Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyses

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

Thanks for posting the link so soon after the article release. I wonder if this a research area of interest for you, or if not, how it came to your attention? Just curious.

It is not central for me, so I had to poke around a bit. The cited PSMSL cite indeed appears to be the 'go-to' place for global tide gauge data. It is much more extensive than the data that Houston and Dean (above) examined. To my reading they did not clearly indicate why they focused on the US data subset. But that of course is their choice.

A chapter in a 2010 book

Modern Sea-Level-Change Estimates, in Understanding Sea-Level Rise and Variability (eds J. A. Church, P. L. Woodworth, T. Aarup and W. S. Wilson), Wiley-Blackwell, Oxford, 2010

did a similar analysis using more PSMSL data. That chapter is for free here

ftp://www.grdl.noaa.gov/pub/laury/The%20Book/c05.pdf

Which incidentally is a US govt funded NOAA website. For that matter, so is PSMSL. Not sure how that might fit with the tags you've chosen for this thread. But anyway, what Mitchell et al. found seems a lot like what Houston and Dean did. Namely, the number of tidal gauge stations showing increasing trends is nearly equal to those showing decreasing trends. Both studies state this (outside the section you excerpted). So if it's news, it's 2010 news.

For Houston and Dean this means no global acceleration, while for Mitchell et al., that more effort in tidal monitoring, and new attention to the satellite data are warranted. Personally I am more inclined towards the latter.

The satellite data is short, but it does exceed tidal gauges in both studies over the same time period.

The extrapolation method that Houston and Dean use is driven by air temperature trends, and thus explicitly excludes the possibility of future 'tipping points' accelerating ice melt non-linearly. Reading the ice dynamics literature will help both of us consider the possibility of such - what would you say about that idea?

In any case I am always glad to be pointed to the diversity of results being published in earth system science. Value of 'climate-change' supported data, uncensored publication of results, etc.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

"First, they did not focus solely on US data as you indicated"
See the station list in Table 1.

"I'm not sure what your point was in bringing in the other reference"
1. The conclusion that accelerations nearly equal decelerations is not news (if indeed you were implying so)
2. The Mitchum et al. 2010 book chapter (I called it Mitchell; my mistake) was a more complete analysis of PSMSL data, not referenced by Huston and Dean. References get missed; I'll not blame them for that, but presuming you are interested in the topic, it is reasonable that you'd want to read it.
3. Mitchum is free to download and this is usually not the case for book chapters. Saved you a trip to the library, but I cannot guess from your response that you've had the chance to read it yet.

"extrapolation trends"
1. This one goes to you. Huston and Dean did not attempt to extrapolate future tidal gauge height based on predicted future air temperatures.
2. They did develop the idea in the introduction, citing Meehl. In the conclusions they said this "It is essential that investigations continue to address why this worldwide-temperature increase has not produced acceleration of global sea level over the past 100 years". However, your study of the scientific literature is very different from mine if it leads you to the conclusion that this is not what they are talking about.
3. If this study has no connection with global temperature trends, why would they have written those things?

"fantasy such as "tripping point""
1. Now, at last an editorial comment that can be rightly described as such. But most commonly "tipping point'.
2. Now we get to the important stuff. If Huston and Dean are only reporting 20th century US patterns, I wouldn't be very interested. But their abstract begins "

Without sea-level acceleration, the 20th-century sea-level trend of 1.7 mm/y would produce a rise of only approximately 0.15 m from 2010 to 2100" So they are interested in the future. Me too. So they offer a linear extrapolation to the end of this century, which is small. If that's the only thing that could happen I think we'd do better to focus our minds elsewhere.
3. But it's not the only thing that could happen if current temperature trends accelerate, or if grounded ice melt rates accelerate. I'm sure it's tedious and excessive to refer here to the very large numbers of studies that address those two matters.
4. If the research on future climates and ice dynamics is all fraud (thread tag), then +0.15 meters is probably a fair guess for year 2100.​

"You apparently find an approach which continues to search for data to prove a preconception rather than accepting the historical data to be more appropriate."
1. Not at all. I am interested in knowing the most likely range for future sea levels over a reasonable range of the factors that drive sea level change.
2. Historical tide gauge data contributes to this, so we rightly turn to this study of few stations, Mitchum's study of more, and the other studies that both have cited.
3. I desire that the number of tidal gauges be increased, and that satellite data be increasingly integrated. The papers that we have both cited do as well.
4. If we follow Houston and Dean's linear extrapolation, we will abandon a large body of other research and perhaps not achieve the goal I stated in 1.​

Finally you have asserted that I have a preconception, and thus implied that you know what it is. So I am bound to ask you about yours, if any. Do you preconcieve that the bulk of climate dynamics research is fraud, snake-oil, or something along those lines?​

Huston and Dean 2011 is none of those (which makes the thread tags perplexing). It is a small piece of a large puzzle, and once again, I thank you for introducing us to it here.​

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

How high will it rise and how fast.

IMHO the artic ice cap is going to melt, and there is no stopping it. Its time to get ready for the flood.

Will sea level rise only 0.15 meters by 2100? Could be. That seems to be a reasonable guesstimate. What about that inconvenient estimate of 20 feet? That seems a little crazy. Well to me that was a huge exaggeration. I've read that 1 meter rise by 2100 is in the realm of possibility when looking at the land based ice melting. Multiple sources but best source is church. That would provide a mechanism for acceleration. If you are on a low island I would be worried. Is this the valid range 0-1 meters sea level rise by 2100?

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

It is worth reading at least the abstract of the Douglas article that was cited in this article, then looking at the standard errors of the main results in this article.

This entire line of analysis is about whether or not you will be able to pick up the predicted acceleration of sea level rise using tide gauge records. Or, at least, that was Douglas' work. His answer was a pretty definite no, you will need decades of additional data before the predicted acceleration will be detectable, owing to a lot of factors. In this more recent paper, by a couple of retired civil engineers, you will note that the measured deceleration is not even close to statistically significant -- it's about one standard error away from zero (that's the bottom line of Table 2 in the paper).

(Aside: Had this been a paper in my field, the authors would not have been allowed to discuss the result as it was done here, because it's not statistically significant. In other words, every mention of "small deceleration" would have had to have been phrased as "no evidence of acceleration or deceleration".)

A picture is worth a thousand words here. By eye, it's clear there's no significant quadratic component to sea level rise over the long time period studied. (The paper itself explains why that long time period is necessary, for the tidal gauge data). So it's not too surprising that you can't generate one out of the data. The data look close to linear over that period, and, sure enough, the quadratic term in their fit is not statistically significantly different from zero.



(Aside: Though, if Table 2 really is their results, I have to say that their statistical analysis is pretty crude. Maybe I should read the paper more closely and see if they did anything smarter than just taking the average of the tidal gauges. For example, they need to have corrected for the geographic correlation of their locations -- two gauges located close to one another do not provide two statistical degrees of freedom. On the other hand, they didn't actually bother to pool their results in any way, and they didn't seem to present any formal statistical tests, so ... maybe that just doesn't matter here.)

The real questions are:

Does this put the satellite altimeter data in doubt? The satellite data appear to show faster rise in the last couple of decades, something that (if I read Douglas correctly) you couldn't resolve out of the tidal gauges in any case. So that's a significant research issue.

But the more important question is: OK, everybody agrees that sea level has been rising. Who, exactly, says that the rise should have been accelerating fast enough to have been able to have been resolved over this period using tidal gauge data? Who says this information contradicts what is otherwise believed about sea level rise?

In short, is this actually new, contrary information?

Well, why not take two minutes and check?

Answer, no.

I went back to read the chapter in the IPCC 4th TAR. It is very good and is worth reading, if you have any strong interest in understanding sea level changes.

http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch5s5-5.html

For example, I had forgotten how variable sea level rise, geographically. Reconciling the land-based gauges with the satellite altimeter readings is not as simple as taking the average of the land-based gauges (as here), because they don't measure sea level out in the middle of the ocean. The land-based gauges take a thin, geographically clustered set of measurements (coasts) out of a large area (ocean). The substantial regional variation in sea level rise means that it takes a fair bit of work to reconcile one set against the other, when you are trying to detect something fairly subtle out of a set of noisy data.

Again, one picture is worth a lot. Here's the geographic variation in sea level change (you have to go to the page to get the caption of the graphic). Note that you have large areas of both increase and decrease that are located well away from the coasts, and thus not captured by the tide gauge data.

Figure 5.16 - AR4 WGI Chapter 5: Observations: Oceanic Climate Change and Sea Level

I looked in the 2007 IPCC 4th TAR for their take on 20th century sea level rise acceleration. Here's what they said:

"Interannual or longer variability is a major reason why no long-term acceleration of sea level has been identified using 20th-century data alone (Woodworth, 1990; Douglas, 1992). Another possibility is that the sparse tide gauge network may have been inadequate to detect it if present (Gregory et al., 2001). The longest records available from Europe and North America contain accelerations of the order of 0.4 mm yr–1 per century between the 19th and 20th century (Ekman, 1988; Woodworth et al., 1999). For the reconstruction shown in Figure 5.13, Church and White (2006) found an acceleration of 1.3 ± 0.5 mm yr–1 per century over the period 1870 to 2000. These data support an inference that the onset of acceleration occurred during the 19th century (see Section 9.5.2)."

All you really need to see is " no long-term acceleration of sea level has been identified using 20th-century data alone".

So the upshot is this: Its well known that you don't find any statistically significant acceleration in the 20th century data using tidal gauges. Or, at least, nothing you can hang your hat on. It appears that the reasons for that are well known. And, probably, it will be quite a while before you will be able to resolve any acceleration out of the tide gauge data. This new work extends prior work by a couple of decades (the authors thank Douglas for providing data), and demonstrates the same result as before: no statistically significant quadratic term. A completely worthwhile thing to do.

The only thing in the paper that was even moderately out-of-line was in the conclusions, where in passing, they compare what they found in the past century to the increase projected for the next century. They appear to be trying to make a lot out of the fact that the projected acceleration for next century is much larger than acceleration, measured using tide gauges, for the past century. I'm not sure that's a logical thing to do, but it's not beyond the pale to bring that up.

I imagine some denialists are in a tizzy about this. I see that it's posted on Watts Up With That, which is a pretty good indicator that it's being flacked pretty hard. But as far as I can tell, this isn't some earth-shattering discovery, it's not even new, it's just a bit of normal science.

But, it has additional value to some people, because most people can't be bothered to read what the mainstream understanding of the issue is. Even when it's posted on-line in a very readable form. So it can be hyped as some great bit of contrary data. You'd think that trick would have worn thin by this point, but, apparently, not.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

One strange question comes to mind looking at this data. Is the rise all due to ice melting? Given that tide gauge data could have some land mass rise or fall component, is this completely eliminated as a factor? Could there be any other factor?

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

Houston and Dean excluded stations they considered sensitive to earthquake-induced changes of sea level.

Isostatic rebound (that is glacial mass unloading) is a big deal ove the longer term, but is presumed to not change at the decadal time scale, so it should not alter acceleration/decelleration patterns on the time scale.

Factors that do (promptly) change sea level are melt-rate changes of grounded ice, net change in the balance of precipitation over land vs. evaporation over the seas, and thermal expansion of sea water.

IMO the Mitchum 2010 book chapter discussed them all pretty well. Chogan2 liked the AR4 section on that, so those could make a start for your reading list.

BTW I give Chogan2 a -1 for playing the "retired" card. Ad Hom. I hope he/she gets better treatment when publishing in the 'golden years'.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

The reason for asking was based on some interferometric SAR images I saw long ago. Supersensitive to small changes over time. That would seem to be a good thing to consider extracting land mass changes.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

The tipping points literature is quite large; certainly beyond what I can assimilate. Those who wish to dip in might begin at Lenton's website

Tipping points in the Earth system | ResearchPages.net

Overpeck and Cole 2006 is for free here:

http://academic.evergreen.edu/z/zita/teaching/CClittell/readings/Jan31_Overpeck_and_Cole_2006.pdf

Kattenberg and Verver did a lit review in 2009:

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.163.2104&rep=rep1&type=pdf

Again, this is a tiny sample, but it introduces both 'natural' and anthropogenic tips. The hard-core tipping lit. has a lot of chaos and bifurcation theory.. don't say I didn't warn you.

(edit) new in GRL, evidence from perturbation runs with 'standard' climate change models that the arctic sea-ice albedo 'tip' may not exist. tease:

"Our results suggest that anomalous loss of Arctic sea ice during a single summer is reversible, as the ice–albedo feedback is alleviated by large-scale recovery mechanisms."

Tietsche et al. found that the water warmth accumulated in summer is lost quickly the following winter. Full pdf available by PM. GRL would fry me if they knew I was holing their paywall, but in the interests of full disclosure, and not hiding any results adverse to a 'climate-change' point of view, etc. (end edit)

Also new in Geophysical Research Letters is Rignot et al. about ice-melt acceleration. Here's the tease:

"Notably, the acceleration in ice sheet loss over the last 18 years was 21.9 ± 1 Gt/yr^2 for Greenland and 14.5 ± 2 Gt/yr^2 for Antarctica, for a combined total of 36.3 ± 2 Gt/yr^2. This acceleration is 3 times larger than for mountain glaciers and ice caps (12 ± 6 Gt/yr^2)."


And I can send the pdf along at request. I'd be remiss not to point out that a gigaton of water represents a tiny fraction of a mm of sea level.

Beyond that, I'm happy for Octane to have the last word here.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

The main factor is that water expands as it warms. Then comes melting.

But when it gets down to something as slight as this, even the net increase in water held behind dams, or the net withdrawal of ground water for irrigation (e.g., drawing down the Oglala Aquifer in the US Midwest) matters. In the long run, you'd even have to factor in the runoff from the melting of the permafrost.

But here are the main factors. This is the IPCC's graphic, not really clear-at-a-glance, but the brown bars are the last decade, the blue bars are the last 40 years. The width of the bar represents the uncertainty of the measurement.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

Neat. To me this would seem to be the best evidence showing global warming (regardless of cause). Water expansion is pure physics, is "self averaging" over most all factors, and is directly measurable. (e.g. If someone claims that there is no "warming", how could they explain the ocean rise?) That is not an explanation of what is causing it, but helps move the discussion level to the next topic. The "derivative coefficients" are really a small issue here for the simple reason that they do not explain the cause, just the effect. That linear coefficient is the monster to be understood first.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

This morning I just finished reading a Rutgers science article (Science-18MAR11-pg 1404) about the two North American Volcano eruptions 201 million years ago. Part of what you say is true. Nature was able to handle the 1000 ppm to 4000 ppm CO2 spike caused by the volcanos. Natural weathering of silicates was able to reduce the CO2 surge with a reduction averaging about 2500 ppm over 750,000 years. The only problem is that these events coincide with Mass Extinctions.....which nature obviously recovered from.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

Having looked up the surface area of the global ocean, I can say that 1 mm of sea-level rise due to new net input from land sources equal 360 gigatons. That could be added as a second horizontal axis to Chogan's IPCC blue/brown graphic above. But it would not apply to the thermal expansion bars, because that is not net new water. It is just slightly bigger water. If you see what I mean.

The original cited paper indicated that there had been temperature increase and there had been sea level rise in the 20th century. The IPCC chart above (from Bindoff et al. 2007 also cited by Houston and Dean 2011) indicates that this sea level rise can be attributed mostly to thermal expansion and less to net new water.

This is not my main area of interest, so I have been obliged to study up on the topic. Thanks all, for the stimulus to do so.

 

Re: Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyse

Our oceans are not at risk of freezing over. Floating ice does not raise the level of the rest of the ocean.