Biomarker approaches for determining past changes in peatland vegetation, pH and biogeochemistry: how they inform the future of tropical peatlands

Lipid biomarkers are now routinely used in palaeoclimate and palaeoenvironmental investigations of peatlands.  However, these applications are mainly focused on vegetation reconstruction, i.e. using n-alkane distributions as a tracer for Sphagnum mosses, and based on biomarker tools largely developed in (Northen hemisphere) boreal and temperate peatlands, especially acidic ombrotrophic bogs. Here, we examine and confirm the applicability of these proxies in a wide variety of subtropical and tropical peatlands. We also identify and highlight the potential of underutilised proxies for environmental reconstruction, i.e. those sensitive to pH, and introduce new proxies for tracing biogeochemical cycling, including methane cycling.

In some cases, the expansion to tropical peatlands complicates the use of well-established biomarker proxies. In particular, the diverse distributions of n-alkanes in peat-forming graminoids complicates vegetation reconstruction in many tropical peatlands. Further complication arises from the contributions of above ground- vs below ground-derived organic material.  These issues can be partly resolved via macromolecular characterisation, including lignin monomer distributions. In other cases, biomarker proxies clearly have under-exploited potential. We have previously shown that the stereochemistry of bacterial-derived hopanes and the distribution of bacterial branched glycerol dialkyl glycerol tetraethers exhibit strong relationships with pH, but these proxies are not yet commonly employed in peatland palaeoecological interpretation. We confirm their applicability to tropical settings, as well as their coherent behaviour, which allows cross-validation of their palaeoecological interpretation and encourages wider application. 

We illustrate the coupled application of vegetation and microbial biomarkers using peatland archives from the Democratic Republic of Congo (DRC), Uganda and Panama, many of which reveal sensitive ecological tipping points. For example, in DRC peatlands, Holocene dry intervals are associated with biomarker-inferred shifts from forest- to graminoid-dominated peatland and a concomitant increase in pH. Other sites exhibit pronounced past pH changes despite relatively stable vegetation – or vice versa – suggesting that focussing exclusively on one parameter obscures more nuanced palaeoenvironmental change. We suggest that new insights into tropical peatland development and history can be obtained through holistic biomarker analyses, especially when coupled to other approaches, and that these will better inform our understanding of future responses of these crucial carbon stocks to changing climate.