Getting ready for PalaeOpen: Multi-proxy data synthesis of Holocene vegetation and fire dynamics in the tropical Andes

Tropical high mountain ecosystems, including páramo grasslands, are highly vulnerable to climate change and important providers of biodiversity, carbon storage and water regulation. Paleoecological reconstructions that rely on environmental proxies such as pollen, charcoal or fungal spores preserved in lake sediments offer valuable insights into how ecosystems have responded to past climate changes and human activities. The Holocene, a relatively stable interglacial with significant human impact, serves as a key interval for investigating vegetation, climate and fire regime dynamics.

This study as a test case for the new EU COST Action PaleoOpen (https://www.cost.eu/actions/CA23116/), explores the potential of using the open-access and community-curated Neotoma palaeoecological database to integrate multiple local proxy datasets into cross-regional reconstructions of vegetation and fire dynamics in the montane forest and páramo ecosystems in the Ecuadorian and Colombian Andes.

We find that Holocene vegetational dynamics exhibit regionally divergent trends between Ecuador and Colombia, contradicting the initial hypothesis of coherent cross-regional climate signals within the tropical Andes. These results suggest either diverging regional climatic trajectories, potentially driven by Intertropical Convergence Zone (ITCZ) and El Niño-Southern Oscillation (ENSO) variability, or limitations due to insufficient spatiotemporal data coverage and high topographic complexity within the Andean region. Some fluctuations likely reflect diverse anthropogenic land use. These results highlight the need for enhanced spatial data coverage, alongside temporally and taxonomically harmonized datasets (a key challenge for PaleoOpen) to provide a better understanding of tropical high mountain ecosystem dynamics – essential for developing effective conservation and adaptation strategies facing accelerating climate change.