Vulnerability of North American Boreal Peatlands to Interactions between Climate, Hydrology, and Wildland Fires

Laura L. Bourgeau-Chavez, Michigan Tech Research Institute, lchavez@mtu.edu (Presenter)
Eric S. Kasischke, NASA/University of Maryland, eric.s.kasischke@nasa.gov
Merritt Turetsky, University of Guelph, mrt@uoguelph.ca
William de Groot, Canadian Forest Service, bill.degroot@nrcan-rncan.gc.ca
Liza Jenkins, Michigan Tech Research Institute, lliverse@mtu.edu
Sarah Endres, Michigan Tech Research Institute, slendres@mtu.edu
Michael Billmire, Michigan Tech Research Institute, mgbillmi@mtu.edu
Brian Brian, Florida Atlantic University, bbenscot@fau.edu

Boreal peatlands store tremendous reservoirs of soil carbon that are likely to become increasingly vulnerable to fire as climate change lowers water tables and exposes carbon-rich peat to burning. Although peatland soils were believed to be too wet to burn in the past, recent evidence is showing that to be untrue. North American boreal peatland sites of Alaska, Alberta Canada, and the southern limit of the boreal ecoregion (Michigan’s Upper Peninsula) were the focus of a recently completed project to better understand the fire weather, hydrology, and climatic controls on boreal peatland fires. The overall goal of the research was to reduce uncertainties of the role of boreal peatland ecosystems in the global carbon cycle and to improve carbon emission estimates from boreal wildfires.

Landscape scale quantification of wildfire effects in boreal peatlands required development of three important mapping and monitoring capabilities specific to peatlands including: 1) methods for mapping of peatland type (e.g. bog vs. fen) and level of biomass (herbaceous, shrub, forest dominated); 2) algorithms to map burn severity to the peat surface; and 3) algorithms for mapping pre- and post-burn soil moisture with SAR data. While generalized maps exist on the distribution of peatlands for Canada, they lack the details on type. The hydrologic and water chemistry differences between bogs, poor fens (treed), moderately rich (shrubby), and rich (e.g. sedge) fens give rise to different plant community composition. These differences in hydrology and biomass result in different fire behavior and fuel consumption. Methods were developed to use a fusion of multi-date, multi-sensor SAR and optical data in the machine learning Random Forest classifier to distinguish bogs from fens and level of biomass with high accuracy (>91%).

The peatland type and burn severity map products were intersected for four different fires to assess the extent of fire and degree of burn occurring in bogs and fens versus upland forests. The results of the comparison showed: 1) bogs burned proportionally more area than uplands within a given burn perimeter; and 2) in general bogs experienced greater burn severity and fuel consumption to the organic soil layers than uplands. However, initial results of CanFIRE model runs show C emissions as lower in the bogs than the uplands primarily due to the lower aboveground biomass in the bogs.

This project resulted in development of remote sensing and field methods to assess peatland vulnerability to wildfire and it demonstrated that peatlands can burn as frequently as uplands. However, the project was not able to fill all data gaps that need to be addressed to improve carbon emissions estimates from peatlands. Further work is needed to more fully quantify and understand the seasonal effects of wildfire on peatlands including extreme dry conditions that typically occur in late summer. During the previous project only early fire season (May and June) burns, when water tables are typically high, were available for study which limited the analysis. Additional research is also needed to better understand wildfire effects in fens which were less prevalent in the fire-affected study regions than bogs. Addressing some of these data gaps is a focus of a one year NASA Rapid Response project to collect time-critical data on the 3.4 million hectares of record breaking wildfire that affected Northwest Territories in 2014. The 2014 wildfires burned in all seasons across several ecoregions, permafrost zones and ecosystem types, including a diversity of peatlands. This work is vital to extend the results of the previous project and it will allow us to capture the variability in fire effects across diverse gradients. Increasing wildfire activity in peatlands could cause a shift in post-fire trajectories, and peatlands may become net sources of C to the atmosphere, which is likely to have large influences on atmospheric carbon concentrations through positive feedbacks that enhance climate warming.

Presentation: 2015_Poster_BourgeauChavez_113_139.pdf (442k)

Presentation Type:  Poster

Session:  General Contributions   (Tue 4:35 PM)

Associated Project(s): 

• Bourgeau-Chavez, Laura: Vulnerability of North American Boreal Peatlands To Interactions Between Climate, Hydrology, and Wildland Fires ...details

Poster Location ID: 113