DETERMINING SOURCES OF NITRATE AND RESIDENCE TIMES OF GROUNDWATER DISCHARGING FROM SPRINGS IN A KARST SYSTEM DURING VARIABLE RECHARGE CONDITIONS

During 1997-2000, water samples were collected annually from five first- and second-magnitude springs along the Suwannee River and analyzed for d¹⁵N-NO₃, dissolved gases (Ar, N₂, N₂O), chlorofluorocarbons, and ³H/³He, as part of a study of nitrate contamination sources and residence time of groundwater in northern Florida. Springs were sampled during low-flow conditions for the Suwannee River; however, there was an excess of 69.55 cm of rainfall relative to the long-term monthly averages during January 1997 through October 1998, and a rainfall deficit of 78.46 cm during November 1998 through July 2000. In response to lower rainfall, spring flows decreased in 2000 by 24% (Manatee Spring) to 54% (Little River Spring) compared to flows in 1997 and 1998. Correspondingly, during 1997-2000, nitrate-N concentrations in the five springs decreased by 0.2 (Telford and Manatee Springs) to 1.3 mg/L (Troy Spring); d¹⁵N-NO₃ values increased by 0.1 (Manatee Spring) to 1.8 per mil (Mearson Spring); dissolved O₂ concentrations decreased by 0.3 to 1.2 mg/L, and apparent ages (CFC-12, CFC-113, ³H/³He) increased. These changes all would be consistent with the presence of a fraction of relatively old, reduced and denitrified water that increased in relative abundance as spring flow decreased. Alternatively, these changes also are consistent with the presence of older, more reduced non-denitrified water containing nitrate originating from more of an animal waste source. Denitrification is considered to be minor in the aquifer based on low DOC concentrations (<0.2 mg/L) and persistence of O₂ (0.5 to 3.2 mg/L). Furthermore, low excess N₂ values (less than 1 mg/L) were estimated based on measured concentrations of N₂ and Ar in spring waters that were consistent with atmospheric equilibration (21±2 ^oC) during groundwater recharge. Nitrous oxide was not detected in spring waters sampled in 2000, indicating that active dentrification was not occurring, but denitrification some time in the past cannot be ruled out, as small amounts of N₂O were detected in Mearson Spring samples in 1998 and 1999. Based on historic county-wide estimates of nitrogen inputs to groundwater, higher proportions of land-applied poultry and cow manure relative to synthetic fertilizer in the early 1970's may account for the slight increase of d¹⁵N-NO3 in spring waters sampled in 2000.