2018 ESA Annual Meeting (August 5 -- 10)

COS 77-6 - Multi-decade succession of the understory plant community following the 1988 fires: Assessing patterns of recovery

Wednesday, August 8, 2018: 3:20 PM
333-334, New Orleans Ernest N. Morial Convention Center
Andrew J. Andrade1, Diana F. Tomback1, Sabine Mellmann-Brown2 and Timothy R. Seastedt3, (1)Integrative Biology, University of Colorado Denver, Denver, CO, (2)Northern Region, US Forest Service, Missoula, MT, (3)Department of Ecology and Evolutionary Biology, University of Colorado at Boulder, Boulder, CO
Background/Question/Methods

The 1988 Yellowstone fires represent the most significant fire event of the 20th century within the Greater Yellowstone Ecosystem (GYE), consuming >250,000 ha. The understory plant community (forbs, graminoids, shrubs) responds rapidly after fire. Within lower subalpine elevations, previous studies have indicated rapid increases in understory cover during the first decade post fire, followed by gradual declines associated with new conifer growth. However, it is unclear whether this same pattern holds at upper subalpine elevations, where the harsher climate limits plant growth. In 1990, DFT initiated a long-term study of upper subalpine succession at Henderson Mountain, Montana and established permanent plots at study sites characterized by burn status (burned, unburned) and moisture regime (mesic, xeric). Plots were measured in 1990, 1992, 1994, 2001, and 2016 for total plant cover and species richness. Trends in community composition were assessed by NMDS ordinations. We asked the following questions: 1) How have plant cover, richness, and Shannon-evenness changed over the course of the study and how do values compare among sites in 2016; 2) Have sites become more homogeneous in community composition with time; and 3) Have burned and unburned sites converged in community composition nearly 30 years after the fires?

Results/Conclusions

Preliminary results indicate low richness in burned sites in 1990, with yearly increases until 2001. By 2016, mean richness was greater within burned than unburned sites, as well as mesic than xeric sites (17.18 ± 0.72 xeric-burned, 7.64 ± 0.85 xeric-unburned, 26.36 ± 0.82 mesic-burned, 18.32 ± 1.46 mesic-unburned). Similarly, mean cover was initially low in 1992 (23.18 ± 2.51 xeric-burned, 42.05 ± 4.11 mesic-burned, 35.45 ± 3.01 mesic-unburned, 23.86 ± 2.03 xeric-unburned) but peaked by 2001 and stabilized through 2016. Bootstrapped 95% high density intervals suggest no difference in mean cover among study sites in 2016 (xeric-burned [41.59, 49.32], mesic-burned [49.09, 61.05], xeric-unburned [49.32, 62.50], mesic-unburned [51.45, 62.95]). Although Shannon index values peaked in 1992 in unburned study sites, values did not peak in burned sites until 2001. Bray-Curtis NMDS of the study plots in 2016 produced a stable solution with two dimensions (stress = 0.1554), indicating persistent separation in multivariate space between burned and unburned study sites. Nearly 30 years after the 1988 fires, burned sites appear to have stabilized with regards to species accumulation and plant cover, but have not transitioned to a more forested state and remain distinct from unburned sites in terms of community composition.