Miconia (Miconia calvescens DC) was introduced to the East Maui Watershed (EMW) a half-century ago with more than 25 yr of management recorded. Using a historical spatiotemporal data set, we constructed a leptokurtic dispersal kernel with 99% of progeny confined to within 549 m of the nearest maternal source and the remaining 1% dispersed out to 1,636 m. Seedbank persistence, based on postdated recruitment, displayed an exponential decay projecting extinction beyond 20 yr. These parameters are highly congruent to independent interpretations of M. calvescens in Australia and Tahiti. In a simulated stage matrix model, we projected management efforts to locally eradicate a small incipient propagule bank wherein optimal management was achieved with an annual harvest rate that eliminated all juvenile recruits before reaching maturity, until extinction. Based on current pricing for helicopter herbicide ballistic technology (HBT) operations, the optimal, variable cost to locally eradicate this incipient propagule bank was estimated to be less than US$42,000, with ~90% of the effort searching for the most distant 1% of the progeny expended within the first 9 yr after the mature discovery. This variable cost was sensitive to seedbank size, recruitment rate, and dispersal range, but was most sensitive to harvest rates between suboptimal and excess. In a scenario prioritizing the upper region of EMW, we retroactively analyzed past HBT efforts eliminating satellite M. calvescens and determined that 27% of the total effort resulted in 87% of the total protection to this priority asset, with every US$1 invested potentially avoiding US$184 in future costs. Management outside the priority area was less economical, with returns in protection diminishing with distance from the priority upper region. Miconia calvescens is currently not eradicable in the EMW, and full containment of the invasion would require a substantial increase in stable, long-term funding. With limited resources, local eradication of satellite M. calvescens could be the most cost-effective alternative to protecting uninvaded areas prioritized for critical ecosystem functions.

Black swallowwort [Vincetoxicum nigrum (L.) Moench] and pale swallowwort [Vincetoxicum rossicum (Kleopow) Barb.] are perennial vines of European origin that invade natural areas and perennial cropping systems in the northeastern United States and southeastern Canada. Both species reproduce via wind-dispersed seeds in the form of achenes with comas, but little is known about the extent of dispersal of these seeds. We studied the relationship of seed release height (0.75 m, 2 m), wind speed at the time of release, seed mass, and settling rate on distance traveled. Vincetoxicum nigrum and V. rossicum seeds traveled up to 72.1 and 79.6 m, respectively. Seeds of both species released from 2 m traveled greater distances than seeds released from 0.75 m, which fell within 20 m of the release point. Release height was the most important factor influencing long-distance dispersal events. Wind speed also strongly interacted with release height for long-distance dispersal of V. nigrum. Vincetoxicum nigrum seed mass was greater and settling rates faster than for V. rossicum. Increasing seed mass generally increased settling rate, which in turn decreased distance traveled, except in V. rossicum, for which longer distance–dispersing seeds had a faster settling rate. Our findings suggest that management efforts focus on reducing the presence of these two vines, especially if there is potential for them to climb up taller vegetation such as trees. Seeds released from these greater heights are more likely to travel far from source populations and initiate new populations. Preventing seed production in small, nonclimbing patches will also help suppress the expansion of these two vines, as seeds can still disperse up to 20 m away from parent plants.

Many perennial bioenergy grasses have the potential to escape cultivation and invade natural areas. We quantify dispersal, a key component in invasion, for two bioenergy candidates:Miscanthus sinensis and M. × giganteus. For each species, approximately 1 × 106 caryopses dispersed anemochorously from a point source into traps placed in annuli near the source (0.5 to 5 m; 1.6 to 16.4 ft) and in arcs (10 to 400 m) in the prevailing wind direction. For both species, most caryopses (95% for M. sinensis and 77% for M. × giganteus) were captured within 50 m of the source, but a small percentage (0.2 to 3%) were captured at 300 m and 400 m. Using a maximum-likelihood approach, we evaluated the degree of support in our empirical dispersal data for competing functions to describe seed-dispersal kernels. Fat-tailed functions (lognormal, Weibull, and gamma (Γ)) fit dispersal patterns best for both species overall, but because M. sinensis dispersal distances were significantly affected by wind speed, curves were also fit separately for dispersal distances in low, moderate, and high wind events. Wind speeds shifted the M. sinensis dispersal curve from a thin-tailed exponential function at low speeds to fat-tailed lognormal functions at moderate and high wind speeds. M. sinensis caryopses traveled farther in higher wind speeds (low, 30 m; moderate, 150 m; high, 400 m). Our results demonstrate the ability of Miscanthus caryopses to travel long distances and raise important implications for potential escape and invasion of fertile Miscanthus varieties from bioenergy cultivation.