SEA LEVEL: Rise and Fall – Part 3 – Computational Hubris

Guest Essay by Kip Hansen

Sea Level Rise:  Measured from Space?

In Parts 1 and 2 of this series (posted here at WUWT some time ago now — if you missed them, I recommend reading them first), I offered my opinion, and evidence to support it, that:

Part 1:

1. SLR is a real imminent threat to coastal cities and low-lying coastal and near-coastal densely-populated areas.
2. SLR is not a threat to anything else — not now, not in a hundred years — probably not in a thousand years — maybe, not ever.

Part 2:

1. Tide Gauge data is invaluable for localities in determining tide states, sea surface levels relative to the land, and the rate of change of those levels — the only Sea Level data of concern for local governments and populations. However, Tide Gauge data, even the best station data from the GLOSS network, is only accurate to ±2 centimeters. All derived averages/means of tide gauge data including daily, weekly, monthly and annual means are also only accurate to ±2 centimeters.  Claims of millimetric accuracy of means are unscientific and unsupportable.
2. Tide gauge data is worthless for determining Global Sea Level and/or its change unless it has been explicitly corrected by an on-site CORS-like GPS reference station data capable of correcting for vertical land movement. Since the current standard for Tide Gauge data, the PSMSL GLOSS, is not corrected for vertical land movement, all studies based on this uncorrected PSMSL data producing Global Sea Level Rise findings of any kind — magnitude or rate-of-change — are based on data not suited for the purpose, are not scientifically sound and do not, cannot, inform us reliably about Global Sea Levels or Global Sea Level Change.

[I have also written about sea level rise here:  here, here, here, here and here.  These previous essays are not prerequisites but are interesting specific local examples.]

# # # # #

WARNING:  After struggling with this piece for more than a month, I find it is far longer than most readers will find comfortable.  Again, you have my apologies, but for those seriously interested in the question of Sea Level Rise, I hope it will be worth your time.  Bookmark it and come back to it when you have settled for the day into your most comfortable chair with your favorite evening beverage.

Those with less interest or less patience can page down, look at the graphs and illustrations, and read only the Series Take Home Messages and the Author’s Opinion at the very end.  — kh

# # # # #

There have been so many very good essays on Global Sea Level Rise by persons all of whom have a great deal more expertise than I.   Jo Nova hosts a dozen or so excellent essays, which point at another score of papers and publications, for the most part clearly demonstrating that there are two contrarian positions on sea level rise in the scientific community:  1) Sea level has risen, is rising and will continue to rise at a rate approximately 8-12 inches (20-30 centimeters) per century — due to geological and long-term climatic forces well beyond our control;  and 2a) Other than explicit cases of Local Relative SLR, the sea does not appear to be rising much over the last 50-70 years, if at all.  2b) If it is rising due to general warming of the climate it will not add much to position 1.

For review, the generally accepted SLR graph for the last 130 years is this from Church and White, updated in 2015.  My previous essays covered the Tide Gauge data;  this essay focuses on the short satellite record, shown in green on the upper right, with my annotations.

To bring us into the satellite era, I suggest starting with this essay at Jo Nova’s blog:  “Are sea-levels rising? Nils-Axel Mörner documents a decided lack of rising seas” and work your way through the links to other posts and papers.  Nils-Axel Mörner, past head of the Paleogeophysics and Geodynamics Institute in Stockholm, Sweden and one of the world’s leading experts on the rising and falling of seas,  who has over 500 published peer-reviewed papers spread over the 70 scientific fields that ResearchGate lists as his areas of expertise,  holds to the second opinion for the most part, allowing the first to be possible.

The venerable Wiki on Mörner includes a statement that “Mörner’s claim that sea levels are not rising has been criticised for ignoring correctly calibrated satellite altimeter records all of which show that sea levels are rising.^”  The reference for this statement is a Comment published in “Global and Planetary Change 55”: “Comment on ‘Estimating future sea level change from past records’ by Nils-Axel Mörner”  by R.S. Nerem and others.

Their comment says everything that is really necessary to say about satellite sea level measurements.  [We’ll get to what Nerem et al. says in a bit.]

Note that R. Steven Nerem, of the CU Sea Level Research Group at the University of Colorado, and his group of co-authors,  waited three years to comment on Mörner’s 2004 paper (which was submitted Nov 2001, accepted  May 2003, and published in Global and Planetary Change 40 (2004).)

BIAS WARNING:  You should know that Nerem is one of the contributing author’s to this WaPo, May 2016,  piece “10 things you should know about sea level rise and how bad it could be”, in which it is blandly stated “Scientists estimate that if it warms by about 4 to 5 degrees Celsius (7.2 to 9 degrees Fahrenheit), which is projected to happen by the end of the century if we don’t act on climate change, then all the ice will eventually melt. That’s 230 feet of sea level rise.”

You should consider this when reading his comment to Mörner 2004 and whenever you see anything about sea level out of the SL Research Group @ the University of Colorado — their blog actually contains a link to this newspaper article, as if it were a scientific paper or something to be proud of.

Mörner’s 2004 paper contains the conclusions of the May 2001 INQUA Commission on Sea Level Changes and Coastal Evolution,  which were:

“When we (the INQUA Commission on Sea Level Changes and Coastal Evolution) consider past records, recorded variability, causational processes involved and the last centuries’ data (Figs. 1, 2 and 4), our best estimate of possible future sea level changes is + 10 ± 10 cm in a century [4 to 8 inches] [or 4 inches per century (10cm) with possibilities down to zero or up to 8 inches] or, maybe, even +5 ± 15 cm [2 to 8 inches] [2 inches per century with the possibility of minus 4 inches to plus 8 inches].” [h/t Bob Shapiro for pointing out my conversion errors – kh]

Mörner’s Figures 1, 2 and 4:

Nerem et al. basically say:  “But,  but,  but…..”  :

“Mörner was apparently oblivious to the corrections that must be made to the “raw” altimeter data in order to make correct use of the data.

 Satellite altimetry is somewhat unique in that many adjustments must be made to the raw range measurements to account for atmospheric delays (ionosphere, troposphere), ocean tides, variations in wave height (which can bias how the altimeter measures sea level), and a variety of other effects. In addition, the sea level measurements can be affected by the method used to process the altimeter waveforms, and by the techniques and data used to compute the orbit of the satellite. Early releases of the satellite Geophysical Data Records (GDRs) often contain errors in the raw measurements, the measurement corrections, and the orbit estimates that are later corrected through an on-going calibration/ validation process defined by the T/P and Jason Science Working Team.” [in other words “ongoing ad hoc adjustments by us!”]

In reading the above, the key point is to count how many sources of error and uncertainty exist in the “raw” data.  I count at least eight.  Of course, you can see these clearly as error bars on Nermer’s current graph on his CU SLR page:

Oops, there aren’t ANY error bars, no uncertainty shown at all, not even statistics-package generated Standard Deviation fake-error-bars, despite the claim that there are lots of sources of error and uncertainty in the collection and calculation of the data.  How large are these error sources?  Nerem says there’s one that’s 1.2 mm, another that’s 5mm, the Jason technical manuals show error ranges for wave height scatter in the tens of centimeters, original measurement error in the altimetry function of Jason’s 1, 2 and 3 of +/- 3.4 cm. [Ref: Jason-3 Products Handbook, pg 10]

Note:  That’s centimeters, not millimeters — to be clear, the original measurement accuracy defined for the Jason series, including the current Jason 3 altimetry function is +/- 3.4 cm, a whole order of magnitude greater than the claimed change, delta, in the Global Average SLR.

For comparison, here’s NOAA’s latest graph of satellite SLR:

Note that the Jason-3 portion (purple, upper right) is flat — showing no significant change in sea level since the beginning of its mission.  Of note as well is the fact that NOAA describes the long-term trend with a single straight line, a simple linear trend, unchanged for 25 years, quite unlike Nerem’s acceleration curve.

The Nerem comment goes on with:

“Based on our experience with these issues, and the shape of Fig. 2 in Mörner’s paper, we believe that he used the original release of the T/P GDRs with no attempt to correct for two significant errors. One of the errors is caused by a drift in the TOPEX Microwave Radiometer (TMR). It was first observed in sea level via a comparison to tide gauges (Chambers et al., 1998; Mitchum, 1998)….  [note that these pre-1998 “tide gauges” are all uncorrected for Vertical Land Movement and only measure Local Relative Sea Level change  — thus they are not fit for comparison purposes – kh]

When care is taken to make these corrections, the rate of sea level change over the entire T/P mission is 3.0± 0.4 mm/year, 3.3 mm/year when corrected for the change in ocean volume due to glacial isostatic adjustment (Tamisiea et al., 2005).”   [the extra 0.3 mm addition for change in ocean volume is an interesting study in itself — and has nothing whatever to do with actual changes in sea surface height — kh]

Let’s try to set some of the record straight here.  The paper, Mörner (2004) was written by Nils-Axel Mörner — but its conclusions are those of the INQUA Commission on Sea Level Changes and Coastal Evolution — which consisted of 102 academics from nearly every country in the world.  [see the link listing the members and their institutions].  Nerem writes as if Mörner (2004) is the work of one sloppy misguided post-doc at a city college somewhere.

Now, I can’t blame Nerem and his cohorts for being upset — I ’m believe they have worked their entire careers, most of them, on the satellite data sets — and done their best to make sense out of them.  What they can’t seem to overcome, though, is the effect of the persistent enforced consensus view that sea levels must be rising and that they must be rising faster now that “the Earth is warming dangerously due to human activity”.   Then along comes this guy who was already world famous for his expertise on the seas and ocean dynamics when they were all still undergrads and he says “it ain’t necessarily so.”  That’s quite a blow.

Mörner, however, is rather patient with them in his 2008 response to Nerem: “Comment on comment by Nerem et al. (2007) on ‘Estimating future sea level changes from past records” by Nils-Axel Mörner (2004)’”. [DOI: 10.1016/j.gloplacha.2008.03.002 — Global and Planetary Change 62 (2008) 219–220]

“First let us clear up the origin of my Fig. 2 (in Mörner, 2004). [figure included in this essay above – kh] It is a product of the CLS Company printed in MEDIAS (2000, Fig. 1.2; also available on the net at Aviso, 2000). My curve was a redrawing of this graph. This curve, spanning the time from October 1992 to April 2000, does not record any sea level rise; only a variability plus one (or more) ENSO signals. That was the point of my picture. It should be noted that this graph includes the technical adjustments (including the drift factor of Mitchum, 2000, Fig. 10) illustrated by the lower arrow in Fig. 1.”  [ figure below – kh]    [the AVISO links in Mörner’s comment are no longer working but AVISO is currently here. — kh]

“Later the same graph [as Mörner (20040 figure 2 mentioned above] re-appears with a strong tilt (Aviso, 2003;cf. Leuliette et al., 2004; Mörner, 2005). Why is that?” [referring to the following two examples – kh]

Mörner (2008) (the reply to Nerem’s comment) goes on:

“It is because of the introduction of additional calibrations (Mitchum, 2000; Leuliette et al., 2004; Cazenave & Nerem, 2004) — and those “calibrations” are subjective interpretations (Fig. 1; upper arrow); not objective readings. Consequently, they are opinion-dependent. “We adopt the rate given by Douglas (1991,1995) of 1.8±0.1mm/yr”, Mitchum (2000) states. This rate, however, is widely debated and far from generally accepted. Especially not in the geological sea level community (cf. Mörner, 2004, Fig. 4), where we realize that the tide gauges quite frequently are installed in unstable local position, and I quote (from Mörner et al., 2004): “Tide gauge records, however, do not provide simple and straight-forward measures of regional eustatic sea level. They are often (not to say usually) dominated by the effects of local compaction and local loading subsidence. With this perspective, our multiple morphological and sedimentological records appear more reliable and conclusive” (cf. Mörner, 2007). This makes the requited “correction” for land motion at the tide gauge very delicate and subjective.”

Note that the ”calibrations” are added in on top of the assumed tide-gauge-based  —  “We adopt the rate given by Douglas (1991,1995) of 1.8±0.1mm/yr”, Mitchum (2000) states.” — pre-determined, not observed —  “ 1.8±0.1mm/yr” figure for long-term (100 year) global sea level rise.   It is this point on which the whole SLR meme rests.

All satellite sea level rise calculations start with the a priori  assumption that the global seas are rising at 1.8mm/yr as a “base rate” — based on uncorrected tide gauges that measure not global sea level changes but local relative sea level changes — all of this data uncorrected  for tide-gauge-site Vertical Land Movement and entirely unfit for this purpose.

To the raw instrumental record, Mörner and the INQUA Commission’s 102 other academics say the current method of determining Global SLR adds “interpretational calibrations” that are entirely “opinion-dependent” to first agree with the “accepted” base rate of 1.8/mm/yr and then adjusting for other interpretations of satellite based-data — making the overall result also “opinion-dependent”.

We must acknowledge that groups such as CU Seal Level Research Group are quite sure that their methods are valid and based on proper, scientifically supportable opinions — their own expert opinions based on their experience and interactions with other groups working on the same problems.  They are entitled to their opinions — but the rigors of science demand that they separate out “opinion” from “scientific observational findings” — and present them separately. None of the Sea Level Groups do so.  None of them present their data with any uncertainty estimates for the satellite based series nor any acknowledgement of original measurement uncertainty.

Why do I use Mörner’s papers as an example? 

In the controversy over Sea Level Rise most of the squabbling is about rate of change and acceleration.  But, in my opinion, the real problem lies much, much deeper in the whole approach to subject.

In Parts 1 and 2, I looked at the questions of how sea level was measured on a local basis and illustrated why Tide Gauge data (our longest available data set concerning sea levels) was unsuited to determining long-term Global Sea Level change.  I remind readers that even the most sensitive and modern acoustic Tide Gauges have an instantaneous accuracy of +/- 2 cm for each individual reading, and that that uncertainty range remains unchanged when these readings are averaged into 6-minute official records — the uncertainty still is recorded officially as  +/- 2 cm.  This uncertainty range is ten to twenty times the claimed change rate of global sea levels.  The uncertainty derives from the fact that the sea does not lie still like the water in an unoccupied bathtub.  It has ripples, waves, tides, wind chop, reflected waves, boat wakes — all of these are of differing magnitudes and frequencies and they add to and cancel out one another in chaotic patterns.  When 180 instantaneous readings are averaged (after rejecting 3-sigma outliers)  the result still can only be claimed accurate to +/- 2 cm.  Please remember, this is a direct (-ish) measurement of a single location — and still, the result cannot achieve a precision greater than +/- 2 cm (plus or minus twenty millimeters).

The data from NOAA’s CORS project illustrates that even with the highest quality GPS receivers continuously recording readings from  the latest set of Global Navigation Satellites, it takes several years of data to accurately determine the rate of vertical change in position (Vertical Land Movement – VLM),  even of these nominally stable, unmoving points.  VLM of CORS sites, many of them now purposefully being deployed on the same structures as important Tide Gauges, is of same magnitude — single digit millimeters — as the claims made for Global Sea Level change.

Now, please permit me to give a bit of anecdotal evidence.  I have lived on the seas of this planet for a full one half of my adult life [full disclosure — ~ thirty years].  On the open ocean, the surface of the sea is always in motion — it is a rare and wonderful sight to see a flat and glassy sea to the horizon — giving the illusion that it is really smooth and flat.  Even these glassy seas have ripples and waves on the scale of inches (>2.5 cm)….even when the wind will not move the wind vane at the mast-top to show wind direction (Beaufort Wind Scale 1), the ripples on the sea surface range up to 1 foot [up to 30 cm or 300 mm].  As the wind increases, more and more energy is transferred to the water surface, creating larger and larger waves of varying magnitudes and frequencies.  At the pleasant condition called a “Fresh Breeze – Force Five” — wind speeds 17-20 knots, good sailing weather for my 42-foot catamaran — long-form wave heights are 6-9 feet [2-3 meters] and “many white horses [white caps in US English] are formed; chance of some spray”.  Here’s what it looks like from the deck of a sailboat – Force 2 is “not enough wind to sail a boat this size” and Force 5 is “Let’s go!”:

Note that the smallest irregularities in the sea surfaces pictured above would be measured in inches — in Force 5 conditions, they are measured in feet and meters.

In case you might think that rough seas are the exception — they are not, they are the rule — here is an image of global wind speeds, 16 Dec 2017 (as I write):

Every square foot of the sea surface pictured above is Force 3 or greater — the upper right portion north of San Francisco is experiencing a “Fresh Gale” — Force 8  (red and brown wind flags) — winds 30-40 knots, waves 18-25 feet [5.5 to 7.5 m]. At the bottom right, there is “Gentle Breeze” blowing –Force 3 — winds 10 knots, waves 2 to 3.5 feet [.5 to 1 m].  [KTS to MPH conversion chart]

The Jason series satellites have an orbit height of 830 miles (1,336 kilometers).  Their altimetry beam hits an area measured in kilometers squared.  It hits sea surfaces that look like the images above (and worse).  It records the time it takes for the beam to bounce back and the amount of “scatter” in the returned beam.  The beam is affected by atmospheric conditions, such as humidity (and others).  The data of the Jason altimetry system is thus primarily two bits — the time for the signal to return and some datum concerning the amount of scatter (noise) in the returned signal.  Since Jason 3 has two bands of altimetry data, this data stream can be said to be doubled.  The claim is that the condition of the sea surface can be corrected for “scatter” and precisely determine the sea surface height for those conditions — truthfully, it is no better than a random guess.

This data is then corrected for “estimated” tides, the tidal waves [not to be confused with tsunamis]  that travel across the seas arriving in your local port as High Tide and Low tide.   How accurate are these estimates?  NOAA states: “On average, the heights of observed and predicted high waters are within 0.147m (0.48 ft.); low waters are within 0.135m (0.44ft.); and hourly heights are within 0.143m (0.47ft.).“   The correction, the “calibration”, for tides is done with data known to be “on average” only accurate to around half a foot [ > +/- 140 mm].

As a side-check of the accuracy of the sea height data, NOAA uses a satellite pass over Corsica and checks their Tide Gauge data (accurate only to +3.5 ± 15.0 mm and unadjusted for VLM) as a ground-truth.  [The +/- 18.5mm range is very close to NOAA’s +/- 2cm  or 20mm range for tide gauges]  As the satellite’s orbit is not perfect, it has slight irregularities, a laser-based system on the island helps to check the true orbit height to enable further calibrations.  They also side-check against floating buoys at sea equipped with GPS receivers — buoys that are rising and falling, bouncing and jouncing, on those seas pictured above — buoys, which to be accurate to the millimetric scale, would have to be stationary and continuously operating at that same exact point for several years to determine their vertical location and its change rate.  (ref:  NOAA CORS Project).

To  a realistic, logical scientific mind, knowing the accuracy of Jason series altimetry is limited to +/- 3.4 centimeters (possibly as good as +/- 2 cm), the actual physical conditions of the object of measurement (the sea surface) and the magnitudes of the other multiple uncertainties inherent in the altimetry data set — many of which are orders of magnitude greater than the original measurement uncertainty — it is apparent that the claimed precision for annual change in Global Sea Level — a incredibly more complex and complicated measurement than the VLM of a single stable CORS site — is the result of massive computation — computation  seemingly removed entirely from any careful, logical scientific consideration of the real world uncertainties involved.  This result, precise only because it is the result of digital computation far more complex than that involved in Global Average Surface  Temperature calculations, must, according to Nerem et al and other sea level teams, be “calibrated” — with “calibrations” themselves far far larger than the claimed annual change in sea level —  to agree with pre-conceived opinions on what that change “should” look like.

Nerem et al. confirm this [inadvertently]  in their Comment.

This is Computational Hubris

I refer to this as a result of:

Computational Hubris: “An inordinate and unjustified trust, or faith if you like, in the power of advanced computational machines and processes to produce highly accurate and extremely precise results from relatively inaccurate and imprecise, highly uncertain data using techniques and methods that have not been tested nor verified to be suited to, or sufficient for, the purpose.” — Kip Hansen

In the case of Sea Level measured from space, methods used to measure the altitude of stationary and relatively flat surfaces (land surface, for instance) and changes in vertical location of stationary points (CORS-like data) are being used to measure an “average” altitude of a vast area (71% of Earth’s surface) of fluid,  non-stationary, ever changing [in fact,  chaotically dynamic] ocean surface.  The tools to do so are known to be only accurate to the scale of several centimeters (inches) under ideal conditions (not actual field conditions) and the confounding conditions for those measurements, which must be corrected for, being highly uncertain on a scale of at least one order of magnitude, in some cases, such as estimates of tides, two or more orders of magnitude,  greater than the magnitude of the claimed result — furthermore, processing techniques and calibrations are opinion-dependent thus prone to being affected by the prevailing bias in the field.

We see this today in much of science — not only in climatology — but in epidemiology,  the detection of exoplanets,  scientific cosmology and international economics.

Are they cheating?

Are sea level groups like Nerem’s cheating?  Cooking the books?  Fudging the results?

No, not actually.  They are serious scientists trying to do a job and do it well.  [Though some, like Nerem,  also engage in unseemly advocacy and fanatical CAGW propaganda — as evidenced by Nerem’s Washington Post piece and his group’s Comment on Mörner’s  (2004) paper.]  The majority are doing what they have been trained to do, using the tools and data available.  They have been trained to engage in what I have call “Computational Hubris”,  and to accept its results without question.

The problem, as I see it, is as described early in the computer era by Feynman and others — a scientific field gets rolling, builds on early results, and goes off down some garden path that becomes impossible to backtrack.  No prestigious sea level group could possibly agree with Mörner and the INQUA  Commission — it would mean that everything they’ve done, everything they’ve based their careers on, was off the mark.  It is a rare (and precious) person that can do that.   The vast majority will stay comfortably in the fold of the current consensus position, with which they honestly agree.    John P. A. Ioannidis found the same situation in bio-medical research and published his findings as “Why Most Published Research Findings Are False”, concluding “for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias.” I fear that Sea Level Change research is one of those fields.

# # # # #

AUTHOR’S OPINION:  In preparing this series of three essays on Sea Level, I have reviewed more than a hundred journal articles and as many  essays on the topic — as well as technical documents of the Topex/Poseidon/Jason missions.  And, while I would be hard pressed to point to any single data source, I have come away with a general, personal opinion about today’s ongoing sea level rise.  If I were to present a Global Sea Level change graph with the minimum actual, evidence-based uncertainty ranges, it would look something like this, in general agreement with the work of Mörner and the INQUA Commission’s 2001 conclusions:

I would place the rate of change anywhere from a low of minus .023mm/yr to a high of plus 2.3 mm/yr over the last 25 years (the satellite measurement era), with the most likely rate settling in just under a millimeter per year at around 0.8mm/yr — fairly close to Mörner/INQUA’s  “+ 10 ± 10 cm [4 to 8 inches] in a century”.

I could be wrong now, but I don’t think so!  (h/t Randy Newman)  Check back with me in a hundred years and we’ll see how close I got.

# # # # #

Series Take Home Messages:

1. Overall, the seas have been rising, slowly and inexorably, since the end of the last Ice Age. In general, they will continue to do so for the foreseeable future — at somewhere between 4-12 inches [10-30cm] per century.  This rate is an imminent threat to populated areas built nominally at today’s existing sea level.
2. It does not seem that sea levels are rising dangerously or rapidly, nor is the rise accelerating, on a global scale — though our ability to measure global changes at these very small (millimetric) magnitudes is highly questionable.
3. If Morner (and his 101 colleagues of the INQUA Commission) are correct, there has not been much actual sea level rise at all over the last 25 years (the satellite altimetry era).
4. All of the above could mean that the constant drumbeat of doom regarding rising sea levels is based on the same sort of Computational Hubris that has brought us “average global temperature anomaly” in place of average global surface temperature. [ see here and here ] Or as Nils-Axel Mörner reports anecdotally “At the Moscow global warming meeting in 2005, in answer to my criticisms about this “correction,” one of the persons in the British IPCC delegation said, “We had to adjust the record, otherwise there would not be any trend.”

# # # # #

Author’s Comment Policy:

I am always happy to read your comments and to try and answer your on-topic questions.

Sea Level Rise is an ongoing Scientific Controversy.  This means that great care must be taken in reading and interpreting the past data, new studies and especially media coverage of the topic [including this series!] — bias and advocacy are rampant, opposing forces are firing repeated salvos at one another in the journals and in the press and the consensus may well simply be “an accurate measure of the prevailing bias in the field.”  (h/t John Ioannidis)

Sea Level Rise is being blatantly used as a scare tactic by advocates of Catastrophic Anthropogenic Global Warming/Climate Change in an attempt to keep the research dollars flowing and to bring about their desired social and political goals.

I’m sure I’ve written things that you agree with and things you just hate.  Let me know which is which.  Keep it civil please and if you want a reply, specifically address your comment to me “Kip….” so I am sure to see it.

# # # # #