One giant clam shell shows northern Great Barrier Reef summers ran up to 1.2C cooler around 1800

The comparison has its own control. The modern reference comes from instruments in the same lagoon ^[12], so the old and new numbers describe one patch of water. The sampling is also dense enough to earn the month-by-month label. Spread over about 30 years of growth ^[9], the drill holes come to roughly 26 samples a year, a little over two a month ^[1]. Microscopic daily lines from the cycle of day and night, and broader annual bands from the seasons ^[2], fix each sample's place in the clam's life.

On the researchers' figures, the gap between the coolest and warmest months at this site is roughly 0.9 to 1.2C wider now than it was in the clam's lifetime ^[2]. For judging reef heat stress, I'd expect the warm end of the year to matter most, and this record puts the change at that end. "A single average would miss all this detail," the authors wrote ^[16].

The calendar dates are less precise than the sampling. The radiocarbon window is 42 years wide, for an animal whose roughly 10,900 daily lines add up to about 30 years ^[4]. The account does not say where in that window the 30 years fall.

The thing this doesn't tell you is how typical those decades were. They came near the end of the Little Ice Age, a preindustrial cooling period ^[8]. The clam's own record shows the climate changing during its life ^[15]. Warm, dry El Nino-like episodes were more common in the late 1700s. Cool, wet La Nina-like conditions became more frequent in the early 1800s and included the strongest events in the record ^[15]. A baseline drawn from a cool interval could make modern warming look larger than it would against a longer preindustrial average. I'd put more weight on the seasonal contrast than on its exact size. An error that shifted every month by the same amount would move summers and winters together, and it would not produce the asymmetry on its own.

Longer records are possible in principle. Tridacna gigas can grow to 1.2 metres and live for more than a century ^[1], so an older shell than this 58-centimetre one ^[6] could cover decades this clam never saw. The study, published in Global and Planetary Change, was done in partnership with the Walmbaar Aboriginal Corporation and the Nguurruumungu Indigenous Corporation ^[5]. Jiigurru is a culturally important place for the Dingaal and Ngurrumungu Aboriginal communities, and excavations at shell midden sites there show at least 6,500 years of Aboriginal occupation ^[7].