Theses and Dissertations at Montana State University (MSU)
Permanent URI for this collectionhttps://scholarworks.montana.edu/handle/1/733
Browse
1 results
Search Results
Item Nonindigenous plant species distributions : modeling the role of human disturbances and predicting management responses(Montana State University - Bozeman, College of Agriculture, 2012) Bridges, Melissa Elaine; Co-chairs, Graduate Committee: Lisa J. Rew and Bruce D. Maxwell; Bruce D. Maxwell and Lisa J. Rew were co-authors of the article 'Human disturbance regimes influence the transferability of nonindigenous plant distribution models' in the journal 'Biological invasions' which is contained within this thesis.; Bruce D. Maxwell and Lisa J. Rew were co-authors of the article 'The role of current and historic land uses on the local occupancies of nonindigenous plant species' in the journal 'Landscape ecology' which is contained within this thesis.; Bruce D. Maxwell and Lisa J. Rew were co-authors of the article, 'Nonindigenous population and off target plant community responses to management along an environmental suitability gradient' in the journal 'Journal of applied ecology' which is contained within this thesis.The current paradigm of nonindigenous plant species (NIS) management assumes all NIS populations are invasive and ignores that different populations of the same species have different dynamics and respond differently to perturbations in dissimilar environments. Species distribution models (SDM) can predict spatial patterns of NIS environmental suitability and form a link between management objectives and species distributions. The objectives of this dissertation were to evaluate the utility of SDMs for NIS management. Specifically, the spatial transferability of models, the importance of land uses in explaining NIS distributions, and the relationships between management efficacy and SDM predictions were assessed. The first objective evaluated the transferability of SDMs for two NIS among neighboring regions representing a three-point gradient of human disturbance intensities. The models did not adequately transfer between the two management units representing the least and greatest intensities of human disturbances. This suggested NIS might be distributed differently in response to human disturbances. The second objective compared the relative roles of environmental, current land use, and historical land use variables on explaining occupancies for six NIS. Historical land use explained greater amounts of the variation in NIS occupancies as compared to only environmental variables or environmental plus current land use variables. Land uses currently or previously irrigated for agriculture increased predicted probabilities of occurrence for multiple NIS, including an introduced perennial forage species. The final objective assessed the applicability of SDMs to prioritize NIS populations for management treatments. Herbicide treatments were applied to populations of two NIS located along a gradient of their respective SDM predictions. The effect of herbicide treatment on NIS densities varied either positively or negatively with predicted environmental suitability depending on the specific species. Thus, NIS population responses may be predictable and treatment prioritized using SDM predictions; however, contrasting responses between the two NIS evaluated suggested management should be adapted for species and site specific conditions. This study showed that human disturbance history can affect how NIS are distributed, and, thus, SDMs should be generated from site specific data. Further, SDMs can guide managers as to which NIS populations should be prioritized for management.