Function over Form: Predicting Disturbance Responses with Functional Traits at ESA 2016
A guest post from PLOS Ecology Reporting Fellow, Uma Nagendra, on research from the Ecological Society of America Scientific Meeting in Ft. Lauderdale, Florida, August 7-11, 2016.
In April 2010, the Deepwater Horizon oil rig exploded, releasing millions of barrels of crude oil into the Gulf of Mexico. Despite attempts to slow the spread, oil would inevitably blanket the gulf coast marshes and beaches. These vulnerable areas provide crucial ecosystem services as well as home base for a vibrant seafood industry. After the disaster, environmentalists and fishermen alike asked ecologists to tell them: What’s going to happen to this place? How will the ecosystem change after the sudden coating of oil?
For most situations, however, ecologists still struggle to precisely forecast community trajectories. The scuffle to survive, compete for resources, and flourish after a sudden event differs by so many factors that accurately modelling exactly which species will return in which proportions is a nearly impossible task.
One session, OOS 20: Quantifying Responses of Functional Community Assemblages to Disturbance: A Predictive Tool in a Changing World, at The Ecological Society of America proposed an alternative predictive tool: functional traits. The session— organized by Wendy Leuenberger (SUNY-ESF), Ross Boucek (Florida International), and Ashley Asmus (UT-Arlington)—focused on characterizing a community not only in terms of the species it contains, but also the roles and functions those species can do. By searching for patterns in organism function that emerge after disturbances, they aim to improve our ability to forecast ecosystem responses in our ever-changing world.
This is precisely how Luis Miguel Rodriguez (GA Tech) set out to examine how the Deepwater Horizon oil spill would affect gulf coast microbial communities. He found that, although the number of microbial species decreased after oil reached gulf coast beaches, the species that did persist actually performed a wider variety of functions. While oil was still on the beaches, generalist oil-degrading microbes flourished. Within a year, however, the microbial community returned to nearly pre-oil conditions.
Several studies presented at the session highlighted how resilient ecosystem function could be, even when the community itself appears to be completely different. Mariya Shcheglovitova (University of Maryland Baltimore County) found surprising functional resilience in urban Baltimore streams, for instance. She visited waterways from downtown Baltimore to a secluded rural creek, collecting logs and branches that had fallen into the water. The slick, slimy biofilms coating the wood are thought to play an important role in sucking up nitrogen and pollutants from the water. Despite broad differences in stream quality between her sites, Shcheglovitova found that the biofilms in each area could still perform many of the same functions. This high level of functional redundancy between urban and rural areas indicated that wood-surface biofilms may be resilient to the stresses of urbanization—maintaining their ability to transform pollutants and recycle aquatic nitrogen even in the heart of Baltimore.
Fan Li (University of Houston) used functional traits to assess the fate of Altamaha River marshes in the face of salty tides. As sea levels rise, periodic pulses of salt water start to encroach on freshwater marshes, leaving plants salt-stricken and stunted. Li wanted to know whether freshwater species would persist after fresh water returned— or if even a brief salty spell would permanently change the community to a salt marsh. By constructing artificial marsh mesocosms with carefully manipulated saline “floods,” Li was able to determine that freshwater plants can rebound even after moderate salinity pulses. Once doused with high salinity water, however, the marsh remained dominated by salt-tolerant species only—fresh water plants didn’t return even after nearly a year of only receiving freshwater. Despite this change in community composition, plant productivity remained high. Salt water encroachment may change how a marsh looks and feels, but perhaps at least some of the community functions may stay the same.
Each presentation in this session contributed an example of how functional traits could potentially help researchers predict communities respond to disturbances. But together, they uncovered numerous nuances of how to identify, measure, and interpret a functional trait. For instance, both Kaitlin Kimmel (University of Minnesota) and Robert Shriver (Duke) were interested in how plant communities would respond to altered global precipitation patterns. Like many other functional trait researchers, Kimmel gleaned data from long term ecological research (LTER) sites and classified each species based on a published database of plant traits. This commonly-used method assigns each species a value for each trait, such as root length or leaf longevity. By averaging out regional trait variation and environmental plasticity, the method allows researchers to take advantage of large datasets where individual measurement is just not possible.
Shriver, on the other hand, focused on one region and a smaller number of species. This allowed him to go out and measure how the traits he was interested in (photosynthetic and transpiration rates) varied in a range of conditions.
Kimmel’s results didn’t show large functional shifts in two out of three LTER sites, even with strong experimental treatments. Shriver’s study showed clear functional tradeoffs for plants surviving in dry and wet conditions. Although the difference in their results can also be explained by the many other differences between their studies, the contrast between the two approaches brings up an issue that functional trait researchers will need to address. When must you measure a trait directly, and when can you bring in the big data?
As human and wild worlds continue to collide through urbanization, fragmentation, and a changing climate, we are experiencing more and more novel disturbance combinations. The ability to anticipate community-level responses to these disturbances will be key to management, restoration, and conservation efforts. As this ESA session showed, functional traits can be a powerful tool to bring out broad community patterns. In order to accurately use functional traits as predictive tools, however, researchers will have many nuances to untangle. Thankfully, the organizers and researchers of this session are on the case. Even now that the conference has ended, they are continuing a conversation on functional traits. Keep an eye out for these early career researchers—they’ll have a lot to share.
Uma Nagendra is a PhD candidate in the Department of Plant Biology at The University of Georgia. Her research investigates the impacts of tornadoes on plant-soil interactions in the Southern Appalachian Mountains. She uses a combination of greenhouse and field experiments to examine relationships between tree seedlings and specialist soil organisms in wind-damaged forests. When she is not crawling under fallen trees in North Georgia, Uma collaborates with aerialists and dancers to create vibrant, interactive dance pieces about scientific concepts. Umaalso writes and edits for the Athens Science Observer, where she covers natural disturbances, plant ecology, and art+science collaborations. Follow Uma through her research website or on twitter at @atinytornado