Abstract Introduction Inventory of Data Case Studies Summary References Tables PDF Version

Well data were inventoried and compiled for all wells at existing and historical ASR sites in southern Florida, and cycle test data (also available for many sites) were synthesized. Consulting reports on the construction and testing of wells and on cycle testing provided much of these data. The consulting reports used to compile these data are listed in the selected references section at the back of this report.

Historical and current ASR sites are listed in table 1 along with the utility or operator of the site, the aquifer being used for the storage zone, site status, type of source water used for injection, and number of wells drilled at each site. The locations of these sites are shown in figure 1. The number of injection (storage) wells at each site ranges from one to five, and most sites have at least one monitoring well in the storage zone.

The type of source water used for injection in southern Florida has included treated drinking water, raw ground or surface water, and reclaimed water (table 1). Treated drinking water is the most common source water type, but raw ground water also is used, or has been proposed for use, at a number of sites on the east coast. The source water planned for the CERP ASR program is raw or partially treated ground water or surface water (table 1, Western Hillsboro Canal, site 1). Special permits, obtained through the FDEP Underground Injection Control program and the U.S. Environmental Protection Agency, are required to inject raw surface or ground water because these waters sometimes exceed maximum contaminant levels for primary or secondary drinking water standards for some constituents.

Construction and testing data were compiled into three main categories. These categories are well identification, location, and construction data; hydraulic well-test data; and ambient formation water-quality data.

For the purpose of this study, all ASR storage and associated monitoring wells were assigned a USGS number, and data from these wells have been stored as part of the USGS Ground-Water Site Inventory (GWSI) database. Well identification, location, and construction data are given in table 2 (at end of report). The construction information includes total hole depth, ending date of construction, casing depth and diameter, type of each casing string set in the well, and the completed (constructed) open interval and its diameter. In most cases, the completed interval is open hole, but a gravel-packed screen was installed in a few wells. At many sites, the first well drilled was plugged back to the selected storage zone after being drilled deeper to test other potential zones or to determine water-quality changes with depth. In many instances, the latitude and longitude provided herein were obtained from the construction permit, and this location is representative of the storage well only; however, in some instances, the latitude and longitude were more precisely determined for all wells at a site by the use of a hand-held global positioning system (GPS) (see footnote 1 in table 2 at end of report).

Figure 8. Thickness of open interval in storage wells at aquifer storage and recovery sites in southern Florida. All wells represent first storage well at each site (see table 2). [larger image]

The thickness of the open interval ranges from 45 ft at the Marco Lakes (well C-1206) and Marathon (well MO-189) sites to 452 ft at the West Well Field site (well G-3706) (fig. 8; table 2 at end of report). Open intervals for ASR wells in the Floridan aquifer system average 172 ft thick. The diameter of the open interval ranges from 5.125 in. at the St. Lucie County site to 29 in. at the West Well Field site in Miami-Dade County. Large diameter open intervals are constructed for the purpose of obtaining a high rate of flow. Each of the storage wells at the West Well Field site is designed for a pumping rate of up to 5 Mgal/d.

Reported data describing hydraulic tests were compiled for ASR well systems. The data include the reported results of packer tests conducted during drilling, step drawdown tests, single-well constant rate recovery tests, and multiwell constant rate tests (table 3 at end of report). Tests of other permeable intervals at a site that are shallower or deeper than the interval selected to be the storage zone are also included (table 3 at end of report). Water-level data were not analyzed as part of this study; rather, all of the analytical results given in table 3 (at end of report) came from consulting reports in the selected references listed at the back of this report.

Some tests reported in table 3 (at end of report) are single-well step drawdown tests run to determine the specific capacity of a well. These tests provide insight into the productive capacity of a well and are used to determine the size and depth of a pump to be used in the well for a multiwell test or for long-term operation. At some sites, the transmissivity of the tested interval was estimated from a step drawdown test using the specific capacity at each step. At the Marco Lakes ASR site, transmissivity was determined during a step drawdown test of ASR-3 by analyzing the resulting drawdown data from nearby wells using the Cooper and Jacob (1946) solution. Transmissivity was estimated at the Boynton Beach East WTP site from a step drawdown test of ASR-1 using the Cooper and Jacob (1946) solution, but without any monitoring wells. Specific capacity determined from step drawdown tests of storage zones range from 2.7 gal/min/ft at the Marathon site to 390 gal/min/ft at the West Palm Beach site, well ASR-1 (table 3 at end of report). Specific capacity was reported to be 1,600 gal/min/ft on the basis of a multiwell test at the Taylor Creek/Nubbin Slough (Lake Okeechobee) site.

Packer tests are tests of open-hole intervals conducted during drilling using inflatable packers set on a string of drill pipe for the purpose of isolating the interval to be tested. Often, only specific capacity data are reported for packer tests (table 3 at end of report). However, transmissivity can be estimated either from the specific capacity results, or from analysis of the recovery of water level after a period of constant rate pumping during a packer test. This latter method, known as the Theis (1935) residual drawdown or recovery analysis, gives a more reliable estimate than the specific capacity method. Packer test results can be unreliable because of partial penetration, a low pumping rate, a short pumping period, or incomplete isolation of the interval tested (leaky packers).

Hydraulic properties determined from a multiwell, constant rate, drawdown test include transmissivity, storage coefficient, and leakance. Solutions commonly used to analyze water-level data from this type of test include Theis (1935) and Cooper and Jacob (1946) for confined aquifers and Hantush and Jacob (1955) and Walton (1962) for semiconfined, leaky aquifers. Depending on the amount of drawdown (pumping rate) and the degree of background variations in water level, such as tidal fluctuations, background water-level measurements should be made for at least 1 day prior to the beginning of the pumping test, and these measurements should be subtracted from the drawdown water-level data collected during the test. Single-well constant rate tests usually provide only an estimate of transmissivity, and solutions used to analyze the recovery water-level data from these tests include the Theis (1935) solution for residual drawdown and the Cooper and Jacob (1946) solution.

Multiwell constant rate tests of the storage zone were performed at 16 of the ASR sites (table 3 at end of report), and not including packer tests, single-well constant rate recovery tests of the storage zone were run at 4 sites, 2 of which also had a multiwell-test run. Constant rate test results could be affected by pretest well treatment designed to increase specific capacity. Acidization of the ASR well prior to the multiwell test was done at the Springtree WTP and West Well Field sites. The Western Hillsboro site planned recharge well (EXW-1) also was acidized after the reported step drawdown test (table 3 at end of report).

Hydraulic properties determined from tests of storage zones may apply only to the storage zone or to a thicker interval if the aquifer containing the storage zone is thicker than the storage zone. In the case where the aquifer is thicker than the storage zone, the hydraulic conductivity of a storage zone will be less than that obtained by dividing the transmissivity determined from a test by the thickness of the storage zone. However, in the Upper Floridan aquifer where thick zones of relatively low permeability separate flow zones, tests of part of the aquifer are typically not influenced by the entire thickness of the aquifer. Thus, the value of transmissivity obtained is less than the total transmissivity of the aquifer (Wedderburn and Knapp, 1983).

Figure 9. Transmissivity determined for storage zones in the Floridan aquifer system at aquifer storage and recovery sites in southern Florida. The production well used for the test at all sites was ASR-1, except for sites 15 and 27 where MW-1 was used. [larger image]

For 18 sites where the storage zone is in the Floridan aquifer system, the most reliable or representative values for transmissivity from storage zone tests were selected and then plotted on a map of southern Florida (fig. 9). In most cases, these values came from drawdown analysis of constant rate multiwell tests; if performed, the leaky aquifer solution was used. The storage zone is in the Upper Floridan aquifer in all cases, except at the Taylor Creek/Nubbin Slough (Lake Okeechobee) and the West Well Field sites. At the Lake Okeechobee site, this zone is in the Lower Floridan aquifer (Quiñones-Aponte and others, 1996), and at the West Well Field site, some of the mid-Hawthorn aquifer in addition to the upper part of the Upper Floridan aquifer is included in the storage zone. Transmissivity values range from 800 ft²/d at the Lee County site to nearly 590,000 ft²/d at the Lake Okeechobee site. The highest value in the Upper Floridan aquifer is 108,000 ft²/d at the West Palm Beach WTP site. The average value for sites in the Upper Floridan aquifer is 21,100 ft²/d, and values greater than 30,000 ft²/d are considered to be high in this study.

The high transmissivity estimate at the Lake Okeechobee site (fig. 9) is a function of the large thickness of the open interval and the dominant lithology in this interval, which is dolomite. The storage zone contains several highly permeable flow zones that may have secondary fracture permeability. The open interval in the ASR well for the Lee County WTP site is confined to the lower Hawthorn producing zone of the basal Hawthorn unit (Reese, 2000). A second ASR site was later constructed at the same location (Olga WTP site). The Olga WTP site storage zone is deeper in the Upper Floridan aquifer and is contained within the Suwannee Limestone, about 150 ft below the top of this formation (Water Resources Solutions, Inc., 2000a). The estimated transmissivity for the Olga storage zone is 9,400 ft²/d (fig. 9; table 3 at end of report).

Leakance of the tested aquifer was determined at eight sites in the Floridan aquifer system by multiwell aquifer tests, and values are higher than expected (table 3 at end of report). Leakance is a measure of the degree of aquifer confinement and is defined as the vertical hydraulic conductivity of a confining unit, divided by the thickness of the confining unit. However, leakance determined from an aquifer test applies to both the upper and lower confining units of the aquifer, unless it is known that one of the confining units is nonleaky. Leakance estimates ranged from 3.9 x 10^-5 1/d at the West Well Field site to 6.3 x 10^-2 1/d at the Deerfield Beach West WTP site. Leakance estimates less than 1 x 10^-3 1/d have been used to indicate confining conditions in the surficial aquifer system in southern Florida (Reese and Cunningham, 2000). Of the eight values determined for leakance (table 3 at end of report), five exceed this limiting value. Leakance was greater than 4.0 x 10^-2 1/d at the Deerfield Beach West WTP, Olga WTP, and the St. Lucie County sites. Leakance may also be high at the West Palm Beach WTP site. The confined aquifer Theis (1935) solution was used to analyze the multiwell-test data collected at this site, despite a large observed departure below the type curve during the latter part of the test indicating a leaky aquifer.

The high leakance estimates from the Upper Floridan aquifer are probably best attributed to leakage from below the tested interval rather than from above because of the good confinement generally accepted as being present above the aquifer in southern Florida (Bush and Johnston, 1988). This leakage either originated from intervals lower in the Upper Floridan aquifer or from the middle confining unit of the Floridan aquifer system.

Ambient water-quality data were collected from storage and monitoring wells at ASR sites (table 4). The inventoried data describe formation water salinity and include the sampled interval, sample date, specific conductance, dissolved chloride concentration, dissolved solids concentration, temperature, and dissolved sulfate concentration. The sampling methods, listed in order of increasing reliability, include (1) collected during drilling by the reverse-air rotary method, (2) collected from packer tests, (3) collected from a pump out test of an open interval below casing before final construction of the well, and (4) collected from a completed open interval. Intervals sampled include the storage zone, intervals deeper and shallower than the storage zone, and intervals that include more than the selected storage zone (table 4). Upper Floridan aquifer ASR sites in southwestern Florida were usually sampled from shallower permeable zones of the intermediate aquifer system.

The chloride concentration of ambient water in ASR storage zones in the Floridan aquifer system is shown on a map of southern Florida (fig. 10). Samples used for this map were selected from table 4 based on the most reliable sampling method as described above. Chloride concentrations ranged from 500 mg/L at the Lee County WTP site to 11,000 mg/L at the Englewood South Regional WWTP site. At most sites, the chloride concentration ranged from about 1,000 to 3,000 mg/L, and the average concentration was about 2,300 mg/L. Storage zones containing water with 3,000 mg/L or greater were considered to have high chloride concentration in this study. The highest value found in the east coast area was 3,600 mg/L at the Springtree WTP site. The highest chloride concentration found in the upper part of the Upper Floridan aquifer in southern Florida based on three previous studies was 8,000 mg/L in northeastern Palm Beach County; the lowest concentration found was 400 mg/L in Lee County (Reese, 1994; Reese, 2000; and Reese and Memberg, 2000).

Cycle test information was obtained from consulting reports, other published reports, monthly operating reports (MOR) required by the FDEP as part of the permitting process during operational testing, and in several cases, from daily records provided by a WTP. These data were compiled and are given in table 5. All of the test data given are only for the first storage well (ASR-1) at a site, except for the West Well Field site. Only 18 of the 27 ASR sites listed in table 1 are included in table 5; other ASR sites have not initiated operational testing or test data were not available. Cycle testing at the Olga WTP and North Reservoir sites was postponed due to inadequate treated drinking water supplies that will be used for recharge. The number of days of storage in table 5 includes only the time between the recharge and recovery periods; it does not include days during the recharge period in which injection ceased due to a lack of source water or other operational problems. The MOR provided insufficient data to calculate recovery efficiencies at some ASR sites because the water quality of recharged and recovered water was not reported; these data are not required by the FDEP in the report.

Two recovery efficiency numbers were determined for each cycle (table 5). The first is total recovery efficiency, and it is the percent recovery at the end of the cycle. The chloride concentration of the recovered water at this point is also given in table 5. The chloride concentration at the end of the cycle is usually in the range of 250 to 400 mg/L. The second recovery efficiency number is the potable water recovery efficiency. It is the percent recovery when the chloride concentration of the recovered water reaches only 250 mg/L. Potable water recovery efficiency numbers (potable recovery efficiencies) are used in this report for performance comparisons between sites.

Chloride concentrations of recharged and recovered water for the West Palm Beach WTP site were not reported or made available, and only the total recovery efficiencies are given in table 5. At the West Well Field site, two storage wells were active during the second cycle and all three storage wells were active during the third cycle. However, water was not recovered from well ASR-3 during the cycle 3 recovery period. For cycle 3, recovery efficiencies were determined for individual storage wells and also for all three wells combined (table 5).

The Boynton Beach East WTP site underwent 16 recharge-recovery cycles (table 5). The Marathon site had 11 cycles, and the Marco Lakes and Springtree WTP sites had 5 cycles each; the number of cycles was 4 or less at all other sites. Additional cycles were conducted at the Manatee Road site, but were not reported. Recharge volume per cycle ranged from as low as 0.6 Mgal for cycle 1 at the Lee County WTP site, to as high as 714.33 Mgal during cycle 3 at the West Well Field site. The longest storage period was 181 days for cycle 3 at the Hialeah site.

The highest reported first cycle potable recovery efficiency was 47 percent for the Boynton Beach ASR site. The first cycle recovery efficiency of the Corkscrew WTP site is greater but is not considered here due to the potable nature of water in its storage zone. Except for the Jupiter site where no potable water was reported to be recovered on the first cycle, the lowest potable recovery efficiency was 2 percent at the San Carlos Estates site. Of the 16 sites in table 5 with potable recovery efficiencies calculated, 9 sites had a potable recovery efficiency of well over 10 percent during the first cycle. The seven exceptions include Fiveash WTP, Manatee Road, North Reservoir, San Carlos Estates, Lake Okeechobee, Jupiter, and St. Lucie County sites. Two of these, the Manatee Road and Jupiter sites, showed improvement to a level substantially higher than 10 percent in succeeding cycles. The Fiveash, San Carlos Estates, and Lake Okeechobee sites did not; however, few cycles were conducted at these three sites (two, two and four, respectively). Only one cycle was run at the North Reservoir and St. Lucie County sites.

Ten sites achieved a potable recovery efficiency exceeding 30 percent during at least one cycle; however, at the Shell Creek WTP site, the recovery efficiency diminished to 9 percent during the third cycle when the recharge volume was greatly increased. The highest potable recovery efficiency of 90 percent was during cycle 4 at the Boynton Beach East WTP site, but the recharge volume reported for this cycle could be too low. This recharge volume is based on flow totalizer equipment readings, but calculation of the recharge volume based on reported daily flow rates gives a higher number. The second highest recovery efficiency was 84 percent for cycle 16 at the Boynton Beach site. Recovery efficiency was 72 percent for cycle 4 at the Marathon site; however, the storage zone at this site is within a siliciclastic sandstone aquifer. Because of lower dispersive mixing, recovery from a siliciclastic aquifer may be larger, having only intergranular porosity as compared to carbonate rock storage zones that probably also have secondary, conduit type porosity (Merritt, 1985).