Lichenometric dating of little ice age glacier moraines using explicit demographic models of lichen colonization, growth, and survival

Abstract

Contemporary variants of the lichenometric dating technique depend upon statistical correlations between surface age and maximum lichen sizes, rather than an understanding of lichen biology. To date three terminal moraines of an Alaskan glacier, we used a new lichenometric technique in which surfaces are dated by comparing lichen population distributions with the predictions of ecological demography models with explicit rules for the biological processes that govern lichen populations: colonization, growth, and survival. These rules were inferred from size–frequency distributions of lichens on calibration surfaces, but could be taken directly from biological studies. Working with two lichen taxa, we used multinomial‐based likelihood functions to compare model predictions with measured lichen populations, using only the thalli in the largest 25% of the size distribution. Joint likelihoods that combine the results of both species estimated moraine ages of ad 1938, 1917, and 1816. Ages predicted by hizocarpon alone were older than those of . pubescens. Predicted ages are geologically plausible, and reveal glacier terminus retreat after a ittle ce ge maximum advance around ad 1816, with accelerated retreat starting in the early to mid twentieth century. Importantly, our technique permits calculation of prediction and model uncertainty. We attribute large confidence intervals for some dates to the use of the biologically variable hizocarpon subgenus, small sample sizes, and high inferred lichen mortality. We also suggest the need for improvement in demographic models. A primary advantage of our technique is that a process‐based approach to lichenometry will allow direct incorporation of ongoing advances in lichen biology.