Record Details
Field | Value |
---|---|
Title | Interactive effects of silvicultural herbicides and cervid herbivory on early seral plant communities of the northern Oregon Coast Range |
Names |
Stokely, Thomas
(creator) Betts, Matthew G. (advisor) Hagar, Joan C. (advisor) |
Date Issued | 2014-06-12 (iso8601) |
Note | Graduation date: 2015 |
Abstract | Intensive forest management (IFM, dense conifer plantings and herbicide applications) may alter the characteristics of early seral plant communities that function as major habitat resources for a host of wildlife species, including cervid herbivores such as Cervus elaphus and Odocoileus hemionus. Such large herbivores can also substantially affect plant community characteristics and succession, especially in disturbed early seral habitats. I hypothesized that the effect of cervid herbivory on early seral plant communities is mediated by the effect of silvicultural herbicide treatments. If that is the case, intensively treated stands with low plant cover and diversity should be most susceptible to herbivory, as cervids are less selective and herbivory impacts are highly concentrated where forage has been diminished. To test this hypothesis, I experimentally established paired 225 m² cervid Exclusion and Open-Herbivory treatment plots in 28, 12-15 ha early seral plantation stands throughout the northern Oregon Coast Range, USA, representing a gradient in IFM. The gradient included three herbicide treatments and a no-spray Control applied at the stand scale and replicated using a randomized complete block design. I compared estimates of cover, height and diversity for entire plant communities and specific functional groups among herbivory and herbicide treatments using mixed-effects models with a blocked split-plot design. I found convincing evidence that the effect of herbivory was mediated by herbicide treatment. No-spray Control stands were too vigorous, diverse and rich with native perennial herb and deciduous shrub forage to be substantially impacted by cervid herbivory. The herbaceous specific, Light herbicide treatment reduced Shannon diversity and the cover and richness of native-perennial herbs, releasing deciduous shrub height growth where cervids where excluded. Highly selective herbivory suppressed the shrub height response by 20.5 cm, increasing the abundance and richness of introduced herb species. The broad spectrum, Moderate herbicide treatment reduced diversity, forage cover and diminished the cover and richness of deciduous shrubs and native-perennial herbs, favoring the dominance of introduced-ruderal herbs. Herbivory in the Moderate treatment reduced total cover by 17.7 percent cover, moderate-quality forage cover by 13.2 percent cover and native perennial herb richness by 1.5 species, while suppressing the cover of introduced-ruderal herbs by 4.58 percent cover and reducing the height of ferns and introduced-perennial herbs by 19.9 and 17.3 cm, respectively. Plant communities subject to the Heavy treatment were the most depauperate of all and herbivory exacerbated the effect of this treatment on native-perennial herbs only. Average height of dominant vegetation was consistently lower with cervid access across all stands, especially with Moderate herbicide treatment. My results provide evidence that by reducing diversity and the abundance of native forage species, herbicide treatments altered herbivory selectivity and pressure. The alteration of herbivory pressure in turn influenced the outcome of herbicide treatments, resulting in an exacerbated effect with Moderate and Heavy treatments. The effect of herbivory in controlling vegetation in managed stands may have positive implications for conifer seedling growth, constituting a possible 'ecosystem service' by cervids in the Pacific Northwest. This 'service' may be to the detriment of biodiversity and other early seral associates when coupled with common intensive forest management practices. |
Genre | Thesis/Dissertation |
Access Condition | http://creativecommons.org/licenses/by-nd/3.0/us/ |
Topic | cervid |
Identifier | http://hdl.handle.net/1957/50309 |