Event Detail

Event Type: 
Mathematical Biology Seminar
Date/Time: 
Wednesday, October 17, 2018 - 16:00 to Thursday, October 18, 2018 - 16:45
Location: 
STAG 113

Speaker Info

Abstract: 

Both speakers will share the hour.

Caitlin's talk is titled:
Quantifying the role of salt marshes in Oregon Coast coho salmon production

and her abstract is:
Oregon’s iconic salmon runs are threatened by factors such as degraded habitat, overharvesting, and climate change. One of the most vulnerable species of salmon is the Oregon Coast coho (Oncorhynchus kisutch), which have been federally listed as threatened under the Endangered Species Act primarily due to habitat degradation in streams and estuaries. Compared to other species of salmon, it is suspected that coho may be more vulnerable to human perturbation and climate change and that they rely more heavily on estuarine habitats for rearing. However, we know surprisingly little about the factors limiting coho salmon and the potential role of estuaries in their recovery. Working with Dr. Mark Scheuerell (NOAA), we are developing a hierarchical Bayesian model to quantify the relative contribution of various factors, including estuarine salt marsh habitat, in mediating coho salmon population sizes. I will present preliminary findings using 22 years of abundance and age structure data from 21 independent Oregon coho populations collected by ODFW, including variation in potential driving factors across time and space. Ultimately, the model will allow me to examine the effect of restoration scenarios on coho salmon populations.

Zech's talk is titled:
Multidimensional methods to assess rocky intertidal community succession

and his abstract is:
Quantifying long-term changes in community structure is challenging in diverse habitats. Traditional ordination methods such as non-metric multidimensional scaling (NMDS) provide a first approximation of succession, but the reduction in dimensionality comes with the loss of analytical nuance. In contrast to ordination, vector-based calculations of successional rates and community differences preserve the original n-dimensions of species space and transform species space to allow for explicit comparison between community states. I discuss how to apply these innovative vector-based methods to a long-term dataset of rocky intertidal community structure from Oregon and California, as well as review preliminary results from NMDS ordinations of these data. After quantifying successional patterns using these multidimensional methods, I intend to test the relative influence of deterministic and stochastic processes on rocky intertidal community succession.