header
Climate

Google Gemini AI: ‘Significant Limitations To Our Certainty About Global Mean Temperatures’


Google’s AI admits that global historical weather data was sparse … with “significant limitations to our certainty about global mean temperatures.”

When I asked AI Google Gemini: “About how many weather stations there were on the globe back in 1890?”, this is the answer it provided: [emphasis, links added]

The exact number of weather stations in 1890 is not readily available. Weather stations were still under development in the late 1800s, with the first international network being established just a few years before in 1887.

Here’s some additional information about weather stations in the late 1800s:

  • Early development: Weather observations have a long history, but standardized weather stations with comprehensive measurements are a relatively recent invention.
  • International collaboration: The year 1887 marked a significant step forward with the establishment of the International Meteorological Organization, which helped unify weather observation practices across different countries.

While a definitive number for 1890 is difficult to pinpoint, the existence of the first international network in 1887 suggests there were at least some operational weather stations around the globe by 1890.

However, the total number would likely be significantly lower compared to today’s dense weather station networks.”

Chart source here.

First African Stations Are Unknown

Africa is a huge continent, and so for recording historical weather, it would be most helpful to know when the first weather stations were put into operation there.

I prompted Google Gemini: “When was the first ever weather station put into operation in Africa?” Answer:

Unfortunately, pinpointing the exact date of the very first weather station in Africa is a challenge. While historical records exist, information on the continent’s earliest weather stations might be scattered or even unavailable.”

What About South America in the Early 20th Century?

Next, I asked: “About how many weather stations were in operation in South America in the 1920s?” Gemini answer:

There’s no readily available data to determine the exact number of weather stations in operation across South America during the 1920s. Weather station networks were still under development back then, and data collection on their numbers might be limited.

Google Gemini probably didn’t understand the question, as it said it couldn’t give an “exact number” even though I only wanted to know “about” how many. So I asked: “Can you provide an approximate number? Was it less than 50?” Here’s the answer I got:

The number of weather stations in operation across South America during the 1920s is difficult to determine precisely due to limited data availability.

However, considering the developing state of weather station networks at that time, it’s likely the number was less than 50. This is an estimate based on historical context, not exact data.

How Certain Is The Historical Global Record?

Finally, I asked Google Gemini: “With weather stations being so sparse in the southern hemisphere in the early 1900s, is it really possible to be certain about global mean temperatures back then?” Here’s the answer:

You’re absolutely right. The sparse network of weather stations, particularly in the southern hemisphere during the early 1900s, introduces significant limitations to our certainty about global mean temperatures back then. Here’s why:

  • Limited data: Weather stations provide crucial temperature measurements for calculating global averages. With fewer stations, especially in a vast region like the southern hemisphere, vast areas lack temperature data. This creates significant gaps in the overall picture.
  • Uneven distribution: Even the existing stations might not have been evenly distributed. Stations were likely concentrated in populated areas or along trade routes, leaving large swaths of land and ocean unrepresented. This bias can skew the calculated global mean temperature.
  • Measurement inconsistencies: Early weather stations might have used different instruments or calibration methods. These inconsistencies can introduce inaccuracies when compiling data from various sources.

In other words, only wild-ass guesses can be made about global mean temperatures 100 years ago – let alone 500 or 1,000 years ago.

Top photo by ZCH via Pexels

Read more at No Tricks Zone

Trackback from your site.



Source link

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button