Vegetation response to wildfire and climate forcing in a Rocky Mountain lodgepole pine forest over the past 2500 years

Abstract

<jats:p> Wildfire is a ubiquitous disturbance agent in subalpine forests in western North America. Lodgepole pine ( Pinus contorta var. latifolia), a dominant tree species in these forests, is largely resilient to high-severity fires, but this resilience may be compromised under future scenarios of altered climate and fire activity. We investigated fire occurrence and post-fire vegetation change in a lodgepole pine forest over the past 2500 years to understand ecosystem responses to variability in wildfire and climate. We reconstructed vegetation composition from pollen preserved in a sediment core from Chickaree Lake, Colorado, U.S.A. (1.5-ha lake), in Rocky Mountain National Park, and compared vegetation change to an existing fire history record. Pollen samples ( n = 52) were analyzed to characterize millennial-scale and short-term (decadal-scale) changes in vegetation associated with multiple high-severity fire events. Pollen assemblages were dominated by Pinus throughout the record, reflecting the persistence of lodgepole pine. Wildfires resulted in significant declines in Pinus pollen percentages, but pollen assemblages returned to pre-fire conditions after 18 fire events, within c.75 years. The primary broad-scale change was an increase in Picea, Artemisia, Rosaceae, and Arceuthobium pollen types, around 1155 calibrated years before present. The timing of this change is coincident with changes in regional pollen records, and a shift toward wetter winter conditions identified from regional paleoclimate records. Our results indicate the overall stability of vegetation in Rocky Mountain lodgepole pine forests during climate changes and repeated high-severity fires. Contemporary deviations from this pattern of resilience could indicate future recovery challenges in these ecosystems. </jats:p>

Fire‐regime variability and ecosystem resilience over four millennia in a Rocky Mountain subalpine watershed

“…Pollen abundances were summarized as percentages of the terrestrial sum. Pollen types present in at least 50% of samples and with a maximum abundance of at least 1% were considered major taxa (Chileen et al, 2020); these taxa were further grouped into three categories: conifer (trees), broadleaf (trees and shrubs) and herbaceous (understory grasses and forbs). Pollen percentages were used to calculate squared‐chord distances, and a stratigraphically constrained clustering analysis using the CONISS method was applied to identify distinct pollen zones ( rioja package in R).…”

“…Pollen percentages of each sample reflect the average plant assemblage over the period recorded by the 0.5-cm sediment slice (c. 5 years, the median sample resolution). Time-since-fire values were determined as the age difference between the sediment sample from which pollen was extracted and the closest prior sediment sample with an identified charcoal peak (Chileen et al, 2020); see below for charcoal peak detection methods. Pollen samples that were within 10 years before a charcoal peak or up to 40 years after were considered fire-coincident ('fire', n = 15), to account for potential imprecision in identifying the exact timing of charcoal peaks (which can span multiple samples).…”

Disturbance ecology and the problem of n = 1: A proposed framework for unifying disturbance ecology studies to address theory across multiple ecological systems

“…Palaeoecological studies of disturbance regimes, such as forest fires (e.g. Chileen et al, 2020) are an excellent parallel to the microbial community disturbance studies mentioned above (e.g. Jacquet & Altermatt, 2020).…”

“…It is true that not all categories fit as comfortably as others, depending on the study; for example, while functional groups are reported and interpreted in Chileen et al (2020), they vary continuously across the time period. This makes reporting simple metrics like functional group diversity very difficult.…”

Fire in paradise

“…A large body of work focuses on simulating wildfires, often with the goal to establish predictive models [Monedero et al 2017;Pastor et al 2003;Richards 1990], to simulate fires for different biomes [Cheney et al 1993;Dupuy and Larini 2000], to understand smoke properties and the ignition of wildfires [Anand et al 2017;Gustenyov et al 2018], to predict high-fidelity flows around strongly simplified trees [Mendoza et al 2019], or by specifically focusing on the coupling of wildfires and the atmosphere [Coen 2005;Sun et al 2009]. Finally, researchers also investigate the long-term growth response of vegetation to wildfires [Chileen et al 2020]. Similar to our work, many of these approaches aim at defining accurate models for wood combustion or physically-accurate solvers for wildfires.…”