Carbon dynamics following landscape fire: influence of burn severity, climate, and stand history in the Metolius Watershed, Oregon Public Deposited


Attribute NameValues
Abstract or Summary
  • Fire is a fundamental disturbance that drives terrestrial and atmospheric carbon dynamics. Previous studies have quantified fire effects on carbon cycling from local to global scales but have focused nearly exclusively on high-severity, stand-replacement fire. Since 2002, variable-severity wildfires have burned more than 65 000 ha across the east slope of the Oregon Cascades, including 4 large fires that burned ca. 50% of the forested area within the Metolius Watershed in 2002 and 2003. This thesis integrates data from 64 field plots, remote-sensing, and an ecosystem process model to investigate the effects of low-, moderate-, and high-severity fire. The primary research objectives were to: (a) quantify combustion and mortality effects on carbon pools, postfire net ecosystem production (NEP), and potential regeneration trajectories at the stand scale; (b) introduce novel remote-sensing datasets into a modeling framework to assess the importance of low- and moderate-severity fire across the landscape and region. At the stand-scale, the 3 levels of burn severity (overstory tree mortality) resulted in profoundly different impacts on combustion, mortality, postfire carbon balance, and potential regeneration trajectories. Simulated combustion ranged from 16.6 to 32.3 Mg C ha-1, or 13% to 35% of prefire aboveground carbon. C transfers from fire-induced tree mortality were larger in magnitude than combustion, as live aboveground C decreased by >90% from low- to high-severity stands. Despite this decline, total net primary productivity (NPP) was only 40% lower in high- vs. low-severity stands, reflecting a compensatory effect of non-tree NPP. Dead wood respiratory losses were small relative to C uptake (range: 10-35% of total NPP), suggesting important decomposition lags in this seasonally-arid system. Although soil C, soil respiration, and fine root NPP were conserved across severity classes, NEP declined with increasing severity, driven by trends in aboveground NPP. Postfire conifer seedling density was generally abundant and varied over 5 orders of magnitude (study-wide median: 812, range: 0 – 62 134 seedlings ha-1). Seedling density was negatively correlated with overstory mortality, whereas shrub biomass showed the opposite response, indicating a wide range of potential successional trajectories. Despite substantial combustion and mortality effects on carbon pools and fluxes, the rapid response of postfire vegetation, coupled with conservation of belowground processes, may offset long-term declines in carbon storage, indicating a surprising degree of postfire stability. These stand-scale results describe a broad range of fire effects—a high degree of pyrodiversity—but because burn severity was not evenly distributed across space, the landscape-level fire effects depend on the severity mosaic. At the landscape-scale, moderate- and low-severity fire contributed 25% and 11% of total estimated pyrogenic carbon emission, respectively (0.66 Tg C total, or ca. 2.2% of statewide anthropogenic CO2 emissions equivalent from the same 2-year period). Moderate- and low-severity fire accounted for 23% and 5% of landscape-level tree mortality, respectively, which resulted in the transfer of 2.00 Tg C from live to dead pools. This carbon transfer was ca. 3-fold higher than the one-time pulse from pyrogenic emission, but it will likely take decades for this dead wood to decompose via heterotrophic respiration. The inclusion of moderate-severity fire reduced postfire (2004) mean annual NEP by 39% compared to the high-severity only scenario; low-severity fire influence on NEP was small (additional reduction of 11% in mean NEP), likely because of high tree survivorship and the relatively lower areal coverage of low-severity fire. One year postfire, burned areas were a strong C source (net C exchange across 53 000 ha: -0.065 Tg C y-1; mean ± SD: -123 ± 110 g C m-2 y-1) vs. a prefire mean near C neutral (1997-2001 mean NEP ± SD: -5 ± 51 g C m-2 y-1). The model has been known to underestimate carbon uptake in mature and old semi-arid forests, so the prefire value is likely underestimated. Despite the resurgence of wildfire across western North America, including a substantial increase in the proportion of high-severity fire in the ecoregions studied here, low- and moderate-severity wildfire accounts for the majority of burned area in the Pacific Northwest region. This non-stand-replacement fire has important consequences for carbon loss and uptake at landscape- and regional-scales. The results from this thesis suggest that by accounting for the full gradient of fire effects, carbon modelers can substantially reduce uncertainties in key components of regional and global carbon budgets, particularly pyrogenic emissions, mortality, and NEP. Understanding the effects of disturbance variability on terrestrial carbon cycling will become increasingly important in the context of emerging regional and global carbon policies.
Resource Type
Date Available
Date Copyright
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Committee Member
Academic Affiliation
Non-Academic Affiliation
Rights Statement
Additional Information
  • description.provenance : Rejected by Julie Kurtz(, reason: rejecting so can replace with revised copy to fix a few typos. Open the item that was rejected. Replace the attached file with the revised file and resubmit. Thanks, Julie on 2009-07-10T15:49:37Z (GMT)
  • description.provenance : Approved for entry into archive by Linda Kathman( on 2009-07-21T22:43:17Z (GMT) No. of bitstreams: 1 ^Meigs_Thesis.pdf: 12675538 bytes, checksum: 5a8fe3a5d35cfb6537e5d76c40a3d64a (MD5)
  • description.provenance : Approved for entry into archive by Julie Kurtz( on 2009-07-21T17:43:01Z (GMT) No. of bitstreams: 1 ^Meigs_Thesis.pdf: 12675538 bytes, checksum: 5a8fe3a5d35cfb6537e5d76c40a3d64a (MD5)
  • description.provenance : Submitted by Garrett Meigs ( on 2009-07-05T04:41:29Z No. of bitstreams: 1 Meigs_thesis_final.pdf: 12587684 bytes, checksum: b4149da09ba29274b92f1bac8dbd7fa1 (MD5)
  • description.provenance : Submitted by Garrett Meigs ( on 2009-07-10T17:00:29Z No. of bitstreams: 1 ^Meigs_Thesis.pdf: 12675538 bytes, checksum: 5a8fe3a5d35cfb6537e5d76c40a3d64a (MD5)
  • description.provenance : Made available in DSpace on 2009-07-21T22:43:17Z (GMT). No. of bitstreams: 1 ^Meigs_Thesis.pdf: 12675538 bytes, checksum: 5a8fe3a5d35cfb6537e5d76c40a3d64a (MD5)


In Administrative Set:
Last modified: 08/01/2017

Downloadable Content

Download PDF

EndNote | Zotero | Mendeley