This Is AuburnElectronic Theses and Dissertations

Generation and evaluation of forest aboveground biomass and irrigation products for the Southern United States using Earth Observation data




Islam, Md. Mozahidul

Type of Degree

Master's Thesis


Forestry and Wildlife Science

Restriction Status


Restriction Type


Date Available



The conservation of natural resources in the southern United States plays a pivotal role in addressing climate change, safeguarding its rich forest resources, maintaining ecological integrity, ensuring food sources, and sustaining the national economy. This region, characterized by rapid population growth and agricultural production, demands special attention for its holistic development. In light of this, the study encompasses two interrelated studies: the first deriving high-resolution spatially continuous aboveground biomass (AGB) maps using NEON’s airborne lidar, while the second delves into identifying zones with varying irrigation water demands to enhance agricultural productivity. Chapter 2 underscores the critical need for accurate high-resolution AGB mapping to support ecological integrity, carbon dynamics, and natural resource management. Despite the challenges associated with AGB estimation due to limited data, the National Ecological Observatory Network (NEON) offers a valuable resource with quality-controlled, consistent datasets at various spatial scales. This study leverages NEON's airborne lidar point clouds and plot-level vegetation structure data from Ordway-Swisher Biological Station (OSBS) in Florida, Talladega National Forest (TALL) in Alabama, and Oak Ridge National Laboratory (ORNL) in Tennessee. By utilizing lidar-derived canopy metrics, Sentinel imagery, and ancillary data, AGB is predicted using multiple linear regression (MLR) and Random Forest (RF) models. MLR outperforms RF across all sites, yielding R² values of 0.91, 0.52, and 0.63 for OSBS, TALL, and ORNL respectively, compared to RF's 0.67, 0.20, and 0.40. The %RMSE is 31.32%, 32.7%, and 30.75% for MLR, as opposed to 71.44%, 44.21%, and 45.47% for RF, respectively. Sites with steep slopes and complex structures (e.g., ORNL and TALL) exhibit lower accuracy, highlighting the need for an increased number and distribution of field plots to enhance AGB estimation accuracy. In summary, this study sheds light on the potential and limitations of NEON's datasets in generating spatially continuous AGB estimates. Chapter 3 shifts focus towards optimizing agricultural efficiency and ensuring the sustainable allocation of water resources by comprehending the specific water requirements unique to the region, particularly the dynamics of root zone soil moisture. This research concentrates on three key southern states- Alabama (AL), Florida (FL), and Georgia (GA), recognizing their substantial contributions to agricultural production, the national economy, and the regional ecological equilibrium. Departing from traditional manual data collection, this study employs a data-driven approach to delineate irrigation potential zones. The methodology entails monthly averaging of daily evapotranspiration, precipitation, and root zone soil moisture during the cropping season (May to October) from 2015 to 2023. A comprehensive correlation analysis unveils the intricate relationship between water demand (expressed as potential evapotranspiration – precipitation) and root zone soil moisture, exhibiting a negative correlation pattern spanning from -0.06 to -0.78. The correlation map was categorized into five distinct classes, highlighting varying water requirements in different regions. Analysis of variance (ANOVA) establishes the high overall significance of the correlation categories and varying irrigation required regions (F = 38.59, p < 0.05), and Tukey's Honest Significant Difference (HSD) post hoc test confirms the statistical significance of all categories within the p < 0.05 range. These research findings equip stakeholders with invaluable insights to conserve resources and promote sustainable agricultural practices. By seamlessly integrating data science, remote sensing datasets, and agricultural resource management, this study emerges as a pivotal catalyst for optimizing irrigation practices and ensuring the sustainability of precision agriculture in the region.