COMPARISON OF UNIVARIATE AND TRANSFER-FUNCTION MODELS OF GROUNDWATER FLUCTUATIONS

Seasonal autoregressive integrated moving average (SARIMA) univariate models and single input-single output transfer function (SARIMA with e xternalities or SARIMAX) models of groundwater head fluctuations are d eveloped for 21 Upper Floridan aquifer observation wells in northeast Florida. These models incorporate empirical relationships between rain fall input and head response based on historical correlations and cros s correlations between these two time series. The magnitude of the for ecast error terms indicates that the SARIMA and SARIMAX models explain an average of 84-87% of the variation observed in the monthly piezome tric head levels for 1-month lead forecasts. Thus the models account f or the dominant processes which affect temporal groundwater fluctuatio ns. Both the SARIMA and SARIMAX models provide unbiased forecasts of p iezometric head levels; however, the SARIMAX models produce more accur ate forecasts (i.e., smaller forecast probability limits) than the SAR IMA models, particularly as lead time increases. Modeling efforts reve al consistent model structures over the study region, with local hydro logic and geologic conditions causing site-specific variability in the time series model parameters.