A multiproxy approach to long-term herbivore grazing dynamics in peatlands based on pollen, coprophilous fungi and faecal biomarkers

Abstract

Herbivory plays a significant role in regulating many contemporary terrestrial plant ecosystems, but remains an imperfectly understood component of past ecosystem dynamics because the diagnostic capability of methods is still being tested and refined. To understand the efficacy of a multiproxy approach, we compare the sensitivity of pollen and coprophilous fungal spores (CFS) to changes in grazing intensity over the last 100–150 years in six peat cores from three UK upland areas, and apply faecal lipid biomarkers to two of the cores, using agricultural census data to calculate an independent record of herbivore density. Rising sheep density adversely affected moorland ecology over the last century, which therefore provides a suitable period to test the sensitivity of these proxies. In particular, we assess whether CFS can be used to track variations in large herbivore densities over time, since this has received less attention than their ability to identify high grazing levels. At selected sites, we test whether faecal lipid biomarkers can be used to identify which herbivore species were present. Our results highlight the differential sensitivity of each proxy, demonstrating on peat- and moorlands (i) that peak CFS abundance is a more consistent indicator of ecologically influential (high) herbivore levels than variations in animal density through time; (ii) when recorded with high CFS values, the decline or disappearance of grazing-tolerant pollen taxa is a reliable indicator of high herbivory; and (iii) at low herbivore densities, faecal lipid biomarkers are not an effective indicator of herbivore presence or identity. Quantitative reconstructions of past herbivory and identifying grazer species therefore remain challenging. However, our findings indicate that pollen and CFS provide complementary evidence for high intensity grazing, and emphasise that studies using CFS should aim to define ‘high’ herbivore levels in terms of the grazing sensitivity of the ecosystem, rather than relative animal abundance.