An allelopathic invasive species increases prolonged dormancy in a forest perennial

Background/Question/Methods

Invasive species are considered major threats to the stability of native populations.  In Eastern deciduous forests, the invasive biennial garlic mustard (Alliaria petiolata; hereafter “GM”) can dominant understory communities and is thought to contribute to the decline of native perennials, though much of this evidence is correlative.  It has recently been shown that GM can suppress physiological function and carbon acquisition in native perennials through its release of anti-fungal allelochemcials.  Here, we test whether these physiological effects scale-up to affect life-history transitions in a long-lived perennial herb, Maianthemum racemosum (hereafter “MARA”).  From 2003-2012 we monitored the size and vegetative, flowering or dormancy states of individually marked MARA plants in five experimental 12 m x 12 m plots in a mature deciduous forest in southwestern Pennsylvania invaded by GM.  Beginning in 2006, all GM adults were removed each year from one-half of each plot.  Life history transitions between states were modeled using Cormack-Jolly-Seber (CJS) and multi-state (MS) capture-mark-recapture (CMR) models from 2003 to 2010.  CJS models were used to model the effect of GM removal on prolonged vegetative dormancy rates, and MS-CMR models were used to determine the effects of GM removal on all life history transitions. 

Results/Conclusions

The best CJS models of dormancy included the GM removal treatment (ΔAIC = 6.26).  In 2007, the year after GM removal began, the probability of dormancy in the GM removal treatment was significantly lower than in the control plots (probability of dormancy in GM removal: 0.14 [95% CI = 0.085-0.23]; probability of dormancy when GM present: 0.25 [0.16-0.38]).  Dormancy rates in the GM removal plots were also lower in 2008 (GM removed: 0.11 [0.065-0.19]; GM present: 0.21 [0.13-0.31]). Our MS-CMR model was in agreement with the CJS results and indicated that the probability of dormancy was reduced where GM was removed.  However, other life history transitions were not affected. 

Our results indicate that the presence of GM dramatically increases dormancy rates in MARA.  Moreover, prolonged dormancy appears to be a life history transition that is particularly sensitive to the presence of GM.  Evidence from other species indicates that dormancy rates increase when plants lack sufficient carbohydrate reserves to emerge.  In our prior studies with MARA, the release of allelochemicals by GM was shown to significantly to decrease carbon acquisition and storage. Thus, the altered carbon balance is likely mechanism that induces the higher dormancy rates in Maianthemum racemosum seen in this experiment.