To burn or not to burn Oriental bittersweet: A fire manager's conundrum

Abstract

Oriental bittersweet ( Celastrus orbiculatus ) is an introduced liana (woody vine) that has invaded much of the Eastern United States and is expanding west into the Great Plains. In forests, it can girdle and damage canopy trees. At Indiana Dunes, we have discovered that it is invading non-forested dune habitats as well. Anecdotal evidence suggests that fire might facilitate its spread, but the relationship between fire and this aggressive invader is poorly understood. We investigated four areas important to fire management of oriental bittersweet, each of which we will briefly summarize here. 1) What fire temperatures cause seed mortality? For seeds, temperatures above 140°C for three minute or more kills the embryo. For fruits, temperatures above 140°C kill the seeds inside after five minutes. While oriental bittersweet fruits ripen in October and November, the seeds are not dispersed until later in the early to mid December. Thus fall fires will not have any impact on the seeds unless perhaps if they are near the ground. Late winter and early spring fires are likely to kill seeds in the top litter at least. Thus spring fire can reduce the pool of seeds available to germinate. 2) Does fire modify habitat susceptibility to invasion? We found that post fire environment had no effect on the emergence and survival of oriental bittersweet, except that the tallest plants, after two years since sowing, were in the control plots. Highest establishment occurred in mesic silt loam prairie and oak forest. Survival was greatest in mesic prairie and greatest biomass occurred in the oak forest. 3) Both fire and cutting can cause oriental bittersweet to resprout and root sucker. Does the resprouting response differ between these two treatments and can a combination of cutting and pre- or post-fire treatment facilitate its removal? Cutting sometimes increased stem density between one and two times, but burning increased density by two or more times depending on the maximum fire temperature and duration. Cutting in early July reduced total nonstructural carbohydrates by 50% from normal July levels and 75% below dormant season levels. Thus burning established populations will only serve to increase their local density. 4) How does oriental bittersweet abundance vary with fire regime and can we predict the abundance of this species in a fire mosaic landscape based on fire return interval and time since last fire? At the landscape scale, we can predict the presence and abundance of oriental bittersweet, but have less success predicting its cover and distribution. The presence of oriental bittersweet was significantly negatively influenced by canopy closure, burn frequency, and distance to roads and railroads. In plots where C. orbiculatus was present, abundance was significantly greater in plots with low to moderate burn frequency, and marginally (p = 0.056) lower in plots with greater canopy cover. Both cover and distribution of C. orbiculatus was not significantly affected by the measured variables. These results suggest the frequent fire may be effective in preventing the establishment of oriental bittersweet.