Quantifying Spatiotemporal Vegetation Change in the Southeastern United States

Abstract

Shifts in vegetation affect ecosystem functioning related to flood intensity, climate regulation, and biodiversity support, to name only a few. In the Southeastern United States (SEUS), historical vegetation decrease, quantified by vegetated landcover (LC), has mostly been associated with agricultural land area. However, literature suggests that in recent years vegetated LC decrease is more associated with expanding urban development. Thus, the purpose of this study is to quantify where vegetated LC changes have been occurring while understanding major factors (i.e., agriculture/urban development) affecting vegetated LC pattern shifts across time. Understanding patterns of vegetated LC change is critical for decision-making that supports ecosystem functions provided by vegetation. This research utilizes the 1985 to 2023 annual National Land Cover Dataset (NLCD) and spacetime cube spatiotemporal analyses to determine statistically significant areas of vegetated LC change and correlated LC changes in the SEUS. Results showed spatial variation across the study area. The majority of the SEUS had higher concentrations of vegetated LC that significantly decreased over time and were inversely correlated with developed LC. However, urban areas such as Huntsville, Alabama, had low vegetated LC concentrations, but similarly experienced vegetated LC decrease and inverse correlation with developed LC. Spatial patterns corresponding to the Black Belt region occurred across analyses. Black Belt patterns diverged from SEUS patterns, with vegetated LC increase and inverse correlation to agricultural LC, potentially indicating agricultural abandonment, land repurposed for timber, or exurban expansion. Another area of significant vegetated LC increase was the Florida Green Swamp conservation project, which demonstrated successful wetland and forest conservation. The results of this study offer insights for environmental planning, including rural conservation or urban green infrastructure efforts. This research used novel methodological techniques applicable to other areas of interest, and through quantifying spatiotemporal vegetated LC patterns could assist in future decision making that supports vegetation conservation.