A statistical method for varve verification using seasonal pollen deposition

We developed a fast, inexpensive, statistically rigorous method of varve verification that uses prospective varve layer-splitting and seasonal pollen deposition. This method can be used on any sediment that contains seasonally deposited pollen, and avoids the need for radio-isotopic, optical, or thermoluminescence dating. The method uses a χ² test and non-parametric regression, together with recorded plant bloom times from pollen traps, to assess the differences in pollen abundances in light/dark (i.e. summer/winter-spring) sediment layers. A statistical test is required because such seasonally deposited data are inherently noisy, with a low signal-to-noise ratio. To illustrate this approach, visible laminations of Lake Mina, Minnesota, USA, were assessed in two separate regions of a core spanning 900 years. Results show that the laminations were deposited annually.