Author Type

Graduate Student

Date of Award

Spring 4-14-2026

Document Type

Dissertation

Publication Status

Version of Record

Submission Date

April 2026

Department

Geosciences

College Granting Degree

Charles E. Schmidt College of Science

Department Granting Degree

Geosciences

Degree Name

Doctor of Philosophy (PhD)

Thesis/Dissertation Advisor [Chair]

Caiyun Zhang

Abstract

Accurate quantification of atmospheric carbon dioxide (CO₂) fluxes in wetland ecosystems is essential for understanding their role in both regional and global carbon dynamics, particularly in the context of climate change. However, the spatial and temporal heterogeneity of wetlands presents major challenges for developing reliable upscaling models. This research developed and validated a comprehensive framework to upscale CO₂ fluxes across the Everglades National Park (ENP) and Big Cypress National Preserve (BCNP) in South Florida through the integration of multi-source datasets, including AmeriFlux eddy covariance (EC) tower measurements, NASA’s BlueFlux airborne CO₂ data, and multispectral satellite imagery from Landsat 8 OLI and Sentinel-2 MSI sensors. The framework incorporated Object-Based Image Analysis (OBIA) and a suite of sophisticated machine learning (ML) algorithms, including Random Forest (RF), Support Vector Machine (SVM), k-Nearest Neighbor (KNN), Extreme Gradient Boosting (XGB), and the traditional Multiple Linear Regression (MLR). Weighted and Meta-Ensemble Analysis (EA) techniques were also applied to further improve prediction performance. Three consecutive seasons from December 2021 to April 2023 were analyzed to capture seasonal variability in wetland carbon dynamics.

Results demonstrated that integrating airborne BlueFlux data with Sentinel-2 MSI imagery produced the highest prediction accuracies (r up to 0.95), outperforming both EC tower–based models and Landsat 8 imagery. Among all algorithms, Weighted and Meta-Ensemble Analysis consistently achieved superior performance across all datasets and seasons, highlighting their robustness in integrating heterogeneous predictors. The study revealed distinct seasonal CO₂ flux patterns, with positive fluxes (carbon source) during dry seasons and negative fluxes (carbon sink) during wet seasons, reflecting the hydrological control on carbon exchange. Key environmental drivers identified across all seasons include water table level and air temperature. These variables collectively explained much of the seasonal variation in CO₂ fluxes, emphasizing the dominant roles of hydrology and temperature in regulating carbon emissions from subtropical wetlands. The superior performance of the Sentinel-2 MSI imagery was attributed to its higher spatial resolution (10 m), frequent revisit intervals, and enhanced spectral sensitivity, which improved detection of fine-scale vegetation and moisture variations across the heterogeneous Everglades landscape.

This research provides a scalable and transferable framework for CO₂ flux upscaling that effectively combines ground-based, airborne, and satellite observations using advanced ML and ensemble modeling. The findings highlight the significant potential of NASA’s BlueFlux airborne campaign and Sentinel-2 MSI data for regional carbon monitoring and lay the groundwork for future applications of this framework to other wetland systems worldwide.

Download

Share

COinS