Wednesday, March 15, 2017

A New Way of Analyzing The Depths - 3

Fig. 1 Layer Nine
In the series The Layered Approach To Big Water (The Layered Approach To Big Water, 2, 3, 4, 5, 6) we talked about the 18 layers in the WOD database (9 in each hemisphere).

Those layers have boundaries of latitude lines at their top and bottom (Fig. 1 shows Layer Nine).

That series only dealt with one way of looking at these layers, so today we begin the series that uses three views with which to contemplate each ocean layer.

Fig. 2a
And we are looking at those three views in terms of temperature and in terms of salinity. beginning with the year 1962 and ending with the year 2017 (applies to all graphs today).

And as promised, I am showing those three views in both light background and dark background (you are welcome Mark).

Fig. 2b
The first view graphed today is the "day-of-year" (DOY) view.
The DOY view is graphed from one of three SQL table types in an SQL database.

I explained the first stage in (Databases Galore - 18), containing close to a billion rows of data constructed straight from the WOD "PI format" files I downloaded from the World Ocean Database.

Fig. 3a
The second stage tables in that database have a structure (one is shown in Fig. 8) designed to condense those ~billion measurements into a condensed version (e.g. seven depth layers instead of ten thousand).

The third stage, from which the graphs are produced, are the most condensed.

Getting on with today's graphs, the first part of the DOY view (salinity) is shown in Fig. 2a and Fig. 2b.

Fig. 4a
The second part of the DOY view (temperature) is shown in Fig. 3a and Fig. 3b.

The DOY view is constructed by taking all of the WOD (CTD & PFL) measurements recorded over time and placing them into the day-of-year (1-366) format.

All measurements are placed into the DOY which they were taken, and then a mean average for that DOY is calculated.

Fig. 5a
The different view this produces, along with the other two, will often times give us clues about ocean dynamics we could have otherwise missed.

The hypothesis concerning a layered approach is that these bands within ten degree latitude lines spanning the globe are reasonably similar to be the source of a reasonable mean average.

Fig. 6a
That is to say, as professor Mitrovika indicated, "mean average covers up a lot of reality."

For example, notice how smooth the graphs at Fig. 3a and Fig. 3b are in comparison to Fig. 7a and Fig. 7b.

Fig. 7a
The salinity graphs have the same effect, as shown in Fig. 4a and Fig. 4b compared with Fig. 6a and Fig. 6b.

Remember that this is the same exact data, the same exact measurements taken from 1962 through 2017 (~5.5 decades).

Yet that same data in the DOY view, compared to the Month and Year views, gives helpfully different patterns.

I mean head-scratching helpful.

It is always a helpful challenge to explain some of these things, because we are too accustomed to the "global mean average cop-out.

The salinity dips are of particular interest (see comments to the previous post in this series here).

Fig. 8

In that case there was a sharp drop in salinity around Greenland at about the 240 day mark of the DOY graph.

One one hand it was hypothesized to have taken place because of rain intensity during that time of year.

On the other hand it was hypothesized to have taken place because the most intense melt of the Greenland Ice Sheet impacts the surface waters around Greenland at about that time of year.

Today, we have a repeat of that perplexing pattern in Layer Nine at about DOY 130 (Fig. 2a, Fig. 2b).

In addition, we have the Month view which adds spice to the inquiry (Fig. 4a, Fig. 4b).

"The rain did it" hypothesis has an edge with respect to Layer Nine, because there is no ice sheet there to ponder.

I wonder if it would be helpful to consider the Year view to be like taking a person's blood pressure, while considering the DOY and Month views to be more like taking a look at a person's genetics?

They are bound to produce different patterns, yet they are from the same person's internals.

Let's keep working these hypotheses as we traverse the 18 layers of the vast oceans in upcoming posts.

It is likely that we will see more of this, and it will enhance our ability to falsify or establish the validity of the hypotheses.

There is always the chance that both hypotheses are valid, but not at the same zone or layer.

Where there is no ice sheet near enough to impact local salinity that can be ruled out.

Likewise, remembering that the ice sheets of Greenland and Antarctica are classified as deserts, in terms of annual precipitation, rain or snow are less suspicious than melt is.

Think about it to prepare for the other layers that are in the works.

The next post in this series is here, the previous post in this series is here.

1 comment:

  1. Good stuff Dredd! That mean average tells 'a' story but not the 'whole' story. Analogous to the Calculus as t > 0. Now pondering the salinity spike-hope its from evaporation? Thanks for dark backgrounds-they look more 'commanding'.