Fire and the invasive annual grass Microstegium vimineum in eastern deciduous forests

Authors

S. Luke Flory, University of FloridaFollow
Keith Clay, Indiana UniversityFollow
Sarah Emery, University of LouisvilleFollow
Joseph Robb, Big Oaks National Wildlife RefugeFollow

Date of this Version

2012

Document Type

Article

Citation

Flory et al. JFSP 08-1-2-01 Final Report

Comments

US government work.

Abstract

Non-native plant invasions have the potential to change natural and prescribed fire regimes by increasing fuel loads, continuity of fuels, and fuel composition, which may alter fire intensity, damage native species, and promote further invasions. In this project we sought to evaluate the interaction between fire and the invasive annual grass Microstegium vimineum in eastern deciduous forests. Our goal was to determine if invasions enhance fire intensity, including fire temperatures, flame heights, and fire duration, and negatively affect tree regeneration, and stimulate further invasions. We also sought to determine how prescribed fires and the timing of fires affect the density and demography of Microstegium and we tested the pre and post-fire management options for controlling the post-fire spread of Microstegium invasions. At Big Oaks National Wildlife Refuge in southeastern Indiana, we conducted large-scale prescribed fires to evaluate fire intensity in invaded and uninvaded areas and the response of experimental and naturally regenerating trees. In small-scale plots we manipulated the timing and frequency of fires and applied herbicide treatments to evaluate demographic responses of Microstegium. Our results show that maximum fire temperatures were on average 57% greater in Microstegium-invaded than uninvaded control areas. In addition, fires burned at temperatures over 300 °C for nearly twice as long and flame heights were 98% higher in invaded compared to uninvaded habitats. Microstegium invasion reduced survival of experimental trees by 37% in areas exposed to prescribed fire compared to uninvaded areas and tree survival in invaded, burned plots was 53% lower than invaded, unburned plots. Exposure to prescribed fire increased natural tree regeneration overall but there were 60% and 57% fewer tree seedlings in burned and unburned invaded plots, respectively, compared to control plots with the same treatments. Prescribed fire increased Microstegium biomass by five-fold the following growing season. Experimental spring fires significantly reduced Microstegium seedling numbers by ~75% immediately after the burn, but this did not result in reduced seed production at the end of season or seedling numbers the year following a burn. Burning for two springs in a row similarly reduced seedling numbers during each of the years when the burns were conducted, but this effect did not carry over to reduce Microstegium seedling numbers the following year. Similarly, fall fire reduced seedling numbers by ~50% the following spring, but this reduction also did not result in reduced seed production at the end of that season. The significant effects on seedling numbers but lack of effect on Microstegium seed production was likely due to growth compensation by the surviving plants. Grassspecific, post-emergent herbicide applied without fire was very effective at reducing population numbers, almost eradicating Microstegium populations, but fire reduced herbicide effectiveness. The results of this research demonstrate significant effects of a non-native grass invasion on fire intensity, tree regeneration, and subsequent invasions in eastern deciduous forests, an ecosystem where this phenomenon has not previously been observed. Fire was not useful as a management strategy for Microstegium invasions and interfered with an otherwise effective post-emergent herbicide. To avoid the damaging effects of intense fires in invaded areas, we recommend land managers use herbicides or other treatments to remove invasions prior to the application of prescribed fires.