COS 111-4 - Long-term (11 year) effects of thinning, prescribed burning, and deer browsing on woodland composition and structure

Background/Question/Methods

The failure of oaks to regenerate is shifting woodland composition across much of the United States. Oak regeneration failure is characterized by a lack of mid-sized oak saplings where mature oaks are common, and may be caused by 1) widespread fire suppression, which has favored fire-sensitive competitors over more fire-tolerant oaks, and 2) browsing by over-abundant white-tailed deer. Little is known about how the effects of these two factors may interact to suppress oak regeneration. We compared the long-term effects of two treatments on woodland composition and structure in oak-juniper woodlands in central Texas: selective thinning followed by prescribed burning (the ‘burn’ treatment), imposed factorially with deer exclusion (the ‘fence’ treatment). In 2009, twenty woodland plots were established, including 5 plots per burn-fence treatment combination (i.e., burn-only, fence-only, burn-fence, and control). The burn treatment consisted of first thinning Juniperus ashei (Ashe juniper) saplings in 2010, followed by prescribed burns in December 2010. Deer exclusion plots were also fenced in 2010. The abundance of every woody species was measured in each plot in 2009 (pre-treatment), 2011, 2012, and 2021. Canopy openness at 0.5m above the ground was measured in 2011 and 2021.

Results/Conclusions

In 2011, the canopy was more open in the burn plots regardless of fencing. At that time, both treatments had a positive effect on the abundance of oak re-sprouts, with the greatest positive effect in the burn-fence plots, but no detectable recruitment from seeds. By 2021, the canopy was still most open in the burn-only plots, but had become closed in the burn-fence plots due to woody growth protected from browsing. Independent Quercus buckleyi (Texas red oak) saplings, apparently recruited from acorns, were detected and were most abundant in the burn-fence plots. Saplings of Ilex decidua (possumhaw) were also most abundant in the burn-fence plots. J. ashei sapling density remained lower in the burn plots, with or without fencing. These results suggest that reduced competition from J. ashei, combined with protection from deer herbivory, is necessary for Q. buckleyi recruitment from both seeds and re-sprouts. However, the continuing effectiveness of this treatment combination may depend on whether Q. buckleyi can compete with I. decidua. Our results demonstrate the importance of long-term field experiments for measuring woodland community dynamics and are being used to guide sustainable management of oak woodland systems that provide habitat for imperiled wildlife.