Light-associated habitat factors aid prediction of Miscanthus occurrence outside managed areas

Background/Question/Methods

Effective monitoring and early detection will be critical to minimizing the invasion risk from perennial bioenergy crops.  However, it is difficult to prioritize management schemes before invasions occur, and there are few opportunities to track invasions from their inception.  We used field collected data to identify spatial factors and habitat variables important for predicting the occurrence of Miscanthus.  Specifically, we asked whether distance from source populations and the distribution of surrounding habitat could provide preliminary estimates of high risk areas for Miscanthus establishment outside of cultivation.

We conducted field surveys for the presence and absence of Miscanthus sinensis, a candidate biofuel species and parent species to other candidates, in two sites with established feral populations.  Within sites, we walked survey transects radiating out from areas with escaped populations to record the location and habitat of Miscanthus plants, as well as random points used to characterize the general habitat.  We used CART and spatial analysis to evaluate the relative importance of the collected data for predicting Miscanthus occurrence.

Results/Conclusions

Light associated variables, i.e., light transmittance through the understory, canopy openness, and relatively high elevations, were important predictors of Miscanthus occurrence, though light alone, measured as PAR, was not.  This result may be in part explained by the additional importance of variables describing the surrounding understory community.  Miscanthus appears to be associated with open environments, but may be restricted by competition in high light environments such as forest openings.  Preliminary results suggest that spatial factors may also play a role at a smaller scale (within site areas), but may be less important than habitat factors at larger scales.  These results suggest that landscape-level variables, such as forest age and disturbance, may be used to identify high priority monitoring areas.  Studies such as these are critical to help guide managers and producers tasked with containing invasions from perennial bioenergy production.