Surveying, Monitoring and Predicting the Occurrence and Spread of Native and Non-Native Plant Species at Idaho National Laboratory Site
The goal of this study is to determine the current distribution of nonindigenous plant species (NIS) and rare plant species (RPS) on the INL Site and predict the potential spatial and temporal metapopulation dynamics of these species to help inform management and future development decisions. Specific objectives include:
- Evaluate existing data on NIS and RPS at the INL Site and assemble spatial environmental data for further modeling exercises
- Conduct an NIS and RPS fi eld survey of all NIS and RPS in the INL Site Conservation Management Plan Development Zone
- Develop probability of occurrence models for NIS and RPS and generate maps from these models
- Repeat transects in multiple years to calculate Markov transition probabilities and predict further invasion or extinction dynamics of NIS and RPS throughout the INL Site Development Zone
- Simulate metapopulation dynamics for a range of development scenarios at the INL Site using the multistate Markov transition probabilities.
The presence/absence of NIS and RPS were surveyed along 114 10-m wide belt transects extending 2 km away from roads or facility boundaries in 2009 and 2010. Twenty NIS were observed on the transects and no federally or state protected RPS were found on the transects.
These data were imported to a geographic information system in order to test two modeling methods that could result in predictive models relating individual NIS occurrence to predictor variables. Those predictors were associated with disturbance, dispersal, and environmental conditions. The study showed no differences in the predictive performance of the two models tested, and predictive maps will be generated for both models to assess the potential differences between them for management interpretation.
A simulation model was also evaluated to determine sample size requirements for generating predictive models at the spatial resolution needed to meet INL Site land management needs. Results of these simulations indicate that adequate sampling has been done to generate reliable predictive maps for informing management decisions.
Mean monthly precipitation was the most explanatory variable causing similarity in species among the sampling site. Geographic distance was the next most important variable suggesting a biogeographical effect on local community composition in sagebrush steppe. The next most important predictor variable was whether or not an area had experienced disturbance. The similarity of burned area to areas that were unburned and undisturbed in sagebrush steppe suggests that in the absence of disturbance (e.g. due to overgrazing, road construction, routine herbicide application along roads, etc.), burning of sagebrush steppe does not necessarily change the abundance or biodiversity of native species other than the removal of specifi c species such as big sagebrush. The results also suggest that road construction that does not also involve routine blading, weed control, and similar disturbances can be re-invaded by the native plant biodiversity otherwise limited to undisturbed sagebrush steppe.