Home Introduction Hydrogeology -Hydrostatigraphy -Aquifer Prop. -Water-Budget -Fresh/Salt- water Simulation of GW Discharge Conclusions References Appendix 1 Appendix 2 Appendix 3 Appendix 4 Plates PDF Version

Hydrogeology of Southeastern Florida

Freshwater-Saltwater Transition Zone

Parker and others (1955), Kohout (1960a, 1964), and many other investigators have shown that ground-water flow is affected by differences in ground-water density. In coastal settings, there are variations in ground-water density because the density of seawater is 2.5 percent higher than the density of freshwater. As fresh ground water flows toward the coast, it meets saline ground water that originated from the ocean, and the density differences affect the ground-water flow paths. Kohout (1960a, 1964) studied coastal ground-water flow by installing numerous monitoring wells along a transect perpendicular to the coast in the Cutler Ridge area (near Miami) of southeastern Florida. He used chloride concentration as a proxy for salinity and fluid density. A cross section showing lines of equal chloride concentration for September 18, 1958, indicates that a tongue of relatively dense, saline ground water extended inland from the coast at the base of the Biscayne aquifer (fig. 8). The cross section also indicates no apparent sharp interface between fresh and saline ground water; instead, the saltwater interface was a broad transition zone.

To illustrate the effects of the saltwater interface on ground-water flow, Kohout (1964) performed a flow-net analysis (fig. 8) that shows fresh ground water mixing with saline ground water as it discharges to Biscayne Bay. From this analysis, Kohout (1964) estimated that about 12.5 percent of the ground water discharged to Biscayne Bay was recirculated seawater. By linearly extending the zero horizontal gradient line, which marks the location where all ground-water flow is vertically upward (fig. 8), it appears that all ground-water discharge occurred within about 130 m of the shore.

Figure 9. Lines of equal chloride concentration for the Silver Bluff area, November 2, 1954 (modified from Kohout, 1964). [larger image]

Contours of ground-water salinity also were prepared by Kohout (1964) for the Silver Bluff area (near Miami) for data collected on November 2, 1954 (fig. 9). At Silver Bluff, the saltwater tongue extended inland from the coast about 3.7 km, more than 3 km farther than at Cutler Ridge.

Results from the ground-water sampling performed as part of this study are shown in table 2. Monthly measurements of specific conductance were converted to chloride concentrations using the relation established earlier in this report. Average values of chloride concentration, which were calculated from the monthly values of chloride concentration from March 1998 to February 1999, range from 30 to 19,470 mg/L. During the period from March 1998 to February 1999, the maximum range in chloride concentration (5,556 mg/L) was observed at G-3755. The observed range in chloride concentration for the remaining wells sampled during this period did not exceed 2,669 mg/L.

The chloride concentration of Biscayne Bay water may affect the chloride concentrations in the aquifer beneath Biscayne Bay. Chloride concentrations in the bay, however, are not continuously measured at the three transects. For this reason, the results from a hydrodynamic circulation model (John Wang, University of Miami, written commun., 2000), which simulates the period from January 1995 to December 1998, may provide insight into the temporal variation in chloride concentration at the three transects. The finite-element model has a spatial resolution of about 500 m, which means that simulated concentrations are representative within about 250 m of shore, but the concentrations are probably representative up to about 1,000 m. The relation between salinity and chloride concentration, as previously discussed, was used to convert simulated values of salinity to simulated values of chloride concentration. For the 4-year period simulated by the hydrodynamic circulation model, average values of chloride concentration at the Coconut Grove, Deering Estate, and Mowry Canal transects are 18,200, 16,200, and 12,400 mg/L, respectively. The ranges of simulated chloride concentration for the Coconut Grove, Deering Estate, and Mowry Canal transects are 8,000, 14,900, and 17,500 mg/L, respectively.

Figure 10. Lines of equal chloride concentration from March 1998 to February 1999 for the (A) Coconut Grove, (B) Deering Estate, and (C) Mowry Canal transects. [click on images above for larger versions]

Ground-water salinity data (table 2) and results from the TDEM soundings (table 3) were used to construct plots showing lines of equal chloride concentration for the Coconut Grove, Deering Estate, and Mowry Canal transects (figure 1 and figure 10). At Coconut Grove, the saltwater tongue appears to extend at least 4 km inland from the coast, although there is not enough field data to adequately characterize the inland portion of the transition zone (fig. 10). Offshore, the chloride data for Coconut Grove suggest that ground water discharges to the bay. A monitoring well (G-3654) just offshore has an average chloride concentration of 11,670 mg/L (table 2). Farther offshore, however, the monitoring wells have chloride concentrations similar to the average chloride concentration in the bay. This suggests that most fresh ground-water discharge at the Coconut Grove transect occurs within the first 300 m of the shoreline.

At Deering Estate, the inland extent of the saltwater tongue is less than 1 km from the coast. Offshore data indicate that fresh ground water is discharging to the bay because measured chloride concentrations beneath the bay are less than the average chloride concentration of the bay (fig. 10). A shallow monitoring well (G-3647) about 100 m offshore has a chloride concentration of about 3,810 mg/L (table 2), which suggests that brackish ground water is discharging to the bay 100 m from shore. The chloride data for Deering Estate also suggest that most fresh ground-water discharge to the bay occurs within about 500 m from shore (fig. 10).

At Mowry Canal, TDEM soundings were used to help determine the inland extent of the saltwater tongue. Resistivity values less than 10 ohm-m (ohm-meters) typically are considered indicative of saline ground water; however, the exact concentration cannot be determined from this method. Based on this assumption, the saltwater tongue extends inland about 6 km (fig. 10). Of the three transects, the chloride data suggest that Mowry Canal has the least amount of ground-water discharge to the bay, because chloride concentrations in the aquifer exceed 18,000 mg/L at distances of only 300 m from shore. Near shore, however, there is some fresh ground-water discharge to the bay; a chloride concentration of 9,195 mg/L was measured in a shallow well (G-3639) about 35 m from shore.

Figure 11. Location of saltwater intrusion lines in southern Florida based on previous studies, the Ghyben-Herzberg relation, and a geophysical survey. Differences in line location do not necessarily indicate movement of the saltwater intrusion line; differences may be due to interpretation, method, or data availability. [larger image]

When drawn on a map, the inland extent of the saltwater interface at the base of the Biscayne aquifer is referred to as the saltwater intrusion line. In an effort to protect ground-water resources in Miami-Dade County, the saltwater intrusion line is periodically located and mapped. Parker and others (1955) provide a thorough description of how saltwater in the aquifer migrated inland in response to the draining of the Everglades. More recently, the 1984 position of the saltwater intrusion line was mapped by Klein and Waller (1985). Later, Sonenshein (1997) used data from monitoring wells and TDEM to map the position of the saltwater intrusion line. Fitterman and Deszcz-Pan (1998) used airborne geophysical methods to map the position of the saltwater intrusion line for southern Miami-Dade County. These three data sources are combined in figure 11 to illustrate the inland extent of the transition zone. Differences between saltwater intrusion lines do not necessarily indicate movement of the saltwater interface; differences between saltwater intrusion lines may be the result of interpretation of different data sources. Also included in figure 11 is the saltwater intrusion line predicted by the Ghyben-Herzberg relation (Konikow and Reilly, 1999). The Ghyben-Herzberg relation is based on the balance of forces for a static ground-water system that is composed of fresh ground water overlying saline ground water. The relation states that the depth to the interface between fresh and saline ground water will be at a depth equal to 40 times the freshwater head value. For many areas, the Ghyben-Herzberg relation is a good indicator of the position of the saltwater intrusion line; however, for northern and southern Miami-Dade County, figure 11 suggests that the Ghyben-Herzberg relation is not valid, because it suggests that the interface should be located up to 10 kilometers inland of the observed saltwater intrusion line.

Figure 12. Ground-water monitoring wells in the Biscayne aquifer used to monitor the location of the saltwater intrusion line in Miami-Dade County. [larger image]

Parker and others (1955) described how anthropogenic changes have affected the position of the saltwater intrusion line. Today, the saltwater intrusion line is carefully monitored by a salinity-monitoring network (fig. 12), consisting of ground-water monitoring wells located on or near the saltwater intrusion line. A list of these monitoring wells is given in appendix III. Plots of chloride concentration relative to time were evaluated for each salinity-monitoring well to identify any long-term changes in the saltwater intrusion line that could have affected rates of fresh ground-water discharge to Biscayne Bay from 1989 to 1998. For most of the plots, which use a logarithmic axis for chloride concentration, there was no discernible trend in chloride concentration. Temporal plots of chloride concentration for the five wells that showed substantial changes in chloride concentration are shown in figure 13. Although wells G-432 and G-901 show increased chloride concentrations over time, concentrations tended to stabilize between 1,000 and 2,000 mg/L, suggesting the saltwater intrusion line moved slightly inland at both wells. A westward movement of the saltwater interface also could correspond with a decrease in ground-water discharge to Biscayne Bay. Based on this temporal analysis of chloride concentrations, it seems that the position of the saltwater intrusion line, for the most part, did not appreciably change. This observation is used later in the report to support development of the regional-scale ground-water flow model.

Darcy's law states that ground-water flow is linearly proportional to the hydraulic gradient, which means that ground-water flow to Biscayne Bay is affected by the water-table elevation and the stage in the bay. Average values of hourly, daily, and monthly stage for Biscayne Bay are shown in figure 14. The average stage values were calculated using the downstream monitoring station at structure S-123 (figure 4 and figure 5), which is located less than 1 km from the coast and in the central part of the study area. To ensure that the downstream stage values are not significantly affected by canal discharges or other potential influences, tide data from the Virginia Key station (fig. 1) also are included in figure 14 and the records match. Figure 14 suggests that the stage of Biscayne Bay can greatly affect ground-water discharge. Over a 12-month period, the average monthly stage of the bay can change by 0.4 m, as was the case in 1992. This is a considerable change considering that the range in the water-table elevations is about 3 m over the study area. Average daily stages and hourly stages also exhibit fluctuations up to 0.4 m, which may also affect ground-water discharge over shorter time periods.