Guest essay by Philip Lloyd
There is a general belief that the observed sea level rise of about 1.5mm per year (according to tide gauges) and 3mm per year (from satellite measurements) is partly due to heating of the oceans and their consequent expansion.
If, however, you look at the estimates of sea temperatures, you soon find things that don’t add up. For instance, the latest IPCC Assessment Report gives:
Five different data sets agreeing so well? Miracles happen, but this seems unlikely. A bit more digging, and you discover that most of the pre-1960 data was taken by bucket sampling. It was found that the sample cooled by about 0.2oC due to evaporation, so the data were adjusted upwards. From 1960 to the 1980s, the sample was water drawn in to cool the ship’s engines, but that was found to be warmed by conduction by about 0.6oC, so it was adjusted downwards. Finally we settled on electronic thermometers carried by buoys, with modern versions able to take depth profiles. So the five data sets are the result of lots of adjustments, which perhaps explains why they agree so well.
You will notice that satellite measurements are absent. The reason is that satellite instruments can only measure the temperature of the top few millimetres of the ocean’s surface. When it is calm, the sun can heat the water to over 40oC, when a metre deeper it is 30oC, so the satellite reads high. When it is stormy, cold water mixes with the surface water, and the satellite reads low.
This means it is essential to record the depth at which the temperature is measured. Over the upper 100m, the temperature drops quite rapidly with depth, then drops more slowly. The profile changes with the season, a change which is more pronounced in the higher latitudes. From about 500m the temperature drops more sharply with depth, but the profile is reasonably constant with latitude. It reaches a minimum of close to 0oC below about 1 500m.
The IPCC chart of average ocean temperatures does not reflect the depth at which the “surface temperature” is measured, so this is a further reason to be suspicious. But even if we take the chart at face value, does it look as if it could be a major driver of sea level?
To test this hypothesis, I looked at tide gauge records. A typical one is from the German port of Wismar; monthly data are continuous from 1848:
The linear regression on the raw data indicates an average rate of increase of 1.42±0.08mm per year. Shown in red is the linear regression for the period 1910 to 1950, and in yellow for the period 1980 to 2015, both of which are barely distinguishable from the long-term regression shown by the black line.
Yet the IPCC tells us that the oceans were warming quite rapidly during these periods. If so, there should have been some acceleration in the measured rate of rise of the sea level. None was observed.
A recent study, based upon the solubility of rare gas isotopes, suggests that the oceans have warmed by less than 0.1oC in the past 50 years. The evidence against increasing global temperatures being the cause of sea level rises is growing.
 Scripps Institution of Oceanography 2018. New study identifies thermometer for past global ocean temperatures. https://scripps.ucsd.edu/news/new-study-identifies-thermometer-past-global-ocean and Nature 553, 39–44 (04 January 2018) doi:10.1038/nature25152