Electronic Theses and Dissertations

Date

2024

Document Type

Thesis

Degree Name

Master of Science

Department

Earth Sciences

Committee Chair

Youngsang Kwon

Committee Member

Dorian Burnette

Committee Member

Gary Stinchcomb

Abstract

This study aims to investigate the spatial distribution of tree diversity in the forests of eastern 31 states of US, primiarily focusing on tree species richness (TSR) and the functional group diversity (FD), and how their patterns are shaped by climatic variables, soil properties, and topographic factors. Assessing both SR and FD is necessary as it provides a better understanding of the biodiversity and the ecosystem functionality. In the presented study, species richness is defined as “the number of different species present in a particular ecological community or region,” while functional diversity is defined as “the variety of traits possessed by the plants in a particular geographical location.” The analysis is conducted by utilizing data from the Forest Inventory and Analysis (FIA) program which covers tree communities across the eastern 31 states of the US. FIA plot-level data is post-processed and aggregated into a 20 km by 20 km grid system to investigate the broad-scale spatial patterns of TSR and its FD. Results showed both TSR and FD increase from coastal regions to the Appalachian Mountains; however, high TSR did not always correspond to high FD, and vice versa. For an instance, the Midwest region exhibits higher FD compared to TSR, while the north-western corner shows comparatively higher FD than TSR. Spatial autocorrelation analyses, including Global Moran’s I test, and hotspot/coldspot analysis were conducted to assess the spatial distributions of these two diversity measures. The results highlight the overall patterns of TSR and FD, exhibiting that both the diversity measures are dependent upon the geography of the study area, with the TSR being more spatially dependent than the FD. The most significant hotspots for both the metrics are located in the Appalachian Mountain Region, likely due to diverse topography in this region creating numerous microhabitats. Hotspots are also observed in the Northern Hardwood Forests and the Northern Lakes and Forest regions, representing an ecotone between boreal and temperate forests. On the contrary, southern portion of the country, particularly Florida, emerged as coldspots for both the metrics, possibly due to the peninsula effect, where the diversity seem to decline towards the peninsula's tip due to reduced plant colonization rates and increased plant extinction rates. The differences in the spatial patterns of TSR and FD are primarily observed to be driven by variations in climate, soil, and topography, corresponding to the species’ habitats. A Random Forest machine learning approach, supplemented by SHAP (Shapley Additive Explanations) analysis, was utilized to further explore the observed patterns of TSR and FD. This analysis identified warm temperatures and aridity as the two most important variables influencing TSR. For FD, warm temperatures and topographic factors, particularly RDEM (relative digital elevation model), were the most significant predictors. These findings suggest that, while TSR and FD are influenced by similar environmental factors, the plant communities may respond to these factors in distinct ways.

Comments

Data is provided by the student.”

Library Comment

Dissertation or thesis originally submitted to ProQuest.

Notes

Open Access

Download

Share

COinS