Timescales for nitrate contamination of spring waters, northern Florida, USA

Residence times of groundwater, discharging from springs in the middle Suwannee River Basin, were estimated using chlorofluorocarbons (CFCs), tritium ((3) H), and tritium/helium-3 (H-3/He-3) age-dating methods to assess the chronology of nitrate contamination of spring waters in northern Florida. During base-flow conditions for the Suwannee River in 1997-1999, 17 water samples were collected from 12 first, second, and third magnitude springs discharging groundwater from the Upper Floridan aquifer. Extending age-dating techniques, using transient tracers to spring waters in complex karst systems, required an assessment of several models [piston-flow (PFM), exponential mixing (EMM), and binary-mixing (BMM)] to account for different distributions of groundwater age. Multi-tracer analyses of four springs yielded generally concordant PFM ages of around 20 +/- 2 years from CFC- 12, CFC- 113, H-3, and He-3. with evidence of partial CFC- 11 degradation. The EMM gave a reasonable fit to CFC- 113, CFC- 12. and H-3 data, but did not reproduce the observed He-3 concentrations or H-3/He-3 ratios, nor did a combination PFM-EMM. The BMM could reproduce most of the multi-tracer data set only if both endmembers had H-3 concentrations not much different front modern values. CFC analyses of 14 additional springs yielded apparent PFM ages from about 10 to 20 years from CFC- 113, with evidence of partial CFC- 11 degradation and variable CFC-12 contamination. While it is not conclusive, with respect to the age distribution within each spring, the data indicate that the average residence times were in the order of 10-20 years and were roughly proportional to spring magnitude. Applying similar models to recharge and discharge of nitrate based on historical nitrogen loading data yielded contrasting trends for Suwanee County and Lafayette County. In Suwance County, spring nitrate trends and nitrogen isotope data were consistent with a peak in fertilizer input in the 1970s and a relatively high overall ratio of artificial fertilizer/manure whereas in Lafayette County, spring nitrate trends and nitrogen isotope data were consistent with a more monotonic increase in fertilizer input and relatively low overall ratio of artificial fertilizer/manure. The combined results of this study indicate that the nitrate concentrations of springs in the Suwannee River basin have responded to increased nitrogen loads from various sources in the watersheds over the last few decades, however, the responses have been subdued and delayed because the average residence time of groundwater discharging from springs are in the order of decades. (C) 2001 Published by Elsevier Science B.V.