Testing Indicators of State Flips Along Spatial and Temporal Gradients of Environmental Conditions

Natural ecosystems such as oceans, grasslands, and deserts can flip abruptly between different states in the face of gradual change, but predicting and understanding these flips is difficult because they occur unexpectedly. Without knowing when such flips occur, it is hard to study them. This research will provide novel tests of theories suggesting that clues exist which predict when state flips are approached using photo-transects of shoreline plant and animal communities that show clear, predictable flips across space, rather than time. The work will also extend and analyze unusually long and parallel data series to test the predictive efficacy of state flip indicators. If the theories are successful, they could form the basis for useful quantitative biotechnologies to manage fisheries, agriculture and forestry. The next generation of scientists will be trained, ranging from high school students to post-doctoral fellows and biological technicians. Local stakeholders will participate in the research. Recent theory suggests that the temporal dynamics of ecosystems encode leading indicators that state flips are about to occur, such as a rise in variance and the strength of temporal autocorrelation. Understanding state flips is challenging to study because we don’t know whether a particular system will undergo state flips, when they will occur, and a lack of replication in state flips. To circumvent these limitations, this project empirically tests these ideas in the strong zonation patterns across space classically exhibited on rocky shores. This zonation represents multiple abrupt state flips across a gradual spatial gradient of tidal immersion, allowing the study of ecological dynamics as we approach known, replicated critical transitions and the exploration of whether different behaviors arise as stress increases or decreases. The research will involve Artificial Intelligence and adjacent Bayesian/Hidden Markov applications to image analysis and pattern exploration. Temporal series of ecological dynamics across replicated permanent transects spanning a tidal gradient will test the effectiveness of proposed leading indicators of where state flips will occur. Multi-decadal data series of physical, chemical and biological parameters will be probed for evidence of state flips through time. Together, extensive spatial and temporal data series will substantially advance our understating of the reliability of proposed leading indicators of state flips in ecosystems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2607252 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: John Wootton | Institution: University of Chicago, CHICAGO, IL | Award Amount: $990,905 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2607252 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2607252.html