The Science

What’s so special about old weather in logbooks?

These days we can find out the temperature anywhere in the world by clicking on our closest device, so why do we need weather observations from over 100 years ago? How could weather records written down by a 19th Century sea captain possibly be useful in the 21st century?


Predicting the future is easier with the past

Anyone who’s prepared for a cyclone or made farming decisions based on long-term weather forecasts will understand the value of accurate predictions on what the weather will do in the future.  

And while these long-term weather predictions may appear definitive and insightful, they’re only as good as the past data that they’re based on. That’s how forecasting works, it’s predicting – or modelling – the future using what’s happened in the past.

So gaining as much information about past weather is really useful.

We have a good idea of the Earth’s climatic history over thousands of years – mostly from geological sources – but this doesn’t provide the level of detail required for forecasting weather.

“The quality and quantity of the data that you’re putting in is essential to the sort of product you’re putting out, because if you don’t have enough you’re not going to be able to produce anything meaningful,” says Rob Allan, meteorologist with the UK Met Office, and co-creator with Philip Brohan of Old Weather, the UK version of Weather Detective.

Sea surface temperatures are important

Sea surface temperature is actually a measurement of the temperature of the ocean, not the temperature on board the boat. It’s measured by sticking a thermometer into a bucketful of ocean water.

Because the ocean is more homogenous than land, a single sea temperature reading can give an excellent indication of the sea temperature for a large area. Since the oceans cover 70% of our planet, observations at sea are very important for understanding and predicting weather.

Land-based temperature records are also important. However, microclimates on land can lead to large variations in temperatures within a small area – for example, a shaded valley may be cooler than an open plain – which means that historical records can be unrepresentative.

That doesn’t mean the scientists won’t use the land surface temperatures – they will! Just that sea surface temperature is particularly valuable.

“There had been quite an ongoing effort for a long time in developing sea surface temperature sets,” says Allan “and we have a long set of sea surface temperatures so this project is to see if we could recover more sea surface temperature data.”


The power of citizen science

The amount of work involved in delving into the logbooks to uncover the weather observations is huge! The work involved is way too much for a small team, but with the power of citizen science, the information can be unlocked by sharing the load.

Unfortunately, it’s not the sort of thing that a computer can be trained to do. Reading handwritten text is a skill that people and not computers excel in. Humans are also better at identifying important information.

What it will be used for

The weather observations found by our citizen science weather detectives will add to our understanding of our planet’s weather history.   They’ll go into a database called Atmospheric Circulation Reconstructions over the Earth – ACRE – which will be available to anyone!

“The idea was to get a bigger better database of the weather, where we pick up more events like El Niño, La Niña or storms. So instead of 40, 50 years of recent data you could get 150 maybe longer years of data. And then you suddenly get this much more valuable tool to feed into whatever you want to use it for,” says Allan.

Ambitiously, ACRE aims to put together a full history of our planet’s weather back to 1850 – providing weather details all over the globe for 200 km x 200 km resolutions for every 3-6 hours. The weather details will then be used to reconstruct a 3D picture of what was happening with air masses and air pressure systems at the time. It’s really like something out of a science fiction film!

It’s an incredibly ambitious project that is well underway and should reap some very useful results – from improving weather forecasting, to understanding climate change to re-analysing serious past weather events, such as the Knickerbocker Storm of 1922 in the U.S.

A significant use, and one of the reasons the project was started – is to verify seasonal forecast models which are used by farmers and primary producers for crop production.

On a regional level, it could be really useful for understanding how particular climatic phenomena – such as El Niño – may respond to climate change. And that extra length of records may give us more information about how likely extreme events, such as heatwaves or floods, are to occur in the future.

Back to top