Fire, vegetation and ecosystem processes in Yellowstone National Park
Contacts
Keywords
Lodgepole pine, disturbance, Yellowstone National Park, forest fire, nitrogen cycling, carbon dynamics, coarse wood, spatial heterogeneity, soils processes, microbial community composition
Research Overview
Our current research on fire, vegetation and ecosystem processes includes two main projects
- Influence of disturbance-generated landscape patterns on the spatial dynamics of ecosystem processes
- Carbon cycling at the lanscape scale: the effect of changes in climate and fire frequency on age distribution, stand structure and net ecosystem production (NEP).
described below. We have learned a tremendous amount about nature’s mechanisms for recovery from what people consider “catastrophic” disturbances. Wilderness areas like Yellowstone permit scientists to study ecosystems that have been minimally impacted by humans. Fires are likely to increase in number and size with global warming, and long-term studies may help scientists and land managers anticipate what may happen in the future. For a summary of key findings geared for a general audience, read “Rising from the Ashes” from the Summer 2008 issue of On Wisconsin.
Influence of disturbance-generated landscape patterns on the spatial dynamics of ecosystem processes
With funding from the Andrew W. Mellon Foundation, we have studied interactions between vegetation and ecosystem processes, with particular emphasis on nitrogen (N) cycling, carbon dynamics and the role of post-fire coarse wood, following stand-replacing fire. This study maintained a key focus on areas burned by the 1988 fires, but we also initiated studies in areas burned during 2000 and 2003 and within a chronosequence of stand ages. In addition, we studied within-stand spatial heterogeneity in soil processes and microbial communities in an Alaskan boreal forest that burned in 2001. Analyses to date have revealed important interactions between the post-fire vegetation, newly fallen coarse wood, and nutrient cycling; we briefly summarize some of our findings. View PDF of our June 2009 Final Report to the Mellon Foundation...
Tracking the fate of the stands burned in 1988
 |
| Figure 1. By 2003 (15 years after the 1988 fires), postfire lodgepole pine trees were already producing numerous pine cones. |
Among the forests burned in 1988, the variation in postfire lodgepole pine density remains pronounced. The wide variation in tree density appears to influence soil N availability, with a significant negative relationship observed between inorganic N availability and tree density (Levitt 2006). The high-density stands have a much larger pool of N in the foliage (Turner et al. in prep.), consistent with the nutrient-conservative nature of lodgepole pine. We have found no significant changes in tree density as of yet (i.e., no self-thinning), although there is substantial variation in tree size and growth rates. Individual trees are bigger and more productive in low-density stands, but at the stand level, productivity increases with tree density (Turner et al. 2004, Levitt 2006). The young trees are already producing cones, with cone densities ranging from 1,082 to 1,015,665 cones ha-1 in 2003 (Turner et al. 2007b). Furthermore, serotinous cones were observed in stands at lower elevations, irrespective of tree density. The abundant canopy seed bank in young post-fire stands suggests substantial resilience of this forested landscape to extensive stand-replacing disturbance.
 |
| Figure 2. Most trees killed by the 1988 fires have now fallen, with some resting on the ground and others resting on top of each other. This abundant “dead and down” coarse wood influences decomposition and nutrient cycling in the soil. |
The abundance of fallen trees is a conspicuous feature of the forests that burned in 1988. Most of the trees killed by the 1988 fires have now fallen, although many of these are resting on top of each other and not yet in contact with the ground. Combined with the dense and rapidly growing young lodgepole pine trees, these elevated logs make maneuvering through many of these forests challenging! In studying litter decomposition rates within the 1988 fires, we found that decomposition was significantly slower under the elevated logs in comparison to other microsites (Remsburg and Turner 2006). The soils under the elevated logs are relatively dry, and the logs seem to intercept precipitation then direct it down the bole, thereby acting as an “umbrella” over the soil underneath. The effect of elevated logs also scaled up to the landscape, with decomposition rates decreasing significantly with the percent area of a stand that was covered by elevated logs (Remsburg and Turner 2006).
Intensive studies of net nitrogen mineralization also showed that in situ annual rates of net nitrification were significantly lower under elevated logs compared to other microsites (Metzger et al. 2008). We were surprised to observe the highest rates of net nitrification and net N mineralization in exposed bare soil. Furthermore, gross production and consumption of ammonium and microbial community composition did not differ in soils under new or legacy coarse wood, pine saplings, or in bare soil (Metzger et al. 2008). The potential influence of fallen trees acting to exclude elk from some locations and thus protect aspen seedlings from browsing was also explored. However, the data on elk usage and aspen browsing indicated that the elk were still utilizing the forests with abundant downed wood (Forester et al. 2007), and the fallen trees were unlikely to offer much protection to the aspen.
Nitrogen cycling
 |
| Figure 3. Our 2005 review article in Ecosystems noted that surprisingly few studies of fire and nitrogen cycling have been conducted following severe, stand-replacing fire. |
Our review of the state of understanding of N dynamics following severe stand-replacing fire revealed a paucity of data—most post-fire N studies were conducted following prescribed or low-severity fires (Smithwick et al. 2005b). As nitrogen is a soil nutrient that may limit productivity of lodgepole pine forests, we were interested in exploring how fire may affect spatial and temporal heterogeneity of soil nitrogen availability. With the occurrence of more stand-replacing fires in Greater Yellowstone in both 2000 and 2003, we initiated new studies to examine post-fire N cycling. In summer 2001, we established 10 study plots in areas that burned during summer 2000. Vegetation development, aboveground productivity and inorganic N availability were studied through summer 2001. In addition, we intensively sampled for soil microbial communities in summer 2002. Annual net mineralization rates were largely negative from 2001-2004, indicating substantial immobilization of ammonium (Turner et al. 2007a). Although net nitrification rates were positive, annual net nitrogen mineralization (net ammonification + net nitrification) remained low. Aboveground net primary production (ANPP) increased from 0.25 to 1.6 Mg ha-1 yr-1 from 2001 to 2004, but variation in ANPP among stands was not related to net nitrogen mineralization rates. In areas burned during 2003, we established a study of net N mineralization very soon after the fires were over. Our results indicated that immobilization of ammonium was very pronounced during the first post-fire year (Turner et al. 2007a). Our results suggest a microbial nitrogen sink for several years after severe, stand-replacing fire, confirming earlier hypotheses about post-disturbance succession and nutrient cycling in cold, fire-dominated coniferous forests. Post-fire forests in Yellowstone appear to be highly conservative for nitrogen, and microbial immobilization of ammonium plays a key role during early succession.
Additional studies conducted in and near the fires of 2000 also provided opportunities for studying how foliar N changes in response to fire and determining whether N was limiting plant growth during early succession. Studies of foliar N concentrations in 2002 in the Glade and Moran fires of 2000 revealed a five-fold difference in foliar N among 14 species, from 0.77 % in the native grass Calamagrostis rubescens, to 3.4 % in the native N-fixer Lupinus argenteus and the non-native forb Lactuca serriola (Metzger et al. 2006). We also observed higher foliar N in the burned stands for four of six species that occurred in both burned and unburned areas. However, total biomass and foliar N showed no relationships with site, fire severity, or net N mineralization (Metzger et al. 2006).
 |
| Figure 4. Inorganic nitrogen availability was studied intensively in areas burned by fires in 2000. The flagging seen in this photo (from 2002) shows the location of annual measurements of net N mineralization within this stand. |
To test directly for evidence of inorganic N limitation 3-5 years after the stand-replacing fires of 2000, we experimentally manipulated N availability for 4 common native plant species (Romme et al. in review). Granular reagent grade ammonium nitrate was added around individual plants at a rate equal to the natural background rate of net N mineralization and at 10x this rate. The grass C. rubescens exhibited clear evidence of inorganic N limitation: above-ground biomass and shoot:root ratio increased with the high-fertilizer treatment. Nitrogen:phosphorus (N:P) ratio in unfertilized C. rubescens plants was <14, also consistent with N-limitation, but N:P ratio shifted to >16 in the high-fertilizer treatment, suggesting the onset of P limitation. The upland sedge Carex rossii and seedlings of lodgepole pine were not limited by inorganic N: neither species showed any growth response to N fertilization; N:P ratios were only slightly <14; and foliar N concentrations were greater than critical values reported for mature lodgepole pine. The N-fixing forb Lupinus argenteus was not limited by N, for it showed no growth response to fertilization; rather its N:P ratio of 21 indicated P limitation. In this study, to our knowledge the first experimental evaluation of N limitation in subalpine coniferous forests following wildfire, N limitation was seen in only one of four species tested (Romme et al. in review).
How post-fire N cycling changes over the long term is also not well known. Building upon an established chronosequence of forest stands in Yellowstone (Kashian et al. 2005), we explored patterns of soil N mineralization among mature forest stands that varied in age and tree density. Soils and vegetation were analyzed in 20 lodgepole pine (Pinus contorta) forest stands, varying in age from 50 to 350 years, that initiated following stand-replacing fire (Smithwick et al. 2005c).
 fig5.jpg) |
| Figure 5. Closeup showing the resin cores, one of the methods for estimating net N mineralization in the field. We have also used pool dilution methods and free resin bags. |
Results indicated that soil microbial community composition and carbon availability may partially explain variability in soil nitrogen availability among mature forest stands. The microbial community composition of stands in the 300-350 age class was distinct from stands in younger age classes. At p<0.10, net NH4+ mineralization was significantly higher in the 300-350 age class compared to the 125-175 age class. None of the measured variables significantly explained NH4+ consumption and net mineralization patterns. However, gross NH4+ mineralization rates were best explained by information on microbial community structure (i.e., lipids). Variation among stands within a given age-classes was high, indicating that predictions of N cycling across landscapes must account for substantial heterogeneity among mature stands (Smithwick et al. 2005c).
Selected publications:
Forester, J. D., D. P. Anderson and M. G. Turner. 2007. Do high-density patches of coarse wood and regenerating saplings create browsing refugia for aspen (Populus tremuloides) in Yellowstone National Park (USA)? Forest Ecology & Management 253:211-219.
Levitt, E. A. 2006. Sources of variation in soil nitrogen availability among post-fire lodgepole pine stands in Yellowstone National Park. MS Thesis, University of Wisconsin, Madison.
Metzger, K. L., W. H. Romme and M. G. Turner. 2006. Foliar nitrogen in early postfire vegetation in the Greater Yellowstone Ecosystem (Wyoming, USA). Forest Ecology and Management 227:22-30.
Metzger, K.L., E. A. H. Smithwick , D. B. Tinker, W. H. Romme, T. C. Balser and M. G. Turner. 2008. Influence of coarse wood and pine saplings on nitrogen mineralization and microbial communities in young post-fire Pinus contorta. Forest Ecology & Management 256:59-67.
Remsburg, A. J. and M. G. Turner. 2006. Amount, position and age of coarse wood influence litter decomposition within and among young post-fire Pinus contorta stands. Canadian Journal of Forest Research 36:2112-2123.
Romme, W. H., D. B. Tinker, G. H. Stakes, and M. G. Turner. Does inorganic nitrogen availability limit plant growth 3-5 years after fire in a Wyoming lodgepole pine forest? (In review).
Schoennagel, T., E. A. H. Smithwick and M. G. Turner. Landscape heterogeneity following large fires: insights from Yellowstone National Park, USA. International Journal of Wildland Fire (In press).
Smithwick, E. A. H., M. C. Mack, M. G. Turner, F. S. Chapin III, J. Zhu and T. C. Balser. 2005a. Spatial heterogeneity and soil nitrogen dynamics in a burned black spruce forest stand: distinct controls at different scales. Biogeochemistry 76:517-537.
Smithwick, E. A. H., M. G. Turner, M. C. Mack, and F. S. Chapin, III. 2005b. Post-fire soil N cycling in northern conifer forests affected by severe, stand-replacing wildfires. Ecosystems 8:163-181.
Smithwick, E. A. H., M. G. Turner, K. L. Metzger, and T. C. Balser. 2005c. Variation in NH4+ mineralization and microbial communities with stand age in lodgepole pine (Pinus contorta) forests, Yellowstone National Park (USA). Soil Biology and Biogeochemistry 37:1546-1559.
Turner, M. G., E. A. H. Smithwick, K. L. Metzger, D. B. Tinker and W. H. Romme. 2007a. Inorganic nitrogen availability following severe stand-replacing fire in the Greater Yellowstone Ecosystem. Proceedings of the National Academy of Sciences 104:4782-4789.
Turner, M. G., D. M. Turner, W. H. Romme and D. B. Tinker. 2007b. Cone production in young post-fire Pinus contorta stands in Greater Yellowstone (USA). Forest Ecology and Management 242:119-206.
Carbon cycling at the lanscape scale: the effect of changes in climate and fire frequency on age distributarion, stand structure and net ecosystem production (NEP).
With funding from the Joint Fire Sciences Program, we are examining carbon balance across the entire Yellowstone landscape, how it might change under future climates and fire regimes, and how cycles of carbon and nitrogen interact. Climate, fire (frequency and intensity), and forest structure and development are strongly linked, but our knowledge of the interactions of these factors is poor. We lack the ability to make robust predictions about how changes in climate will alter these interactions and change the carbon balance of a landscape. Our objective is to estimate how changes in fire frequency, pattern, and intensity will alter the distribution of forest age and structure across a landscape and how these changes, in turn, will change the landscape carbon balance. We are determining the current carbon balance for the Yellowstone National Park (YNP) landscape and how much carbon was lost in the 1988 fires. We are also determining how NEP will change for the YNP landscape with changes in climate and fire regimes by calibrating the Century biogeochemical model to assess how changes in climate will alter NEP across stand development, and using models developed in past research to simulate fire frequency, fire spread, and the resulting landscape structure (the distribution of stand age and tree density) for alternative climatic condition. We hypothesize that variation in tree establishment after a fire and the legacy of the prior stand will control the trajectory of NEP through time, and that climate and fire frequency will change the distribution of forest age and structure, and these changes will alter net carbon storage for the landscape. We completed several modeling analyses as well as field data collection; analyses of the empirical data are under way. View PDF of research proposal....
Kashian et al (2006) examined the potential for changes in landscape carbon storage as a result of changing fire frequency in Yellowstone National Park. The heterogeneity of stand ages and structures across this landscape creates resistance to large changes in carbon storage even with drastic increases in fire frequency based on changes in age structures alone. However, our study suggests that widespread changes in postfire stand structure (i.e., especially a substantial reduction in stand density) could move the Yellowstone landscape from a carbon sink to a carbon source (Kashian et al. 2006).
To explore the potential effects of disturbance and climate change on C and N cycling, we modeled (using CENTURY version 4.5) mature and regenerating lodgepole pine (Pinus contorta var. latifolia Engelm. ex S. Wats.) stands that vary in post-fire tree density following stand-replacing fire (Smithwick et al. in press). Results showed that both young (post-fire) and mature stands had elevated forest production and net N mineralization under future climate scenarios relative to current climate. Forest production increased 25% (Hadley) to 36% (Canadian Climate Center), compared to 2% under current climate, among stands that varied in stand age and post-fire density. Although our simulations indicated strong positive responses of lodgepole pine productivity to future changes in climate, C flux over the next century will likely reflect complex relationships between climate, age structure, and disturbance-recovery patterns across the landscape (Smithwick et al. in press).
The most recent climate model projections suggest that, by the end of the 21st C, climate conditions like those of 1988 (the year of the well-known Yellowstone Fires) will represent close to the average year rather than an extreme year. The consequences of a climate shift of this magnitude for the fire regime, post-fire succession and carbon (C) balance of western forest ecosystems are well beyond what scientists have explored to date, and may fundamentally change the potential of western forests to sequester atmospheric C. With recent funding from the JFSP New Initiatives program and in collaboration with Dr. Tony Westerling (UC-Merced), we are extending our research to explore the implications of these current projections for vegetation and C in Greater Yellowstone. We hypothesize that vegetation communities will contribute differentially to future landscape C flux because of different sensitivities to future climate and fire combinations, and the net result could qualitatively change the C dynamics of western forests. View PDF of our research proposal…
Selected publications:
Arcano, R. 2005. Allometric model development, biomass allocation patterns, and nitrogen use efficiency of lodgepole pine in the Greater Yellowstone Ecosystem. MS Thesis, Department of Botany, University of Wyoming, Laramie.
Kashian, D. M., W. H. Romme, D. B. Tinker, M. G. Turner and M. G. Ryan. 2006. Carbon cycling and storage across coniferous landscapes: linking fire frequency, post-fire recovery, and ecosystem processes. BioScience 56:598-606.
Smithwick, E. A. H., M. G. Ryan, D. M. Kashian, W. H. Romme, D. B. Tinker and M. G. Turner. Modeling the effects of fire and climate change on carbon and nitrogen storage in lodgepole pine (Pinus contorta) stands. Global Change Biology (In press).