Measuring Sea Level Is a Suspect Art

A catastrophic rise in sea level is one of the calamities anticipated by those who believe in climate change.  The believers and the skeptics can carry on a scientific discussion about this matter only if they actually know what the current sea level is and how it has been behaving.  Both sides tend to accept that data published by NASA and NOAA regarding measured year-to-year changes in sea level fit the bill.  But are those numbers accurate and scientifically credible?  Conceptual problems associated with identifying sea level combined with the technological limitations on how we go about taking measurements make that unlikely.

NOAA and NASA claim that sea level is currently rising about an eighth of an inch per year.  This is more or less equivalent to the stack height of just two quarters.  However, neither of the two measurement systems upon which estimates of sea level rely can measure so precisely.  The two measurement systems are tide gauges in coastal areas and satellite pings over the oceans.  Neither can record measurements of sea level to a level of precision any better than a few inches.

This problem of instrumental crudity is overcome by assuming that measurement imprecision is unbiased in its distribution and that a large number of measurements taken in a very short span of time can be averaged together to yield an estimate of sea level that is both more precise and more accurate than the actual data obtained from the individual instruments.

With this as introduction, consider some of the problems associated with measuring sea level.  First, the sea does not lie flat.  It never lies flat.  Waves or swells constantly vary the height of the sea surface, and that variation typically exceeds a foot or two at least a handful of times every minute.  Even when the sea appears to be calm, the subtle undulations of passing swells work the same magic.  But usually the sea is not calm, and actual sea level height at any particular point is unpredictable.

Tides are marginally more predictable, but they change the level of the sea perhaps twice as much on average as do waves and swells.  They do so at a slower pace – alternating between peaks and valleys only once or twice a day – but when trying to ascertain the global average sea level for a year, this is still an enormous amount of noise to filter out of the computation. 

The practice of averaging multiple measurements to improve instrumental precision may be appropriate, but it is not appropriate when scientists use the same approach for canceling out the effects of waves and swells and tides.  A computed average (mean) deserves credibility only if the data values are distributed normally (so as to describe the well known bell-shaped curve).  But waves and swells and tides clearly do not vary sea level according to this limitation.  Instead, they linger at the extremes (e.g., high and low tides) and quickly pass by the middle range, where the computed average is bound to fall. 

A second problem for measuring sea level is that there is no way to determine whether it is the same for different locations.  One theoretically logical way of doing so would be to express it as a specified distance from the exact center of the Earth.  But the earth is not a perfect sphere – its diameter at the equator is thirteen miles greater than its diameter measured from pole to pole – and even as an oblate spheroid the earth is less than perfect since the northern and southern hemispheres are not mirror images.

This line of abstract thought may seem unproductive, but to pursue it brings one to the realization that absolute sea level at one location is hard to compare to that of another location.  And yet, if we are to know the global average sea level, do we not need to have some credible method for making such comparisons?

A third measurement problem is that the Earth's land masses are not stationary.  Geologists believe that the Earth's crust floats on a superheated liquefied rock known as the mantle.  The viscosity of the mantle is much greater than that of water but nonetheless retains enough fluidity that the relatively rigid crust floats on it.  This means that a local alteration in the weight of the crust will cause it to float higher or settle deeper in the underlying mantle.  To complicate things, the relative rigidity of the crust means that a shift up or down in one locale inevitably triggers an opposing vertical movement in peripheral areas.

When the northern half of North America lost all its ice at the end of the last glacial period around 12,000 years ago, the weight of the underlying crust was greatly diminished.  The affected region buoyed upward many hundreds of feet over the course of the ensuing millennia.  The adjustment continues even today, although at a much slower rate than it did early on.  This is an example (albeit a dramatic one) of a process that is still ongoing in much of Europe and North America.  Consider what it means for measuring sea level. 

But ice and its disappearance are not the only way that the thickness of the crust gets altered.  In many regions – both on land and under the sea – the crust is being thickened by the forces that make mountains, and everywhere on land weathering and erosion continually operate to grade the continent back down to sea level.  In the process, the removed overburden gets transported some distance and ends up back in the ocean.  In short, the forces of geomorphology are constantly altering the thickness of the Earth's crust most everywhere. 

These additions to and subtractions from the thickness of the crust tend to concentrate in coastal areas, where the question of sea level has its relevance.  Also, the forces of uplift and mountain building can and do occur in the relatively thin crust beneath the sea, so a reconfiguration of the ocean bottom may cause a rise in sea level not due to an increased water volume.

Vertical movements of the crust do occur, and there is no system in place that can measure their direction or magnitude.  In short, even if we can very precisely measure the changing relationship between a coastline and the ocean, we usually cannot evaluate how much of that change is due to a rise or fall of the ocean and how much is due to vertical movement of the crust.

These complications (and many others) challenge the credibility of any numbers claiming to tabulate the exact amount of annual sea level change.  Those official numbers may be correct, but don't bet the farm on it.  And even if they are correct, there is no assurance that they indicate a changing amount of water in the ocean.