I reread Church, White (2011) (PDF) in light of the post concerning the inconsistent calculations of thermosteric sea level change (thermal expansion / contraction).
Like others who were mystified by "the European problem" they flounder, because they do not mention, let alone understand, the gravity of ice sheets (The Gravity of Sea Level Change, 2, 3, 4).
Therefore, neither do they comprehend the Dredd Blog discussions of the phenomenon of ghost water (The Ghost-Water Constant, 2, 3, 4, 5, 6, 7, 8, 9).
Today I want to fulfill the promise I made in a recent post:
"In future posts I will use the same PSMSL tide gauge stations that the authors in journal papers used in their papers, in order to further expand upon the concepts addressed in today's post."
(On The More Robust Sea Level Computation Techniques). In the paper Church, White (2011) they used 491 PSMSL tide gauge station data ("We use ... data ... from ... PSMSL" p. 587), so, that seems like a tall order.
Not to worry, Fig. 1a and Fig. 1b are graphs using the same 491 PSMSL tide gauge stations that they used, excluding "metric" data (I don't use the "Metric" data, as recommended by PSMSL here).
III. The Same Old Story Emerges
Their selection of PSMSL data does nothing to change the reality that if you do not discuss ice sheet gravity dynamics, you don't get it (see Mitrovica video below).
The graphs I provided here (where I made, and today fulfilled, a promise) shout out the same message as today's graphs do (Fig. 1a, Fig. 1b).
That message is the message shouted out for all that consider the four corners of the scenario, which is to say that the assertion indicating "thermal expansion is the main cause of sea level rise in the 19th and 20th centuries" is not supported by robust analysis (On Thermal Expansion & Thermal Contraction, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21).
IV. Conclusion
According to Dr. Mitrovica, a lot of the fault for the problem discussed in this series is the obsession with the mythical bathtub model (The Bathtub Model Doesn't Hold Water, 2, 3, 4).
The next post in this series is here, the previous post in this series is here.
Professor Jerry Mitrovica, Harvard University:
The scientific literature contains some debate. or at least different methodologies, concerning the proper way to calculate the quantity of thermal expansion and contraction of ocean water.
Which means that you can expect, based on the same evidence, different statements about the portion of sea level rise or fall that is considered to be thermosteric (thermal expansion / contraction) in nature, compared to what is considered to be eustatic in nature (scientific literature reflects that unwanted reality).
II. Some Clarification
Thermosteric warming (thermal expansion) does not add atoms to the ocean water, but it does move the atoms further apart from one another, so, the ocean volume increases (even though the number of atoms remains the same).
Thermosteric cooling (thermal contraction) draws the atoms closer together, so, the ocean volume decreases (even though the number of atoms remains the same).
Mass-volume (eustatic) increase is a different dynamic, because it is a function of adding more atoms to the ocean water (via ice berg calving, or by melt water from ice sheets or glaciers on land flowing into the ocean).
Mass-volume (eustatic) decrease is caused, among other things, by evaporation of ocean water into the atmosphere, and the eventual placement of that water on land by rain.
III. Some Variations In Values
I won't belabor the issue of variation of analysis of steric vs eustatic in the published literature (because there is a lot of it), but I will quote from two papers which are at odds as to "who dunnit" (steric man or eustatic man):
"We examine the relationship between 50-year-long records of global sea level (GSL) calculated from 1023 tide gauge stations and global ocean heat
Fig. 1a All Zones, Constant Volume
content (GOHC), glacier and ice sheet melting. The lack of consistent correlation between changes in GOHC and GSL during the period 1955–2003 argues against GOHC being the dominant factor in GSL as is often thought. [IOW eustatic man dunnit] We provide clear evidence of the substantial and increasing role in GSL from the eustatic component (47%) compared with the contribution from increasing heat content (25%), suggesting that the primary role is being played by the melting glaciers and ice sheets. There remains about 1/4 of GSL rise unaccounted for by the best estimates of both eustatic and thermosteric effects[BTW that is the ghost water constant]. This fraction also exhibits large variability that is not readily associated with known causes of sea level variability. The most likely explanation of this unknown fraction is underestimated melting, climate-driven changes in terrestrial storage components, and decadal timescale variability in global water cycle. This argues for a concerted effort to quantify changes in these reservoirs" (JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, D08105, doi:10.1029/2007JD009208, 2008, PDF). ...
"Over the years 2002–2014, we find a global mean steric trend of 1.38 ± 0.16 mm/y, compared with a total trend of 2.74 ± 0.58 mm/y[steric man dunnit]. This is
Fig. 1b G23 Zones, Constant Volume
significantly larger than steric trends derived from in situ temperature/salinity profiles and models which range from 0.66 ± 0.2 to 0.94 ± 0.1 mm/y. Mass contributions from ice sheets and glaciers (1.37 ± 0.09 mm/y, accelerating with 0.03 ± 0.02 mm/y2) are offset by a negative hydrological component (−0.29 ± 0.26 mm/y). The combined mass rate (1.08 ± 0.3 mm/y) is smaller than previous GRACE estimates (up to 2 mm/y), but it is consistent with the sum of individual contributions (ice sheets, glaciers, and hydrology) found in literature" (Revisiting the contemporary sea-level budget on global and regional scales, 2015, PDF).
The ongoing exercise to unify the research into which factors contribute most to ocean volume change (steric or eustatic), must be bolstered with unification of both research and analysis.
One effort in that direction is the TEOS-10 toolkit, which I use.
But, the major factor in unification will be to unify the conceptual framework IMO.
IV. Some Procedural Inconsistencies
One paper expands upon the proper techniques and procedures involved in steric vs eustatic analysis:
"A common practice in sea level research is to analyze separately the variability of the steric and mass components of sea level. However, there are conceptual and practical issues that have sometimes been misinterpreted, leading to erroneous and contradictory conclusions on regional sea level variability. The crucial point to be noted is that the steric component does not account for volume changes but does for volume changes per mass unit (i.e., density changes). This indicates that the steric component only represents actual volume changes when the mass of the considered water body remains constant."
(JOURNAL OF GEOPHYSICAL RESEARCH: OCEANS, VOL. 118, 953–963, doi:10.1002/jgrc.20060, by Gabriel Jordà and Damià Gomis, 2013; @p. 953, 954, emphasis added). One way to remember mass volume compared to steric / spatial volume is that mass volume is how many atoms the water column contains, but steric / spatial volume refers to how far apart from one another those atoms are.
V. A Dredd Blog Solution
In light of ("the steric component only represents actual volume changes when the mass of the considered water body remains constant") I decided to modify the software module to calculate both situations.
Fig. 2a All Zones, Variable Volume
That is, to calculate based on both the mass remaining constant, as well as the mass quantity changing.
(Of course the mass is constantly changing in the real world, but I digress.)
I decided to do both the constant scenario and the variable scenario because it would be helpful for detecting situations where presentations are at odds as a result of the use of these two different techniques as if they were the same.
Fig. 2b G23 Zones, Variable Volume
Notice Fig. 1a and Fig. 1b, which graph the situations in both the Golden 23 as well as All Zones as to the situation where the mass remains constant.
They both use the constant mass volume technique.
Then compare those two with Fig. 2a and Fig. 2b, which use the variable mass volume technique to graph those situations in both the Golden 23 as well as All Zones.
VI. Conclusion
The situation where the mass remains constant generates less thermal expansion and contraction than the variable mass situation does.
The constant mass is more accurate, in terms of calculating actual thermosteric volume change, than the variable volume scenario is.
That is because using the variable volume assumes a volume amount that has been increased or decreased because of an increase or decrease in the total amount of water in the oceans, rather than being based only on the temperature and salinity changes.
Thermosteric change can only be isolated by using a fixed quantity of water that experiences changing water temperatures.
Using the varying mass-volume quantity each year will mix steric with eustatic to produce a conflated thermal expansion analysis.
It is, in that respect, the same as using a combination of zones that are biased toward either sea level fall or sea level rise.
One must choose both carefully with an unswerving goal of having a balanced input of data.
In future posts I will use the same PSMSL tide gauge stations that the authors in journal papers used in their papers, in order to further expand upon the concepts addressed in today's post.
The PSMSL recently updated their down-loadable datasets (both monthly and annual datasets were offered), so I updated my SQL database to include those new tide gauge station records.
Fig. 1a
At the same time I adjusted the date bench-mark for the volume of the oceans to 2010 from 2000, and calibrated my relevant software modules to:
I also ran across a definition of "thermal expansion", a term we see in sea level related text around the Internet:
"The 'thermosteric component of sea level change' represents the change of sea level due to warming or cooling of a column of sea water. Warming of a sea water column results in higher sea level and cooling of a sea water column results in lower sea level."
(Definition of Thermosteric, Page 3, PDF). That was refreshing because we very rarely come across the mention of "oh by the way, when that warmed water cools it shrinks").
I also ran across a bit of information which I have not really concerned myself with to tell you the truth:
"Authors have typically achieved higher levels of education than the average reading level and tend to write at the same reading level as other authors in their niche. So where does that leave the actual reader? According to many reports (including the U.S. National Center for Education Statistics’ 1992 Adult Literacy survey), the average reading level is the 7th or 8th grade. Combine that with reports of increasingly low-attention spans of Internet users who require even milder language and you’re looking at a reading level of the 6th or 7th grade."
I have full confidence that if regular readers (who are quite savvy) want clarification on any issue, they know how to get it (including "what the heck did you mean by that Dredd?").
II. Using The Updates
What is important in the context of thermal expansion/contraction (thermosteric volume change) is the original volume of the ocean at the beginning of the calculation sequence.
In this context, that date is 1880.
I conformed all the data (GISS, PSMSL, and WOD) to the year 1880 as the beginning date.
Since PSMSL and GISS have in situ measurements for that year, but WOD does not, I had to calculate the values, then project them into the past.
That exercise must be applied to generate the ocean mass-volume in 1880 as well.
To do that I first isolated the mass-volume value described in the paper linked-to in Section I. above (1.332370930 x 109 km3), together with the year the mass-volume value was calculated (2010).
Next, I isolated the global mean sea level in that year, as measured by "the PSMSL golden 23 tide gauge stations."
With those values in hand it was easy to calculate backwards from 2010.
If the mean sea level fell or rose in any year I calculated the percentage of change in the PSMSL sea level, then adjusted the ocean mass-volume up or down based on the sea level change percent (5% drop in sea level = 5% drop in ocean mass-volume; 7% increase in sea level = 7% increase in mass-volume).
I proceeded until I (the software module) reached the year 1880, at which time I saved the value and used it for TEOS thermal expansion calculations going forward (Golden 23 Zones Meet TEOS-10).
The same technique is used to calculate historical ocean temperatures back to 1880, except GISS temperatures are used as the guide instead of PSMSL values.
Instead, I use the percent of GISS temperature value changes, then apply 93% of that GISS change to the WOD values.
The WOD values for CTD and PFL datasets are robust beginning circa 1967, so I start with the GISS surface temperature changes, and use 93% of that change.
That is because the current thinking is that about 93% of the heat from global warming ends up in the oceans.
III. Graphs Generated From The Updates
There isn't much change in the graphs generated after implementing the updates.
You can compare Fig. 1a and Fig. 1b with graphs generated prior to these changes (see The World According To Anomalies - 2, at Fig. 2 and Fig. 3).
However, there is a lot of difference between these recent thermal expansion/contraction (thermosteric) graphs and older ones, when I first wrestled with learning to use TEOS-10 and earlier formulas.
IV. Conclusion
I have a handle on it now, at least as reasonable a handle as the established scientific community has.
So, relax and don't expect changes in the way that is calculated, or in the results (unless readers point out an error in that process).
