How sure can we be that global warming is an existential threat?

During her 2019 address to the United Nations, Greta Thunberg spoke of a looming biological catastrophe: ‘Entire ecosystems are collapsing. We are in the beginning of a mass extinction.’

In July 2023, UN Secretary-General António Guterres made headlines by declaring: ‘The era of global warming has ended; the era of global boiling has arrived.’

How seriously should we take these extreme claims?

Let’s be clear about one thing: the true driver of the Earth’s climate is the world’s ocean system. While extreme air temperatures and other weather events dominate media headlines about climate change, it is actually the oceans that control long-term temperature changes.

The ocean’s domination of climate is evident from a comparison of the local climates of Alice Springs and Norfolk Island, which are similar distances south of the equator.

The inland climate of Alice Springs suffers huge temperature ranges: from oppressive daytime heat in summer to freezing overnight cold in winter. The daily temperature range, of over 17°C superimposed on a similar seasonal range, gives average daytime summer maximum temperatures of about 37°C and average overnight winter minimums of only 4°C.

On the other hand, the oceanic climate of Norfolk Island is mild all year, with average daily maximums matching the local sea surface temperatures, ranging from about 24°C in summer to 19°C in winter. Overnight minimums are only a few degrees lower. We need to look at ocean temperatures for clues about global warming.

Global ocean surface temperature trends based on currently accepted data are available from the National Oceanic and Atmosphere Administration, National Centres for Environmental Information, and Climate at a Glance Global Time Series from 1850 to 2026. Temperature trends (or anomalies) are shown as departures from the average temperature over the century from 1901 to 2000.

In the ocean-dominated southern hemisphere, the ocean surface temperature anomaly has risen in recent years to about 0.7°C above the long-term average. In the land-dominated northern hemisphere the rise is a little greater.

We need to bear in mind that, over the past 175 years of the available temperature records, different instruments and techniques have been used in different periods.

From 1850 to about 1940, sailors measured ocean water temperature by placing a thermometer in a bucket of water collected from over the side of the boat. From about 1940 to 1980, temperatures of intake water in ship engine cooling systems were measured automatically. Since about 1980, satellites have measured ‘skin’ temperatures (within a fraction of a millimetre below the surface) using infrared and microwave sensors.

Since about 2000, temperatures now accepted as the ‘gold standard’ are measured at a depth of about 1 metre with precision instruments in floating buoys.

The accuracy of temperature measurements is limited by various factors.

Firstly, the coverage of the world’s oceans varies greatly. Bucket and engine intake temperatures were recorded only along major shipping routes, with plentiful observations in some areas while other areas remained unobserved. Buoys provide extensive but sparse coverage, with some 1,500 buoys for 360 million square kilometres of global oceans – one buoy for an ocean area the size of Victoria. Only satellite measurements provide comprehensive coverage of the whole globe.

Secondly, different techniques measure temperatures at different water depths. This matters because of the thermocline: the water temperature typically decreases with depth. At a ship’s intake depth (5 to 10 metres) the water temperature can be up to 3°C cooler than at the standard 1 metre depth. Moreover, the thermocline temperature difference varies with time of day and geographical location.

Satellites measure the ‘skin’ temperature in the top 10 micrometres, which can be cooler than at 1 metre depth by up to about 0.2°C. This is due to evaporation, which depends on wind speed and other factors. Correction of historical ship intake measurements and recent satellite values depends on numerous assumptions, that could leave residual uncertainties of some 0.5°C.

Thirdly, the different measurement techniques and instruments have inherent uncertainties. Canvas or wooden bucket samples of water brought on deck are generally cooler than water at 1 metre depth due to evaporative cooling. However, such samples can also be warmer due to solar heating or other effects. The resulting uncertainties can be up to 0.5°C.

Engine intake temperatures are generally warmer than the surrounding water, due to proximity of the engine. And the variation between different ships results in uncertainties that can be as much as 1.8°C.

Satellite measurements of ocean surface (‘skin’) temperatures since the 1980s have the advantage of providing extensive coverage of the world’s oceans. But they are subject to atmospheric interference, sensor degradation and calibration issues, which lead to temperature uncertainties of about 0.5°C. Satellite measurements near buoys can be adjusted to match the buoy temperatures. But uncertainties remain in other locations.

Much oceanographic research has been directed towards identifying predictable biases in ocean surface temperature measurements, so that corrections can be applied to remove these errors. However, significant uncertainties remain due to natural unpredictable variations in oceanic properties and behaviour. Inherent instrumental limitations contribute further uncertainties.

The combined effect of all these uncertainties appears to be between about 0.5 and 1.0°C. The reported rise in ocean surface temperature anomalies above the long-term average of 0.7 °C lies within the range of these uncertainties. But, if the biases have been corrected, and the residual variations are random, averaging over many measurements can reduce the uncertainties. The reported average rise would then be significant – and real.

Less widely reported is the fall in ocean surface temperature anomalies of about 0.3°C from 1850 to about 1910. Falling temperatures, increasing arctic ice cover and unusual weather led to concerns in the early 1970s that another ice age was imminent. In 1974, Time magazine ran a story titled, Another Ice Age? But it was not to be, for temperatures rose again.

European history over the past two millennia shows warm periods and cold periods. Perhaps the fall and rise in ocean surface temperature over the past 170 years is part of another slow thermal cycle.

However, it may take another century for us to be sure.

Dr David Phillips is a former research scientist and founder of FamilyVoice Australia