How does variation in life history strategies effect long term population trajectories of eelgrass?
Stephanie Thurner, '17
Senior Mathematics Major
Mathematics
Albion College
Seagrasses are important habitat forming marine angiosperm in coastal ecosystems. Contemporary
declines in seagrass habitats worldwide warrant understanding factors that may allow managers to
predict change and recovery in these habitats. In the Pacific Northwest, eelgrass (Zostera marina) forms
these critically productive meadows and has experienced instances of localized population decline.
These site specific declines are often persistent, lacking natural recovery. In an attempt to combat these
declines, human aided restoration is endeavored with worldwide success rates around 30%. Eelgrass
reproduces using two different life history strategies, asexual and sexual reproduction, with varying life
history strategies between populations. It is unknown how variation in life history strategy affects the
long term population trajectories of eelgrass. In this study we develop a stage-based matrix population
model, parameterized by field data collection, previous experiments, and data mining to map the effects
of life history variation on population growth. The eelgrass lifecycle was mathematically defined as three
stages (vegetative shoots, flowering shoots, and seeds) as well as by the vital rates describing the
transitions between these stages (branching rate, flowering rate, fecundity, germination rate, and
seedling survival rate). By analyzing populations with variations in sexual and asexual reproduction and
validating the model through a comparison with a long term field study experiment, we saw that the
model is a highly conservative estimate for solely sexually reproducing populations, and an over
estimate for populations with asexual reproduction. When the field recovery experiment was analyzed,
we could also see that recovery of a population back to initial population levels after a disturbance is
different when density is determined spatially rather than by looking at the entire area. Further data
collection and refinement of vital rates as well as the addition of other environmental conditions will
increase the accuracy of the model and help inform management and conservation strategies.