hitting some unfortunate individuals in New Zealand in the pocket. A nephew recently completed a elevation survey for a City Council so they could zone regions “at risk”. Insurance companies are already studying these and planning adjustments accordingly. Government has also absolved themselves from any liability should property prices be influenced – as I am sure they are or will. The authorities just don’t want to know about the real science.
NZ is an active tectonic country. Until GPS or other surveying apparatus are installed along with every tide gauge we can’t know what the heck is going on. This is the sort of injustice that makes me livid.
Dean P January 17, 2017 at 8:57 am
Hia “scientific” choice is to extend a power curve out a hundred years? Sounds like Tammy, all right.
Doesn’t surprise me. Ever since he kicked me off his website a decade ago or so for asking inconvenient questions Tammy has been doing his trivial best to bite my ankles … hasn’t succeeded yet.
w.
Willis, there seems to be an inconsistency between your charts of annual rate rise and your fig 3.
You seem to have plotted the instantaneous rates at the future dates whereas for the ‘observed rate’ is the trend since 92, examination of the data shows that the current observed rate is more like ~4mm/yr.
To make an ‘apples to apples’ comparison surely it would be appropriate to plot rates based on the same basis?
Phil, thanks for checking the numbers. However, Table 1 in the cited paper says that under the “NOAA Low” estimate the level goes up by 0.93 feet in 109 years, from 1992 to 2100. This is 2.6 mm/year.
Since I am comparing this to the other NOAA estimates in Figs 3 & 4, it is indeed apples to apples. All the numbers I used were from Table 1.
w.
Willis Eschenbach January 17, 2017 at 1:32 pm
Phil, thanks for checking the numbers. However, Table 1 in the cited paper says that under the “NOAA Low” estimate the level goes up by 0.93 feet in 109 years, from 1992 to 2100. This is 2.6 mm/year.
Since I am comparing this to the other NOAA estimates in Figs 3 & 4, it is indeed apples to apples. All the numbers I used were from Table 1.
You’re welcome. Thanks for the Table ref I’d just been eye-balling the graphs, however I still get a discrepancy.
For example NOAA High in 2100 shows 6.89 ft
So 6.89x12x25.4/109 gives 19.3 mm/yr
NOAA Int Low in 2025 shows 0.38 ft
So 0.38x12x25.4/33 gives 3.5 mm/yr which is less than the value for Nov 2016.
I can’t see an error?
Phil, if I understand you I think you may be confusing average change with annual change. For NOAA High, for example, the annual change in 2100 will be much larger than the average change over the period. They are only the same for the straight line case.
w.
I agree, that’s what I said in my first question but you said that you used the data in Table 1 which only allows you to calculate the average. In order to calculate the annual change in 2100 you’d need the gradients of the projected fits. In which case in the 2016 plot the ‘observed value’ in your table is not correct and you should have another column labelled ‘linear fit’ (2.66/yr).
I did use the data in Table 1. The “NOAA Low is an extension of the observed long-term trend, starting in 1992 and extending for 100 years. That’s what I meant by “observed”, sorry if it wasn’t clear.
w.
No statistics here. No analysis to back up the claims. Just loads of attitude & ridicule. Eyeballing of graphs. How about a demonstration that a linear function DOES give a better fit to sea level rise than higher-order, and that it is statistically significant? Hmm? Or is making defensible claims not what this site is about?
Thanks, Patrick, but I must object. “Eyeballing of graphs”? What on earth are you talking about? I gave you the link to the paper. I referred you to Table 1 in the paper which contains the data I used. I explained what I did.
If you think that my analysis is incorrect, if you want to get traction you’ll have to do what I asked and QUOTE THE EXACT WORDS THAT YOU OBJECT TO.
In addition, I’ve given you the location of all of the data involved. Let’s see your figures for what you think is going on.
Do you really think I got the numbers in Figs. 3 & 4 to one decimal place from eyeballing a graph? Following the form of their data, I simply fit a power curve of the form y = m x b to the data in Table 1 for each of the NOAA scenarios.
You have the data. I await your analysis …
w.
Well, Willis, Excel is a powerful tool where you don’t have to type data manually.
Just adapt Exels grid size, lengths and hights, to said graph.
Take that graphic for Excels background picture and calculate the formula aka ‘model’ you need to get that graph in Ecxel.
Ce’ voila, excel stems of lotus.123, use your magic wand and get while the getting is good.
Of course don’t forget to x your excel file:
set an ‘x’ into every cell your underlying graph has to pass before doing that .123 magic wand thing so excel knows what’s up.