Skip to main content

Oak Savanna Restoration and Pollinator Communities (Lettow)

Temperate savannas are among the earth’s least conserved ecosystems.  In Michigan, high quality remnant oak savannas and barrens are virtually absent, creating uncertainty regarding the best methods for restoring degraded remnants.  Many of these degraded remnants exist in the form of closed canopy forests as opposed to a historically open canopy with scattered trees.  This change in savanna conditions has occurred largely due to fire suppression and disruption, and has resulted in an associated shift towards forest biota at the expense of the savanna community. 

In 2010, along with Dr. Lars Brudvig in MSU’s Plant Biology department, we initiated a long-term restoration study on a closed-canopy oak savanna remnant at MSU MacCready Reserve in Jackson County, MI.  Specifically we are comparing approaches of gradual cutting of non-oak fire-intolerant woody species coupled with fire (cut+burn), fire alone (burn), and unmanaged reference plots.  We are generally interested in how these different restoration approaches affect the structure, diversity, and function of these oak savanna remnants. 

Furthermore, I am interested in understanding how restoration approaches affect pollinator communities and their floral interactions.  Oak savannas have a high proportion of flowering plant species in their understory, and creating this diversity is a goal for many restoration practioners. This makes the function of pollination in recovering oak savannas a critical factor in their recovery. As such, I am investigating the pollinator species richness, abundance, community composition, and pollination function among restoration treatments along with determining which environmental variables may be most important in their effect on these communities and their function.

I am collecting pollinator data using a “bee bowl” technique of pan-trapping, using small dishes of three colors to adjust for pollinator preferences, which has been shown to catch a fairly wide range of bee taxa. In order to capture information about pollinators that bee bowls fail to capture, such as large butterflies and bumblebees, I am using timed observations with aerial netting of these pollinators that fly into our various treatments. I am gathering pollinator function data by using three species of flowering plants native to savannas, which would be likely to respond to our restoration treatments from increased light and a reduction of the depth of leaf litter that impedes their germination. Taking potted individuals of the same species from the greenhouse, we pair two plants with flowers covered by mesh with two plants with flowers exposed to flower visitors, leave them in our treatments for four days, and then cover and retrieve them to later examine seed viability.

This research will help us begin to better understand how restoration approaches affect pollinators that could alter plant community trajectories and overall restoration success.

Mitch and Ian with treebags

The Nature Conservancy hillside burn

Hillside savanna