Can we really say some day was the warmest in 100,000 years? Premium

Some headlines proclaimed recently that a particular day in July was the warmest in more than a 100,000 years. It is not scientifically possible to make such a claim. This article, by Prof. Raghu Murtugudde, explains why.

Some headlines proclaimed recently that a particular day in July was the warmest in more than a 100,000 years. It is not scientifically possible to make such a claim. Here is why.

Temperature estimates from before thermometers were invented are derived from “palaeo proxies”. These are biological and chemical signatures of the temperature somewhere having been warmer or colder than a specific baseline temperature. Such a baseline is typically from the modern times, when thermometer records have existed. These measures are called “proxies” because they do not directly measure temperatures. Instead, they are simply the responses of physical, biological, and chemical processes to temperatures at that time having been warmer or colder than the baseline value.

Another thing we need to make claims about temperatures of a time in the past are some isotopes that undergo a steady rate of radioactive decay. Knowing this rate, and the expected quantity of the isotope X years ago, scientists can estimate how long it took to diminish to its present quantity. Based on the length of time one needs to go back to, the isotopes could be of carbon or lead, based on their half-lives (5,000 to more than 10 million years).

A major assumption required to make the “paleo proxy” technique workable is that the processes that produced the proxies have operated similarly back then as they do today. More specifically, and crucially, for a period of hundreds of thousands of years, proxies – which are typically buried in the ocean and lake sediments – can only record temperature anomalies, i.e. deviations from the baseline, on time scales of centuries, if not thousands of years.

They are mixed by the ocean water above and the microbes within, smoothing out the information they contain over such long timescales. From this object, it is almost impossible to estimate even decadal or annual changes in long-term temperature, forget daily temperature.

Scientists derive estimates of temperature anomalies over shorter time scales from tree rings, corals, and the shells of marine and terrestrial organisms. But even here, the best of the “palaeo proxies” only provide weekly or seasonal timescale temperature anomaly estimates.

Similarly, in the spatial sense, all temperature proxies are only local or regional estimates of historical temperature anomalies. Reliable local temperature anomalies also come with fairly significant uncertainties – even for the Holocene epoch, the period in which we evolved as modern humans, that began in around 9700 BC. Global estimates, which are based on averaging all local proxies, have even higher uncertainties.