Among Southern States, Alabama is also experiencing negative consequences such as habitat degradation, ecological instability, and loss of biodiversity, due to invasive tree species. We have studied the extent and spread of six major invasive tree species in Alabama: Chinese tallow tree, Silk tree, Princess tree, Tree-of-Heaven, Chinaberry tree and Russian olive. USDA Forest Service Forest Inventory and Analysis (FIA) data were used to collect subplot level information which included more than 5000 permanent FIA plots in the state measured from 2001 to 2018. Within the selected time frame, most of the plots were re-measured for three times. Invasive tree count per acre was calculated for each subplot and those counts were aggregated to the 308 hydrological units in the State. Spatial and temporal trend of the invasive trees were modeled using generalized additive model (GAM) and Bayesian hierarchical regression model. Zero-inflated Poisson function was used in both models as most of the subplots and aggregated hydrological units had excessive zero responses. GAM and Bayesian models were compared with both root mean square error using the 10 fold cross validation.
Results/Conclusions
Our results suggested that the presence of invasive tree species doubled during the period (from 0.77 to 1.95 percent). Both models suggested that location, basal area, stocking, and treatments were significant factors for predicting the likelihood and number of invasive trees. Predictions from the Bayesian model reflected true values, which show a growing extent of invasive trees abundance in the state over time. Furthermore, spreading was concentrated near the invasive tree sources in previous cycles.