Wednesday, July 12, 2017

On Thermal Expansion & Thermal Contraction - 21

Fig. 1 WOD Zone map
I. Background

Water temperature and heat content in bodies of water can be mentioned or used by writers in a manner that does not connote the actual dynamics taking place.

In other words, misconceptions can be unintentionally created, whether wanted or not.

So, as regular readers know, I have been trying to clarify some of the greater and more widespread misconceptions that have been generated by research practices that only skim the surface.

In that pursuit, I recently reprocessed my partial copy (CTD & PFL datasets) of the World Ocean Database (WOD) again, in order to make it even more useful.

The exercise I went through was to convert tables in Appendix 11 in the WOD Manual (pages 132 - 137; PDF pages 142  - 148; WOD Manual PDF Link) into SQL tables which I could then access from various software modules.

The manual explains that appendix as follows:

The range values provided has range values for temperature, salinity, oxygen, phosphate, silicate, nitrate, pH, chlorophyll, and alkalinity. The range values in
Fig. 2 World Ocean Areas
the tables are used to help identify the most obvious questionable values for these variables. Please note that ranges are given on 33 standard levels (+ one for depths deeper than 5500 m). All standard depths in between given standard depths have the same values as the nearest standard depth shown (for example, 90m standard depth uses 100m range values. If a standard depth is equidistance between two shown standard depths, the ranges values will be the same as the shallower shown standard depth (i.e. 5 m range values will be the same as 0 m shown values, not 10 m shown values).
(ibid). As you can see, the different ocean areas have different maximum and minimum values per se, and further, those maximum and minimum values change with depth.

II. Conditions

In my work, I only use temperature, salinity, and sea pressure values from the WOD (excluding "oxygen, phosphate, silicate, nitrate, pH, chlorophyll, and alkalinity" values).

Fig. 3
One reason I do so is that I am mainly exploring the hypothesis that "thermal expansion is the main cause of sea level rise for the past 100 years" or so (On Thermal Expansion & Thermal Contraction, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19).
Fig. 4

Fig. 5
That hypothesis is based mainly on ocean temperature, ocean salinity, and sea pressure at relevant depths.
Fig. 6

Fig. 7
The conventional application of that naive hypothesis is that "water expands when heated."

That is not absolutely true, because water also contracts (shrinks) when heated, depending on the temperature of the water at the time the heat is applied (ibid).
Fig. 8

More importantly, water also contracts (shrinks) when heat leaves that water (again depending on temperature of the water at the time the heat is removed).

III. Application of the Tables

Prior to having those tables in an SQL table, the only way I had to analyze the validity of measurements of temperature and salinity was to reject measurements which the general error flags offered.

If the data package had any error condition other than zero (no errors) I rejected the measurements outright.

Now, I can use measurements regardless of useless errors ("the scientist is ugly") by seeing if the measurement is within the maximum and minimum parameters of the ocean area where the measurements were taken.

In addition to that, I can conform the measurement to the maximum or minimum.

If it is too high I can modify it to the maximum, or if it is too low of  a value, I can modify it to the minimum, and thereby preserve some of the valuable measurement data that way.

IV. Complications

The zone map at Fig. 1 can lead one to think "this will be simple because the zones all look like similar square thingys."

Fig. 9
Take a look at Fig. 2 and compare it to the Appendix 11 values and you realize at once how different they all are, and how difficult it is when a billion records are screaming past, to determine exactly which ocean a zone is in (and other difficulties).

So, I reprocessed all of the nearly one billion measurements, applying the maximum and minimum value constraints on the temperature and salinity values if needed.

Interestingly, the PFL datasets (produced by ARGO submarine-like machines) were remarkably accurate, and the CTD measurements were not bad either (overall, less than 1% needed modification to conform to the Appendix 11 restrictions).

V. The World According To Measurements

Another concern I had was with the TEOS-10 toolbox which I have begun to use (Golden 23 Zones Meet TEOS-10).

Some may think that because that toolbox conforms measurements to well studied conclusions, the measurements are being ignored.

So, the module I wrote uses all PSMSL tide gauge station records, and all WOD zone records (not limited to the "golden 23"), and keeps track of the in situ measurements separate from the TEOS calculated values.

Compare the in situ measurement graph for temperature (Fig. 7) with the TEOS-10 toolbox generated "conservative temperature" (Fig. 4) and notice that they have no major discernible differences.

But the Absolute Salinity graph (Fig. 8) compared with the in situ measurements of salinity (Fig. 5) does reveal discernible differences.

That is no surprise in the sense that the previous 40 years of measuring salinity in ocean water used a less reliable technique (TEOS-10 Home).

VI. The Other Graphs

The graphs, today, that bear most upon my argument against the "thermal expansion is the main cause of sea level rise for the past 100 years," are Fig. 3, Fig. 6, and Fig. 9.

The calculation of thermosteric ocean volume change (Fig. 9) compared with tide gauge records of sea level change (Fig. 6) are at odds with the thermal expansion hypothesis.

Putting them all on one 4-panel graph (Fig. 3) further supports the notion that the thermal expansion hypothesis is at odds with the world of measurements.

The graphs show that. during the same 1967-2016 time frame. sea level was rising as thermosteric volume fell.

VII. Conclusion

The Larson C ice-shelf calving is closer to the truth than the thermal expansion hypothesis is.

That is, the Greenland, Antarctic, and other ice concentrations in the Cryosphere are and have been the single greatest source of sea level change from day one of climate change caused by global warming.

"The ice is melting, the ice is melting."

The previous post in this series is here.


  1. The June 5 discussion with spherical volume representations of Earth's water ( fresh and salt) has been haunting me since seeing it the first time.

    Freshwater is just 2.5% of all Earth's water and of that:

    ~68.7% of Earth's fresh water is ice and the remainder ~30.1% is in the ground.
    The very tiniest amount of freshwater (1.2%) is on the surface trying to support life.

    Most I've talked with have NFI about the above but when its examined, only a fool ( or paid activist) would suggest that rising sea levels could possibly be caused by ANYTHING other than the surface ice.

    Keep up the great work Dredd! Many thanks for sharing the knowledge.

    (Note: this is a very effective teaching tool and nearly in the same league for challenging convention as the Copernican revolution.)

    1. Mark,

      Indeed, there is more work to be done when researchers get wind of this and therafter respond.

      Your input, as well as input from other regular readers, will be as important then as it is now.

      I can't see every angle and detail alone, so I look forward to your ongoing input, should we ever strike up a useful back and forth with them.

      In at least one sense, the delay in response to this challenge to the TE hypothesis is understandable.

      I use nearly 1 billion measurements, which if a researcher reads one each second, it would take about 192.9 days to process just that one type of cursory glance.

      Not to mention comparing them to the Appendix 11 maximum minimum values, then combining it into graphs and writings.

      So, some of them will be at a disadvantage, at least to the extent that they cannot "create assistants" (specialized software modules, not models) to analyze millions of measurements per second.

      I am willing to be patient, in that sense, because they have an added burden of communicating with a programmer and transferring one type of scientific nomenclature to another via English or some other non-programming language.

      I don't have that disadvantage because I am a professional software developer (retired).

      Nevertheless, I am going to continue to study the matter and report results to the Dredd Blog readership.

      Thank you.

  2. "Scientific reticence" has its consequences says New Yorker Magazine link