Second warmest July ends a 13-month record streak

But global temperatures remain far higher than any year prior to 2023

Global surface temperatures have been on a tear since June 2023, shattering prior records by hereto unforeseen margins (0.3C to 0.5C above prior monthly records) in the latter half of 2023 and more modestly (0.1C to 0.2C) so far in 2024. All in all, we saw 13 new records set from June 2023 to June 2024, the second longest streak of new monthly records in the modern era (after the super El Nino event of 2016/2017).

Margin by which each month from June 2023 to June 2024 beat the prior monthly record in the ERA5 dataset. Figure from Carbon Brief.

However, the remarkable extent of the records set in summer 2023 set the bar was so high that it is becoming increasingly difficult for 2024 to set new records – particularly as El Nino conditions fade and the world potentially moves toward La Nina.¹

With 30 of the 31 days now available from the Copernicus/ECMWF ERA5 reanalysis dataset, we can say with certainty that July 2024 ended up with cooler monthly global temperatures than July 2023, by a small margin (~0.05C):

Monthly global surface temperature anomalies relative to preindustrial (1850-1900) in the ERA5 dataset.

Here is another way to look at the data, showing every July in the ERA5 dataset since 1940 with the latest (2024) value in red:

July global surface temperature anomalies relative to preindustrial (1850-1900) in the ERA5 dataset.

While it did not set a new record compared to 2023, July global temperatures in 2024 were still around 0.3C warmer than any July prior to 2023.

An exceptionally warm July

We knew relatively early in the month that July 2024 was unlikely to beat 2023 and set a new record. It started out quite a bit cooler than 2023 for the first two weeks; global temperatures had a remarkable spike in the start of 2023 that was not seen in 2024, though a more modest spike over the past two weeks did help some individual days beat 2023 to set a new all-time absolute global daily temperature record.²

Daily global surface temperature anomalies relative to preindustrial (1850-1900) in the ERA5 dataset.

The elevated global temperatures over the past two weeks of July ended up making it a closer contest than I expected at the beginning of the month. It also raises an interesting question: will the warmth of the past two weeks persist and potentially lead to a record-setting August, or will they fade back to the (relatively) cooler conditions that characterized early July?

If we look at the distribution of global temperatures, we see that the global temperature spike in mid-July was entirely due to a jump in global temperatures in the Southern Hemisphere, as shown in the figure from Karsten Haustein’s eminently useful climate tracking site below.

Daily global and hemispheric temperatures in the NCEP GFS reanalysis product, including a 7-day forecast. Anomalies shown relative to a 1981-2010 baseline.

It turns out that this Southern Hemisphere warming spike is, in turn, driven by a huge anomaly over Antarctica that has resulted in temperatures around 28C above normal for large regions of the continent.

Distribution of global temperatures in the NCEP GFS reanalysis product over the past 7 days (July 26th-August 1st). Anomalies shown relative to a 1981-2010 baseline.

Antartica has, historically, had far more short-term temperature variability than much of the rest of the world. While its possible that hot conditions there may persist (and global weather models expect them to for at least the next week), its just as likely that Antarctic temperatures will become less extreme over the course of August simply due to internal climate variability.

For that reason, my expectation at this point is that August 2024 will probably end up also falling a bit behind 2023 as the second warmest August on record.

Uncharted territory?

Earlier this year NASA’s Gavin Schmidt and I separately wrote that the evolution of global temperatures in 2024 would be important to tell us if the “gobsmacking” conditions we saw in the latter half of 2023 represented a new persistent condition for the climate or more of a temporary phenomenon.

Gavin suggested that we would have a better sense by August if conditions were stabilizing or the climate was heading into “uncharted territory”.

With August almost upon us we remain in something of a liminal space. Both June and July were notably warmer than I expected earlier in the year (coming in 0.4C and 0.3C respectively above the last big El Nino year of 2016). At the same time, we have moved out of record territory, and we still expect some additional cooling influence from fading El Nino conditions and potential La Nina development.

So I think we will still have to wait and see, though if the spike in temperatures over the past few weeks persists to push August 2024 to set a new record it would be a worrying sign.

1. The development of a La Nina event is far from certain at this point, with about half of the models expecting formal La Nina conditions (Nino 3.4 region < -0.5C) to develop. Though those models expecting La Nina tend to be dynamic models (rather than statistical), which did much better in predicting the development of El Nino in 2023.

2. I do think we’ve somewhat overstated how meaningful a single absolute daily temperature record is. We’ve seen far higher anomalies (>2C) for some individual days in late 2023 and early 2024, and the fact that a spike in temperatures happened to occur in the few weeks when the global temperature climatology peaks (due to larger NH land area) is somewhat coincidental and less remarkable than earlier daily record anomalies (e.g. departures from normal).