Aquatic invertebrates found in mountain streams — crayfish, stoneflies and mayflies, among others — are important to ecosystems because they are part of the natural food web and are often used by state agencies as indicators of freshwater health.
Soon, land managers will be able to track the behaviors of these invertebrates using a computer model developed by a research team that includes a Virginia Tech aquatic ecologist.
The model, supported by a National Science Foundation grant, will simulate different possible natural responses to environmental changes by considering the location and shape of a river network and the types and behavior of invertebrate species within it. Land managers will then be able to use it when deciding how to restore watersheds after a disturbance, such as a flood.
“We’re thinking about this in terms of river networks, but the model would really apply to any set of populations or communities on the landscape,” said Bryan Brown, an associate professor of biological sciences in the College of Science and a Global Change Center at Virginia Tech affiliate. “What we envision is for land managers to plug in their scenarios and tweak the parameters so the model responds like nature.”
To develop the model, the team will first perform a large-scale analysis of past research that explores how communities of freshwater invertebrates change and disperse over time around rivers and streams. These aquatic communities are made of different species, each with their own distinct life cycles and movement patterns that tie them to the landscape. Mayflies, for example, begin life in headwater streams where they live as larvae for up to two years in stream-bottom sediment. During this stage of their life cycle, they feed on algae and move downstream by drifting. Once larvae complete metamorphoses, the winged adults fly back upstream to lay eggs.
These movement patterns are affected by the shapes of rivers and tributaries within watersheds, connections between forest fragments, and environmental conditions, such as prevailing winds. Taken together, these invertebrate movements and landscape conditions will be used to create models or "networks" to predict how ecological systems respond to human and natural disturbances. The model will also be used to identify what areas may be most sensitive to those disturbances.
Beginning in the spring, Brown will lead a series of field experiments at the Coweeta Long-Term Ecological Research site in Franklin, North Carolina. There, he and lab members, including new biology Ph.D. student and Interfaces of Global Change Fellow Sara Cathey, will set up artificial streams to compare invertebrate abundance in headwaters to larger intermediate zones. In the process, they will induce natural disturbances to see how invertebrates respond and in what locations the species communities are the most stable.
In a study published in July, Brown and collaborator Chris Swan at the University of Maryland found that the most effective areas to restore are the headwaters, likely because these are isolated sections of watersheds. Although these areas tend to be the most vulnerable to environmental change, they are more likely to sustain stable communities when restored compared to intermediate zones – larger streams that have more well-connected populations of species.
The Coweeta field experiments and initial meta-analysis will then be merged and computerized, making one “grand synthetic model,” said Brown.
Along with three collaborators, including Virginia Tech alumnus Eric Sokol, now at the University of Colorado-Boulder, Brown will simulate networks of hundreds of species communities continuously. The computer model will then allow users to explore possible scenarios based on different watershed shapes, organism types, and novel disturbances that cannot be studied with current research. Users will also be able to consider different parameters, such as temperature and rainfall.
“Understanding how things are responding in different areas or how different shapes of networks respond to different sorts of disturbances, such as increased precipitation, also helps us get a handle as climate starts to fluctuate more and more,” Brown said. “Hopefully this [model] can help us predict what areas are going to be more sensitive than others as the climate changes.”
The project is funded for three years by NSF’s Division of Environmental Biology with grant DEB-1655927.
Written by Cassandra Hockman