Explainer: How tropical oceans and El Niños heat the land worldwide

A robust feature of global warming is that the surface of the land warms more than the surface of the oceans. We call this the land/ocean temperature contrast.
The reason is partly because the oceans are slow to warm up compared to land; the oceans have a much larger heat capacity than land which means lots more energy is required to change their temperature. But that’s not the whole story.
If we were to warm earth’s oceans by \(1^{\circ}\)C and then stop (in our real world “experiment” we haven’t stopped yet), the land would warm by about \(1.5^{\circ}\)C and then stay warmer than the oceans, even after the ocean has ’caught up’ to the land. This is due to the relationship between the surface and the atmosphere above it; the air in the few kilometers above land is much dryer than the air above oceans. This dryer air is more sensitive to change so its temperature increases more than the air above oceans, leading to a land/ocean temperature contrast. The land and ocean surface temperatures have a similar relationship in the warming and cooling of year-to-year variability.

Observed temperature change in the 20C. Note the increased warming over the continents. Figure from (Compo 2008)

From one year to the next, the global temperature fluctuates up and down around an average temperature. If we compare the fluctuations of land surface temperatures to the fluctuations of the ocean surface temperatures, we see that the land temperature fluctuates more than the ocean. Why? We might expect that the oceans are slower to respond to changes than the land. This is certainly part of the story but we can do an experiment to test how important this is.

In our experiment, we take a global climate model and we specify the ocean surface temperatures; we make them vary up and down each year and see how the land responds. The results show that the land varies more than the ocean temperatures. The ocean variability – which we have input into the model – is being amplified over the land. Interesting! But let’s pull it apart a bit more.

We know that the El Niño-Southern Oscillation (ENSO) is the main source of year to year temperature variability in the oceans and atmosphere. Through changes in winds and ocean currents it has the ability take up or release heat to the atmosphere. It doesn’t really change the long-term temperature but can make one year hotter (El Niño) or cooler (La Niña) than the next. ENSO occurs in the tropical Pacific but is strong enough to have a global influence.
We can take our climate model again but instead of assigning the whole ocean temperature we just put in a “fake” El-Niño/La-Niña. We make the tropical Pacific ocean temperatures go up and down - over a period of four years - and let the rest of the ocean, and the land, respond. The period of four years is related to the actual period ENSO, although the real world is much less regular than our model.
In this experiment, the global land surface temperature still responds with amplified variability relative to the ocean; if the ocean surface temperature increases or decreases by \(1^{\circ}\) the land temperature will increase or decrease by almost \(1.5^{\circ}\). Essentially, we can control the global land temperature by changing the temperature of the tropical Pacific ocean.