Can You Count Years in Mud? Each Summer Leaves a Different Layer

In a lake fed by meltwater from a glacier, the summer melt carries a large volume of silt into the lake, and the coarser grains settle out quickly to form a pale layer. When the surface freezes over in winter the inflow stops, the water becomes still, and the finest clay particles that stayed suspended all summer finally settle, forming a thin dark layer on top. The pair together is one year. The same annual pairing occurs in lakes without glaciers where seasonal algal blooms, floods or changes in oxygen produce a regular alternation, and in some marine basins that are cut off from mixing.

Each pair is a record as well as a tick of a clock:

• •The thickness of the summer layer indicates how warm and wet that year was
• •Pollen trapped in a layer records what was growing that year
• •Volcanic ash in a layer dates the eruption exactly
• •Charcoal marks a fire in the catchment
• •Chemistry of the sediment tracks changes in the water
• •A missing or doubled pair marks an unusual year rather than an error

The whole method depends on nothing disturbing the sediment after it settles, which is a demanding condition and rules out most lakes. Burrowing worms and insects churn the top of ordinary lake sediment thoroughly and destroy any layering within a few years, so the method works only where the bottom water lacks oxygen and nothing can live there to burrow. That requires a lake deep enough or sheltered enough that it does not mix from top to bottom each year. Currents, slumping of the lake sides and human disturbance all break the record. Cores are therefore taken carefully, kept intact, and frequently scanned by X-ray or examined in thin slices under a microscope rather than split open.

Counting tens of thousands of layers by eye is as tedious and error-prone as it sounds, and the methods have been improved considerably. A core is split, photographed under controlled light, and the images analysed to detect the alternation automatically, with the software flagging ambiguous sections for a person to judge. X-ray imaging of the intact core detects density differences without opening it. Scanning for chemical elements along the core length reveals seasonal alternations invisible to the eye. The critical safeguard is counting the same sequence independently more than once, by different people or methods, since a systematic miscount accumulates, and published sequences state an uncertainty that grows with depth.

The technique gives an exact count of years but only from the top of the sequence downwards, so it needs anchoring to the calendar, and its most valuable role has been calibrating other methods rather than dating things directly. A lake in Japan yielded a continuous sequence of more than fifty thousand annual pairs, each containing plant fragments that could be radiocarbon dated, which provided a direct year-by-year comparison between radiocarbon results and true calendar years across that whole span. That comparison is now built into the standard calibration that every radiocarbon date uses. Sequences in Scandinavia were used from the early twentieth century to date the retreat of the ice sheet year by year.

A coarse pale summer layer and a fine dark winter layer make one year, and counting pairs gives an exact year count in lakes where nothing burrows into the sediment. Layer thickness, pollen, ash and charcoal make each pair a record of that year as well. A Japanese lake supplied more than fifty thousand annual pairs with datable plant fragments, which calibrated radiocarbon dating against true calendar years.