It's a bit unusual, but there have been longer gaps during the last 2000 years.

There is a 2024 CRS report that notes that FEMA is now handling 150% more major disasters per year than in the first decade after it was founded.

That might have been misconstrued as FEMA doing 150% more now, though it isn't "faster" exactly.

Often early weather data exists only on paper records in some obscure government or university storage facility.

Finding them and digitizing them for further use helps to open up new insights and data that would not be available in other ways.

6/6

The best way to reduce those large uncertainties would be to bring in more historical data for better global coverage.

Data rescue and digitizations efforts like HCLIM, can play a big role in trying to make more early weather data available.

5/

Even with large uncertainties, early instrumental reconstructions can be useful in probing climate states that are different from what we experience today.

bsky.app/profile/raro...

4/

In very early instrumental reconstructions (pre-1850), one is often only sampling a small area, e.g. ~15% of the Earth.

Using modern weather patterns, one them estimates how different the other ~85% is likely to be from the part you can see, giving rise to an uncertainty.

3/

Though accurate liquid-in-glass thermometers were invented in early 18th century, weather monitoring remained sparse for the next ~150 years.

2/

More generally, the heightened volcanism of the late 18th and early 19th century likely played a large role in why this era is remembered as the height of the Little Ice Age.

Since then fewer volcanoes and increasing carbon dioxide have helped to produce warming.

3/3

Large volcanic eruptions inject sulfur and sulfur oxides into the stratosphere, where they reflect some sunlight and create a temporary cooling effect.

The eruption of Tambora in 1815 famously led to what some Europeans described as the Year Without A Summer in 1816.

2/

It doesn't get much attention from Americans or Europeans, but the Middle East has some of the worst rates for recent regional warming.

That figure, from January, predates Hansen et al.

Your ERA5-land curve has very different absolute temperature values than than either my curve or the climate reanalyzer one. That suggests there might be an averaging problem.

One of the things to know in this context is that ERA5 has none of the weather station temperatures in its input dataset.

So its surface air temperature is going to depend heavily on how well it can translate things like pressure fields and satellite data into Antarctic surface temperature.

One sometimes sees claims that Antarctica isn't warming at all, but that's never been true in the Berkeley Earth analysis.

It is however the slowest warming continent, and the one subject to the most uncertainty along with large year-to-year variability (annual averages shown).

2/2

Here is our October time series for Antarctica. Record warm on a long-term upward trend, but the new record is not by a huge margin.

This is specifically masked to the Antarctic land mass. I think some of your plots are based on a wider area including sea ice, so that would be a difference.

Returning to this, here is our monthly anomaly map based on in situ weather station data.

Still record warmth for the Antarctic average, but notably less extreme in places than ERA5.

berkeleyearth.org/october-2025...

The long-range forecasts from IRI/CPC expect that this La Niña will be weak and short-lived, dissipating early in 2026.

The development of a new El Niño event is possible in mid-to-late 2026.

iri.columbia.edu/our-expertis...

3/3