Integrating Systems Models and Remote Sensing to Explore Aquatic Ecosystem Vulnerability to Global Change in Lake Huron
Clementina Calvo, Texas State University, tmu22@txstate.edu (Presenter)
Michael Battaglia, Michigan Tech Research Institute, mjbattag@mtu.edu
Laura Bourgeau-Chavez, Michigan Tech Research Institute, lchavez@mtu.edu
William Currie, University of Michigan, wcurrie@umich.edu
Kenneth Elgersma, University of Northern Iowa, kenneth.elgersma@uni.edu
David Hyndman, University of Texas, Dallas, hyndman@utdallas.edu
Anthony Kendall, Michigan State University, kendal30@msu.edu
Michael Sayers, Michigan Tech Research Institute, mjsayers@mtu.edu
Jason Martina, Texas State University, jpmartina@txstate.edu
The integrity of the Great Lakes (GL), Earth's largest surface freshwater reserve, is intricately connected to its wetlands. Aside from direct anthropogenic threats, GL wetlands are under increasing pressure from climate change, nutrient loading, invasive plants, and rapid water level changes. These pressures could cause coastal wetlands to switch from being nutrient sinks to sources, affecting submerged aquatic vegetation and phytoplankton production in the nearshore areas. This project aims to quantify the function and role of GL coastal wetlands in controlling the vulnerability of nearshore aquatic ecosystems to disturbances. To achieve this, we are improving the linkage between Mondrian, a patch-level wetland ecology model, and Landscape-Hydrology-Model, a large-scale hydrology model. To overcome large computational and scale discrepancies between them, statistical meta-models are being created that will allow us to simulate ecosystem- and location-specific wetland function across large domains. We completed 100K Mondrian simulation runs to explore the abiotic and biotic parameter space of wetlands in the region, and are training a variety of machine learning algorithms to replicate these computer-intensive Mondrian runs. Additionally, electro-optical satellite image data and specific bio-optical algorithms are being used to generate time-series observations of water quality parameters and assess linkages between coastal wetlands and nearshore aquatic ecosystems. We utilized multi-source satellite remote sensing data to generate circa 2018 land cover maps identifying coastal wetlands and invasive species of interest, and are generating time-series inundation extent maps. Finally, we created the Coastal Ecohydrology Training Undergraduate Program, a 2-year cross-institutional cohort of six underrepresented undergraduate students per year working directly with the grant’s PIs to gain training in remote sensing, data science, and simulation modeling, while getting field experience in the GL. The 2022-cohort successfully completed their research projects and presented at regional conferences. Our research will help elucidate the interactions between the GL and coastal areas, and how upland hydro-ecological processes influence coastal ecology and near-shore water quality.
Associated Project(s):
Poster Location ID: 2-59
Presentation Type: Poster
Session: Poster Session 2
Session Date: Wed (May 10) 5:15-7:15 PM
CCE Program: OBB