Skip to main content

Unraveling Coastal Wetland Root Dynamics Through Imaging and Deep Learning


EMSL Project ID
61028

Abstract

Coastal wetlands are valuable ecosystems. They improve water quality, provide wildlife habitat and biodiversity, sequester carbon, and very importantly protect coastal communities from hurricanes by dampening waves, distancing the urban centers from open water, and reducing storm surge heights. Coastal ecosystem models of wetland platforms need to consider how the root productivity and turnover rates evolves over space and time by eco-geomorphic processes, and how, in turn, this evolution impacts their resilience towards transient disturbances. Our specific research objectives include the following: (1) Perform macro and micro XCT, OCT, and SEM scans in tandem with tensile strength tests on individual live to dead roots; and (2) Develop and validate AI/ML model to fuse multimodal and resolution data types for segmenting live, dead, and decaying roots, root architecture, pore structure, and sediment density. This will be achieved through a comprehensive and synergistic collaboration with EMSL. We will specifically leverage the x-ray computed tomography and scanning electron microscopy capabilities at LSU, and we will use the optical coherence tomography capabilities at EMSL. This limited scope proposal will build our case for a large-scale proposal.

Project Details

Project type
Limited Scope
Start Date
2024-03-25
End Date
N/A
Status
Accepted

Team

Principal Investigator

Navid Jafari
Institution
Louisiana State University

Co-Investigator(s)

Robert Twilley
Institution
Louisiana State University

Leslie Butler
Institution
Louisiana State University

Team Members

Andre Rovai
Institution
US Army Corp of Engineers Engineer Research and Development Center

Mohamed Hassan
Institution
Louisiana State University

Kyungmin Ham
Institution
Louisiana State University