COS 69-9
Can experimental manipulations of fire and herbivory be utilized to restore populations of an endangered terrestrial orchid in Texas post oak savannas?

Wednesday, August 13, 2014: 10:50 AM
314, Sacramento Convention Center
William E. Rogers, Ecosystem Science & Management, Texas A&M University, College Station, TX
Carissa Wonkka, Agronomy & Horticulture, University of Nebraska, Lincoln, NE
Dirac Twidwell, Agronomy and Horticulture, University of Nebraska - Lincoln, Lincoln, NE
Fred E. Smeins, Ecosystem Science and Management, Texas A&M University, College Station, TX
Martha C. Ariza, Ecosystem Science and Management, Texas A&M University, College Station, TX
Background/Question/Methods

Fire and herbivory are widely recognized as important in savanna ecosystems, however, little is known about their effects on Spiranthes parksii, an endangered terrestrial orchid endemic to Texas post oak savannas. We designed a full-factorial, completely randomized field experiment to assess fire and vertebrate herbivory on S.parksii. Thirty-two 1.5x1.5m plots were established in areas of high orchid abundance. Treatments included i) burned and fenced, ii) burned, not fenced, iii) fenced, not burned, and iv) unburned, unfenced. Pre-treatment data were collected and 2-m tall fences with 10-cm wire mesh were erected in May 2009. Fires were conducted in July 2010 using a steel burn-box. We routinely collected data on orchids, herbaceous cover, light, and soil moisture in each plot through spring 2014.  We then used these data to develop stage-structured population models for each plot to predict an asymptotic annual growth rate for each plot.

Results/Conclusions

Fenced plots had fewer flowering stalks in 2009-10 (p=0.008) and 2010-11 (p=0.03). Both fenced and unfenced flowers experienced herbivory damage suggesting small herbivores may be more important than previously appreciated.  Flower densities were markedly suppressed in all plots during 2011-12 likely due to unprecedented drought conditions. There were no significant effects of fences on flowering stalks from 2011-2014. Burning significantly reduced the number of flowering stalks in 2010-11 (p=0.01), but flowering stalk density was significantly higher in burned plots in 2012-13 (p=0.04). Rosette density was not significantly affected by either the fire or herbivory treatments during any growing season.  Rosette emergence occurred significantly earlier in burned plots than unburned plots in 2011-12 (p=0.03). Rosette area was greater in burned plots than unburned plots (p=0.03) in 2013-14, but there were no differences among treatments for any other growing season.  An understanding of S.parksii ecology with an emphasis on establishment, recruitment, and maintenance is essential for the conservation of this endangered species.  Specifically, long-term assessments of responses to fire and herbivore grazing, as well as the use of both experimentation and demographic modeling will provide a more complete understanding of the implications of these disturbances on population dynamics.