First, to be clear, “global average temperature,” as a metric, has no real meaning. It is made up. Global average temperature is unrelated to the health of the planet, since the Earth, per se, being not a living entity, has no “health.” Nor does it have any relation to the health or survival of life on Earth, since life has thrived in much hotter temperatures — think the age of dinosaurs — and much colder temperatures — think various “ice ages.” Whatever the temperature, the Earth abides, and some forms of life on it thrive. There is no ideal temperature, except perhaps from a God’s-eye point of view, the latter of which we don’t have knowledge of. All we can say for sure, based on history, is that humans and certain other species thrive better in warmer climates than in colder ones. That’s a fact established by the history of civilization.
Second, as detailed in various Climate Change Weekly posts, there is a serious problem with the temperature-measuring system. Too many ground-based temperature stations are polluted by the urban heat island effect. The vast majority of stations violate the government’s own standards for producing an objective record. In addition, investigations have revealed that many “stations” allegedly “reporting” temperatures no longer exist, and haven’t for years. For the sake of producing a continuous record, the government is just making up data reported from many locations by averaging temperatures recorded at other locations in the region.
As a result, claims of record-setting 2026 summer heat are suspect. In the United States, 38 out of 50 U.S. states recorded their all-time high temperature records before 1955. The single decade of the 1930s recorded four times as many state high-temperature records as were recorded since 2000, during the supposed hottest two and a half decades on record. In fact, the U.S. Climate Reference Network, which does not require corrective adjustments to account for urban heat islands, has recorded no significant increase in the number of temperature anomalies since its creation in 2005.
The rest of the story wouldn’t be complete without mentioning how overrepresented highly developed, largely industrialized, small parts of the globe are in the overall temperature dataset.
Leave aside the woefully inadequate coverage of temperatures for the world’s oceans, accounting for 71% of the Earth’s surface; of the roughly 25,000 to 26,000 active monthly land temperature stations globally, 70% to 80% are located in developed countries or regions, like North America, Europe, Japan, and Australia, leaving just 20% to 30% to cover the entire continents of Africa, Asia, South America, and Antarctica. This means developed nations are grossly over-represented in the “global average temperature” calculation. That number is not representative of the globe.
For example, Antarctica is 39 times larger than Germany, yet Germany has 1,800 primary and secondary weather and temperature stations, while Antarctica has 55 to 70 weather stations. New York State alone, which is 99 times smaller than Antarctica, has twice as many weather stations. As a result, temperatures in Antarctica lack comprehensive coverage. A warm year in Europe, when averaging global temperatures, would grossly skew even widespread record cold temperatures covering much more surface area simply because industrialized countries are overrepresented in the sampling.
