While it is thought that long-distance spread of most invasive plants is human mediated, quantification of the vectors involved has been lacking, in part due to the difficulty of documenting long-distance dispersal events. In the case of Microstegium vimineum, Japanese stiltgrass, our previous work showed that natural local-scale spread of the species is several orders of magnitude smaller than observed spread rates through forests. The dirt and gravel roads typically found in forests in Pennsylvania are subjected to frequent maintenance activities such as road grading, which were suspected to move propagules of invasive plants such as M. vimineum.
We conducted a series of grading experiments to quantify the dispersal kernel of road grading. During each trial, approximately 20,000 seed proxies were placed along a roadside at four starting points before road grading operations. Normal road grading, consisting of making three passes over the road, was allowed to proceed. We then relocated as many seed proxies as possible using fixed sampling transects across the road every 5 m up to 50 m, followed by transects every 50 m up to 1 km, in both directions from the starting point. Additionally, we also recorded any other sightings of seeds beyond 20 m by visually scanning the surface of the road and roadside for proxies.
We present the results from three trials in 2008 and eight trials in 2009. Recovery rates ranged from 1.34 % to 5.08 %. Spread occurred in both directions from the start point in all trials; on average, spread was similar in both directions. The mean dispersal distance observed ranged from 6.3 to 33.3 m and the median dispersal distances ranged from 5 to 35 m. The furthest recorded distance traveled by a seed proxy was 205 m. In many cases, the distribution was fat-tailed, with more individuals observed at far distances than might have been expected. Anecdotal evidence suggests that the moisture content of the road can affect the distance traveled, with the furthest distances observed when the road moisture content was high but not saturated. These results highlight the potential for road maintenance to spread to spread propagules of any invasive plant found on roadsides, and to serve as a vector for M. vimineum in particular. In order to limit the spread of roadside invasives, it is therefore necessary to consider the effects of road maintenance activities such as grading in management efforts