Overabundant white-tailed deer (Odocoileus virginianus) have been a concern for land managers in eastern North America because of their impacts on native forest ecosystems. Managers have sought native plant species to serve as phytoindicators of deer impacts to supplement deer surveys. We analyzed experimental data about red trillium (Trillium recurvatum), large flowered trillium (T. grandiflorum), nodding trillium (T. cernuum), and declined trillium (T. flexipes) growth in paired exclosure (fenced) plots and control (unfenced) plots from 2002 to 2010 at the Indiana Dunes National Lakeshore. The latter two species lacked replication, so statistical analysis was not possible. All red trillium plants were surveyed for height-to-leaf, effects of browsing, and presence of flowers.

Data from individuals in 2009 demonstrated a sigmoidal relationship between height-to-leaf and probability of flowering. The relationship on moraine soils was shifted to taller plants compared to those on sand substrates, with respectively 50 percent flowering at 18 and 16 cm and 33 percent flowering at 16 and 14 cm height-to-leaf. On a plot basis, the proportion of plants flowering was influenced by height to leaf, duration of protection, and deviation in rainfall. The proportion of plants flowering increased ninefold in exclosures (28 percent) compared to control plots (3 percent) over the 8 years of protection. The mean height-to-leaf was a function of the interaction between treatment and duration, as well as red trillium density. Changes in height-to-leaf in control plots from year to year were significantly influenced by an interaction between change in deer density and change in snowfall depth. There was a significant negative correlation between change in deer density and snowfall depth. Plants in the exclosures increased in height at a rate of 1.5 cm yr⁻¹ whereas control plants decreased in height by 0.9 cm yr⁻¹. In all, 78 percent of the control plots lacked flowering individuals over the 9 years of study, indicating that red trillium is being negatively affected by deer throughout the East Unit of the park. Of the five deer management zones studied, only one showed pre-impact height-to-leaf and flowering percentages in control plots that then declined after 2005.

The results of this study demonstrate that Trillium species growing in the lands of the Indiana Dunes National Lakeshore are being suppressed reproductively by deer browsing. Specifically, we demonstrate, for the first time, the utility of using red trillium (Trillium recurvatum) height-to-leaf and percentage of flowering as indicators of the impacts of deer browsing. Application of the recommended thresholds demonstrates their utility in adopting red trillium as a phytoindicator of deer impact. Responses of plants to protection from deer suggest that deer culling might be necessary for 6 or more years for red trillium populations and rare trillium species to recover.