Research
Microbial ecosystem modeling
A major focus of our work is to understand microbial ecosystem structure and function on large scales. The animation below shows global surface ocean simulations with interacting microbes, simulated using the Darwin ecosystem model. The observed patterns are the result of many microscopic interactions between different members of the ecosystem. Convection and mixing deliver nutrients to the surface ocean which the phytoplankton and bacteria feed on. Grazers and viruses in turn feed on the phytoplankton and bacteria, which can lead to local recycling of nutrients, but also export of material away from the ocean surface. Our challenge is to understand whether these idealized simulations are reflective of real ocean processes, and to tie microbial interactions with large scale cycling of nutrients and carbon.
The simulations shown above have not yet been published, due to lack of appropriate data to rigorously determine whether the model predictions are a plausible representation of real-world processes. We are working closely with empirical scientists to gather appropriate data to better understand whether the processes represented are reasonable with respect to observations.
In a 2019 paper in the journal Environmental Microbiology, a version of the model shown above was used to explore the competitive ecology of Emiliania huxleyi viruses in the North Atlantic. A detailed description of the study can be found in the article itself (available for download here), and a representation of competition among a 'fast' and 'slow' viruses in contrasting ecological scenarios is shown below. Blue indicates slow viruses are dominant, red indicates fast viruses are dominant. The bottom row shows competitive dynamics along the transect marked by the vertical dashed lines in the top row. Scenarios where the slow virus dominates were shown to be consistent with genetic biomarkers of infection. Our model could only reproduce these realistic dynamics in cases where the fast virus was subjected to a higher loss rate (middle column), or the slow virus could infect a host subpopulation that was resistant to the fast virus (right-column). Both differential loss and resistant subpopulations were shown to be consistent with laboratory and environmental data, pointing to these mechanisms as important regulators of viral competitive ecology in the ocean.
In a 2019 paper in the journal Environmental Microbiology, a version of the model shown above was used to explore the competitive ecology of Emiliania huxleyi viruses in the North Atlantic. A detailed description of the study can be found in the article itself (available for download here), and a representation of competition among a 'fast' and 'slow' viruses in contrasting ecological scenarios is shown below. Blue indicates slow viruses are dominant, red indicates fast viruses are dominant. The bottom row shows competitive dynamics along the transect marked by the vertical dashed lines in the top row. Scenarios where the slow virus dominates were shown to be consistent with genetic biomarkers of infection. Our model could only reproduce these realistic dynamics in cases where the fast virus was subjected to a higher loss rate (middle column), or the slow virus could infect a host subpopulation that was resistant to the fast virus (right-column). Both differential loss and resistant subpopulations were shown to be consistent with laboratory and environmental data, pointing to these mechanisms as important regulators of viral competitive ecology in the ocean.
More recently, we have used highly simplified food web models to explore the relationship between phytoplantkon and microzooplankton in the surface ocean. Examples of the simulated relationship between phytoplankton and microzooplankton is shown in the upper left panel, below. The upper right panel shows simulated change in community composition along a transect in the Atlantic, indicated by the red dotted line on the map showing ocean primary production in the lower left panel. The lower right panel shows depth resolved total planktonic carbon along the same transect. Comparison of these model predictions with environmental data is available in a recent preprint, available here.
|
|
Outreach
Modeling - a 'Tool of Science'. Mathematical modeling is used in diverse areas, from weather prediction, to designing race cars and rockets, and environmental research. To help broaden understanding of how models are used to develop our understanding of natural processes, we teamed up with colleagues at Rutgers University, MIT, and Talapia Film, based in LA. The movie below is part of the Tools of Science project, that will be used to help high school students understand how models, and other tools, are used to develop our understanding of natural systems.
Modeling - a 'Tool of Science'. Mathematical modeling is used in diverse areas, from weather prediction, to designing race cars and rockets, and environmental research. To help broaden understanding of how models are used to develop our understanding of natural processes, we teamed up with colleagues at Rutgers University, MIT, and Talapia Film, based in LA. The movie below is part of the Tools of Science project, that will be used to help high school students understand how models, and other tools, are used to develop our understanding of natural systems.
Expanding your Horizons. A new generation of young women scientists and STEM graduates will be integral to the future and progression of science. In November 2019, post-doc Audra Hinson participated in a STEM (science, technology, engineering, mathematics) Activity Day hosted by National Institute for Mathematical and Biological Synthesis (NIMBioS) and the Expanding Your Horizons Network (EYH) for middle-school girls from around Knoxville and East Tennessee. Audra and other women faculty and staff from the University of Tennessee and Oakridge National Laboratory led STEM hands-on activities for 55 middle-school girls.
http://www.nimbios.org/education/horizons_conf
http://www.nimbios.org/education/horizons_conf
