Purpose and Scope

The purpose of this report is to document the interpretation of water quality and borehole geophysical data and their integration into a map of the location of the saltwater interface at the base of the Biscayne aquifer in Miami-Dade County in 2022. Water-quality data, including chloride concentrations and specific conductance of water samples from monitoring wells, specific conductance profiles in long-open interval monitoring wells, and time-series electromagnetic induction log (TSEMIL) datasets, were obtained from existing data sources. These data were evaluated and entered into geographic information system (GIS) software to construct a map of the inland extent of saltwater intrusion in the Biscayne aquifer in 2022. Changes in the location of the freshwater/saltwater interface at the base of the aquifer from 2018, when the interface was last mapped, to 2022 are described. Potential improvements to the saltwater intrusion monitoring network and strategies are also discussed.

Hydrogeologic Framework of the Study Area

There are two primary aquifer systems in Miami-Dade County, the surficial aquifer system and the Floridan aquifer system. The Biscayne aquifer is part of the surficial aquifer system, covers approximately 10,400 square kilometers, and underlies parts of Palm Beach, Broward, Miami-Dade, and Monroe Counties (Miller, 1990; Renken and others, 2005; Williams and Dixon, 2015; Williams and Kuniansky, 2015). The Biscayne aquifer is wedge shaped and is thickest toward its eastern extent near the coast (approximately 120 m), thinning out toward its inland extent in the west. Fish and Stewart (1991, p. 15–16) defined the Biscayne aquifer as

The surficial aquifer system is separated from the Floridan aquifer system by a thick sequence of clay, silt, limestone, and fine sand, known as the intermediate confining unit. Most of the intermediate confining unit is composed of the Miocene-Pliocene Hawthorn Group, but in some local areas, it includes parts of the Pliocene Tamiami Formation (Riggs, 1979; Fish and Stewart, 1991; Renken and others, 2005). The Floridan aquifer system covers approximately 260,000 square kilometers of the southeastern United States and extends to a depth of about 1,200 m below land surface in southeastern Florida (Williams and Dixon, 2015; Williams and Kuniansky, 2015).

Data Compilation and Processing

The location of the saltwater front at the base of the Biscayne aquifer was mapped as the 1,000-mg/L isochlor (sheet 1). To construct the map showing the location of the freshwater-saltwater interface in the Biscayne aquifer in Miami-Dade County in 2022, well construction information, chloride concentrations, and (or) specific conductance measurements were compiled. These data were collected by the USGS, Florida Power & Light (FPL), the Florida Keys Aqueduct Authority (FKAA), and MacVicar Consulting, Inc. (referred to as “MacVicar” herein where company wells are discussed) and have been published by Zhang and Renshaw (2024). One-hundred and six wells were selected on the basis of their proximity to the previously mapped position of the saltwater interface. The sampling, analysis, and quality-assurance procedures for the water-quality data varied depending on which agency collected the data. Summaries of each agency’s procedures can be found in Prinos (2019).

Seventy-six of the wells used for this mapping effort are sampled by the USGS as a part of Miami-Dade County’s saltwater intrusion monitoring program. The frequency of data collection in these wells varies from site to site and includes monthly, quarterly, or semiannual sample collection, specific conductance measurement, and chloride analysis. Data from samples collected between January 1, 2022, and December 31, 2022, were downloaded from the USGS National Water Information System (NWIS) (U.S. Geological Survey, 2023b) using an R-script (R Core Team, 2022) that automatically selected the maximum chloride concentration for each site during this period. Well construction and location information for all the USGS monitoring sites were also obtained from the USGS NWIS database (U.S. Geological Survey, 2023b). Well construction information is reported as depth below land surface in the NWIS database. These depths were converted to altitude using the land surface altitude, referenced to the National Geodetic Vertical Datum of 1929 (NGVD 29), that was also retrieved from the NWIS database.

The remaining 30 wells used for this project are monitored by the FPL, FKAA, and MacVicar Consulting, Inc., and are located in southeastern Miami-Dade County. Well construction information and chloride concentrations from FPL’s quarterly groundwater quality monitoring program were obtained from their 2022 (June 2021–May 2022) and 2023 (June 2022–May 2023) annual monitoring reports (Florida Power & Light, 2022, 2023). The five FPL wells located nearest to the 2018 maximum inland extent of saltwater intrusion (Prinos, 2019) were included in this analysis. Most of FPL’s other monitoring wells are located closer to the coast near the Turkey Point Nuclear Power Plant cooling canal system, which is hypersaline. Well construction depths in the FPL monitoring reports were measured from top of casing (TOC), with altitude referenced to the North American Vertical Datum of 1988 (NAVD 88). Altitudes for the top and bottom of the screen and bottom of the well are also referenced to NAVD 88. These NAVD 88 values were converted to NGVD 29 values using the software program Corpscon (U.S. Army Corps of Engineers, 2024). The maximum chloride concentration measured in water samples collected from each of the five FPL wells during 2022 was selected to help determine the location of the 1,000-mg/L isochlor (table 5 in Zhang and Renshaw, 2024).

The 2022 saltwater intrusion monitoring data were requested from the FKAA. The FKAA monitors 16 wells for saltwater intrusion and collects a single sample for specific conductance measurement and chloride analysis from 3 of these wells on a monthly basis (G-1603, G-3166, and G-3342) (table 4 in Zhang and Renshaw, 2024). The maximum 2022 chloride concentration in each of these 3 wells was used for this mapping effort. The other 13 FKAA wells have long open intervals. The FKAA collects monthly depth profiles of specific conductance in these wells by lowering a water-quality probe through the open portion of the borehole and taking a specific conductance measurement at 1.5-m intervals. The FKAA also collects monthly samples for depth profiles of chloride concentration from these long-open-interval wells by pumping from sample tubes at 1.5-m depth intervals. Because of concerns about possible mixing caused by pumping-induced vertical flow during sampling, data from these chloride profiles were not used for this mapping effort. Instead, for these 13 sites, chloride concentrations were estimated from the maximum 2022 specific conductance reading at the profile location closest to base of the Biscayne aquifer in each well (table 4 in Zhang and Renshaw, 2024). The linear regression model used to estimate chloride concentrations from specific conductance is described below. The depths for the specific conductance profiles collected by the FKAA were measured from the TOC of the monitoring well. The altitudes of the specific conductance profile sample locations were determined using the altitude of the TOC, referenced to NGVD 29.

Saltwater intrusion monitoring data for 2022 collected by MacVicar Consulting Inc. were obtained from their quarterly monitoring reports (MacVicar Consulting, Inc., 2022a, c), which include depth profiles of specific conductance in nine long-open-interval monitoring wells. Although they measured the depths for the profiles from the water surface, their reports do not reference the water level to a national vertical datum. Therefore, water-level data from a nearby USGS site, G-3356 (sheet 1 and fig. 1D), referenced to NGVD 29, were used to approximate the altitudes of the water-level at the MacVicar wells. To determine which profile location was closest to the base of the Biscayne aquifer in each of the MacVicar wells, the profile depths and depth to the base of the Biscayne aquifer were both converted to depth below TOC. Equation 1 was used to calculate the depth of the base of the Biscayne aquifer below the TOC:

The maximum 2022 specific conductance measurements from the sample depth closest to the base of the Biscayne aquifer were used to estimate the maximum chloride concentration in each MacVicar well (table 3 in Zhang and Renshaw, 2024).

Estimating Chloride Concentration From Specific Conductance

To estimate chloride concentrations from the specific conductance measurements collected by FKAA and MacVicar Consulting, Inc., a linear regression model was developed from chloride and specific conductance data from USGS monitoring wells. Several previous studies have evaluated the relation between chloride and specific conductance in the Biscayne aquifer (Fitterman and Prinos, 2011; Fitterman, 2014; Prinos, 2017; MacVicar Consulting, Inc., 2022a, b, c). For the previous USGS update of the inland extent of saltwater intrusion in the Biscayne aquifer, a linear regression between chloride and specific conductance was created by using data from 249 water samples with chloride concentrations less than 2,000 mg/L. These samples were collected by the USGS from seven monitoring wells in the southeastern part of Miami-Dade County between October 26, 1999, and July 23, 2018 (Prinos, 2019). This previous regression, shown in equation 2, resulted in a coefficient of determination (R²) of 0.9667 and yielded the following relation between chloride and specific conductance (Prinos, 2019):

Figure 2A shows the same data used for the Prinos (2019) regression, with the addition of data from more recent samples that were collected between July 24, 2018, and July 12, 2022, and a line fit through the data with linear regression. Figure 2B provides a closer view of these data and the linear fit at the low end of the plot (specific conductance less than 1,000 µS/cm and chloride concentration greater than 200 mg/L). Samples with specific conductance values below 728 µS/cm or chloride concentrations below 100 mg/L deviate from the regression line. Therefore, instead of only limiting chloride concentrations to below 2,000 mg/L, as in Prinos (2019), this report narrows the chloride concentrations in samples used for the regression analysis to between 100 and 2,000 mg/L. The resulting dataset for the regression analysis included chloride and specific conductance data from 58 water samples collected from four USGS monitoring wells (fig. 3; table 2 in Zhang and Renshaw, 2024) between July 2010 and July 2020. Equation 3 and figure 3 show the resulting relationship between chloride concentration and specific conductance:

The maximum 2022 specific conductance measurements collected from the profile locations closest to the base of the Biscayne aquifer in the FKAA and MacVicar wells were used to estimate chloride concentrations. To avoid extrapolating the regression equation beyond the limits of the data, estimated chloride values less than 100 mg/L or greater than 2,000 mg/L were designated as <100 and >2,000, respectively. The estimated chloride concentrations are shown in tables 3 and 4 in Zhang and Renshaw (2024).

Potential Permeable Layers Below the Mapped Base of the Biscayne Aquifer

Fish and Stewart (1991) constructed a structure contour map of the base of the Biscayne aquifer on the basis of data (aquifer tests and vertical hydraulic conductivity profiles) from approximately 40 test wells throughout southeastern Florida. For all non-USGS wells used to map the saltwater front for this report, the altitude of the base of the Biscayne aquifer was assumed to be the same as that estimated by Fish and Stewart (1991). In some locations, however, borehole geophysical data collected in relatively new USGS monitoring wells indicate that saltwater is readily intruding below the base of the Biscayne aquifer as estimated by Fish and Stewart (1991), particularly in northeastern Miami-Dade County.

It is important to correctly identify the base of the Biscayne aquifer, because a fresh chloride measurement taken from a shallower depth could lead to the misinterpretation of the position of the saltwater interface. In some areas, the base of the Biscayne aquifer is difficult to identify because of interfingering sands and calcareous sediments, and data were only available from four test holes in northeastern Miami-Dade County when Fish and Stewart (1991) created their map. The fact that saltwater is readily intruding below the base of the Biscayne aquifer as estimated by Fish and Stewart (1991) indicates the presence of permeable beds in the upper Tamiami Formation that were not apparent in the data available to the authors of that report. It is beyond the scope of this report to re-evaluate the location of the base of the Biscayne aquifer. However, at wells where borehole geophysical data indicate saltwater intrusion is present below the mapped basal surface of the aquifer, for the purposes of this report, it is assumed that there is a permeable zone in the Tamiami Formation in these locations that is hydraulically connected to the Biscayne aquifer. More discussion on this can be found in the “Changes in the Location of the Saltwater Interface Between 2018 and 2022” section of this report.

Map and Isochlor Position

All of the measured and estimated chloride concentration data, well location, and well construction information were imported into ArcGIS Pro (Esri, Redlands, California) to create the map and isochlor (sheet 1). The location of the 1,000-mg/L isochlor was estimated and mapped by using the maximum measured or estimated chloride concentrations from each well for 2022. In some locations, bulk conductivity data from TSEMILs were used to help interpret the inland extent of saltwater intrusion.

Prinos and Valderrama (2016) determined that a bulk conductivity of about 70 millisiemens per meter (mS/m) represents the threshold between brackish and saline water in the Biscayne aquifer. Therefore, for this report, saltwater was assumed to be present where bulk conductivity from electromagnetic induction logs was greater than around 70 mS/m. The resulting isochlor represents the approximate maximum inland extent of saltwater intrusion at the base of the Biscayne aquifer.

Northern Miami-Dade County

G-3949 Well Cluster

In northernmost Miami-Dade County, the 1,000-mg/L isochlor was interpreted as having advanced inland 0.1 kilometer (km) near the G-3949 well cluster since 2018 (sheet 1 and fig. 1B) but had not yet crossed the cluster. The deep well in this cluster, G-3949D, is screened from −62 to −64 m NGVD 29 (fig. 4), and between 2018 and 2022, the maximum chloride concentration of water samples from this well increased from 247 to 871 mg/L (Prinos, 2019; Zhang and Renshaw, 2024).

Fish and Stewart (1991) estimated the base of the Biscayne aquifer in this area to be around −45 m NGVD 29, and core descriptions from G-3949D indicate that the top of the Tamiami Formation is approximately −31 m NGVD 29. However, the core descriptions also indicate that the rock is generally highly or moderately porous from −1 to −80 m NGVD 29 (Prinos and others, 2014), and bulk conductivity from TSEMIL logs has been increasing in this well in a zone that is between about −55 and −80 m NGVD 29 since 2011 (fig. 4 and table 1). Therefore, for the purposes of interpreting the location of the salt front, it is assumed that, at this location, a permeable zone is present below the base of the Biscayne aquifer, as estimated by Fish and Stewart (1991). The maximum bulk conductivity in this zone in 2022 was 88 mS/m at −62 m NGVD 29. Therefore, both the chloride concentrations and the increase in bulk conductivity indicate that the interface was very close to this well cluster in 2022 and is readily intruding this area.

Table 1.    

Electromagnetic induction log plots and data for U.S. Geological Survey (USGS) monitoring wells.

[Hyperlinks open to USGS GeoLog locator (U.S. Geological Survey, 2023a). (Wells G-5007D and G-5006D have well names G-5007 and G-5006 in data files, respectively.) ID, identifier; TSEMIL, time-series electromagnetic induction log]

Table 1.    Electromagnetic induction log plots and data for U.S. Geological Survey (USGS) monitoring wells.

USGS site number	USGS site name	Clickable hyperlink to access TSEMIL plot or data
255733080195601	G-3949D	Plot  Data
255639080100401	G-5007D	Plot  Data
255453080110801	G-3978	Plot  Data
255415080114401	G-5006D	Plot  Data
254707080155801	G-5005	Plot  Data
254601080150301	G-3977D	Plot  Data
252714080260901	G-3976	Plot  Data
252506080261001	G-3999	Plot  Data
252036080293501	G-5004	Plot  Data

G-3978

At well G-3978 (fig. 1B), which is about 1.5 km north of the Biscayne Canal, the 1,000-mg/L isochlor for 2022 was mapped 0.15 km farther inland than it was in 2018. Although the maximum chloride concentration of water samples collected from well G-3978 increased from only 69 mg/L in 2018 (Prinos and others, 2014) to a maximum of 81 mg/L in 2022 (table 1 in Zhang and Renshaw, 2024), the bottom of the open interval of this well is 9.4 m above the base of the Biscayne aquifer estimated from Fish and Stewart (1991), which is −49 m NGVD 29. The TSEMIL dataset from G-3978 indicated that saltwater has readily intruded the rocks and sediments at this location from 2018 to 2022 at altitudes between about −43 and −80 m NGVD 29 (fig. 5 and table 1), which indicates that there is a permeable zone beneath the basal surface of the Biscayne aquifer estimated by Fish and Stewart (1991). Specifically, the maximum bulk conductivity in the 2022 TSEMIL log was 192 mS/m at −77 m NGVD 29. Therefore, the 1,000-mg/L isochlor was mapped inland of this well, and the interface is shown as having advanced in this area since 2018.

Near Miami International Airport and the City of Coral Gables

G-5005

Near the Comfort Canal Number C-5, close to where it feeds into the Tamiami Canal Number C-4, the 1,000-mg/L isochlor for 2022 was about 0.4 km inland of where it was mapped in 2018 (fig. 1C, sheet 1). This interpretation is based on the data from a new monitoring well, G-5005, installed in October 2019 and first sampled in April 2022 (Zhang and Renshaw, 2024). The maximum chloride concentration of water samples collected in 2022 was 54 mg/L. However, the bottom of the open interval for this well is 7.4 m above the altitude of the base of the Biscayne aquifer estimated from Fish and Stewart (1991), which is −35.7 m NGVD 29. The TSEMIL dataset collected from well G-5005 (U.S. Geological Survey, 2023a) indicated that saltwater is intruding from −31 to −54 m NGVD 29, which indicates that there is a permeable zone beneath the basal surface of the Biscayne aquifer estimated by Fish and Stewart (1991). In 2022, bulk conductivity reached 200 mS/m between −43 and −47 m NGVD 29 (fig. 6). Therefore, the 1,000-mg/L isochlor was mapped inland of this location.

Model Land Area

Southern Miami-Dade County

Improvements in Spatial Coverage

The position of the freshwater-saltwater interface is transient and monitoring wells may lose their relevance as the salt front migrates past them over time. Thus, the addition of new wells inland of the 1,000-mg/L isochlor could improve accuracy in constraining the maximum inland extent of saltwater intrusion in the Biscayne aquifer. Miami-Dade County has been cooperating with the USGS to install new wells to expand the saltwater intrusion monitoring network in the Biscayne aquifer for decades.

While interpreting the data for this report, locations where data gaps remain and where the saltwater front has moved past the existing monitoring well network were noted. On the basis of this information, 14 potential new well locations were identified for Miami-Dade County to consider when expanding its saltwater intrusion monitoring network. New wells at these locations would further fill in the existing gaps in monitoring and constrain the maximum inland extent of saltwater intrusion for future mapping efforts.

Variations in Well Construction

In addition to limited spatial coverage of monitoring wells in some areas, variations in well construction also contribute to uncertainty in interpreting the data. There are a variety of factors to be considered when determining how to construct monitoring wells, including cost and specific objectives of the monitoring program. In this study, monitoring wells constructed to meet different program objectives were used. As a result, these wells were not all constructed in the same way. For instance, some wells have long open intervals, whereas others are screened over shorter intervals. This variation in well construction can affect data quality. For example, pumping to collect samples for saltwater intrusion monitoring in salt contaminated aquifers in long-open-interval wells could lead to vertical mixing (Kohout and Hoy, 1963; Shapiro, 2002; Shalev and others, 2009; Prinos, 2013; Prinos and Valderrama, 2015; Prinos, 2019). There is also a potential for vertical mixing in the borehole in long-open-interval wells because of natural vertical gradients in the aquifer (Church and Granato, 1996; Oki and Presley, 2008). Conductivity profiles from both short-screened and long-screened monitoring wells are used in this study, which could introduce some uncertainty in the data. To reduce uncertainty, short-open-interval wells are recommended for saltwater intrusion monitoring (Church and Granato, 1996). The USGS, in cooperation with Miami-Dade County, has adopted a strategy of installing either individual short-screened wells or nests of short-screened monitoring wells for saltwater intrusion monitoring. Although not ideal, available data from long-open-interval wells remain valuable for filling spatial gaps and mapping the location of the saltwater front.

In addition, some wells currently in the Miami-Dade County saltwater intrusion monitoring network are not deep enough to detect saltwater at the base of the Biscayne aquifer. Examples of such wells are G-3948, G-3949, G-3976, G-3978, G-3999, G-5005, G-5006, and G-5007. In these areas, interpretation of the saltwater interface position was based on chloride values from samples taken well above the base of the Biscayne aquifer, or through electromagnetic induction logs showing clear indications of saltwater presence or absence below a well’s screened interval in the aquifer. Most of the newer monitoring wells installed by Miami-Dade County have been designed to fully penetrate the Biscayne aquifer and extend well below it so that the TSEMIL datasets can indicate any saltwater intruding deeper than the base of the Biscayne aquifer. Using this type of well construction when installing additional monitoring wells could provide more accurate saltwater intrusion monitoring data.

Effects of Canal Leakage

Even though salinity control structures have been installed by the South Florida Water Management District to control drainage, flooding, and saltwater intrusion, saltwater still leaks around the existing salinity control structures and through the porous rock of the Biscayne aquifer (Parker and others, 1955; Kohout and Leach, 1964; Prinos and others, 2014). This saltwater leakage through canals is observed in the occurrence of unusual fluctuations in chloride concentrations in wells near the canals. As described by Prinos and others (2014), chloride concentrations in some wells near canals were observed to rapidly increase for a short period before gradually dissipating over time. This detection of increased chloride concentrations in wells near canals presents challenges in discerning whether such increases originate from the canals or the ocean. As a result, there is a risk of misinterpreting the position of the saltwater interface. This unusual fluctuation in chloride concentrations is observed from wells near canals including the Florida City Canal, a feature known locally and referred to herein as “Card Sound Road canal,” Cutler Drain Canal C-100, Miami Canal, Mowry Canal Number C-103, Princeton Canal Number C-102, Snapper Creek Canal Number C-2, Black Creek Canal C-1, Biscayne Canal, and Comfort Canal Number C-5 (Prinos and others, 2014). Therefore, increasing the frequency in continuous salinity and water-level monitoring near canals could aid in determining the origin of the increased chloride concentration at those locations. Further discussion on the effects of canal leakage on aquifer salinity can be found in Prinos and others (2014).