Biomass-modulated fire dynamics during the last glacial-interglacial transition at the central pyrenees (Spain)

Abstract

Understanding long-term fire ecology is essential for current day interpretation of ecosystem fire responses. However palaeoecology of fire is still poorly understood, especially at high-altitude mountain environments, despite the fact that these are fire-sensitive ecosystems and their resilience might be affected by changing fire regimes. We reconstruct wildfire occurrence since the Lateglacial (14.7. cal. ka BP) to the Mid-Holocene (6. cal. ka BP) and investigate the climate-fuel-fire relationships in a sedimentary sequence located at the treeline in the Central Spanish Pyrenees. Pollen, macro- and micro-charcoal were analysed for the identification of fire events (FE) in order to detect vegetation post-fire response and to define biomass-fire interactions. mean fire intervals (mfi) reduced since the Lateglacial, peaking at 9-7.7. cal. ka BP while from 7.7 to 6. cal. ka BP no fire is recorded. We hypothesise that Early Holocene maximum summer insolation, as climate forcing, and mesophyte forest expansion, as a fuel-creating factor, were responsible for accelerating fire occurrence in the Central Pyrenees treeline. We also found that fire had long-lasting negative effects on most of the treeline plant communities and that forest contraction from 7.7. cal. ka BP is likely linked to the ecosystem's threshold response to high fire frequencies. © 2014 Elsevier B.V. ; This research has been funded by the projects DINAMO (CGL2009-07992) (funding EGPF — grant ref. BES-2010-038593 and MSC), DINAMO2 (CGL2012-33063), ARAFIRE (2012 GA LC 064), GRACCIE-CONSOLIDER (CSD2007-00067). GGR was funded by the Juan de la Cierva Program (grant ref. JCI2009-04345) and JAE-Doc CSIC Program, LLM was supported by a postdoctoral MINT fellowship funded by the Institute for the Environment (Brunel University), AMC is a Ramón y Cajal fellow (ref: RYC-2008-02431), APS holds a grant funded by the Aragon Government (ref. 17030G/5423/480072/14003) and JAE holds a grant funded by the Basque Country Government (BFI-2010-5). We are indebted to Beatriz Bueno and Aida Adsuar for sampling processing and laboratory analyses and to César Morales del Molino for productive discussions on the ideas of this paper. We are grateful to Steven M. Holland for providing the free software to calculate rarefaction (Analytic Rarefaction), which is available at http://strata.uga.edu/software/. ; Peer Reviewed