Using dendroecology to determine the effect of Celastrus orbiculatus Thunb. (Oriental bittersweet) on Liriodendron tulipifera L. (tulip poplar) growth

Background/Question/Methods

Celastrus orbiculatus Thunb. (Oriental bittersweet), an invasive exotic liana introduced to western North Carolina in the late 19th century, has become established in forests across the southern Appalachians.  Lianas are known  to decrease tree growth by constricting trunks, overtopping canopies, and increasing wind and ice damage, but no research has directly quantified decreased productivity caused by C. orbiculatus.  We chose three invaded sites containing Liriodendron tulipifera L. with and without vines to test for C. orbiculatus’ direct and indirect effects on growth.   Invaded sites were paired with three uninvaded sites of similar composition to control for abiotic growth factors.  For trees with vines, we estimated time of invasion by dating the oldest vine, then compared tree growth before and after this date.  We also calculated each site’s mean invasion date as the average age of its oldest vines. We compared growth before and after this date for trees in invaded (both trees with and without vines) and control sites. Mean basal area increments (BAIs) five, ten, and twenty years before and after invasion were calculated for each analysis.   We expected reduced growth in trees, both with and without vines, after invasion by C. orbiculatus relative to control sites.  

Results/Conclusions

Trees with vines had significantly lower mean growth rates after invasion when the oldest vine was used to establish invasion date.  These declines started within five years of invasion (0.0405 ± 0.0295 mm² yr^-1) and were persistent up to twenty years after invasion (0.0418 ± 0.0029 mm² yr^-1) .  When using the mean date of invasion per site, trees in invaded areas without vines and uninvaded trees showed mixed decreases across time.  There were significant decreases in vineless invaded trees at one site ten years after invasion (0.0685 mm² yr^-1 ) and at two sites twenty years after invasion (0.0544 ± 0.0139 mm² yr^-1 ), and significant decreases in uninvaded trees across all sites ten years after invasion (0.0280 ± 0.0061 mm² yr^-1 ) and at two sites twenty years after invasion (0.0721 ± 0.0095 mm² yr^-1 ).  We found no significant differences in growth between invaded and uninvaded sites, after mean site invasion dates.  There was decreased growth in all sites after mean invasion dates, but we were unable to conclusively link this decrease to invasion by C. orbiculatus.  Competition with other components of the plant community might be a stronger factor in this growth decrease.