EPA/520-6-77-010
EFFECTS OF PHOSPHATE MINERALIZATION
AND THE PHOSPHATE INDUSTRY ON
RADIUM-226 IN GROUND WATER
OF CENTRAL FLORIDA
^
322
LU
CD
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RADIATION PROGRAMS
LAS VEGAS FACILITY
LAS VEGAS, NEVADA 89114
OCTOBER 1977
-------
EPA/52Q-6-77-Q10
EFFECTS OF PHOSPHATE MINERALIZATION AND
THE PHOSPHATE INDUSTRY ON
RADIUM-226 IN GROUND WATER OF CENTRAL FLORIDA
Robert F. Kaufmann
James D. Bliss
October 1977
OFFICE OF RADIATION PROGRAMS - LAS VEGAS FACILITY
U. S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
DISCLAIMER
This report has been reviewed by the Office of Radiation Programs--
Las Vegas Facility, U.S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does not constitute
endorsement or recommendation for their use.
-------
PREFACE
The Office of Radiation Programs of the U.S. Environmental Protection
Agency carries out a national program designed to evaluate population exposure
to ionizing and non-ionizing radiation and to promote development of controls
necessary to protect the public health and safety. In-depth field studies of
various radiation sources (e.g. nuclear facilities, uranium mill tailings, and
phosphate mills) provide technical data for environmental impact statement
reviews as well as needed information on source characteristics, environmental
transport, critical pathways for population exposure, and dose model validation
Where possible in terms of programmatic priorities and available staff,
the Office of Radiation Programs laboratories also provide technical assis-
tance to EPA regional offices. In this technical assistance role, staff of
the Las Vegas Facility were responsible for assessing the impacts of the
central Florida phosphate industry on radiochemical quality of ground water.
Available geologic, hydrologic, and water quality data were assembled and
interpreted to determine what adverse impacts, if any, are attributable to the
industry and to compare radiochemical quality of ground water in the study
area to State and national conditions. Finally, recommendations were developed
to mitigate adverse or objectionable situations in terms of preserving environ-
mental quality and public health.
The reader should be aware that two rather diverse viewpoints surround
this study and report. Some have considered the data base too limited to
conclude that no widespread or significant contamination has occurred or is
occurring. Therefore, extensive additional studies are essential. Others,
particularly industry, consider the data and interpretation herein as con-
firmation that contamination has not occurred, that present practices and data
collection requirements are adequate, and that more study and monitoring would
be inefficient. The authors explicitly recognize this schism in that while
n i
-------
the available data are analyzed and conclusions drawn, criticism as to adequacy
and improvements are also clearly stated.
Readers of this report are encouraged to inform the Office of Radiation
Programs of any omissions or errors. Comments or requests for further infor-
mation are also invited.
W. D. Rowe, Ph.D.
Deputy Assistant Administrator
Office of Radiation Programs
IV
-------
CONTENTS
Page
PREFACE iii
LIST OF FIGURES vii
LIST OF TABLES viii
ACKNOWLEDGMENTS ix
ABSTRACT 1
SUMMARY AND CONCLUSIONS 3
RECOMMENDATIONS 7
Improved Waste Disposal 7
Monitoring ?
Water Sampling and Analysis 8
Hydrology and Geology 9
Data Interpretation and Reportinp 10
PROBLEM DESCRIPTION 11
PREVIOUS AND ONGOING INVESTIGATIONS 13
HYDROGEOLOGIC SETTING OF WEST CENTRAL FLORIDA 15
Geology of Central Florida 15
Aquifer Systems and Ground-Water Flow 20
Influence of Mining and Processing 23
SOURCE TERM CHARACTERIZATION 29
SOURCES OF RADIOCHEMICAL DATA 34
TECHNIQUES FOR MONITORING RADIUM IN GROUND WATER 39
Monitoring Objectives 39
Sampling Points and Methods for Sampling Radium-226 in Ground Water 39
Sample Preservation and Handling 40
Significance Relative to Future Studies 42
RADIUM IN SURFACE AND GROUND WATER 44
Concentrations in Continental and Oceanic Waters 44
Florida Ground Water 46
-------
CONTENTS (Continued)
Pa^e
WATER QUALITY EFFECTS OF PHOSPHATE MINERALIZATION AND THE PHOSPHATE 50
INDUSTRY
Statistical Methodology 50
Spatial Variations in Water Quality 52
Water Table Aquifer 52
Upper Floridan Aquifer 55
Lower Floridan Aquifer 56
SARASOTA COUNTY 62
TEMPORAL CHANGES IN WATER QUALITY 71
LOCAL CONTAMINATION 73
ADEQUACY OF INDUSTRY RESPONSE TO THE DRI PROCESS 82
REFERENCES 84
APPENDICES
1. Dissolved radium-226 (pCi/1) in ground water in the central
Florida phosphate district 90
2. Analytical results from the 1966 FWPCA survey of radium-226
in central Florida ground water 98
3. Well numbering system 113
4. Dissolved radium-226 concentration (pCi/1) in ground water in
Sarasota County 114
VI
-------
LIST OF FIGURES
Number Page
1 General location map 12
2 Generalized geologic cross section through southern Polk and 17
northern Hardee Counties
3 Generalized geologic cross section through northeastern Manatee 18
County
4 Generalized southwest-northeast hydrogeologic cross section through 21
Polk and Manatee Counties
5 Interaction of mining operations and the hydrogeologic system 24
6 Location of wells sampled in Polk, Hillsborough, and Hardee Counties 37
7 Location of wells sampled in Manatee County 38
8 Location of counties used to establish background levels of 47
radium-226 in Florida ground water
9 Log-probability plot of background levels of radium-226 in 48
Florida ground water
10 Log-probability plot of USGS data for the water table aquifer 54
in unmined and mined mineralized areas
11 Log-probability plot of USEPA data for the Lower Floridan aquifer 57
in mined and unmined mineralized areas and in nonmineralized areas
12 Component populations of radium-226 in the Lower Floridan aquifer 59
of central Florida
13 Location of wells sampled for radium-226 in Sarasota County 63
14 Log-probability plot of radium-226 in the water table and Floridan 64
aquifers, Sarasota County
15 Location of radium-226 observations in the water table aquifer in 66
Sarasota County
16 Contour map of radium-226 in the Floridan aquifer in Sarasota County 67
17 Plot of dissolved solids versus radium-226 in Sarasota County 70
ground water
18 Gross alpha radioactivity in ground water in the vicinity of 76
C.F. Industries, Inc. gypsum ponds
-------
LIST OF TABLES
Number Page
1 Geologic and hydrogeo"logic units in central Florida 16
2 Summary of principal sources of radium-226 data 35
3 Summary of the occurrence of dissolved radium in water 45
4 Summary of available radium-226 data and statistics 53
5 Comparison of 1966 and 1974-1976 radium-226 data for the 72
mineralized area in Polk, Hardee, Manatee, and Hillsborough
counties
6 Ground-water quality data from monitoring wells in the 78
vicinity of the C.F. Industries, Inc. gypsum pond near
Mulberry
7 Summary of 1973-1976 radium-226 data exceeding 5 pCi/1 80
vm
-------
ACKNOWLEDGMENTS
Particular appreciation is extended to the many reviewers for their
numerous criticisms and suggestions on the draft report. Noteworthy in this
regard are staff of the Office of Radiation Programs—Las Vegas Facility and
the Eastern Environmental Radiation Facility, members of the Florida Phosphate
Council, and especially the chief consultant, Mr. Gordon F. Palm. Thanks are
also extended to Mr. Gene McNeil 1, EPA, Region IV, for his continuing interest
and patience shown in the many months of report preparation. Technical advice
and suggestions were freely given by Barbara Boatwright, Southwest Florida
Water Management District, and by William Wilson, Craig Hutchinson, and James
Cathcart of the U.S. Geological Survey. Mr. Charles R. Russell, formerly with
the Office of Radiation Programs, greatly assisted in the early stages of data
collection and reduction. A special note of thanks is extended to Mrs.
Edith Boyd and Mr. David Ball for their assistance with typing and drafting.
-------
EFFECTS OF PHOSPHATE MINERALIZATION AND THE PHOSPHATE INDUSTRY ON
RADIUM-226 IN GROUND WATER OF CENTRAL FLORIDA
ABSTRACT
Principal U. S. phosphate production is from central Florida where
mining, processing, and waste disposal practices intimately associate the
industry with water resources. Available radium-226 data from 1966 and from
1973-1976 were statistically analyzed to characterize radium in the water
table, Upper Floridan, and Lower Floridan aquifers. Mined and unmined mineral-
ized areas and nonmineralized areas in the primary study area in Polk, Hardee,
Hi 11sborough, Manatee, and De Soto counties were studied. Log-normal probabil-
ity plots and nonparametric statistical tests (Mann-Whitney, Kruskal-Wallis,
Kolmogorov-Smirnov, simultaneous multiple comparison) were used to analyze for
central tendency, variance, and significant difference as functions of time,
depth, and location.
Geometric mean radium-226 content of the water table aquifer in mineral-
ized unmined areas is 0.17 pCi/1, with few observations exceeding 5 pCi/1.
Compliance with the EPA drinking water standard for dissolved radium (5 pCi/1
for radium-226 plus -228) is likely although confirmatory radium-228 data are
needed. This is particularly so in areas containing monazite sands where
thorium-232, the parent for radium-228, is elevated relative to other areas of
central Florida. No significant difference exists in the radium content of
the water table aquifer in mineralized (mined and unmined) areas versus
nonmineralized areas, inferring that mining and mineralization have not caused
a widespread and significant increase in the radium content of this aquifer.
Radium content of the Upper Floridan aquifer is poorly documented. For
mineralized but unmined areas, 1 of 5 observations exceeds 5 pCi/1, compared
to 1 of 10 in mined areas. Simultaneous multiple comparison at the 80
percent confidence level reveals significant difference between mined and
nonmineralized groups, yet mean radium in nonmineralized areas exceeds that in
1
-------
mined areas, implying no adverse marked change in radium-226 in the Upper
Floridan aquifer as a result of mining and waste management. Log-normal
probability plots of radium in the Lower Floridan aquifer for mineralized and
nonmineralized areas are very similar, again indicating that phosphate mineral-
ization or the industry is probably not a factor. Three separate populations
are indicated with geometric means of 0.7, 3, and 10 pCi/1.
Radium in the Floridan aquifer in Manatee and Sarasota Counties is
elevated relative to that in the water table and in other areas of Florida.
The mean content for seven Manatee County wells was 4.52 pCi/1 versus 1.23
pCi/1 in three water table wells. Geometric mean radium content of the water
table aquifer in Sarasota County is 15 pCi/1 versus 7.5 pCi/1 in the Floridan.
Potential radium sources for the water table include shallow phosphate sedi-
ments and monazite sands whereas radium in the Floridan aquifer in this area
may be related to mineralized water in the aquifer and to crystalline basement
rocks or other strata unrelated to phosphatic zones of current economic
interest.
The existing radium-226 data base is marginal in terms of number and
spatial distribution of analyses, particularly for the water table and Upper
Floridan aquifer. Time series data are nonexistent and study objectives and
techniques for the past decade have been rather inconsistent. Therefore the
data are not readily compared. No distinct temporal trend is apparent in
comparing individual or grouped observations made in 1966 and those in the
period 1973-1976. Local contamination associated with specific operations has
occurred and is likely to continue as water development and mining expand.
Natural variability in radium content of ground water complicates determination
of background versus contaminated conditions and underscores the need for more
intensive data collection as an integral part of water and land management.
-------
SUMMARY AND CONCLUSIONS
1. Using available radium analyses for the period 1966 to 1976 from a study
area located in Polk, Hardee, Hillsborough, Desoto, and Manatee Counties
of central Florida, nonparametric statistical tests and graphical tech-
niques were utilized to evaluate radium concentrations in three separate
aquifer systems (water table, Upper Floridan, Lower Floridan), mineralized
and nonmineralized areas, and two time periods: 1966 and 1973-1976.
Geometric mean radium concentration in the water table aquifer in unmined,
mineralized areas is 0.17 pCi/1 compared to 0.55 pCi/1 in mined areas.
For the Upper and Lower Floridan aquifers, average concentratons of
radium-226 are equal or higher in the control areas relative to mineral-
ized or mining areas.
2. Assessment of dissolved radium-226 in the water table aquifer indicates
that no significant differences (Mann-Whitney U test, a = .05) exist
between areas impacted by mining and those with mineralization, but not
yet mined. None of the radium-226 observations available in mined areas
for the water table aquifer exceed 5 pCi/1.
3. Assessment of dissolved radium-226 in the Upper Floridan aquifer indicates
significant differences (Kruskal-Wallis test, a = .05) between data from
nonmineralized, mineralized (and mined) and mineralized, unmined areas.
Using simultaneous multiple comparisons (a = .20), one pair has a signifi-
cant difference. The nonmineralized data have higher radium levels than
data from areas influenced by mining. Clearly, detrimental increase
in radium in the Upper Floridan aquifer is not documented. Average
levels for all Upper Floridan aquifer observations are higher than for
those in the water table aquifer, and similar to that of the Lower
Floridan aquifer.
-------
4. Comparison of data from mined and nonmined but mineralized areas versus
data from nonmineralized areas reveals that radium-226 in the Lower
Floridan is not significantly different (Mann-Whitney U test, a = .05).
In addition, graphical analysis of the data suggests that dissolved
radium in the Lower Floridan may be made up of as many as three separate
populations, the geometric means of which are 0.7, 3.0, and 10 pCi/1.
5. Radium-226 data collected in 1966 by the FWPCA and in 1974-1976 by the
USGS from each of the three aquifers reveals no statistically significant
difference (Kruskal-Wallis test, a = 0.05 and simultaneous multiple
comparisons, a = 0.20) for the decade considered.
6. Incidence of occasional local contamination (laterally to distances of a
mile and to depths of several hundred feet) is likely to continue as water
development and mining expand. Contamination is generally poorly docu-
mented due in part to monitoring deficiencies. Hydrogeologic conditions
favor entrance of contaminants to at least the water table and Upper
Floridan aquifers. Potential contaminant sources include high dissolved
radium in gypsum pond water and suspended radium in slime ponds. Entrance
of contaminants into ground water can occur as a result of sinkhole
collapse (or similar release) and seepage. Siting of waste disposal
facilities in contact with limestone strata of the Hawthorn Formation, in
particular, fosters contamination problems.
7. Radium-226 concentrations in ground water of Sarasota County, the secon-
dary study area, are two orders of magnitude greater than in the primary
study area for the water table and almost an order of magnitude greater
for the Floridan aquifer. Radium in the water table aquifer is signifi-
cantly greater in the coastal area as opposed to an inland area (Mann-
Whitney U test, a = .05). Elevated (relative to Florida and national
averages) radium-226 levels in both the Floridan and water table aquifers
are a result of natural enrichment processes, probably related to radium-
enriched, mineralized ground water deep under the central peninsula but
at shallow depths in coastal areas, and dissolution of radium-226 from
the Hawthorn Formation which crops out or is very near the land surface
-------
in the western portion of Sarasota County. Monazite sands in the coastal
area are a potential source of radium-228 in ground water which warrants
further monitoring.
8. Hydrogeologic conditions in the area of phosphate mineralization are
quite variable and can be favorable or unfavorable for waste disposal.
The principal variables are the distribution and permeability of car-
bonate strata in the Hawthorn Formation and the degree of interconnection
between the Upper and Lower Floridan aquifers. Disposal of acidic
chemical processing effluent may induce solution and collapse of car-
bonate strata and thereby increase contamination. However, the data to
support widespread occurrence of this phenomenon are scanty.
9. The radiochemical quality of water introduced to the Lower Floridan
aquifer via recharge wells has only recently begun to be documented.
Preliminary data from SWFWMD indicate that radium-226 levels are generally
lower than those in the Lower Floridan aquifer. The occurrence of at
least locally high natural concentrations of radium in the shallower
aquifers requires close monitoring and reporting of such data so that the
presence or absence of naturally contaminated water in a given mining
operation is documented by site specific radiochemical data.
10. Three separate aquifers present in the study area require specific
monitoring programs to determine baseline and subsequent conditions
through the reclamation stage. Until recently there has been excessive
reliance on use of existing water wells for monitoring water quality in
the Lower Floridan aquifer. Data collection typically consists of single
grab samples and diverse analytical procedures, both of which present
major limitations to detailed definition of water quality impacts from
the industry, particularly with respect to the water table and Upper
Floridan aquifers.
11. From both temporal and spatial standpoints and in relation to commonly
recognized objectives for monitoring, the existing radiochemical data
base is not adequate. Baseline and contamined water quality in areas of
5
-------
mineralization (with and without mining) is difficult to establish,
particularly for the water table aquifer. Hydrologic and geologic data
collected by industry largely precede mining and processing and thereby
only partially document baseline conditions. Time series data are
needed, analytical procedures should be standardized, and emphasis should
be on sampling wells designed specifically for monitoring. From a public
health study standpoint, existing radium-226 data are probably minimally
adequate to determine the quality of public water supplies. Particularly
in areas of phosphate mineralization and/or in areas containing monazite
sands, limited radium-228 data should also be collected for private and
public water supplies to confirm whether the EPA drinking water standard
is being met.
12. Water budget studies are recommended to document seepage losses from
gypsum ponds. Extensive ground-water studies by industry largely focus
on the Lower Floridan aquifer and are designed to determine effect of
withdrawal on the hydrologic system for purposes of justifying application
for SWFWMD consumptive use permits. Past monitoring of the water table
and Upper Floridan aquifers for contamination was minimal but the situa-
tion is improving, largely due to SWFWMD requirements.
-------
RECOMMENDATIONS
IMPROVED WASTE DISPOSAL
Use of lined ponds for gypsum wastes should be required if such facilities
are found to contribute significant radioactivity to ground water. Regardless,
rainfall exceeds evaporation in Florida, and alternative water removal and/or
recycling may become a necessary part of the control program to prevent the
long-term formation and release of leachate from gypsum piles. The feasibility
of neutralizing acidic wastes to reduce radium solubility and the potential
for solution collapse of gypsum pond substrates should be investigated.
MONITORING
Review of recent discussion of ground water quality monitoring method-
ology (Todd et al., 1976) reveals four basic objectives of monitoring:
ambient trend, source, case preparation, and research. There is little
evidence that any concerted effort with respect to radiochemical contaminants
has ever been put forth in any of the four types. Particularly lacking,
considering the scale of the phosphate industry, are measurements of ambient
spatial and temporal trends and deviations in relation to standards. Moni-
toring of effluent quantity and quality factors as potential sources of
ground water contamination are also noticeably absent.
The reader is referred to extensive discussions by Todd et al. (1976) of
the steps involved in implementing a ground-water monitoring program. Appli-
cation and tailoring of the conceptual steps to the central Florida setting
and the phosphate industry far exceeds the scope of this report. However, the
meager knowledge now existing concerning the effects of the industry on shallow
ground water quality is perhaps the most compelling reason to implement needed
studies and related abatement programs in consonance with section 102(a),
104(a), 106(e), and 502(19) of the Federal Water Pollution Control Act (as
amended) and sections 1424(e), 1442(a)(l), 1442(a)(4), and 1442(a)(5) of the
Safe Drinking Water Act.
-------
As discussed here, monitoring denotes a scientifically designed program
of ongoing surveillance incorporating direct sampling, inventory of existing
and potential causes of change, analysis of cause and effect with respect to
water quality, and prediction of future change. Of key concern with respect
to Florida phosphate mining and processing is long-term prediction of the
extent, or at least the trend of radium contamination.
Recommended monitoring during the operation of a mine and related facili-
ties includes measurement of water levels in piezometers and selected water
wells, water sampling, geophysical measurements, and maintenance of material
balances, particularly water. The reader is referred to Warner (1974) and
Le Grand (1968) for information concerning the kind and location of sampling
points, frequency of sampling, measurement of water levels and geophysical
surveys. Borehole techniques and electrical resistivity (Merkel, 1972;
Hackbarth, 1971; Stellar and Roux, 1975) are deemed particularly suitable for
delineating zones of preferential contaminant migration and for selection of
monitoring points.
WATER SAMPLING AND ANALYSIS
1. Radium-226 data should be routinely collected from the three princi-
pal aquifers in mineralized (mined and unmined) areas to ascertain
temporal and spatial trends in water quality. Measurement of other
parameters such as gross chemistry, pH, fluoride, and suspended/
dissolved solids is also recommended. Concentrations of other
radionuclides should also be investigated.
2. Wells should be repetitively sampled and a concerted effort should
be made to resample wells used in previous studies. In this case,
identical analytical procedures should be followed to reduce differ-
ences due to analytical techniques.
3. Sample collection and handling should emphasize field filtration
followed by acidification. Analysis should be for dissolved and
suspended radium.
-------
4. Sampling of producing wells low in suspended solids is preferable.
Monitoring wells and piezometers not in daily use should be pumped
or bailed and allowed to settle shortly before sampling. Sampling,
per se, should allow minimal suspended solids to be collected.
HYDROLOGY AND GEOLOGY
1. Analysis of lateral and vertical ground-water flow patterns in all
principal aquifers should precede, accompany, and follow mining,
waste disposal, and reclamation operations.
2. Water budget and material balance studies should be undertaken to
confirm the magnitude of seepage from gypsum ponds. Where extensive
seepage can be documented or where disposal areas are geologically
unsuitable, corrective action in the form of lining or alternative
siting is recommended.
3. Ground-water quality monitoring should be based on analysis of local
hydrogeologic conditions and should place principal emphasis on the
water table and Upper Floridan aquifers
4. Geochemical investigations are necessary to document the amount,
kind, and distribution of radioactivity in overburden and matrix
materials prior to and after mining and waste disposal. Predictions
of radionuclide migration from disturbed soil profiles remaining
after mining and reclamation should be made to assess future radio-
nuclide content of shallow sediments and contained ground water.
Kinetics of radium solubility under changed Eh and pH conditions in
the subsurface are poorly known. The transport rate for radium in
the vadose and saturated zones, taking into account dispersion and
sorption, require much additional study relevant to the Florida
phosphate situation.
5. Reasons for and mechanics of sinkhole collapse or other means whereby
massive release of slimes and gypsum wastes to the subsurface should
-------
be documented and corrective active taken, particularly considering
the expansion of phosphoric acid production in Florida.
6. Stabilization of gypsum piles, in particular, also necessitates
control of infiltration from precipitation in order to prevent
leachate production and subsequent migration to nearby surface and
subsurface water bodies. Techniques developed for moisture control
in common sanitary landfills and for land disposal of toxic wastes
should be investigated for their applicability.
DATA INTERPRETATION AND REPORTING
1. Applications for mining and processing and periodic interpretive
written reports submitted by industry should detail monitoring
programs and results and need for corrective action.
2. Periodic documented, reports of the state of the environment should
be prepared by appropriate State agencies based on results of
industry data and other study programs conducted by the State or
State/Federal cooperative efforts. Emphasis should be on confirma-
tion of conditions predicted in DRI applications, maintenance or
improvement of desired environmental conditions, and identification
of short and long-term benefits and impacts stemming from phosphate
extraction and related commitments of land and water resources.
10
-------
PROBLEM DESCRIPTION
Major centers of production for domestic phosphorus are located in
Florida, North Carolina, Tennessee and in the western states of Idaho, Montana,
Wyoming, and Utah. Total U.S. production of marketable (beneficiated) phos-
phate rock in 1976 was about fifty million short tons, about eighty percent of
which was from Florida (Stowasser, 1976). The principal area in Florida and
the subject of this report is the land pebble mining district (Figure 1) which
was selected for study because it is the principal existing and potential
producing area in the State and therefore the scene of extensive mining,
chemical processing of phosphoric acid, and land reclamation. Extensive
ground-water development for irrigated agriculture, the phosphate industry,
and municipal purposes is occurring in the region.
The multiplicity of Federal interests and actions in the Florida phosphate
industry prompted the Council of Environmental Quality to request and fund
preparation of an environmental impact statement under the auspices of Region
IV of the U.S. Environmental Protection Agency (USEPA). This report is in
direct support of the overall effort to document present and expected environ-
mental impacts associated with phosphate mining and ore processing in central
Florida. Other studies by the Office of Radiation Programs address other
aspects including atmospheric releases from processing plants and indoor and
outdoor radiation exposures associated with use of reclaimed phosphate lands
for housing structures.
The U.S. Environmental Protection Agency (1976a) has issued regulations
concerning the amount of radium in public water supplies. In terms of drinking
water regulations, radium-226 is lumped with radium-228 and jointly the two
must not exceed 5 pCi/1. While radium-228 is a beta emitter as opposed to
radium-226, which is an alpha emitter, radium-228 has a chain of alpha-particle
emitting daughters such that the gross alpha particle activity limit for
drinking water is defined to include radium-228. Furthermore, if water
11
-------
:/COUNTIES IN THE
PRIMARY STUDY AREA
GULF
A^«-_
OF
MEXICO
HILLSBOROUGH CO.
POLK CO.
MANATEEm CO
HARDEE CO.
I|SARASbTA"\J
DESOTO CO. j
< ji\SARASOTA COUNTY
(SUPPLEMENTARY STUDY AREA)
Mm''1,''1, AREA OF AVAILABLE RADIUM-226 DATA
APPROXIMATE BOUNDARY FOR SIGNIFICANT
PHOSPHATE MINERALIZATION;
HACHURES ON MINERALIZED SIDE
APPROXIMATE LOCATIONS OF
CITIES AND TOWNS
Figure 1. General location map.
12
-------
contains greater than 5 pCi/1 alpha activity, radium-226 must be determined.
If radium-226 is greater than 3 pCi/1, then radium-228 must be determined.
However, if other information suggests that radium-228 is present, States are
recommended to determine radium-228 and/or radium-226 if gross alpha is greater
than 2 pCi/1 (USEPA, 1976a).
In the study area, land and water use patterns and management are closely
related to the phosphate industry insofar as mining, beneficiation, and
chemical processing have the potential to adversely affect ground-water
quality through a variety of mechanisms. Radiochemical species naturally
present can be concentrated and mobilized as a result of mining activity or
waste discharge from mining, processing, or reclamation activities. Given
this, the objective of this report is to document, using available data,
hydrogeologic and water quality conditions as they passively or actively
relate to the phosphate industry in west-central Florida. Secondly, there is
a need to identify necessary additional studies and data collection efforts.
This report summarizes the hydrogeologic situation in a portion of
central Florida as related to phosphate occurrence and, more importantly, to
the potential for ground-water contamination from phosphate mining and pro-
cessing. Also included is a brief review of the occurrence of radium in
water. Existing radium (in water) data from central Florida are statistically
analyzed to compare radium concentrations in ground water for 1) mineralized
versus nonmineralized areas and 2) mining/processing areas versus mineralized
but undeveloped areas. Monitoring efforts to date and resulting data base are
evaluated and recommendations are made concerning needed improvements.
PREVIOUS AND ONGOING INVESTIGATIONS
Owing to the demands on ground-water resources imposed by the phosphate
industry and other municipal and agricultural water users in central Florida,
there has been greatly increased interest in the last decade to identify
environmental impacts. Since 1973, State and regional agencies within Florida
carefully reviewed applications for mining and other activities affecting the
environment to determine if environmental, social, and economic impacts have
13
-------
been identified and to rule on the acceptability of such impacts (380.06(06),
Florida Statutes). Known as Development of Regional Impact (DRI), the review
process is somewhat analogous to an environmental impact statement and is
initiated by the interests proposing development. Technical documents and
development orders appurtenant to the DRI process constitute site-specific
sources of available information concerning data collection, expected impacts,
and mitigating measures. Numerous geologic and hydrologic investigations
conducted in central Florida also constitute valuable resource material and
are referred to as appropriate.
Water quality effects associated with the phosphate industry and areas of
phosphate mineralization has been mentioned in reports by the Federal Water
Pollution Control Administration (Shearer et al., 1966), Battelle Memorial
Institute (1971), Datagraphics, Inc. (1971), U.S. Environmental Protection
Agency (1973; Guimond and Windham, 1975), and most recently the U.S. Geological
Survey (Irwin and Hutchinson, 1976). The principal radium-226 data upon which
this report is based include the following: 1) Open file analyses from the
U.S. Geological Survey (obtained from R. C. Scott, private consultant,
Atascadero, California, formerly with the USGS and USEPA), 2) Scott and
Barker's (1962) study of the distribution of radium in water throughout the
country, 3) an FWPCA survey in 1966 (Shearer et al. , 1966), Aground-
water sampling by-USEPA from 1973-1976, 5) ground water and surface water
sampling by the USGS from 1974-1976 (Irwin and Hutchinson, 1976). The manner
in which these various data bases are used is explained in the section on
water quality.
Other studies include Osmond's (1964) summary review of the distribution
of uranium and thorium in the rocks and water of Florida. Williams et al.,
(1965) presented gross alpha data for 280 untreated well water samples from
Florida. This included 22 samples from 18 shallow wells in the central
Florida phosphate district.
14
-------
HYDROGEOLOGIC SETTING OF WEST CENTRAL FLORIDA
GEOLOGY OF CENTRAL FLORIDA
Geologic investigations of those portions of Polk and Manatee Counties of
interest herein include U.S. Geological Survey quadrangle reports (Cathcart,
1963a, 1963b, 1963c, 1964, and 1966) and numerous cooperative studies by the
U.S. Geological Survey in cooperation with the Florida Geological Survey
(Peek, 1958; Pride et al., 1966; Stewart, 1966; and Robertson, 1973).
(Note: Florida Geological Survey is now called Bureau of Geology.)
The stratigraphic sequence in the study area primarily consists of
gently dipping carbonate bedrock overlain by a thin sequence of clastic and
phosphatic sediments. Formational names, lithologic descriptions, thickness,
and aquifer makeup are shown in Table 1. Generalized stratigraphic cross
sections through the study area are shown in Figures 2 and 3. In most areas,
the surface material consists of Pleistocene sands, commonly called terrace
sands, containing varying amounts of organic debris. In southwest Polk and
adjacent counties, these sands are underlain by the Bone Valley phosphorite
unit, a complex assortment of reworked phosphatic clay, silt, and sand at
least partly derived from the underlying Hawthorn Formation as a result of
intense lateritic weathering and leaching (Cathcart, 1964). The lowest unit
consists of a basal phosphatic conglomerate much enriched in phosphate (Hoppe,
1976) and usually part of the phosphate ore zone which is locally called
matrix. The Bone Valley unit thins or is absent in northern and eastern Polk
County but is prominent throughout much of southwestern Polk County and
adjacent areas in Hillsborough, Manatee, DeSoto, and Hardee Counties. The
gentle southerly dip (Cathcart and McGreevy, 1959) results in deeper burial in
these latter areas.
Underlying the Bone Valley unit is a phosphatic, interbedded sequence of
limestone, dolomite, sand, sandy clay, and gray clay known as the Hawthorn
15
-------
TABLE 1. GEOLOGIC AND HYDROGEOLOGIC UNITS IN CENTRAL FLORIDA
GEOLOGICAL
AGE
PLEISTOCENE
PLIOCENE
MIDDLE MIOCENE
EARLY MIOCENE
LATE OLIGOCENE
LATE EOCENE
FORMATION
TERRACE DEPOSITS
01
BONE VALLEY
FORMATION
Tb
o>
CL
a.
=J
Tbu
1
Tbl
HAWTHORN FORMATION
Th
TAMPA FORMATION
Tl
SUWANNEE LIMESTONE
Ts
OCALA GROUP
To
LITHOLOGICAL DESCRIPTION
UNWEATHERED
UNCONSOLIDATED QUARTZ
SAND WITH ORGANIC DEBRIS
RED & WHITE MOTTLED SANDY
CLAY: GREY AND TAN SAND.
CLAYEY; SCATTERED PHOSPHATIC
NODULES INCREASING TOWARD BASE
SANDY, CLAYEY GREY & GREY
GREEN PHOSPHORITE: NODULES
SAND TO GRANULE SIZE. LOCAL
BASAL PHOSPHATIC CONGLOMERATE
OLIVE GREEN. GREY BROWN
CLAYEY SANDSTONE WITH
NODULES TO A BUFF. WHITE
LIMESTONE WITH NODULES
AT BASE. DOLOMITIZED
LIGHT YELLOW. SANDY AND
CLAYEY LIMESTONE: BROWN
BLACK PHOSPHATIC NODULES
MAY BE CLAYEY SAND
FOSSILIFEROUS CREAM OR
TAN LIMESTONE: VERY SOFT
GRANULAR DETRITAL
WHITE, GREY. CREAM OR TAN
SOFT GRANULAR LIMESTONE
OF HIGH PURITY: OOLOMITIZED
ALSO SOFT, CHALKY
WEATHERED
LOOSE QUARTZ SAND. SWAMP
DEBRIS
WHITE SAND. TRACE OF CLAY
WITH SCATTERED DULL WHITE
PHOSPHATIC NODULES PARTICULARLY
AT BASE
WHITE SAND. VESICULAR
WITH DULL WHITE
PHOSPHATE GRADING DOWNWARD
TO GREY OR GREY-GREEN SAND
CLAYEY OR SANDY CLAY WITH
NUMEROUS NODULES.
CALCAREOUS SANDY CLAY
WITH BROWN OR BLACK
PHOSPHATIC NODULES WITH
POSSIBLE GRADATION TO LIME-
STONE AT DEPTH
CALCAREOUS SANDY CLAY
CONTAINING SCATTERED
PHOSPHATIC NODULES
NEAR SURFACE OCCURENCE
TENDS TO BE SILICIFIED
SILICIFIEO TO HARD.
GREY TO WHITE CHERT.
THICKNESS
(It.)
0-65
P!:0-25
M3:0-35
ItatalTb)
P:0-35
P:0-130
M:l 50-350
P:10-80
M:150-175
P:80-120
M:150-300
P: 125-270
M:300-325
AQUIFER
wr
WT
UA<
P:UFS
M:UF or
LF6
P:UF
M:UF or
LF
LF
LF
ROLE IN PHOSPHATE
DEPOSITS
NO DIRECT ROLE NOTED
INTENSE LATERITIC WEATHERING
USUALLY DEPLETED IN PHOSPHATE
BASAL CONGLOMERATE GREATLY
ENRICHED IN PHOSPHORUS DURING
INTENSE WEATHERING OF HAWTHORN
PRINCIPAL ORE COMPONENT IN
POLK CO.
PARENT MATERIAL, IN PART. FOR
BONE VALLEY.IRREGULAR UPPER
SURFACE BECOMES IMPORTANT
ORE IN MANATEE CO.
SOME RESIDUAL CLAY PRESENT
FROM WEATHERING EVENTS WHICH
ENRICHED ORE WHEN UPPER
HAWTHORN ABSENT. NOT ORE
NO DIRECT ROLE NOTED
NO DIRECT ROLE NOTED
'NONARTESiAN 'POLK COUNTY 'MANATEE COUNTY 'UPPER MOST ARTESIAN 'UPPER FLORIOAN ARTESIAN 'LOWER FLORIDAN ARTESIAN
-------
^DIFFERENTIATED
FORMATION
-H '—"~r
FORMATION
LIMESTONE
-L-
01 2345 MILES
VERTICAL EXAGGERATION <105
SECTION IN POLK AND HARDEE COUNTIES ADAPTED FROM STEWART(1966) AND
WILSON (1975), RESPECTIVELY. WELLS ARE IDENTIFIED ACCORDING TO THE
DESIGNATION USED IN THESE REPORTS. TRACE OF CROSS SECTION SHOWN ON
FIGURE
Figure 2. Generalized geologic cross section through southern Polk and northern Hardee Counties
-------
MANATEE RIVER TERRACE SANDS
M-8 \
BONE VALLEY FORMATION
Tjnconformlty
NORTH FORK
MANATEE RIVER
HAWTHORN FORMATION
unconformity"
TAMPA FORMATION
unconformity-
SUWANNEE LIMESTONE
unconformity-
OCALA GROUP
unconformity"
AVON PARK LIMESTONE
--MSL
See Figure 7. for trace of cross section. Formation contact elevations were derived from structure contour maps, thickness data, and well logs presented by Peek(1958). Wells
M-12, M-8 etc.are shown for orientation purposes only as these are shallow wells.
Figure 3. Generalized geologic cross section through northeastern Manatee County.
-------
Formation (Cathcart, 1963a) which is generally present south of the Alafia
River and west of the Highlands Ridge. It is present in Polk, Hillsborough,
Hardee, DeSoto, Charlotte, Manatee, and Sarasota Counties (Cathcart and
McGreevy, 1959; Stewart, 1966; Wilson, 1975). A source for part of the Bone
Valley, the Hawthorn has an irregular upper surface and is sometimes suffi-
ciently enriched in phosphate to constitute part of the matrix (Hoppe, 1976).
In Manatee County the Hawthorn contains more clastic materials and is thicker
relative to the section in Polk County (Cathcart, 1963a). From east to west,
it becomes increasingly silty as well. At the Swift Chemical Company phosphate
mine area in northeastern Manatee County, the Hawthorn is predominantly silt
whereas in Hardee County it is characterized by clay.
Southward in Manatee County and particularly in the area of the Manatee-
Sarasota County line, the Hawthorn is exposed or quite shallow, rather clayey,
and the principal source of phosphate ore. Phosphate deposits, presumably
from the Hawthorn, are exposed in the banks of the Intercoastal Waterway Canal
in the area of Casey Paso. Much of Sarasota County is underlain by phosphatic
sediment beneath twenty feet or less overburden which thins or is absent to
the west.
The Tampa Formation which underlies the Hawthorn Formation was originally
identified (Parker et al., 1955) from subsurface data in the Tampa area where
the unit is predominantly limestone and a prominent aquifer in the Floridan
aquifer system. Elsewhere in central Florida and particularly in southern
Polk County and adjacent portions of Hillsborough, Manatee and Hardee Counties,
clastic materials are common in the lower part of the Tampa which acts as a
confining layer. Use of the term "Tampa Formation" was recommended by Parker
(Geraghty and Miller, Inc., personal communication, February 25, 1977) and is
so used herein.
The lithological sequence of the Tampa Formation and its role in the
Floridan Aquifer system in central Florida has been debated for many years.
It is generally considered a limestone in which there occurs a confining
clayey member. Stewart (1966) contended that the clay unit is at the top of
the Tampa. However, Wilson's (1975) study of Hardee and DeSoto Counties
19
-------
revealed a persistent and extensive clay and silt unit at the bottom of the
Tampa Formation. This is generally believed to be true in southern and
southwestern Polk County as well, where the Tampa Formation is present only as
a clayey unit (B. Boatwright, Southwest Florida Water Management District,
personal communication, February 16, 1977). It is absent in northern Polk
County (Pride et al., 1966). In Manatee County and the western third of
Hardee County it is largely made up of carbonates (Wilson, 1975) at least in
part equivalent to the clay and silt member present in Polk and Hardee Counties
In Manatee County the clay and silt member is believed present only in the
northeast corner (based on extension of the trends reported in Hardee County
by Wilson, 1975). The Tampa Formation generally dips south or southeast in
Polk County (Stewart, 1966) and to the southwest in Manatee County (Peek,
1958).
AQUIFER SYSTEMS AND GROUND-WATER FLOW
For the purpose of this report three principal aquifers are defined in
the study area: The Upper and Lower Floridan aquifers which are typically
confined and an unconfined local shallow aquifer. These are schematically
shown in Figure 4 which also depicts the principal geologic units present.
The degree of interconnection between aquifers is highly variable with location
and, in effect, creates the large Floridan aquifer system which is made up of
a series of aquifers and confining layers. The most prolific aquifer is the
Avon Park Limestone. The initial discussion which follows is applicable
primarily to Polk County and immediately adjacent portions of Hillsborough,
Manatee, and Hardee Counties.
Unconfined or water table conditions occur in the unconsolidated surficial
sediments and locally may extend to the lower part of the Hawthorn Formation.
The Floridan aquifer system, composed of at least five major stratigraphic
units and individual aquifers, includes the Upper and Lower Floridan aquifers
as defined herein. In most of the study area the Lower Hawthorn and the Upper
Tampa belong to the Upper Floridan aquifer which is sporadically distributed
and variable with respect to head, yield characteristics and water quality.
In northeastern Manatee County at the Swift Chemical Company mine site and at
20
-------
CENTRAL MANATEE CO.
CENTRAL POLK CO.
ro
WATER TABLE
AQUIFER
AQUITARD
UPPER FLORIDAN
AQUIFER
.AQUITARD.
LOWER FLORIDAN
AQUIFER
TERRACE SANDS
BONE VALLEY
FORMATION
HAWTHORN
FORMATION
TAMPA
FORMATION
SUWANNEE
LIMESTONE
OCALA
GROUP
PREDOMINANTLY LIMESTONE AND DOLOMITE
NOT TO SCALE
LEGEND
Direction of ground-water flow
Water table position
Carbonate bedrock
(limestone,dolomite)
Clay, silty clay
Figure 4. Generalized southwest-northeast hydrogeologic cross section through Polk and Manatee Counties
-------
the Phillips site in Desota County, a distinct Upper Flon'dan aquifer is not
present (B. Boatwright, Southwest Florida Hater Management District, written
communication, June 6, 1977). Successively deeper carbonate units below the
Tampa are part of the Lower Floridan aquifer which extends throughout most of
central Florida. The Lower Floridan is typically under artesian pressure and
water encountered in a bore hole penetrating a given aquifer will rise above
the elevation of the stratigraphic boundary between the aquifer and overlying
units. In certain instances the rise is sufficient to allow water to flow to
the land surface but this condition now exists in only a few areas. Because
of heavy pumping, particularly in southwestern Polk County and Southwestern
Hardee County, water levels or head in the Lower Floridan are falling. Under
completely natural conditions they were generally also below those in shallower
aquifers except in areas where there were or are flowing wells developed in
the Lower Floridan. Pumping, therefore, is causing increased downward flow in
much of the study area.
The foregoing explanation is highly simplified and is probably applicable
only to that portion of the study area in Polk County and immediately adjacent
portions of Hillsborough, Manatee, and DeSoto Counties. Aquifer boundary
conditions are largely stratigraphic and therefore variable depending on the
persistence of lateral and vertical successions and structural controls. For
example, stratigraphic relationships of the surficial sand unit and the
lithologic makeup of the Hawthorn are very influential in determining the
amount of recharge to the latter and whether it constitutes an aquifer or a
semi-confining unit. Similarly, development of solution channels and other
karst features in the Hawthorn and Upper Tampa Formations determine in part
the amount of recharge from local precipitation or other hydrologic events at
the land surface. Where the Hawthorn is thick and areally extensive, karst is
not well developed. For Sarasota and Manatee Counties, separation of the
Upper and Lower Floridan aquifers is generally not possible. In large part
this is due to scarcity of stratigraphic and hydraulic information and because
the limited facies data available (Wilson, 1975) indicate different conditions
than the area to the northeast.
Structural elements in Polk and Manatee Counties are not topographically
prominent but may play important roles in ground-water occurrence and movement
22
-------
in the Floridan aquifer system. The principal structural element in central
Florida is the north-trending Peninsular Arch which is east of Polk County.
Along the west flank of the arch is the Ocala Uplift, the south end of which
extends into northern Polk County (Figures 8 and 9, in_ Pride et al., 1966).
The Ocala Uplift imposes both hydrologic and geologic control on ground water
in central Florida. The elevated position of the arch over geologic time
allowed erosion and/or non-deposition, thereby positioning more permeable
carbonate units at or near the land surface. This facilitates increased
recharge. In addition to the stratigraphic framework, lateral and vertical
controls on flow include paleokarst and fault planes associated with uplift,
deposition/erosion cycles, and paleo-weathering features. Flanking the
highland area at depth are less permeable materials which act as semi-confining
units and aid in the creation of artesian conditions. The combination of
increased recharge, elevated topographic position, stratigraphic boundary
conditions, and regional dips generally to the south, southeast, or southwest
gives rise to artesian conditions in the Floridan aquifer, and to a more local
extent, in the basal limestone of the Hawthorn Formation. Most of the fore-
going features are schematically shown in Figure 4.
INFLUENCE OF MINING AND PROCESSING
Strip mining and chemical processing as practiced in Florida's central
phosphate district can affect ground-water as a result of activities associated
with three broad categories: 1) pre-mining site preparation, 2) waste disposal
related to ore extraction/beneficiation/chemical processing, and 3) post-
mining site reclamation and slime pond or gypsum pond maintenance. Figure 5
shows the major operations involved in mining. By convention within the
industry, the term mining includes beneficiation. Overburden consisting of
surficial sand and the Bone Valley unit is stripped with draglines to expose
the matrix. Water pumped from the pits or recycled from the slime ponds,
supplemented with water from the Floridan aquifer, is used to convert the
matrix to a slurry which is pumped to the beneficiation plant. There, screens
and flotations separate the phosphate from the sand and slime tailings which
are piped to disposal ponds. Makeup water is obtained primarily from deep
wells and from water added to the system as precipitation, ground-water
inflow, or matrix moisture. Mining may go as deep as the carbonate unit
23
-------
DEWATERING
WELLS
CONNECTOR (PUMPED)
PRESENT WATER TABLE
TERRACE SANDS OR
QUATERNARY CLASTIC DEPOSITS
QVERBURDEN
ACTIVE INING
NAREA BEING OEWATERED
\PRJOR TO MINING
— - - - -^7-e-j T- ~^J- —T-L—
— - , I £ /—I— i • • •
POTENTIOMETRIC SURFACE
— UPPER FLORIDAN '
. .\\\ J s i^=\
.
T.....-TAMPA FM.
POTEsmOMETRIC SURFACE
' 'LOWER FLORIDAN f
SURFICIALnB WATER
ORE\^ ( AQUIFER UK TABLE
UPPER
FLORIDAN
AQUIFER
LOWER
FLORIDAN
AQUIFER
NOT TO SCALE
Figure 5. Interaction of mining operations and the hydrogeologic system.
-------
of the Hawthorn Formation. After stripping the ore, overburden is replaced
and shallow ground-water levels, drawn down to allow mining, gradually recover
The stratigraphic units affected by excavation are shown in Figure 5.
Also shown are the water table and Upper Floridan aquifers, wells used for
water supply and dewatering, and water table position. Due to nonuniform
limestone distribution in the Hawthorn Formation and differences in the verti-
cal extent of phosphatic matrix, many or even most mines are not likely to
extend to the uppermost beds of the Floridan aquifer. In certain areas of
active mining, particularly in Polk County, shallow confining materials may be
removed and increased recharge to the Floridan aquifer may occur. Another
suspected but poorly documented source of recharge results from occasional
sinkhole collapse beneath slime or gypsum ponds. In the case of the latter,
dissolution of carbonate strata in contact with acidic gypsum pond water may
be a factor. Only one case of such a collapse is known and a few others are
suspect. Cavities are not always present and collapse incidence is believed
to be low. Remote sensing data indicate that seepage through sand tailings
dams containing gypsum ponds is common (Coker, 1971; 1972).
Use of connector wells is typified by the International Minerals and
Chemical Corporation (IMC) installation at the Kingsford mine of more than 50
such wells completed in the surface aquifer and in the underlying Floridan
aquifer (Hoppe, 1976). Cathcart (USGS, personal communication, Februarys,
1977) indicates this practice has become widely used as it is an effective
method to dewater the ore zone and sand cover prior to mining. Dewatering is
not always necessary depending on the depth to water relative to the base
of the ore body and the type of mining operation. Because of declining heads
in the Lower Floridan aquifer, recharge via connector wells is beneficial with
respect to the water supply picture. Credit is given for the amount of water
recharged, thereby allowing more pumping from the Lower Floridan than if only
natural recharge occurred. The IMC recharge program also includes 60 observa-
tion wells and 20 surface monitoring stations to preclude introduction of low
quality waters to the Floridan aquifer (Hoppe, 1976). As head in the Floridan
aquifer decreases in areas of heavy pumping, head differential relative to
shallow aquifers will increase and recharge wells will become increasingly
25
-------
popular. In the past more use was made of pumped wells or pits to effect
dewatering. Water was discharged into adjacent pits or discharged to surface
water courses. The latter practice is not favored because water is exported
from the basin rather than being used to replenish ground-water resources,
particularly the Floridan aquifer system.
Stripping of the overburden and leach zone and subsequent removal of the
phosphate ore or matrix with draglines thoroughly disrupts the natural sequence
of overburden, leach zone and matrix. Although a leach zone is not always
present it is significant from a radiation standpoint because it is believed
to contain the greatest concentration of uranium (Golden, 1968). Formerly,
overburden and leach zone materials were dumped on vacant ground adjacent to
the trench to allow removal of the matrix. As successive parallel trenches
are cut, the overburden and leach zone materials are put into adjacent, mined-
out cuts. In recent years, overburden and leach zone materials are separated
so that overburden is replaced last. Ore is hydraulically disaggregated in an
open pit and pumped to a washer plant as slurry containing about 40 percent
solids. For a typical mine, about 400 acres/year will be stripped, resulting
in the use of 4.5 billion gallons of water for slurry makeup.
Water utilization, primarily from the Floridan aquifer, has caused
significant decline in the potentiometric surface of the Floridan aquifer
between 1964 and the present. In central Florida approximately sixty percent
of the water pumped from the Lower Floridan aquifer is for irrigation and
about twenty percent for the phosphate industry. Municipal use accounts for
the balance. Introduction of new techniques for beneficiation have reduced
the phosphate water usage and at least one company, Brewster Phosphates, Inc.,
is operating without need for water withdrawal from the Lower Floridan aquifer.
The recent configuration of the potentiometric surface in the Floridan
aquifer is represented in maps prepared by the U.S. Geological Survey and the
Southwest Florida Water Management District (Stewart et al., 1971; Mills and
Laughlin, 1976). Significant potentiometric troughs are centered in the area
due north of Bartow, as well as in north-central Manatee County and north-
western Sarasota County. The latter two troughs developed in the period
26
-------
1969-1975 and are probably related to irrigated agriculture as the phosphate
industry does not operate in these areas.
Over much of the central Florida study area, the vertical flow gradient
has been downward although there are areas along the coast and along the Peace
River where gradients are or were upward. With heavy pumping and steepened
downward flow gradients there is increased potential for downward movement of
shallow ground water. Increasingly active monitoring of shallow and deep
ground water is now required by the Southwest Florida Water Management District
and as a part of the development orders appurtenant to the DRI process.
Drainage water associated with mines is not usually discharged to surface
streams. Although mine water and ambient ground water are intimately associ-
ated, the 1973 USEPA study revealed that recirculated water (tail water from
slime ponds) and pit seepage at two mines in Florida contained only 0.28 to
1.5 pCi/1 dissolved radium-226. In comparison, leachate from gypsum spoil
piles at chemical processing plants contains 60 to 100 pCi/1. It would appear
that potential adverse effects on ground-water quality would primarily be a
result of chemical processing (to produce phophoric acid) and accompanying
waste handling featuring gypsum ponds and piled gypsum waste.
Ore beneficiation utilizes recycled slime pond water supplemented with
deep ground water. Tailings and slime are emplaced in specially constructed
settling basins and mined out pits. Although the ratio varies depending on
ore grade, a general guide is that a ton each of sand tailings and slime is
generated per ton of phosphate rock product. One of the critical problems the
phosphate industry faces is dewatering large amounts of clay slime which now
requires substantial funds for dam construction, maintenance, and monitoring
because solids remain in suspension and/or will not dewater. As a result the
slime ponds not only commit a large amount of otherwise usable water and land,
but also are potential sources of radiochemical and other pollutants that can
affect nearby water resources. Recent and ongoing studies funded by the U.S.
Bureau of Mines, in particular, show that accelerated settlement of slimes is
economically feasible. One of the more promising methods involves mixing sand
and slime fractions to increase settling and water expulsion.
27
-------
As the number and/or capacity of chemical processing plants increases,
attendant gypsum piles and acidic effluent may pose increasing problems with
respect to long-term stabilization of wastes. Both the effluent and the
gypsum are elevated in radium, necessitating long-term stabilization to
prevent leachate formation and migration. No monitoring studies are available
detailing ground-water quality around such piles or stabilization techniques
for the pile proper. One pond near Mulberry failed by sinkhole collapse in
1975 and a study by Zellars and Williams (1977) showed average leakage of 13
percent or 2039 gallons per minute for two years of record at two mines and
one year at a third. The amount of leakage from gypsum ponds is unknown.
28
-------
SOURCE TERM CHARACTERIZATION
"Phosphate rock" is a commercial term denoting a rock with one or more
phosphate minerals and of sufficient enrichment and composition to be usable
as is or after concentration. In Florida, the principal ore is phosphatized
limestone.
The principal phosphorus-bearing minerals in the central Florida phos-
phate deposits are in the carbonate apatite group with the general formula:
Cag (PO,)o(F,Cl ,OH). Phosphate exchanges with small quantities of VO,, AsO*,
and SO, whereas Na, Si, Th, U, and rare earths can exchange with Ca. Such
replacement plus the cryptocrystalline structure gives rise to the term cello-
phane, indicative of a nondefinitive suite of carbonate apatite minerals which
are the essential minerals of phosphorite (sedimentary phosphate rock).
Uranium, vanadium, selenium, chromium, and rare earths can be present in
sufficient concentrations to constitute actual or potential by-products.
Natural uranium, present in Florida phosphate deposits in concentrations
of 0.1 to 0.4 pounds per ton (Stowasser, 1976), consists of approximately 99.3
percent (by mass) uranium-238. Radioactive decay of uranium-238, commonly
known as the uranium series, gives rise to uranium-234, thorium-230, radium-
226, and radon-222, amongst others. For reasons of half life, toxicity and
mobility not all members of the decay series, which ends with the formation of
(stable) lead-206, present equal hazard. Although all radiation exposure is
considered harmful and adverse effects are assumed proportional to dose
(linear, no threshold hypothesis). Radium-226 is singled out because of its
known occurrence in areas of phosphate mineralization. It is long lived (1600
years), has relatively high transfer from the gastrointestinal tract to the
blood, and has an affinity for bone where it replaces calcium and is toxic dye
to high energy alpha decay characteristics. Study is also facilitated by the
data base from sampling efforts in the last decade and particularly from 1973-
1976. Uranium and principal progeny are in secular equilibrium in the matrix
29
-------
(Guimond and Windham, 1975) and presumably in the leach zone as well. Elevated
levels of radon in structures built on reclaimed land (U.S. Environmental
Protection Agency, 1975; and 1976b; Fitzgerald et al., 1976) suggest that
uranium and radium distribution in shallow depths may be increased relative to
pre-mining levels. Presumably this occurs if leach zone materials are mixed
with other overburden materials as part of overall reclamation. What effect
weathering and leaching from rainfall may have on redistributing radioactivity
in the subsurface has not been determined and is of legitimate concern, partic-
ularly with respect to long term water quality in the water table aquifer.
Before discussing various liquid and solid wastes associated with phos-
phate mining and processing, some description of basic operations may be
useful. After the overburden is stripped, ore is excavated by draglines, put
into slurry form with hydraulic jets, and pumped to nearby washing plants for
size separation. Minus-200 mesh particles or slimes are discharged to slime
ponds for settling to allow reuse of water. Slimes amount to about one-third
of the original ore volume and represent about one-third of the total mineral
values extracted. In addition they are costly to handle and store. Slime
production from Florida alone is estimated at 36.3 million short tons in 1973
(Guimond and Windham, 1975). Estimated radium content is 1480 Curies. Slimes
are potentially damaging to water resources as a result of entrance to ground
water reservoirs or dike failure and release to streams. Reported radium-226
in Florida slimes is 45 pCi/g versus 42 pCi/g in phosphate products (Guimond
and Windham, 1975).
After separation from the ore, sand tailings are 1) piped to sand piles
as a slurry, 2) used to build up dikes around slime ponds, and 3) placed in
mined out areas. Because sand tailings yield water readily, there is increased
interest in using them to assist in slime stabilization or to bury them in
mined out areas where their permeability can be used to store and transmit
ground water for use in ore transport and beneficiation. Amine flotation
water used for sand separation is discharged to mine pits and recycled.
Radium content of sand tailings is rather low, averaging 7.5 pCi/g. Radio-
activity primarily is associated with the phosphate fraction, one-third of
which is slime waste and two-thirds product. Approximately 380 Curies of
30
-------
radium-226 in 55.8 million short tons of tailings were generated in Florida
alone in 1973 (Guimond and Windham, 1975).
Phosphate rock separated from the ore is either exported or sent to a
chemical processing plant for conversion to superphosphate by treating finely
ground phosphate rock with sulfuric acid:
tricalciurn sulfuric monocalcium gypsum
phosphate acid phosphate
The product of the above reaction is superphosphate. Monocalcium phos-
phate is soluble in water and therefore plant available. Gypsum remains in
the product and dilutes it. Phosphoric acid is produced by adding additional
sulfuric acid and separating phosphoric acid from the gypsum according to the
reaction:
Ca1()F2(P04)6 + 10 H2S04 + 20 H20 = 10 CaS04- 2H20 + 2 HF + 6 H3P04
Phosphoric acid can be reacted with additional phosphate rock to yield triple
superphosphate which contains 45 to 48 percent PO^C- ^n the above reaction,
large piles of gypsum and ponds of acidic effluents result. Production of
other fertilizers such as diammonium phosphate, urea-ammonium phosphate, or
"complete" fertilizers consisting of phosphorus, nitrogen, and potassium
compounds is increasing due to market demand and favorable transportation
economics.
Radionuclide concentrations in product, slime, and sand tailings mater-
ials from Florida phosphate deposits were determined in previous USEPA studies
(1973; and Guimond and Windham, 1975). These, in turn, were expansions of the
work begun by Spalding (1972) of the Texas A & M Oceanography Department.
Data presented by Guimond and Windham (1975) show that beneficiation processes
do not alter the isotopic ratios, which are essentially in secular equilibrium
for the uranium, actinium, and thorium decay series. However, redistribution
of radionuclide concentrations occurs among the three principal fractions
31
-------
cited. Uranium occurs as a trace element in the phosphate-bearing minerals,
hence it tends to remain with the fertilizer, whereas most of the radium is
concentrated in the solid and liquid wastes. Phosphoric acid is enriched in
radium-226, uranium, and thorium relative to the concentration of phosphate
deposits. The sand tailings fraction contains about 12 percent of the total
radioactivity, whereas the slime and product fractions contain 48 percent to
40 percent, respectively. Overall, about 60 percent of the activity, which in
1973 amounted to over a thousand Curies each of radium-226, uranium-234 and -
238, and thorium-230, is contained in the slime and sand tailings (Guimond and
Windham, 1975, Table 2).
Radium concentrations are greatest in gypsum wastes associated with
phosphoric acid plants (U.S. Environmental Protection Agency, 1973; Guimond
and Windham, 1975). Whereas seepage into mine pits and recirculated mine
water contained 0.28 to 1.5 pCi/1 dissolved radium-226, gypsum water at four
separate fertilizer plants ranged from about 50 pCi/1 for plants without
recycling and 90 to 100 pCi/1 in those that did (U.S. Environmental Protection
Agency, 1973). By-product gypsum from central Florida acid plants contains 21
to 33 pCi/g radium-226 compared to 45 pCi/g in washer plant slimes and 42
pCi/g in phosphate rock product. Gypsum solubility and high rainfall (50
inches per year) indicate continuing measures will be necessary to control
leachate production and migration from such piles. It is recommended that
techniques developed for moisture control in common sanitary landfills and for
land disposal of toxic wastes, including uranium mill tailings, be investigated
for their applicability to gypsum piles. Additional measures are necessary
for treatment and immobilization of the liquid fractions.
Dissolved radium-226 in slime pond influents and discharges consistently
averages less than 5 pCi/1 and typically less than 2 pCi/1 (U.S. Environmental
Protection Agency, 1973; Guimond and Windham, 1975). Radium in the suspended
fraction of the influents is more variable, ranging from 9.8 to 72.6 pCi/g
(mean 33.5) on a weight basis and 10.2 to 2248 pCi/1 (mean 673) on a volumetric
basis. The latter is highly dependent on the amount of suspended solids in
the slime discharge. In effect, radium content on a per unit weight (of
undissolved solids) basis varies by a factor of seven, whereas the quantity of
undissolved solids varies by a factor of 220. Total radium that potentially
32
-------
could contaminate ground water is therefore quite variable and suspended
solids content is the more significant contributor. Analysis for "dissolved"
versus "undissolved" radium is highly dependent on the pre-analysis methods.
Acidification followed by filtration, wet chemistry, and counting will give a
higher content of dissolved, radium and a lower content of undissolved radium
relative to methods calling for filtration first. More importantly, the
amount of suspended and settleable solids in a given sample can greatly affect
the analytical results. Thus it is apparent that ground-water monitoring for
suspected or actual contamination in the vicinity of a gypsum pond or slime
pond, or monitoring of ground water from mineralized stratigraphic intervals
calls for different procedures than in the case of a public health related
survey involving water supply wells. This point will be developed further in
a subsequent section.
On the basis of available grab sampling data,radium solubility does not
appear to increase as a result of residence time or other conditions in slime
ponds. Dissolved radium in discharges ranged from 0.02 to 2.2 pCi/1
and total radium (dissolved + undissolved, pCi/1) never exceeded 3.0 pCi/1 at
any one facility. Seepage to ground water, however, may contain higher total
radium depending on the transport route (conduit versus intergranular seepage)
and the suspended solids content of the wastewater.
In summary, seepage, overflows, and accidental releases from the various
basins and pits put the wastes in actual or potential contact with adjacent
water resources. Although one can only speculate on the basis of very limited
available data, mining practices probably introduce marked changes in the
chemical and hydrologic stability characterizing the ore body and associated
overburden and leach zone. The effects, if any, on increased leaching of
radionuclide-bearing minerals are unknown. From a ground-water protection
standpoint, intergranular seepage from slime ponds would not be expected to
grossly change radium concentrations in native ground water, particularly in
the unconfined aquifer. However, movement of slimes high in suspended solids
into solution cavities and other secondary permeability features could result
in contamination. Sampling of such contamination using methods whereby acidi-
fication is followed by filtration will give anomalously high results for
33
-------
dissolved radium. Measurement of the suspended solids content and radio-
chemical character thereof as well as dissolved radium should be incorporated
in monitoring studies of slime pond seepage to ascertain the degree of natural
filtering and the prevalence of turbid seepage, if any. Wells completed in
the upper limestone unit and situated at or very near pond perimeters are
recommended.
Field and laboratory studies by Guimond and Windham (1975) of phosphoric
acid plant waste treatment using basic solutions containing quick lime,
hydrated lime, limestone, or dolomite indicate the feasibility of removing at
least 94 percent of soluble radium-226. Use of the double liming procedure to
markedly reduce dissolved radium was also recommended in a previous USEPA
(1973) study. Coprecipitation of radium in calcium-radium sulfate appears to
be the removal mechanism which is facilitated by abundance of calcium, sulfate
and phosphate ions, reduced solubility of radium sulfate at neutral or near-
neutral pH, and settling of precipitates. In this way, process water contain-
ing 60 to 90 pCi/1 of dissolved radium can be treated so as to reduce concen-
trations to between 1 and 3 pCi/1. Neutralization of acidic effluents would
greatly reduce soluble radium content. Whether such effluents actually
dissolve carbonate strata beneath gypsum ponds and thereby increase incidence
of solution cavity collapse/development is unknown. If additional study
confirms such dissolution, there is additional reason for treating the wastes.
SOURCES OF RADIOCHEMICAL DATA
Analytical data for radium-226 in water were primarily obtained from the
following sources: 1) U.S. Environmental Protection Agency, 2) Federal Water
Pollution Control Administration and 3) U.S. Geological Survey These data
are tabulated in Appendices 1, 2, and 4. Radium and gross alpha data were
also obtained from the Southwest Florida Water Management District and from
preciously published reports on radium in Florida and the United States.
Table 2 summarizes the data sources, number of analyses, time of collection,
and how the data are used in this report. Only an approximate tally of
analyses actually used is possible. For example, the 1966 survey by FWPCA
produced data for 105 sampling points, all of which are listed in Appendix 2.
Only about 40 of these had sufficient well information or were properly located
34
-------
TABLE 2. SUMMARY OF PRINCIPAL SOURCES OF RADIUM-226 DATA
SOURCE
USEPA
USGS - Irwin and
Hutchinson, 1976
USGS - Scott and
Barker, 1962
USEPA (for
Sarasota County
Health Department)
FWPCA - Shearer
et al. , 1966
TIME OF
CULLECTION
1973-1975
1974-1976
1974-1976
1954-1957
1954-1957
1954-1957
1975-1976
1966
UTILIZATION IN
PRESENT STUDY
Analyze spatial
variation
Analyze spatial
variation
Analyze spatial
variation
Analyze spatial
variation
Analyze spatial
variation
NUMBER OF
OBSERVATIONS
71
64
5
500
86
35
49
105
PREANALYSIS
TREATMENT 2
A/F
U/F
A/F
U/F
U/F
U/F
A/F
U/F
LOCATION OF DATA
IN REPORT
Appendix 1
Appendix 1
Not tabulated
Mot tabulated
Figure 9
Appendix 4
Appendix 3
AREA COVERED
Primary study area
Primary study area
United States
Atlantic and Gulf Coast Plain
Florida, excluding study
areas, (see Figure 8)
Supplementary study area
in Sarasota County
Primary study area
1. Radium-226 dissolved in water
2. U = unpreserved, A = acidified, F = filtered
3. Desoto, Hillsborough, Hardee, Manatee and
Polk Counties (shown in Figure 1)
-------
relative to the study area to allow use in the statistical analysis. In other
instances, gross alpha or stable chemical data from previous studies were
utilized. These are not included in Table 2.
The FWPCA data were originally developed by Shearer et al. (1966) as part
of a 1966 survey by the Technical Advisory and Investigations Branch, Cincinnati,
Ohio. Radium-226, radon-222, natural uranium, thorium-230, polonium-210,
gross alpha, and gross beta were determined together with gross chemical
analyses on 105 public and private water supply wells in central Florida. EPA
data (category 1 above) were collected from 1973 to 1976 by the Office of
Radiation Programs, in cooperation with county and state health departments,
and analyzed by the Office of Radiation Programs, Eastern Environmental
Radiation Facility, Montgomery, Alabama. Also included in Appendix 1 are
selected USGS data from a 1974 to 1976 survey by Irwin and Hutchinson (1976).
Open file U.S. Geological Survey data from Florida were collected in the
course of a nationwide assessment of radium in ground water by Scott and
Barker (1962). These data were provided by R. Scott (private consultant,
Atascadero, California).
The locations of wells sampled by the USEPA, the USGS, and FWPCA in Polk,
Hillsborough, Hardee, Manatee, and Sarasota Counties are shown in Figures 6,
7, and 13.
36
-------
R21E
R22E
R23E
R24E
R25E
R 26 E
R 27 E
R 28 E
P-12.
V X HAINES CITY
•P-13 LAKELAND
P-49,P-50,P-51,P-52
754-151-45 I^BARTOW
P 47. f \ • • P-52A
'" \P-.55. .p.57
P-61 .',A'P-59'P"58«P-57A
P-60 .\752-150-l
P-62 '
P-81 FORTMEADE
P-74. p-84- PiBZ-»)''p-79
.p.84(.
P-87, P-88 • • p-86
P-85A-
HB-ll.HB-12
. HILLSBOROUGH CO.
MANATEE CO
POLK CO. .P-88F
OBOWLING GREEN
• H-B
FOR LOCATIONS OF WELLS
SEE FIGURE 7.
BOUNDARY OF MINERALIZED AREA;
HACHURES ON MINERALIZED SIDE
OGIC CROSS
SECTION SHOWN IN FIG.3
WELL IDENTIFICATION NUMBER
STEWAHT(1966| OR WILSON
LOCATION OF WELL EP-53-15
(SEE APPENDIX 1)
DS-l 15 MILES SOUTH .
R21E
R22E
R28E
Figure 6. Location of wells sampled in Polk, Hillsborough, and Hardee Counties
37
-------
(A
m
CO
V)
t
(A
in
ro
l-
(A
«O
CO
R 19E
•M-
R20E
JL.
R21 E
MANATEE COUNTY
R22E
M-4*
M-3A.
OPARRISH
• M-2a
>M-3
M-5'
M-9a-M-10»
M-11
M-8
01234
SCALE:MILES
LEGEND
\
A'
M-10
BOUNDARY OF MINERALIZED AREA;
HACHURES ON MINERALIZED SIDE
TRACE OF GEOLOGIC CROSS
SECTION SHOWN IN FIG. 2.
GROSS ALPHA <2.0 pCi/l
LOCATION OF WELL GM10A-1
(SEE APPENDIX 1)
oo
*»
(A
M-15a
8QMYAKKA CITY
.A
FLORIDA
GEOLOGICAL
SURVEY WELL
W-2595
M-18(
00
Ul
(A
CO
0)
CA
R19E
R20E
R21 E
R22E
Figure 7. Location of wells sampled in Manatee County.
38
-------
TECHNIQUES FOR MONITORING RADIUM-226 IN GROUND WATER
MONITORING OBJECTIVES
Two commonly recognized objectives for monitoring radium-226 in ground
water include public health investigations, (i.e.,the radium content relative
to established drinking water standards), and environmental quality studies to
ascertain ambient quality or changes therein, typically as a result of human
activities. Both objectives are quite necessary but unfortunately data
collected for one purpose can be rather inadequate with respect to the other.
Ideally, both needs should be served. Some of the findings and difficulties
encountered using available data in the present study may help improve future
investigations and for this reason are elaborated upon.
SAMPLING POINTS AND METHODS FOR SAMPLING RADIUM-226 IN GROUND WATER
By far most radium-226 analyses of ground water collected in Florida up
to the present time are based on one-time grab samples collected at a well
head or spring orifice, or they are raw or finished water samples from a water
supply system, most of which are supplied by a number of wells pumping into a
common reservoir. Both sampling objectives cited in the previous paragraph
utilize these various sampling points. Obviously, health-related studies are
essentially restricted to wells and springs used for potable supply whereas
ground-water monitoring studies will include and may solely use wells installed
or at least utilized for monitoring purposes. Ultimately, trends in potable
status would also be of concern. Certain characteristics of these sampling
points, sampling methods, or physical attributes of the water sample, per se,
necessarily interrelate to the objectives of a given study and the potential
for reaching definitive conclusion.
Samples collected from public or domestic water supply wells in the study
area are, with rare exception, equipped with pumps and the wells are actively
39
-------
used. Suspended solids are typically quite low and the water samples are
likely to be representative of the aquifer(s) tapped. Because such wells are
commonly completed with relatively large intervals or production zones, the
water sample is a composite of quality in this interval. Wells used solely
for monitoring, particularly shallow ones are generally not equipped with
pumps and sampling is done with a bailer or thief sampler. If a pump is
present, it is used intermittently for sampling purposes only. As a result,
water quality in the borehole may not be representative of the aquifer(s) and
suspended solids are likely to be higher than a well in daily use. Wells used
solely for monitoring are used intermittently, typically are not equipped with
pumps, and are completed with less care than supply wells. As a result they
produce more settleable and suspended solids. Introduction of a bailer or
thief sampler or startup of a pump in a dormant well can easily raise the
level of suspended or settleable solids to concentrations not commonly found
and certainly not acceptable in a potable supply well. Proper sampling of
such wells and subsequent sample handling can markedly influence the analytical
results and more importantly, the importance of the data relative to the study
objectives.
SAMPLE PRESERVATION AND HANDLING
Water samples to be analyzed for radium-226 are usually either acidified
and filtered before analysis or they are simply filtered. Marked differences
in analytical results using the two methods commonly occur, due in large part
to the amount and composition of settleable and suspended solids in the
sample. Acidification immediately after sampling is favored in health-related
surveys because it is reasoned that the water sampled is representative of the
potable supply and that acidification to pH of about two duplicates conditions
in the stomach. Another reason for acidification is to prevent sorption of
radium onto the walls of the sampling container.
Samples collected for hydrologic investigations or as part of surveys to
ascertain ambient and contaminated levels of radium in ground water are
filtered as soon as possible after collection and preferably in the field.
Dissolved radium is defined as that passing through the filter. Acidification
to prevent plating out is optional. The USGS does not favor acidification for
40
-------
naturally occurring isotopes in the uranium and thorium decay series and
believes that other factors such as Eh and oxidation state may be more signifi-
cant, particularly for fission products (Thatcher et al., 1977). Dissolved
radium may be markedly lower with this approach of filtering first and then
acidifying as compared to acidification first, followed by mixing, digestion,
and filtering. For example, in five of the six data pairs shown in Table 4,
the USEPA data means range from 1.06 to 2.88 (mean 2.04) times higher than
USGS data. The significance of this difference' is weakened by the fact that
the samples are not from the same wells. However, splits of three surface
water and two ground-water samples collected by Irwin and Hutchinson (1976),
were analyzed for radium-226 first using filtered raw water and then acidified
raw water (followed by filtering) with the following results:
Dissolved Radium-226, pCi/1
Sample Station Untreated Samples (1) Acidified Samp1es(2)
1. Lithia Springs near Lithia 0.68 ± .07 0.80 ± .08
2. Alafia River at Lithia 0.06 ± .012 0.53 ± .05
3. Peace River near Fort Meade 0.12 ± .024 0.58 ± .006
4. Well - Upper Floridan 0.24 ± .048 20.00 ± 1.4
5. Well - Lower Floridan 0.06 ± .012 0.14 ± .028
Mean for five samples 0.23 4.41 ratio = 0.052
Mean for four samples 0.23 0.51 ratio = 0.45
Two sigma error terms shown include counting and analytical error; estimated as
follows:
<.5 pCi/1 : 20%; 0.5 to 2 pCi/1 : 10%; 2 to 10 pCi/1 : 7% (L. Schroder, USGS,
personal communication, April 11, 1977). Ratio = untreated r acidified.
They concluded that little difference was attributed to sample prepara-
tion but rather to the presence of particulate material which was thought to
be high in Sample 4. From the comparisons presented by Irwin and Hutchinson,
we conclude that dissolved radium-226 in acidified samples is 1.17 to 83 times
higher than in untreated samples, and particulate material can greatly increase
dissolved radium when acid preservative is used. The mean radium-226 content
of the acidified USEPA samples is 2.85 pCi/1 versus 1.84 pCi/1 in filtered but
unacidified samples in the Geological Survey data base (data contained in
Appendix 1). If plating out occurs, samples with no appreciable suspended
solids may show a lesser concentration of dissolved radium when the acidification
41
-------
step is omitted. However, this plating out hypothesis is not well supported
in the literature. It is apparent then that analytical differences do exist
as a function of sediment content and sample preservation. Although differences
are unidirectional, they are rather variable and not subject to simple correc-
tion.
SIGNIFICANCE RELATIVE TO FUTURE STUDIES
The authors conclude that the available radium-226 data base for central
Florida ground water is fragmented because of the different objectives and
sampling/analytical methods incorporated in past studies. Future sampling
efforts focused clearly on ascertaining public health significance of radium
in ground water seem bound to utilize sampling methods involving acidification
followed by filtration. Application of this method, in conjunction with
sampling of monitoring wells using bailers, thief samplers, or very inter-
mittent pumping is likely to generate spurious data for evaluating environ-
mental levels of dissolved radium-226. Acidification of turbid water samples
from public or private potable water systems is also likely to produce higher
concentrations of dissolved radium than if the water is filtered before
acidification. Techniques involving acidification in the field, followed by
filtration in the laboratory, should be reserved for health-related surveys of
water systems that are in use. In terms of settleable or suspended solids,
samples must be representative of the water being consumed. Mere collection
of samples, with little regard as to the amount and composition of suspended
solids, combined with acidification prior to filtering, may produce analytical
results with little technical value for public health or environmental monitor-
ing purposes.
Monitoring studies to determine ambient conditions, temporal, or spatial
trends, contaminant migration, or equilibria between dissolved and suspended
fractions and between these and host strata with respect to radium require
different sample handling procedures. Actual sampling points, be they wells
or springs, can be either representative of a small vertical interval or
volume of the subsurface or they can be points that are expressive of some
average condition within an aquifer or an aquifer system. Depending on the
objectives for a given study, one or both types of sampling points are of
42
-------
value. With respect to filtration and acidification, samples should be
filtered in the field, if at all possible, or at least upon arrival at the
laboratory. Filter papers should be saved for analysis of the activity in the
solids. Acidification after filtering is recommended to prevent or reduce
plating out.
Studies or at least data collection concerning radium in ground water of
central Florida will continue. Misapplication of study objectives to the
types of sampling points available, e.g., public health study objectives and
sample handling applied to monitoring and connector wells, should be studiously
avoided by regulatory agencies and industry alike. There is genuine need to
study both environmental quality and public health in a multiple-aquifer,
multiple land use setting. Cost effective and information-effective data
collection programs are essential lest the next 30 years of mining activity
and related monitoring simply mirror the past shortcomings.
43
-------
RADIUM IN SURFACE AND GROUND WATER
Although it is beyond the scope of this report to give a detailed assess-
ment of radium occurence in surface and ground water throughout the world and
United States, a short discussion is included to put the observations in
Florida and in the specific study area of this report into better perspective.
It should be noted that some of the values refer to a composite of all radium
isotopes whereas in the detailed part of this study only radium-226 is
addressed. However, radium-226 is the most common isotope and is therefore
believed to be a fair approximation of total radium. There are some important
exceptions to this generalization and these will be noted in a later section.
Table 3 is provided to summarize the text that follows.
CONCENTRATIONS IN CONTINENTAL AND OCEANIC WATERS
Radium concentrations in ocean and surface waters tend to be quite low.
Koczy (1958) observed that near bottom water varied from 0.08 pCi/1 in the
Indian Ocean to 0.51 pCi/1 in the Pacific Ocean, with surface ocean water at
about 10~ pCi/1. Tokarev and Shcherbakov (1956) considered 0.1 pCi/1 as the
mean for ocean water, compared to 0.1 pCi/1 in fresh water lakes, and 0.2 pCi/1
in rivers. Miyake et al. (1964) observed 0.08 pCi/1 dissolved radium in
Japanese rivers. Hursh's (1953) tabulation of 42 radium-226 observations made
in water supply systems dominated by surface water as a source showed a
geometric mean (GM) of 0.04 pCi/1. We conclude that surface water values for
radium-226 are less than 0.1 pCi/1.
Ground water is generally higher in radium. Water supply systems in
which excessive radium is present are typically ground-water dominated
(Samuels, 1964). With respect to hydrogeologic factors and their influence on
radium in ground water, Tokarev and Shcherbakov (1956) distinguished between
sedimentary and siliceous igneous rocks and between circulating and stagnant
aquifer systems. In sedimentary rocks, mean radium concentration in a cir-
culating system is 2 pCi/1 as opposed to 300 pCi/1 in a stagnant system.
Where siliceous igneous rocks are predominant, radium varies from 2 pCi/1 in
a circulating system to 4 pCi/1 in a stagnant one.
44
-------
TABLE 3. SUMMARY OF THE OCCURENCE OF DISSOLVED RADIUM IN WATER
MEDIUM
WATER
Ocean
Bottom
Surface
Mean
FRESH WATER LAKES
RIVERS
Japan
POTABLE WATER SUPPLY
Surface Water
GROUND WATER
Sedimentary Rocks
Circulating Water
Stagnant Water
Igneous Siliceous Rocks
Circulating Water
Stagnant Water
UNITED STATES
Conterminous
Atlantic and Gulf
Coastal Plain
Florida - Exclusive of
central P04 District
Central Florida
CONCENTRATION (pCi/1)
0.07*
0.08 - 0.15
.0001
0.1*
1.0*
0.2*
0.08
0.04**
2.0***
300***
2
4
0.1 - 720
0.15**
0.2**
0.2 - 12
1.0
1.0 - 15**
REFERENCE
Koczy, 1958
Koczy, 1958
Tokarev and Shcherbakov, 1956
Tokarev and Shcherbakov, 1956
Tokarev and Shcherbakov, 1956
Miyake et al., 1964
Hursh, 1953
Tokarev and Shcherbakov, 1956
Tokarev and Shcherbakov, 1956
Tokarev and Shcherbakov, 1956
Tokarev and Shcherbakov, 1956
Tokarev and Shcherbakov, 1956
Scott and Barker, 1962
Present study; calculated
from Scott and Barker, 1962
Scott and Barker, 1962
Scott and Barker, 1962
Present study
* Estimated mean
** Geometric mean
*** Arthmetic mean
45
-------
Scott and Barker (1962) observed that radium in ground water in the
conterminous United States ranged from less than 0.1 pCi/1 to 720 pCi/1.
Subdividing the United States into ten regions based on similarity of geologic
physiographic, and ground-water conditions, they found that the geometric mean
of six regions (Atlantic and Gulf Coastal Plain, Appalachian Orogenic Belt,
Eastern Stable Region, Western Stable Region, Rocky Mountain Orogenic Belt,
and Pacific Orogenic Belt) with adequate data varied from <0.1 pCi/1 to
0.6 pCi/1. A log-normal probability plot of over 500 observations of radium
data for ground water throughout the country reveals a geometric mean concen-
tration of 0.15 pCi/1. The Atlantic and Gulf Coastal Plain Region (Plate I,
Scott and Barker, 1962), which includes Florida, has a geometric mean of
0.2 pCi/1 of radium. In general, ground-water values as determined for the
United States have a geometric mean of about an order of magnitude larger than
concentrations in surface waters.
FLORIDA GROUND WATER
Radium-226 data from areas outside the central Florida land pebble
district were observed in 35 locations in 12 counties throughout the state
(Figure 8). The plotted data are from Irwin and Hutchinson (1976) and open
file data of the U.S. Geological Survey (Figure 9). The latter data, largely
developed in the course of the Scott and Barker (1962) study, were supplied by
R. C. Scott (Atascadero, California, written communication). Most of the
Florida data reported by Scott and Barker are from flowing wells and are
therefore not likely to be affected by wastewater or other surface sources of
contamination. No attempt was made to separate the data according to aquifer
or aquifer system. It is assumed that the radium present is from natural
causes. No correlation of radium, with chloride, floride, uranium, or well
depth is apparent.
The geometric mean radium concentration for background dissolved radium
in Florida ground water is 1 pCi/1, with essentially all values falling in the
range of 0.2 to 12 pCi/1. A near linear fit of the data on a log-normal
probability plot suggests that the observations are part of the same popula-
tion.
46
-------
COUNTIES IN WHICH GROUND WATER SAMPLES
ANALYZED FOR RADIUM-226 ARE LOCATED
NUMBER OF RADIUM-226 OBSERVATIONS LOCATED
IN INDICATED COUNTY
OUTLINE OF PRIMARY STUDY AREA
OUTLINE OF SUPPLEMENTARY STUDY AREA
Figure 8. Location of counties used to establish background levels of
radium-226 in Florida ground water.
47
-------
100-
50-
20
o
a
X
tO
10-
6.0
cc
Q
(A
Q
2.0
1.0-
.5
.2
GEOMETRIC MEAN= 1 pCi/l
GEOMETRIC
STANDARD DEVIATION = 4.4
2% 5 10 15 20 30 40 50 60 70 80 85 90 95 98%
PERCENTAGE
Figure 9. Log-probability plot of background levels of radium-226
in Florida ground water.
48
-------
Assuming these data are reasonably representative of the radium-226 in
Florida ground water exclusive of the central Florida land pebble phosphate
district study area, the geometric mean is about one pCi/1 or about an order
of magnitude higher than the national geometric mean of 0.15 pCi/1, calculated
from radium data presented by Scott and Barker (1962). Arithmetic and geo-
metric means for radium in ground water for the study area in central Florida
are higher yet, with averages typically ranging from 1.5 to 15 pCi/1. Since
portions of Florida contain economic deposits of monazite sands which are
enriched in thorium, radium-228 may also be elevated in associated ground
water. Unfortunately, no radium-228 data are available for Florida ground
water. In summary, there is good evidence that Florida ground water has
naturally higher radium-226 levels than either the nation as a whole or the
Atlantic and Gulf Coastal Plain region. With this introduction established,
we will now focus on radium in ground water of central Florida.
49
-------
WATER QUALITY EFFECTS OF PHOSPHATE MINERALIZATION AND THE PHOSPHATE INDUSTRY
As previously mentioned, data bases developed by FWPCA, USERA, and USGS
are available. Because of differences in analytical procedures, or time of
sampling, the data bases were treated separately to analyze for spatial and
temporal changes in ground-water quality. With respect to spatial variation,
data availability as determined by aquifer and mineralization or mining
status resulted in decidedly uneven coverage in certain categories. In order
to make fullest use of the available data for 1973-1976 and also maintain
analytical comparability, the data group with maximum observations, either
USEPA or USGS, was selected for any given comparison of water quality according
to aquifer, mining/nonmining, and mineralization criteria.
Three separate aquifers (water table, Upper Floridan, Lower Floridan)
were considered for three separate land status categories (mined, unmined, and
nonmineralized). Aquifer information was obtained by comparing well depth and
casing information relative to hydrogeologic conditions, and from information
presented in Irwin and Hutchinson (1976). Land use information was obtained
from existing topographic maps, geologic reports, and mineral leasing records
(W. Lancaster, Texas Instruments, Inc., written communication, April 1, 1977;
Wayne Thomas, Inc., 1976). Aquifer and land use information for each sample
is presented in Appendix 1.
STATISTICAL METHODOLOGY
Basic approaches taken to analyze the data include characterization of
distributions, central tendency (mean, geometric mean), and variance (geometric
standard deviation, standard deviation). With this information, nonparametric
tests were selected to compare observations for significant difference as a
function of time (1966 versus 19/4-1976), space (mineralized, nonmineralized,
etc. ), and depth (aquifer). Primarily because of the limited number of
observations and therefore uncertainty concerning population distribution,
nonparametric methods were used to analyze and compare variance and central
tendency.
Mineralized areas are defined as those within the pebble area (i.e., 55
percent or more BPL) of phosphate mineralization as mapped by Mansfield
(1942, Plate 5). Locations of mined areas and plant sites appear on a
detailed map by Wayne Thomas, Inc. (1976). Wells located in such areas
or within one mile or less thereof are classified in the "mined" category.
50
-------
Statistical analysis techniques often assume that the population or
grouping of observations has a distribution which is known. If data are
highly skewed with numerous small observations and several large ones, it may
have a log-normal distribution (Koch and Link, 1970), i.e., the logarithms of
the observations have a normal distribution. Trace elements, for example, are
log-normally distributed (Ahrens, 1957) as are many types of environmental
data (Denham and Waite, 1975). Therefore, it would be reasonable to expect
trace amounts of radium-226 in ground water to be approximated by the log-
normal distribution and a plot of such data is expected to have a
linear appearance on log probability paper. The geometric mean equals the
50th cumulative percentile line, and the geometric standard deviation is
defined by the line slope. At a geometric standard deviation of about 1.3 or
less, skewness is sufficiently small that the population might be treated as
normal without transforming the logarithms. Failure of a set of observations
to approximate a log-normal distribution can be inferred using the Kolmogorov-
Smirnov goodness-of-fit test. Denham and Waite (1975) suggest that when the
plotted data have a linear trend, a single and presumably background popula-
tion is depicted. Several line segments on a single log-normal probability
plot suggest several overlapping populations, the distribution of which is
unknown when few observations define a given line segment.
Log-normal probability plots of radium-226 data were extensively used to
characterize several data sets in terms of geometric mean and geometric stan-
dard deviation for each aquifer or land category. In order to test possible
relationships between the various groups of data which generally were not from
a single, clearly defined distribution, "distribution-free" techniques of
hypothesis testing were used. Two nonparametric statistical tests were used,
the Mann-Whitney U test and the Kruskal-Wallis one-way analysis of variance
(by ranks) test. The Mann-Whitney U test is a nonparametric analog of the
student t test and can be used to determine if two independent groups were
selected from the same population. A variant of the Mann-Whitney test (Springer,
1976) is used to validate a basic assumption of the test, i.e., whether both
data sets are similarly dispersed. The Kruskal-Wallis test is a nonparametric
analog to the parametric one-way analysis of variance or F test used to test
dependency between three or more independent samples. A logical extension of
51
-------
the Kruskal-Wallis test, particularly when the ruling hypothesis is rejected,
is to determine which groups are significantly different from one another.
Gibbons (1976) suggests that simultaneous multiple comparisons is the best
method for this, particularly as described in Dunn (1964). Since several
simultaneous statements of difference are made, the level of significance used
is larger so that a single difference is more likely to be detected. It is
also quite possible that several groups may not be equal as indicated by the
Kruskal-Wallis test but that simultaneous multiple comparisons may not find
any pair with significant differences. Siege! 0956) and Gibbons (1976) give
excellent examples and explanations on nonparametric statistics and should be
consulted by those interested in exploring further into this subject.
SPATIAL VARIATIONS IN WATER QUALITY
Table 4 summarizes the number of observations by source (USEPA, USGS),
range, mean, and standard deviation for the land categories and aquifers
considered. The USGS data were used to analyze the water table aquifer and
Upper Floridan aquifer and USEPA data were used for the Lower Floridan
aquifer. Temporal changes in water quality are discussed in a subsequent
section.
Water Table Aquifer
Unfortunately, radium-226 data for the water table aquifer is restricted to
mineralized areas with no information available in the nonmineralized portions
of the study area (Table 4). For 23 observations in the mineralized unmined
area the geometric mean (GM) is 0.17 pCi/1 and the geometric standard deviation
12.9. The observations show pronounced linearity on log probability paper
(Figure 10), suggesting a log-normal population. This, however, was rejected
by the Kolmogorov-Smirnov goodness-of-fit test at the 95 percent confidence
level. Three of the 23 observations exceed the 5 pCi/1 combined radium-226
and radium-228 standard for community water systems (U.S. Environmental
Protection Agency, 1976a).
52
-------
TABLE 4. SUMMARY OF AVAILABLE RADIUM-226 DATA AND STATISTICS
AQUIFER
1973-1976 EPA DATA
\
en
CO
1974-1976 USGS DATA
DATA SET
USED IN ANALYSIS
LOWER FLORIDAN
COMBINED
EPA DATA
-------
100-
50-
20
10-
o
a
to
3 5.0
D
O
£ 2.0
O
(0
1.0-
.5
WELLS LOCATED IN UNMINED AREAS (*
WELLS LOCATED IN MINED AREAS •
./ * V
2% 5 10 15 20 30 40 50 60 70 80 85 90 95 98%
PERCENTAGE
Figure 10. Log-probability plots of USGS data for the water table
aquifer in unmined and mined mineralized areas.
54
-------
For 12 observations in the mined mineralized area, the GM is 0.55 and GSD
is 3.3, indicating a population which is much less skewed than in the mineral-
ized unmined area (Figure 10) but is similar in having a strong linear trend.
That radium in mined, mineralized areas is log-normally distributed can not be
rejected by the Kolomogorov-Smirnov goodness-of-fit test at the 95 percent
confidence level. None of the observations exceed 5 pCi/1 for radium-226. No
radium-228 data are available to confirm whether combined radium-226 plus -228
exceeds the 5 pCi/1 limit established by USEPA for public water supplies.
In order to determine whether these two data sets are in fact from the
same population, a "distribution-free" method of comparison was used. For
testing whether two groups are equally likely to have been taken from the same
population, the nonparametric Mann-Whitney U test is quite effective (Siege!,
1956). Using a method described by Springer (1976), the observations in both
data sets were first tested for similar dispersion, a prerequisite assumption
for use of the Mann-Whitney test. Dispersions or variances are similar at the
95 percent confidence level. The test value or "Z" statistic generated by the
Mann-Whitney test was 1.01. This value is too small to reject the null
(ruling) hypothesis at the 95 percent confidence level. Therefore, the two
samples appear to have been taken from the same population. Based on these
data and the analysis described, no impact on the water table aquifer appears
to have occurred as a result of mining.
Upper Floridan Aquifer
As the USGS data base has observations in all three land classification
groups, it was selected to characterize radium-226 in the Upper Floridan
aquifer. All groups had 10 or less observations and no attempt was made to
plot on log probability paper. Statistical parameters used to express central
tendency and standard deviation are arithmetic (Table 4). In areas of miner-
alization without mining, one out of the five observations is greater than
5 pCi/1. In areas subject to the impact of mining, one out of ten observa-
tions is greater than 5 pCi/1. Of the three observations in the mineralized
55
-------
area, two are greater than 5 pCi/1. To test whether these three groups are
part of the same population, the Kruskal-Wallis one-way analysis of variance
test was used, yielding a test value for H of 52.37 . The probability of H
this large due to random chance is less than 0.001, inferring that the three
groups are not equal. Using simultaneous multiple comparisons, at the 80
percent confidence level, only the mined versus nonmineralized groups are
significantly different whereas for the nonmineralized area, mean radium-226
is greater than that in mined areas. Based on these data and analysis, no
impact on the Upper Floridan aquifer seems to have occurred as a result of
mining and related waste management.
Lower Floridan Aquifer
The USEPA data base was used to characterize the radium-226 in the Lower
Floridan aquifer. Two groups, mineralized unmined and unmineralized, have the
greatest number of observations, 24 and 14 (Table 3). A log-normal probability
plot of dissolved radium-226 in the mineralized unmined area exhibit three
distinct line segments. This pattern remains if the six observations from the
mined area are included (Figure 11). They were included because they are
evenly distributed within the range of values for the mineralized unmined
area. A very similar alignment of line segments also characterizes the Lower
Floridan in areas of nonmineralization (Figure 11). This suggests that radium
in the Lower Floridan aquifer is not related to phosphate mineralization.
Although only 4 of the 24 observations in the unmined, mineralized areas
exceed 5 pCi/1 for radium-226, other samples with lesser concentrations might
also exceed the USEPA standard of 5 pCi/1 radium-226 plus radium-228. In the
mined area none of the six samples contained over 5 pCi/1 radium-226 whereas
in the nonmineralized area, one well of 14 sampled exceeded 5 pCi/1. Again,
radium-228 data are unavailable. On the basis of available radium-226 data,
phosphate mining and processing do not appear to have deteriorated the quality
of water in the Lower Floridan.
The visual similarity in the plots for data from mineralized and non-
mineralized areas is inferred by the Mann-Whitney U test, which has a test "Z"
value of 0.08. This has an associated probability of 0.47, clearly much
56
-------
100
50n
20
10-
o
a
to
5
D
O
<
DC
Q
O
W
2.0
1.0-
.5-
.2
WELLS LOCATED IN MINERALIZED AREAS (•)•
(MINED & UNMINED)
WELLS LOCATED IN
NONMINERALIZED AREAS (*|
2% 5
Figure 11
10 15 20 30 40 50 60 70 80 85 90 95
PERCENTAGE
98%
Log-probability plot of USEPA data for the Lower Floridan
aquifer in mined and unmined mineral ized areas and in
nonmineralized areas.
57
-------
greater than the pre-set alpha of 0.05. Therefore, the ruling hypothesis is
accepted, i.e., there is no significant difference in radium content of ground
water in mineralized versus nonmineralized areas.
Plots of dissolved radium-226 data for the Lower Floridan aquifer suggest
the presence of three separate populations. Grouping all the radium-226 into
a single log probability plot and using graphical techniques described by
Sinclair (1974), the curve was partitioned into three parent populations as
shown in Figure 12. The three populations were combined in proportion to
their presence to generate the observed plot. Population B is dominant. It
represents 52 percent of the total observations and has a GM of about 3 pCi/1.
About 10 percent of population B observations are suggested to exceed the 5
pCi/1 limit. Population A is second most abundant, contributing 37 percent of
the observed radium-226 values. With a GM of about 0.7 pCi/1 , this population
will have on the order of one percent of its observations exceeding the 5
pCi/1 limit. Population C, representing about 11 percent of the total observa-
tions, has a GM of 10 pCi/1 and has about ninety percent of its values above
the 5 pCi/1 level. All of population C observations were taken outside the
areas defined to be influenced by mining.
Clearly the Lower Floridan aquifer is a complex system with radium-226
distribution dependent on several processes and/or sources. Much more detailed
information on water chemistry, lithologic composition, and hydrology would be
required in order to attempt to define the underlying reason for these sug-
gested populations. Modification in the chemical and physical processes in
the Lower Floridan aquifer so as to enhance the content of dissolved radium to
the level typified by population C would be highly undesirable. We conclude
there are occasional high radium-226 observations in ground water from the
Lower Floridan aquifer associated with natural factors essentially unrelated
to phosphate mineralization or the central Florida phosphate industry.
Elevated levels of radium-226 in the Lower Floridan aquifer can be due to
a number of natural factors unrelated to phosphate mineralization or waste
management in the phosphate industry. Increased solubility of radium occurs
in ground water enriched in chloride (Tanner, 1964). Kaufman and Dion (1967)
have shown that upwelling of mineralized water occurs along the trace of the
58
-------
100-
50
20
10-
Population (C)
o
a.
i>
-------
southern Peace River. Widespread, pronounced reduction of the potentiometric
surface for the Lower Floridan aquifer is well documented and problems of
water supply and water quality are receiving increasing attention in central
Florida. Sea water encroachment and upwelling of deeper more mineralized
ground water in cones of depression due to pumping from the Lower Floridan
aquifer, particularly the Avon Park Limestone, are of particular concern.
Whether heavy pumping of the Floridan aquifer will actually induce upward
movement of more saline, possibly radium-enriched water is an unanswered
question and probably one which will require much additional data and in-
terpretation.
Dissolved gross alpha radioactivity values for 10 wells in Manatee County,
9 of which are for domestic use, are less than 2 pCi/1 in all cases (See
Figure 7). The wells tap the Floridan aquifer which is undifferentiated as to
Lower and Upper because of the absence of a prominent aquitard at the base of
the Tampa Formation in this area (Wilson, 1975). Because the correlation
between dissolved radium and gross alpha is rather inconsistent, particularly
at low concentrations and because dissolved radon accounts for most of the
alpha activity, the Southwest Florida Water Management District has gradually
increased the gross alpha threshold from 2 to 5 pCi/1 and then from 5 to 15
pCi/1 as the basis for requiring analysis for radium-226 (B. Boatwright,
SWFWMD, personal communication, May 31, 1977). That is, below 15 pCi/1 gross
alpha, radium is believed to be present in acceptable levels of 5 pCi/1 or
less. For this reason, values of 2 pCi/1 gross alpha in the Manatee County
wells for which only gross alpha data are reported on herein, are believed to
contain very low concentrations of radium-226.
Mean radium content in seven Manatee County wells tapping the Floridan
aquifer was 4.52 pCi/1 versus 1.23 pCi/1 from three water table wells. Radium
varied from 0.11 to 3.7 (mean 1.94) in six wells of unknown depth or produc-
tion interval. In general, radium increases with depth. There was no phosphate
mining at the time of sampling; hence, this source is discounted and natural
factors other than the phosphate industry are believed responsible for high
radium values observed. This is reinforced by the fact that half of the wells
with values exceeding 3 pCi/1 are located in areas not considered ore bearing.
60
-------
In summary, the statistical analysis techniques utilizing available data
and the conclusions therefrom are probabilistic in nature. There is neither
"proof", per se, for the conclusions herein nor can they be considered final.
Qualification of the available data is a necessary procedure designed to put
the interpretations into proper context but without necessarily discrediting
the conclusions. The exact extent to which phosphate mineralization causes
elevated radium-226 in the three aquifers considered is not readily discern-
ible. This is due in part to lack of observations in the water table in
nonmineralized areas and the availability of only three observations for the
Upper Floridan aquifer for the same areas. Theoretically, phosphate minerali-
zation should be a radium-226 source for the Upper Floridan aquifer due to the
intimate association of aquifer and ore body. However, there are too few data
to make meaningful comparisons on the basis of land use or mineralization.
Under natural conditions the water table aquifer is perhaps not as
closely associated with phosphate mineralization as is the Upper Floridan.
However, disturbance by phosphate mining directly disrupts the water table
aquifer. Overburden and wastes associated with benefaction and chemical
processing have the potential for introducing radium into both aquifers. Yet,
radium-226 data suggest that mining has not appreciably changed radium-226
levels in the water table. Again, this conclusion is probabilistic, being
based on statistical analysis of available data. To presume no change is
occurring may well be erroneous. Ground-water dynamics are greatly affected
by mining, as is the lithological sequence, the matrix and leach zone portions
of which have a potential for contributing radium-226 to ground water. It is
difficult to believe that change is not underway.
On the basis of the data presented herein, radium-226 in the Lower
Floridan aquifer appears to bear no relation to the effects of land use or the
presence/absence of phosphate mineralization. The distribution pattern of
radium data on log-probability plots suggests a complex system and possibly
multiple populations, probably stimulated by natural factors unrelated to
phosphate deposits or the industry. In general, radium-226 levels below
mineralized areas are low. Possibly radium-226 is depleted as a result of
61
-------
mineralized areas are low. Possibly radium-226 is depleted as a result of
dilution and sorption in overlying strata and aquifers. Although the data are
limited and numerous exceptions are present, radium-226 generally increases
with salinity which, in turn, varies with well depth and position in the flow
regime within the Floridan aquifer system. Radium-226 concentrations of 10-20
pCi/1 may be associated with more mineralized or brackish ground water in the
deeper parts of the aquifer. Whether altered flow patterns are present due to
pumping and whether such patterns can induce upwelling of water with higher
radium content are unanswered questions.
The existing radium-226 data base is marginal in terms of number and
spatial distribution of analyses, particularly for the water table and Upper
Floridan aquifers. Aquifer system complexity and widespread changes in land
use require a greatly increased sampling program if these variable are to be
considered. Inconsistency in study objectives and analytical procedures and
lack of overlap between the USGS and USEPA data bases on a well-by-well basis,
or even by aquifer, restrict reaching precise definition of spatial changes in
radium-226 content of ground water.
SARASOTA COUNTY
Because of its location at the extreme southern end of the Central
Floridan phosphate district and closeness to the discharge portion of regional
ground-water flow within the Floridan aquifer system, Sarasota County radium-
226 data are considered separately. Radium-226 data for untreated water from
public and private wells are shown in Appendix 4. Locations are shown on
Figure 13. Most wells are completed in both the water table aquifer and Flori-
dan aquifers. Clear distinction between Upper and Lower Floridan aquifers on
the basis of available stratigraphic information was not considered feasible
for Sarasota County. All 12 observations in the water table aquifer have radium-
226 concentrations greater than 5 pCi/1 with a geometric mean of about 15 pCi/1
(see Figure 14). In comparison, the Floridan aquifer has generally lower
radium-226 concentrations, although 70 percent of the observations also exceed
the 5 pCi/1 level. Geometric mean radium content is about 7.5 pCi/1. Radium
in the water table and Floridan aquifers probably represents two independent
62
-------
R 17 E
R 19 E
•S-34
GULF OF MEXICO
• S-5
S-16 S-17
lS-18l*S-19
S-46V* *S'38
s-2oYV7s-41
S-35
V
01234
SCALE: MILES
LOCATION OF WELL (SEE APPENDIX 4)
BOUNDARY OF MINERALIZED AREA
HACHURES ON MINERALIZED SIDE
•S-11
©VENICE
IS-28
S-35*
S-31*
• S-44
• S-1
R 20 E
•S-39
S-21. .S-22
S-23* ?S-24
S-25 * S-30
R 21 E
R22E
SARASOTA COUNTY
»S-9
co
01
00
• S-33
Figure 13. Location of wells sampled for radium-226 in Sarasota County.
63
-------
1000-
500-
200
100-
o
a
to
gj 50H
5
D
a
a
UJ 20
o
CO
to
10-
5.0-
2.0
1.0-
2%
WELLS LOCATED IN
WATER TABLE AQUIFER)*
WELLS LOCATED IN
FLORIDAN AQUIFER
10 15 20
30 40 50 60 70
PERCENTAGE
Figure 14.
Log-probability plot of radium-226 in the water table
and Floridan aquifers, Sarasota County.
64
80 85 90 95 98%
-------
populations. After testing the data sets for similar dispersion, the Mann-
Whitney test produced a test value for "Z" of 3.57 with associated probability
much less than the pre-set alpha of 0.05. This indicates that the two groups
are probably significantly different and that the water table aquifer has
significantly greater radium-226 than the Floridan.
Most of the data for the water table aquifer occur in two areas:
1) within a mile of the Gulf Coast, midway between Venice and Sarasota, and 2)
in the area centered in the Myakka Rtver State Park in the northeast part of
the county adjacent to Manatee County (Figure 15). The coastal wells have
significantly greater radium-226 (mean = 20.8 pCi/1) than inland wells (mean =
13.6 pCi/1) on the basis of the Mann-Whitney test at the 95 percent confidence
level. Observations of radium-226 in the Floridan aquifer are largely confined
to a mile-wide strip running parallel to the general trend of the coast.
Isopleths of radium-226 on the basis of a 5 pCi/1 contour interval (Figure 16)
suggest that ground water in areas immediately adjacent and within about one-
eighth mile of the coast contains 5 pCi/1 or less radium-226. Paralleling
this zone is a second one wherein radium exceeds 15 pCi/1. The inferred
radium-226 high appears to trend eastward and away from the city of Sarasota
in the northern part of the county, possibly extending into south-central
Manatee County.
Inland of the coastal zone just described there are insufficient data to
characterize radium in the Floridan aquifer. As shown in Figure 16, levels
east of Venice and southeast of Sarasota are 5 pCi/1 or less but may well be
higher in the area between, as well as in the southern tip of the county.
There are no data for the eastern two-thirds of the county.
Clearly radium-226 in ground water in Sarasota County is considerably
above levels observed in Polk County and surrounding counties of the primary
study area. This is true for the water table and Floridan aquifers. The
Hawthorn Formation, the principal phosphate-bearing formation in the area,
crops out in the northern part of the county and extends toward the coast
(Mansfield, 1942), where it again crops out or is within 20 feet of the land
surface (Sarasota County Health Department, 1976). In terms of phosphate con-
tent, the Hawthorn is quite low, relative to the Polk County and surrounding
65
-------
R 17 E
MANATEE CO.
R18E | R19E
R20E
GULF OF MEXICO
O
R 21 E
R 22 E
N
01234
SCALE MILES
WATER TABLE AQUIFER
VALUES ARE RADIUM-226
IN pCi/l
APPROXIMATE LOCATION OF
CITIES INDICATED
BOUNDARY OF MINERALIZED AREA;
HACHURES ON MINERALIZED SIDE
SARASOTA COUNTY
CHARLOTTE CO.
C/9
CO
CO
O
O
O
LU
O
Figure 15. Location of radium-226 observations in the water table
aquifer in Sarasota County.
66
-------
MANATEE CO.
R 19 E,
FLORIDAN AQUIFER
CONTOUR OF RADIUM-226
CONTOUR INTERVAL = 5 pCi/l
El <5 pCi/l
i pCi/l
R 21 E
R 22 E
SARASOTA COUNTY
CHARLOTTE CO.
|(0
[co
6
O
O
O
c/o
HI
w Q
o>
Figure 16. Contour map of radium-226 in theFloridan aquifer
in Sarasota County.
67
-------
Another potential radium source in Sarasota County is monazite sand.
Overstreet (1967) noted large amounts of localized heavy mineral sands in
beaches near Venice in Sarasota County. Analysis of these sands (Florida
Department of Health and Rehabilitative Services, 1976) as a part of a study
of radiation in Sarasota County revealed 0.05 to 1.5 percent monazite, a
cerium-lanthanum phosphate containing minor thorium-232, the parent of radium-
228. Additional analysis for radium-228 in Sarasota County ground water is
strongly recommended, particularly in the water table aquifer.
Why does Sarasota County ground water have such high radium-226 levels?
A possible answer^may be suggested by exploring some facts about the general
chemical status of the Floridan aquifer, radium geochemistry, and regional
ground-water dynamics. First, what are some of the chemical attributes of the
Floridan aquifer? Back (1969) lumped the Floridan aquifers into a single
system and observed that the system is characterized by low total dissolved
solids near the piezometric high in Polk County, with an increase in all
directions toward the coasts and Lake Okeechobee. In fact, all fresh water in
Florida is underlain by saline water which ranges in depth from 700 meters in
the central part of the peninsula to near sea level along some shorelines.
Ground water becomes increasingly enriched in chloride, sulfate, calcium, and
magnesium as the coastal areas are approached. Back also noted that the
increase in these constitutents is not simply due to mixing fresh water and
ocean water as there is too much calcium and carbonate and too little sodium.
Increasing amounts of limestone are going into solution with the increased
salinity. Similar increases in concentration are observed for magnesium and
sulfate.
How these factors influence radium-226 movement is evident from the fact
that radium is an alkaline earth with behavior similar to that of calcium,
strontium, and barium. However it is somewhat less mobile. Radium solubility
in solutions is enhanced at high and low values of pH. Dissolution of radium
is most enhanced by common cations already in solution. Of these Na is most
+ 2+
important, followed by K and Ca . Tanner (1964) observed that chloride-rich
waters are either enriched in radium or have greater capability for leaching
radium.
68
-------
It is quite possible that increased concentrations of dissolved solids
and calcium in particular, in the ploridan aquifer could help mobilize radium
in the coastal areas. Secondly, the Hawthorn, though not normally viewed as a
major component of phosphate reserves in the specific study area, apparently
contains sufficient phosphate and associated radionuclides that chemi-
cally active ground water can leach out radium-226. Correlation of radium-
226 with undissolved solids is modest (r =0.59) in Sarasota County ground
water. However, the correlation of radium-226 concentrations with total
dissolved solids is quite strong (r =0-82, Figure 17), tending to support a
cause-effect relationship between dissolved solids and radium. Regression
analysis of available Sarasota County data suggests that for every additional
100 mg/1 dissolved solids, radium-226 increases about 1 pCi/1.
Several points should be considered concerning the influence of ground
water movement on radium content of Sarasota County ground water. The county
is clearly an area of discharge as viewed regionally. Ground-water development
is relatively light and flow directions are expected to be upward and toward
the coast. One available analysis of radium from a well completed deep in the
Floridan aquifer contains 21.7 pCi/1 radium and is high in dissolved solids.
This possibly suggests that radium-226 levels in ground water can also be high
without direct leaching of the Hawthorn Formation. Various speculations
concerning the source of radium in deep, mineralized ground water have been
offered. Leaching of crystalline basement rocks is favored but the mechanics
are unknown. It is reasonable to assume that shallow, phosphatic sediments
are not the source of radium in deep ground water.
Utilization of ground water from the Floridan aquifer for domestic and
municipal water supplies in Sarasota County is hampered by the high total
dissolved solids. To what extent radium exceeds 5 pCi/1 and is therefore also
a limitation is unknown and worthy of additional data collection. If radium
concentrations are at or near the 5 pCi/1 limit in those areas of the county
where no data are now available, extensive and rather refined monitoring is
recommended to insure that ground-water quality is preserved and possibly
enhanced as a result of any major land or water use activities and particularly
phosphate mining and chemical processing.
69
-------
2500-
RADIUM-226 = 0.01 (TDS)-0.96
CORP. COEFF. = 0.82
2000-
\
ro
E
1500-
o
CO
Q
UJ
_i
O
CO
CO
Q
1000-
500-
5 10 15
DISSOLVED RADIUM-226 (pCi/l)
Figure 17. Plot of dissolved solids versus radium-226
in Sarasota County ground water.
70
-------
TEMPORAL CHANGES IN WATER QUALITY
Wells sampled in both the 1966 survey and again in 1974-1976 (USGS data
for unacidified samples only; Appendix 1) were checked to ascertain if temporal
changes in water quality could be identified for the three principal aquifers.
The USGS data for unacidified, filtered samples are considered more comparable
to the 1966 FWPCA survey data developed from samples which were filtered and
then acidified. By comparison, the USEPA data shown in Appendix 1 are from
samples which were acidified, mixed, allowed to stand for dissolution of
solids, and then filtered. This is believed to produce generally higher
values for dissolved radium.
Efforts to match identical or similar (in terms of location and producing
aquifer) wells using the 1966 FWPCA survey and the 1974-1976 USGS survey were
disappointing in that only two data pairs involving similar wells could be
identified. Data interpretation of significance for such a small sample set
is deemed of no significance. Comparison of the 1966 survey with the 1973-
1976 USEPA data base revealed eight paired sets involving 18 wells. Only 2 of
the 8 pairs involved identical wells. Unfortunately, the USEPA data are not
considered comparable to the FWPCA data because of differences in pre-analysis
procedures.
For the area considered to be mineralized (see Figures 6 and 7), analyses
were selected from the 1966 survey and from the 1974-1976 USGS survey. Using
the Kruskal-Wallis test (Siegel, 1956), the six data sets as defined by three
aquifers and two sampling periods (see Table 5) were evaluated to determine
whether they were homogeneous. A test value "H" of 291.35 strongly suggests
that the six data sets are not all equal. Is this inequality due to differ-
ences exhibited for an aquifer between the two periods of sampling? Using the
simultaneous multiple comparisons test, (Gibson, 1976) between data pairs
defined by time in each aquifer, no significance was determined at the 80
percent confidence level (Table 5). That is , radium in ground water is not
significantly different in 1974-1976 as, compared to 1966.
At best, such testing and resultant findings should be considered as
indicative rather than conclusive in that only two data sets based on one-time
71
-------
grab samples are available and these are not from the same wells nor are the
analyses by the same laboratory or with the same technique. Particularly
lacking are time series data showing the variability in radium content for a
given well or series of wells. On the basis of monthly sampling for 1975,
Keefer (in preparation) has shown radium in the two municipal wells in the
Afton, Oklahoma area to vary about 4 pCi/1 annually. Similar data to show
variability in central Florida ground water are lacking but if similar vari-
ability is present, temporal comparisons of ground water using the approaches
demonstrated herein may be in error.
No concerted effort has been made to develop time series data for any
aquifer, a deficiency which is most serious with respect to the water table
and Upper Floridan aquifers commonly utilized for single family water supplies.
Data collection has been inconsistent with respect to well locations and pre-
analysis procedures. Data concerning the extent of mineralization and mining
are also a source of error. Finally, subdividing an initially limited data
base according to aquifers is a logical approach but one which reduces the
degrees of freedom. Given this, subtle change in quality might well not be
statistically significant and a much larger data base in terms of the number
of sampling points and replicate sampling may be necessary to establish environ
mental quality trends. The pronounced paucity of data for the water table (36
analyses) and Upper Floridan (36 analyses),the aquifers most likely to be
affected, substantiates the need for a greatly increased data base if meaning-
ful comparisons of temporal change in quality are expected.
TABLE 5. COMPARISON OF 1966 AND 1974-1976 RADIUM-226 DATA FOR THE MINERALIZED
AREA IN POLK, HARDEE, MANATEE AND HILLSBOROUGH COUNTIES*
Significant
(S) or Mo
Aquifer
Considered
Water table
U.
L.
Floridan
Floridan
No. of
1966
6
22
15
Observations
1974-1976
30
14
10
16
2
2
1 9 6
Mean
**
.9
.07
.03
S
29
1
1
6
.D.
.8
.93
.52
1974-1976
Mean
1.9
1.99
2.46
S.D.
4.17
2.35
4.26
Significant
Change
NS
NS
NS
(NS)
* Simultaneous multiple comparisons test, a = .20
** Discounting a maximum value of 76.0 pCi/1, the arithmetic mean is 5.1 pCi/1
and no significant difference is indicated
72
-------
LOCAL CONTAMINATION
Despite a consistent regional or areal pattern of apparent non-degradation
of any of the three principal aquifers in or adjacent to zones of phosphate
mineralization, local contamination of the water table and probably the Upper
Floridan aquifers is probable but the frequency of occurrence and significance
are unknown. "Local" is loosely defined as the general area of a given mine,
slime, pond, processing plant, etc. and extending at most for one or two miles
in any direction and 100 to 200 feet depth. Seepage from gypsum ponds and
slime ponds has occurred both as percolation through berms or dams and as
sudden collapse of pond substrates due to sinkhole development. Sinkhole
collapse and similar acute incidents, although dramatic, are uncommon or at
least not well documented in regulatory agency files. Only a few actual cases
of contamination can be cited.. This might indicate that no serious problems
exist or it may mean that data collection to date is inadequate relative to
hydrogeologic conditions and to land and water use patterns characteristic of
central Florida. Obviously the issue is rather subjective but considering the
magnitude, duration, and areal extent of the phosphate industry in central
Florida, the near absence of historical radiochemical monitoring data and
interpretive studies is regarded as a shortcoming. This lack has been docu-
mented herein. At present and for the foreseeable future, increased monitoring
requirements are being levied on industry and this "is laudable. It is equally
critical that responsible local, regional, and state agencies have the capabil-
ity to review and react to the data provided and supply additional independent
assessment as necessary.
Industry maintains that the substrate beneath gypsum ponds are self
sealing due to precipitation of insoluble minerals, particularly calcium
fluorapatite, by neutralization of acidic effluents coming in contact with
bicarbonate ground water and carbonate-rich host rocks. Opponents to such
waste management practices regard such precipitation and self sealing as
alleged, at best, and that unless ponds are sealed when first installed,
seepage results. Effluent and gypsum associated with gypsum wastes respec-
tively contain about 91 pCi/1 and 20 pCi/g radium-226 (U.S. Environmental
Protection Agency, 1973). It is reasonable to expect that wherever highly
permeable limestone strata of the Hawthorn Formation are present in mined-out
73
-------
pits used for disposal of slimes or gypsum wastes, seepage is likely to
occur. Whether such seepage has associated with it sufficient precipitation
to affect sealing of solution channels, fractures, and other secondary
permeability features is unanswered. By comparison, precipitation and self-
sealing of earthen berms around gypsum ponds does not occur despite intergranular
porosity and permeability (John Barnett, Department of Environmental Regulation,
personal communication, April 13, 1977; remote sensing studies by Coker, 1971,
1972). Proof of contamination is somewhat speculative or hypothe-
tical except in a few actual, documented cases. The writers conclude that
additional field study of gypsum ponds is necessary with key emphasis on
ground-water quality monitoring and development of reliable water budgets for
representative ponds. This would, materially assist in ending the speculation
as to whether seepage is occurring.
Some consensus exists that contamination of the Floridan aquifer is
local at most and that tracing of contamination from known sources is
difficult to impossible (G. Parker, Geraghty and Miller, personal communication,
February 25, 1977 and B. Boatwright, SWFWMD, personal communication,
March 30, 1976). In part, this is attributable to two pronounced but diverse
hydrogeologic characteristics of strata in the study area: 1) extensive
shallow clay and silt units with poor permeability and large sorptive capacity,
and 2) cavernous saturated limestones capable of diluting wastes in extremely
large volumes of native ground water. Undoubtedly both characteristics serve
to attenuate contamination on a local scale. Regional variability in radium
content further obscures the presence and three-dimensional extent of contaminated
ground water.
Perhaps the best documentation of local contamination concerns the C. F.
Industries, Inc. gypsum pile failure in April 1975 as a result of sinkhole
collapse (B. Boatwright, written communication, May 3, 1976). The sinkhole
collapse was first sighted by Southwest Florida Water Management District
(District) staff on May 17, 1975 in the course of aerial reconnaissance of the
Alafia Basin. The District arranged meetings between industry and concerned
public agencies and served as hydrologic advisor to the Department of Pollution
74
-------
Control (DPC) which required remedial action in the form of monitoring and
filling of the sinkhole in a prescribed manner. The West Central Regional
Office of DPC was the lead agency in the matter. Water sampling was done by
the District in cooperation with the Orlando Radiological Laboratory. Analyti-
cal methods were outlined by DPC (Tallahasee).
The stack is located about three miles southeast of Mulberry. Failure
occurred when a sinkhole about 200 feet in diameter formed and allowed 90,000
cubic yards of gypsum and 4.5 million gallons of effluent to recharge the
water table and underlying artesian aquifers in the period April 30 - May 19,
1975. Thereafter, discharge to the subsurface decreased as a result of gypsum
plugging; however, the gypsum continued to dissolve and a more permanent plug
using clay or other nonreactive material was ordered by the State Department
of Pollution Control (1975). Semi-consolidated slimes were used. Aside from
being acid (pH 2) and high in radium, the effluent contained high concentra-
tions of fluoride, phosphorus, and sulfate.
Ground-water monitoring commenced April 30, 1975 with a survey of exist-
ing private and public supply wells. Data from additional wells constructed
by the company specifically for head measurements and water sampling revealed
the presence of primary and secondary artesian aquifers within the Hawthorn
Formation and a generally northwestward flow in the upper aquifer within the
vicinity of the sinkhole (based on information prepared by Richard Fountain
and Associates, consultant to C. F. Industries, Inc.). Understandably,
dispersal of contaminants in ground water downgradient from the sinkhole is
expected and is shown in Figure 10 prepared from samples collected on July 21,
1975 (B. Boatwright, written communication, May 3, 1976). Gross alpha values
east, south, and west of the sinkhole range from 3 to 16 pCi/1 (mean 8.5
pCi/1) and are well below 17 and 35 pCi/1 range (mean 27 pCi/1) evident in
ground water on the north and northwest sides.
Prior to sinkhole collapse, recharge of the water table and deeper
aquifers was associated with the presumed presence of a recharge mound having
downward and outward flow components (Florida Department of Pollution Control,
1975). The State required corrective action including a monitoring plan by
which C. F. Industries, Inc. would measure fluid potential (head) and water
75
-------
STATE ROUTE 60
C.F. INDUSTRIES
GYPSUM PONDS
DOMESTIC WELL
SW
12
A
SE
16
A
A
•
8.0
WELL GREATER THAN
100 FEET IN DEPTH
WELL OF UNKNOWN DEPTH
GROSS ALPHA, pCi/l
Figure 18.
Gross alpha radioactivity in ground water in the
vicinity of C.F. Industries, Inc. gypsum ponds.
76
-------
quality in the water table (surficial sand) aquifer as well as in the upper
and lower secondary artesian aquifers below.
Despite the leakage of contaminants, pronounced changes in water quality,
with the exception of gross alpha are not apparent (Table 6). Rapid neutrali-
zation of acid wastes occurs and probably quickly reduces the concentration of
radium, although radium data are not available to substantiate this. Reduction
of the concentration by sorption is likely, particularly in areas underlain by
poorly permeable strata between the ponds and the uppermost limestone units of
the Floridan aquifer (G. Parker, Geraghty and Miller, Inc., personal com-
munication, February 25, 1977). The gross alpha data indicate elevated levels
of alpha-emitting radionuclides downgradient. Comparison of other
radium and gross alpha data from Shearer et al. (1966) using multiple regression
techniques shows a rather inconsistent relationship particularly at gross
alpha levels below about 12 pCi/1. Unfortunately, at the time of writing, no
radium data were available for the C. F. Industries, Inc. case.
Radium concentrations in ground water within roughly a three to six mile
radius of the C. F. Industries, Inc. gypsum ponds range from 0.58 to 6.0
pCi/1. Wells completed principally in the Lower Floridan average less than
1.8 pCi/1 versus 2.7 pCi/1 for the Upper Floridan. Thus there appears to be
little difference between aquifers. Again assuming that there is some con-
sistent relationship between gross alpha and radium, the downgradient gross
alpha values in ground water affected by the sinkhole collapse at C. F.
Industries, Inc. indicate radium contamination, despite the lack of pronounced
change in the other parameters.
A second case of sinkhole collapse beneath a slime pond occurred in 1968.
Thermal infrared imagery revealed preferential development of sinkholes along
a lineament which extended through the ponds. Despite substrate collapse, no
ground-water monitoring data were collected, hence the extent of contamination
is unknown (B. Boatwright, Southwest Florida Water Management District,
written communication, June 6, 1977). At the Gardinier, Inc. plant south of
Tampa, wells in a shallow aquifer downgradient from gypsum piles are report-
edly also affected (James Pool, Department of Environmental Regulation,
77
-------
personal communication, April 7, 1977). At the time of writing, no additional
information concerning these incidents was available to the authors.
TABLE 6- GROUND-WATER QUALITY DATA FROM MONITORING WELLS IN THE VICINITY OF
THE C. F. INDUSTRIES, INC. GYPSUM POND NEAR MULBERRY
WELL LOCATION
Station*
Parameter**
PH
Dissolved solids
Total solids
Acidity
Alkalinity
Hardness
Ca
Mg
so4
F
P04
Total P (as P04)
Si02
Gross alpha
A D J A C
Domestic
Well
8.0
269
272
10
160
78.7
32
14.3
10.5
0.434
0.986
0.17
5.8
<2
E N T
P2-2
8.1
210
230
10
200
102
12
17.5
2.7
0.8
0.078
0.19
9.0
5
DOWN
Pl-3
8.0
218
240
12
248
124
15
21
1.8
0.47
0.01
0.096
7.9
35
G R A D
Pl-1
7.5
283
298
44
268
155
21
25
8.0
0.55
0.003
0.12
5.8
32
I E N T
P2-2
8.5
167
236
0
154
77
13
10.7
47.5
0.9
0.094
1.01
3.4
28
* Station identifiers located on Figure 10..
** Chemical analysis by Florida Department of Environmental Regulation;
sampling data July 21, 1975
Water losses via seepage from three slime ponds studied by Zellars-
Williams (1977) under contract to the U.S. Bureau of Mines amounted to 10 to
22 percent of the total water used for mining and beneficiation. It is reason-
able to assume that seepage from gypsum ponds also enters the water table at a rate
which in some locations may well equal or exceed that of slime ponds if
dissolution of carbonate strata by acidic waste water occurs. Admittedly this
is speculative because water budget and water quality data to ascertain if
seepage from gypsum ponds occurs are notoriously lacking. To the authors'
78
-------
knowledge, monitoring of shallow ground water around gypsum ponds in central
Florida has been limited to the sinkhole collapse incident near Mullberry. We
conclude that additional monitoring studies of the effects of gypsum ponds on
shallow ground-water quality are needed and that data should be collected to
document changes in quality, if any, that follow mining and reclamation.
Of the 80 privately-owned wells sampled in the 1966 survey, two (wells 26
and 29) contained markedly elevated concentrations of radon (22,700 to 28,800
pCi/1), radium-226 (49 to 76 pCi/1), gross alpha (75 to 97 pCi/1), nitrate
(16 to 26 mg/1), and sulfate (120 to 220 mg/1). Unfortunately well depths
were not stated, hence the aquifer(s) involved is unknown. Well 26 is used
for industrial water supply and is located two miles west of Agricola. The
water may be naturally deteriorated insofar as water quality is highly variable
in the area, particularly in the water table aquifer (B. Boatwright, SWFWMD,
personal communication, March 30, 1977). Surrounding land use is predominantly
rangeland. The nearest phosphate mining and processing activity is within one
mile to the south. Well 29 is located on Bovis Road, two and one-half miles
west of Fort Meade. Again, surrounding land is predominantly grassland with
the nearest phosphate mining approximately one mile south.
Radium-226 concentrations in excess of 5 pCi/1 are summarized in Table 7.
Of the 122 analyses from the 1973-1976 EPA and USGS surveys, only 12 or about
10 percent exceed 5 pCi/1. The two highest values of 22 and 90 pCi/1 are from
a water table well in a mineralized but unmined area. This may indicate that
high radium can also be a natural phenomenon and not necessarily indicative of
contamination. Hutchinson (1975) noted this natural condition but gave no
specific data. Wells in the nonmineralized area contain 5.2 to 14.7 pCi/1
radium-226 which is considered to be a natural condition owing to the location
and depth of the wells. Only two wells in the mined areas contain elevated
levels of radium and these are from the Upper and Lower Floridan aquifers. No
conclusion is drawn as to whether these represent natural or contaminated
conditions. However, the concentrations are similar to peak values reported
for wells of similar depth in mineralized, unmined areas and in nonmineralized
areas.
79
-------
TABLE 7. SUMMARY OF 1973-1976 RADIUM-226 DATA EXCEEDING 5 pCi/1
Principal Aquifer and
Well No.
EP5-22
GP40-34
GP95-22
EP47-6
GP59-4
GP76
EMI 4-2
EP69-18
EP69-14
EP85-12
GP89-14
EP91-17
Total/Cased
Land Use Depth/ Depth
Mineralized, unmined 1801
39/34
27/22
Mineralized, mined -/-
-/-
Mineralized -/-
400/90
Nonmineralized 800/-
200/99
705/-
220/-
200/-
Radium Content, pCi/1
WT UF LF
15.3 ±
7.7
22,90
6.0
14
5.3
12.1 ±
14.7 ±
7.3
7.3 ±
5.2±0.052
10.6+0.106
1.53
0.121
0.147
0.073
Use of connector wells to dewater surficial sand and upper Hawthorn
strata was initially done with little information on the radiochemical quality
of ground water allowed to recharge the deeper aquifers. Recent permits for
consumptive use of water issued by the Southwest Florida Water Management
District require monthly water analyses of production wells and connector
wells. Included in the minimum of 15 parameters for analysis is gross alpha
radiation. If gross alpha exceeds 15 pCi/1 analysis for radium-226 and total
radium is also required. These data are reported to SWFWMD on an annual basis
for production wells and monthly for connector wells. There is provision for
additional testing as necessary and for administrative action if signifi-
cant harm of receiving waters is indicated by the data (B. Boatwright, SWFWMD,
written communication, June 6, 1977). Recent management practices emphasize
stricter controls on the quality of water recharged to the Floridan aquifer
system. This is worthwhile insofar as there is some difference of opinion
concerning the radium content of shallow ground water. Hutchinson (1975)
reported excessive radium concentrations whereas preliminary radium data
supplied by industry for two mines where connector wells are in use show
80
-------
concentrations of 3 pCi/1 or less in recharge water (B. Boatwright, written
communication, June 6, 1977).
Abandonment procedures for connector wells require grouting but there is
open concern by regulatory and management agencies that the rapid turnover of
wells, necessitated by the pace of dewatering, stripping, and reclamation
activities, results in improper abandonment (D. Guthrie, Polk County Depart-
ment of Environmental Control, personal communication, June 2, 1977; B. Boat-
wright, SWFWMD, personal communication, June 3, 1977). There are no require-
ments for monitoring of ground-water quality after stripping and reclamation
is completed. It is also unknown if radium increases in very shallow ground
water as a result of stripping and particularly as a result of disruption of
the leach zone materials and shallow aquifer. Additional geochemical or
hydrogeochemical studies are recommended to determine whether shallow ground
water may possibly become enriched in radium relative to pre-mining concentra-
tions. Although available radiochemical data do not indicate this problem,
the data that have been gathered are decidedly deficient for environmental
monitoring purposes relative to the scale of phosphate industry activity. The
water table in mined areas is poorly monitored in terms of water quality,
although recent SWFWMD permits for consumptive use will require more data to
close this gap.
As presently conceived, the proposed Underground Injection Control
Program of the USEPA would include the practice of using connector wells to
move water from the water table and Upper Floridan aquifers to the Lower
Floridan aquifer. Monitoring of water quality will be a necessary part of
such monitoring. Control over seepage of contaminated, radioactive gypsum
pond water would not be included in these regulations. It is recommended that
the use of connector wells be carefully regulated in terms of the control
program and that seepage from gypsum ponds be carefully studied as to magnitude
and effects to determine if corrective action is necessary.
81
-------
ADEQUACY OF INDUSTRY RESPONSE TO THE DRI PROCESS
Three Development of Regional Impact applications (Borden, Inc., 1975;
W.R. Grace and Co., 1975; Phillips Petroleum Co., 1975) for permission to
conduct surface mining in portions of Hillsborough, Desoto, Manatee, and
Hardee Counties were reviewed for their approach to prediction of radiochemical
impact on ground water as a result of mining and related operations. With
rare exception, qualitative statements based on the limited radiochemical data
existing prior to 1975 and arguments or positions based on theory were advanced
to indicate that mining would have no or little adverse impacts.
Concerning radiation, one report (Borden, Inc., 1975, p. 27-29) discusses
uranium but not radium. Unsubstantiated conclusion is reached that "no problems
with regard to radiochemical pollution of air or water, or of employee exposure,
are anticipated...." Elsewhere (p. 29) the report mentions solution openings
in the Hawthorn Formation having been discovered onsite, yet design and con-
struction of slime ponds only incorporates seepage control measures for dams.
No mention is made of the overall water balance in the ponds, despite the fact
that at least 2.5 square miles of ponds will be floored in limestone, and have
heads at least thirty feet above those in the Hawthorn Formation (p. 110 and
exhibit 76).
Of the three DRI applications reviewed, only one, (Phillips Petroleum
Co., 1975) included a preoperational health physics and environmental study
including a terrestial gamma radiation survey and measurement of radioactivity
concentrations (radium-226, polonium-210, lead-210, natural uranium) in air,
water and vegetation. There is mention (p. 37) of plans for water quality
monitoring through mining and reclamation, yet no details as to parameters,
sampling locations or depths, or frequency are given. The conclusion is
reached (p. 36) that no degradation of water quality in the water table aquifer
should result. Comparison (p. 53-54) of waste water quality relative to that
of adjacent ground water at two settling ponds indicates no deterioration as a
82
-------
result of high quality water in the ponds and very low seepage. Appendix A of
the same report purports to estimate water quality before, during, and after
mining but this is not accomplished and no radiochemical measurements of any
kind are provided. Impacts of mining on the hydrologic system are addressed
only in terms of flows and head declines due to dewatering or pumping. Long
term (post mining and reclamation) effects and radiochemical water quality
impacts are not addressed for the water table aquifer or for either of the two
principal artesian aquifers.
Proposed mining of another large tract (18,685 acres total; 12,845 acres
mined) in Manatee and Hillsborough Counties involves surface water quality
monitoring only to ensure discharge in accordance with pertinent regulations
(W.R. Grace and Co., 1975, p. 23). Although some seepage into the shallow
aquifer is acknowledged (p. 24), the dissolved mineral load is considered
minor and no direct discharge of liquid wastes to ground water is expected
It is implicitly assumed that seepage will contain only dissolved solids,
including radionuclides, although the onsite occurrence of a prominent system
of collapse features is noted (p. 68). No mention of their potential role in
contamination incidents is made. Monitoring of ground water will be limited
to measurement of head and chloride concentration in the water supply wells
tapping the Floridan aquifer.
We recognize that three DRI applications are a rather small sample and
that they contain only a portion of the data reviewed and required by public
agencies such as SWFWMD and the regional planning councils. Additional studies
and data concerning water use and hydrology are increasingly being required as
part of the permitting procedures or development orders issued subsequent to
the DRI application. For example, connector wells arid water supply wells must
be monitored periodically, often monthly, for flow, gross alpha, and other
nonradiochemical parameters. If gross alpha exceeds some limit, initially set
at 2 pCi/1 and more recently to 15 pCi/1, radium-226 analysis is required. In
some cases a 5 pCi/1 gross alpha screening level is used. If radium-226
exceeds 3 pCi/1, radium-228 must also be determined (Gordon F. Palm, private
consultant, written communications, June 16, July 25, 1977 ; Barbara A. Boat-
wright, SWFWMD, written communication, June 6, 1977).
83
-------
REFERENCES
Ahrens, L.H., 1957, Lognormal-type distributions, Pt. 3 of the lognormal
distribution of the elements-a fundamental law of geochemistry and its
subsidiary: Geochim. et Cosmochim. Acta, v. 11, no, 4, p. 205-212.
Back, W., 1969, Chemical hydrogeology of the carbonate peninsulas of Florida
and Yucatan: Ph.D. dissertation, Univ. of Nevada Reno, 71 p.
Battelle Memorial Institute, 1971, Inorganic fertilizer and phosphate mining
industries - water pollution control: Final report to U.S. Environ.
Protection Agency, Washington, D.C., under grant 12020FPD.
Borden, Inc., 1975, Development of regional impact (DRI) application for
development approval for the Big Four Mine, Hillsborough County,
Florida: submitted by Smith-Douglas Div. of Borden Chemicals, Borden,
Inc., Plant City, Florida to Board of County Commissioners and the
Florida Division of State Planning, 212 p.
Cathcart, J.B., 1963a, Economic geology of the Chicora quadrangle, Florida:
U.S. Geol. Survey Bull. 1162-A, 66 p.
_, 1963b, Economic geology of the Keysville quadrangle, Florida:
U.S. Geol. Survey Bull. 1128, 82 p.
, 1963c, Economic geology of the Plant City quadrangle, Florida:
U.S. Geol. Survey Bull. 1142-D, 56 p.
_, 1964, Economic geology of the Lakeland quadrangle, Florida:
U.S. Geol. Survey Bull. 1162-G, 128 p.
, 1966, Economic geology of the Fort Meade quadrangle, Polk
and Hardee Counties, Florida: U.S. Geol. Survey Bull. 1207, 93 p.
Cathcart, J.B., and McGreevy, L.J., 1959, Results of geologic exploration by
core drilling, 1953, land-pebble phosphate district, florida: U.S. Geol
Survey Bull. 1046-K, 298 p.
Coker, A.E., 1971, Remote sensing of phosphates and fluoride pollution in the
karst terrain of Florida: paper presented at National Water Well Assoc.
Ann. meeting, Las Vegas, Nevada.
, 1972, Discrimination of flouride and phosphate contamination in
central Florida for analysis of environmental effects: 4th annual earth
84
-------
resources program review-National Aeronautics and Space Administration,
Manned Spacecraft Center, Houston, Texas, 12 p.
Datagraphics, Inc., 1971, Inorganic chemicals industry profile (updated):
Final report to U.S. Environ. Protection Agency, Washington, D.C.,
under grant 12020EJI.
Denham, D.H., and Waite, D.A., 1975, Some practical applications of the
log-normal distribution for interpreting environmental data: Paper
presented at the 20th Ann. Mtg. of the Health Physics Society, Buffalo,
NY, July 14-17, 1975, 31 p.
Dunn, O.J., 1964, Multiple comparisons using rank sums: Technometrics, v. 6,
p. 241-252.
Fitzgerald, J.E., Jr., Guimond, R.J., and Shaw, R.A., 1976, A preliminary
evaluation of the control of indoor radon daughter levels in new
structures: U.S. Environ. Protection Agency, Office of Radiation
Programs, Criteria and Standards Div., Washington, D.C., Rpt.
no. EPA-520/4-76-018, 62 p.
Florida Department of Health and Rehabilitative Services, 1976, Study of
radiation environment, Sarasota County, Florida: unpublished report
dated May 20, 1976, 10 p.
Florida Department of Pollution Control, 1975, letter of June 12 from
S. Winn, Deputy Executive Director to M.L. Walton, Acting Plant
Manager, CF Chemicals, Inc., Bartow, Florida, 6 p.
Gibbons, J.D., 1976, Nonparametric methods for quantitative analysis: Holt,
Rinehart and Winston, New York, 463 p.
Golden, J.C., 1968, Natural background radiation levels in Florida: Sandia
Laboratory, Albuquerque, SC-RR-68-196 (May), 27 p.
Guimond, R.J., and Windham, S.T., 1975 Radioactivity distribution in phosphate
products, by-products, effluents, and wastes: U.S. Environ. Protection
Agency, Office of Radiation Programs, Criteria and Standards Div.,
Washington, D.C., Tech note ORP/CDS-75-3, 30 p.
Hackbarth, D.A., 1971, Field study of subsurface spent sulfite liquor
movement using earth resistivity measurements: Ground Water, v. 9,
no. 3, p. 11-16.
Hoppe, R.W., 1976, Phosphates are vital to agriculture - and Florida mines
for one-third of the world: Engineering and Mining Jour., May 1976,
p. 79-89.
Hursh, J.B., 1953, The radium content of public water supplies: Univ. of
Rochester, Rochester, NY, Rep. UR-257.
Hutchinson, C.B., 1975, Effects of strip mining on shallow aquifer systems
in phosphate district: U.S. Geol. Survey Prof. Paper 975, p. 89.
85
-------
Irwin, G.A., and Hutchinson, G.B., 1976, Reconnaissance water sampling for
radium-226 in central and northern Florida, Dec. 1974 - March 1976:
U.S. Geol. Survey Water Res. Inv. 76-103, 16 p.
Kaufman, M.I., and Dion, N.P., 1967, Chemical character of water in the
Floridan aquifer in southern Peace River basin, Florida; U.S. Geol.
Survey in cooperation with the Florida Bur. of Geol., Map Series 27.
Keefer, Douglas H., in preparation, Radiation exposure in public ground water
supplies, Ottawa County, Oklahoma: U.S. Environ. Protection Agency,
Robert S. Kerr Environ. Res. Lab., Ada, Oklahoma, 57 p.
Koch, G.S., and Link, R.F-, 1971, The coefficient of variation - a guide to
the sampling of ore deposits: Econ. Geol., v. 66, p. 293-301.
Koczy, F.F., 1958, Natural radioactivity as a tracer in the ocean: Proc. 2nd
U.N. Int. Conf. Peaceful Uses of Atomic Energy, United Nations, NY.
Le Grand, H.E., 1968, Monitoring of changes in quality of ground water: Ground
Water, v. 6, no. 4, p. 14-18.
Mansfield, G.R., 1942, Phosphate resources of Florida: U.S. Geol. Survey
Bull. 934, 82 p.
Merkel, R.H., 1972, The use of resistivity techniques to delineate acid mine
drainage in ground water: Ground Water, v. 10, no. 5, p. 38-42.
Mills, L.R., and Laughlin, C.P., 1976, Potentiometric surface of Floridan
aquifer-May 1975, and change of potentiometric surface 1969 to 1975,
Southwest Florida Water Management District and adjacent areas:
U.S. Geol. Survey Water Resources Inv. no. 76-80.
Miyake, Y., Sugimura, Y., and Tsubota, H., 1964, Content of uranium, radium
and thorium in river waters in Japan: In The Natural Radiation Environ-
ment, Rice Univ. Houston, Texas, Adams, J.A.S., and Lowder, W.M., eds.,
p. 219-225.
Osmond, O.K., 1964, The distribution of the heavy radioelements in the rocks
and waters of Florida: j£ The Natural Radiation Environment, Rice Univ.,
Houston, Texas, Adams, J.A.S., and Lowder, W.M., eds., p. 153-159.
Overstreet, W.C., 1967, The geologic occurrence of monazite: U.S. Geol.
Survey Prof. Paper 530, 327 p.
Parker, C.G., Ferguson, G.E., Love, S.K., and Others, 1955, Water resources
of southeastern Florida: U.S. Geol. Survey Water Supply Paper 1255,
965 p.
Peek, H.M., 1958, Ground-water resources of Manatee County, Florida: U.S.
Geol. Survey Report of Inv. 18, prepared in cooperation with the Florida
Geol. Survey, Board of County Commissioners of Manatee County, and the
Manatee River Soil Conversation District, 99 p.
86
-------
Phillips Petroleum Co., 1975, Development of regional impact (DRI) application
for development approval for phosphate mining and benefication in De Soto
and Manatee Counties, Florida: submitted by Phillips Petroleum. Company,
Bartlesville, Oklahoma to Boards of County Commissioners and the Division
of State Planning, 419 p.
Pride, R. W., F- W. Meyer, and R. N. Cherry, 1966, Hydrology of Green Swamp
Area in Central Florida: U.S. Geol. Survey, in cooperation with the
Florida Geol. Survey, the Florida Div. of Water Resources and Conserva-
tion, and the Southwest Florida Water Management District, 137 p.
Robertson, A. F., 1973, Hydrologic Conditions in the Lakeland Ridge Area of
Polk County, Florida: U.S. Geol. Survey Report of Inv. 64, prepared in
cooperation with the Southwest Florida Water Management District, The
City of Lakeland, Florida, and the Bur. of Geol., Florida Dept. of
Natural Resources, 54 p.
Samuels, L.D., 1964, A study of environmental exposure to radium in drinking
water: j_n_ The Natural Radiation Environment, Rice Univ., Houston, Texas,
Adams, J.A.S., and Lowder, W. M. eds., p. 239-251.
Scott, R. C., Barker, F.B., 1962, Data on uranium and radium in ground
water in the United States: U.S. Geol. Survey Prof. Paper 426; 36 p.
Shearer, S. Smith, B., and Rushing, D. E., 1966, Draft report of radium-226
and radon-222 concentrations in central Florida ground waters:
Appendix B j_n_ Reconnaissance study of radiochemical pollution for phosphate
rock mining and milling, 1973, U.S. Environ. Protection Agency, National
Field Inv. Center-Denver, p. B1-B29.
Siegel, S., 1956, Nonparametric statistics for the behavioral sciences:
New York, McGraw-Hill, 312 p.
Sinclair, A.J., 1974, Selection of threshold values in geochemical data using
probability graphs: Jour, of Geochem. Explor., v. 3, no. 2, p. 129-149.
Spaulding, R.F., 1972, The contemporary geochemistry of uranium in the
Gulf of Mexico distributive provice: Ph.D. thesis, Texas A&M Univ.
Springer, Melvin D., 1976, Research for the development of guidelines for
conducting and analyzing an environmental quality study to determine
statistically meaningful results: Univ. of Arkansas, Water Resources
Research Center Pub. 37, in cooperation with Engineering Experiment
Station Research Rpt. 26, 163 p.
Stewart, H. G., Jr., 1966, Ground-water resources of Polk County: U.S. Geol.
Survey Report of Inv. 44, prepared in cooperation with the Div. of Geol.,
the Board of County Commissioners of Polk County, and the Southwest
Florida Water Management District, 170 p.
87
-------
Stewart, J. W., Mills, L.R., Knochenmus, D.D., and Faulkner, F.L., 1971,
Potentiometric surface and areas of artesian flow, May 1969, and change
of potentiometric surface 1964 to 1969, Floridan aquifer, Southwest
Florida Water Management District, Florida: U.S. Geol. Survey Hydrologic
Inv. Atlas HA-440.
Stellar, R. L., and Roux, P., 1975, Earth resistivity surveys - a method for
defining ground-water contamination: Ground Water, v. 13, no. 2,
p. 145-150.
Stowasser, W. F-, 1976, Phosphate Rock: jjn Mineral Facts and Problems, U.S.
Bur. of Mines Bull. 667, p. 819-834.
Tanner, A. B., 1964, Physical and chemical controls on distribution of
radium-226 and radon-222 in ground water near Great Salt Lake, Utah:
_i_n_ The Natural Radiation Environment, Rice Univ., Houston, Texas, Adams,
J. A. S., and Lowder, W.M., eds., p. 253-276.
Thatcher, L. L., Janzer, V.J., and Edwards, K.W., 1977, Methods for determina-
tion of radioactive substances in water and fluvial sediments: U.S.
Geol. Survey Techniques of Water-Resources Inv. Book 5 (Laboratory
Analysis), Chap. A5, 95 p.
Todd, D. K., Tinlin, R.M., Schmidt, K.D., and Everett, L.G., 1976, Monitoring
ground-water quality - monitoring methodology: Final report to U.S.
Environ. Protection Agency, Office of Research and Development, Environ.
Monitoring and Support Lab., Las Vegas, NV, under contract 68-01-0759,
report no. EPA-600/4-76-026, 154 p.
Tokarev, A.N., and Shcherbakov, A.V., 1956, Radiohydrogeology ("Radiogidro-
geologiya"): Gosgeoltekhizdat. (English trans.: U.S. Atomic Energy
Comm. Rpt., AEC-tr-4100, 1960, 346 p.).
U. S. Environmental Protection Agency, 1973, Reconnaissance study of radio-
chemical pollution from phosphate rock mining and milling: Office of
Enforcement, National Field Inv. Center-Denver, Denver, Colorado, 46 p.
, 1975 Preliminary findings radon daughter levels in structures
constructed on reclaimed Florida phosphate land: Office of Radiation
Programs, Criteria and Standards Div., Tech note ORP/CSD-75-4, 7 p.
1976a, Drinking water regulations (Radionuclides): Federal
Register, v. 41, no. 133, July 9, 1976, p. 28402-28409.
, 1976b, Florida phosphate lands, interim recommendations for
radiation levels: Federal Register, v. 41, no. 123, p. 26066-26068.
Warner, D. L., 1974, Rationale and methodology for monitoring ground water
polluted by mining activities: Final report to U.S. Environ. Protection
Agency, Office of Research and Development, Mat'l Environ. Research
Center, Las Vegas, NV, under contract 68-01-0759, report no.
EPA-600/4-74-003, 76 p.
-------
Wayne Thomas, Inc., 1976, Florida pebble phosphate field land ownership map
(1) northern portion; (2) southern portion: Wayne Thomas, Inc.,
private consultant, Tampa, Florida, 2 sheets.
Williams, E. G., Golden, J.C., Jr., Roessler, C.E., and Clark, y., 1965,
Background r&diation in Florida: Div. of Rad. and Occupational Health,
Florida State Board of Health, RH-00054-02, 60 p.
Wilson, W.E., 1975, Ground-water resources of De Soto and Hardee Counties,
Florida: U.S. Geol. Survey open file report prepared in cooperation with
Florida Bur. of Geol., 189 p.
W.R. Grace and Co., 1975, Development of regional impact (DRI) application
for development approval for the Four Corners mine, Hillsborough and
Manatee Counties, Florida: submitted by W.R. Grace and Co., Bartow,
Florida to the Boards of County Commissioners and the Florida Division
of State Planning, 160 p.
Zellars and Williams, Inc., 1977, Water recirculation system balance of central
Florida phosphate mining: prepared under contract to U.S. Bur. of Mines,
USBM open file report OFR 120-77 by Zellars-Wil1iams, Inc., Lakeland,
Florida, 34 p.
89
-------
o
APPENDIX 1. DISSOLVED RA-226 CONCENTRATION (pCi/1) IN GROUND WATER IN THE CENTRAL FLORIDA PHOSPHATE DISTRICT
Site number prefixes indicate the Agency originating the analysis: E - U.S. Environmental Protection Agency, G - U.S. Geological Survey. The second
letter indicates the county: P - Polk Co., M - Manatee Co., HB - Hillsborough Co. , H - Hardee Co, D-DeSoto Co. The first letter indicates the
presence (M) or absence (N) of mineralization. Mineralized areas are further divided into mined and nonmined classes by use of a second letter,
M or N, as appropriate. See Appendix 3. for procedure used to locate wells. Data from Irwin and Hutchinson (1976) are located by latitude and
longitude only and are additionally located herein by township and range. Aquifers are defined as follows: LF - Lower Floridan aquifer, UF - Upper
riorioan aquire
Site No.
Classification
GP1
N
EP2-15
M-N
EP3-19
M-N
EP4-20
M-N
EP5-22
M-N
EP6-21
M-N
EP7-7
M-N
EP8-8
N
GP8A
N
EP9-4
N
EP10-3
N
GPU
N
N
EP13-11
M-N
EP14-18
M-M
EP15-19
M-M
r, wi - water laoie aquirer.
Name or Latitude and Longitude
USGS (acidified sample)
Trailer park, Lakeland
Trailer park, Lakeland
Trailer park, Eaton
Trailer park, Lakeland
Trailer park, Lakeland
Private well, Auburndale
Trailer park and camp,
Haines City
USGS (acidified sample)
Lake Alfred well 2
Lake Alfred well 1
USGS (acidified sample)
Haines City well no. 4
Trailer park
Central Avenue well 9,
Lakeland
Well 20, Lakeland
T. R. S.
26.27.11
27. 23. lied
27.23.20cc
27.23.28ab
27.23.28bd
27.23.29ad
27.25.21ab
27.26.24cd
27.26.28
27.26.32ad
27.26.32bd
27.27.4
27.27.29ad
28.23.20bc
28.23.25ca
28.23.35ab
Date
Sampled
1/03/75
8/21/73
8/21/73
8/21/73
8/21/73
8/21/73
8/20/73
8/21/73
1/03/75
7/30/74
7/30/74
1/03/75
7/30/74
8/21/73
8/01/74
8/01/74
Total /Depth
depth/ cased
(ft)
362/80
123/-
-/-
206/-
180/-
135/-
150/-
125/-
500/100
550/100
550/95
430/100
810/107
82/-
703/238
-/100
Aquifer
LF
LF
?
LF
LF
LF
LF
LF
LF
LF
LF
LF
LF
UF
LF
UF
Radium-226
1 .8
0.77
1.2 .' .
1.7 > .
15.3 *1.
1.8 i .
1.5 " .
2.4 t ,
1.8
1.8 ±
2.1 ±
.99
0.51 ± .
2.3 ± .
0.73 ± ,
1.6 i .
(pCi/D
007
036
068
53
036
045
.072
.054
.042
,025
.046
.029
.048
-------
GP15A-19
M-N
GP15B
M-N
EP18-6
M-N
EP19-1
M-N
EP20-2
N
GP20A
N
EP21-1
N
EP22-16
N
EP23-6
N
EP24-31
M-M
EP25-20
M-N
EP26-21
M-N
GP27-32
M-M
EP28-22
M-N
EP29-4
M-N
EP30-5
M-N
GP31-33
M-M
GP32-44
M-M
EP33-7
M-M
EP34-23
M-N
USGS monitor well
USGS (acidified sample)
Trailer park, Auburndale
Trailer park, Lakeland
Winterhaven well 7,
Inwood Plant
USGS (acidified sample)
Winterhaven well 8,
3rd street Water Plant
Private well, Lake Hamilton
Dundee wel 1 2
Mulberry'1 '
Drane Airfield Rd. well 16,
Lakeland
Piper Well 35, Lakeland
Mulberry'2'
West Coast Grove, Cornet Road
at Hwy. 60
Trailer park and restaurant,
Mulberry
Private well , school
12)
757156111 - 124^ '
USGS monitor well
Hwy 7 - One mile south of
air base
Multi -family supply well,
Gordonville
28.24.9dd
28.25.11
28.25.17ab
28.25.18cd
28.?5.24bb
28.26.28
28.26.29dd
28.27.28ac
28.27.28ac
29.23.lbb
29.23.4bd
29.23.9ac
29.23.13cd
29.23.30db
29.23.32ad
29.Z3.32ad
29.24.18ba
29.24.20dd
29.25.22cc
29.25.22da
3/04/76
1/03/75
8/20/73
8/21/74
7/30/74
1/06/75
7/30/74
8/21/73
7/30/74
12/04/74
8/01/74
8/01/74
12/04/74
9/10/74
8/21/73
8/21/73
12/04/74
2/13/76
9/10/74
8/20/73
58/31
362/80
110/-
100/-
601/186
312/100
734/-
320/-
755/152
?/?
700/103
550/203
95/53
-/-
168/-
97/-
120/170
37/32
27/14
250/-
UF
LF
UF
UF
LF
LF
LF
LF
LF
LF
LF
LF
UF
?
LF
UF
UF
WT
WT
LF
1.2
1.6
2.8 +
4.4 ±
2.4 <
3.8
1.0 i
1.7 ^
0.6 t
0.13
0.92 i
0.60 ±
0.61
3.0 ±
0.19 ±
0.18 ±
0.16
.64
<2.0
1.2 *
.056
.044
.048
.03
.068
.018
.037
.03
.09
.017
.016
.036
-------
GP35-60A
M-M
GP35B-46
N
GP35C-34
M-N
EP37-3
N
EP38-9
M-M
EP41-23
M-M
EP42-24
M-M
EP43-10
M-M
EP44-25
M-M
GP45-10
M-N
GP45A-2
M-M
EP46-26
M-M
GP47-6
M-M
GP47A-3
M-M
EP48-28
M-M
EP49-9
M-N
EP50-10
M-N
EP51-11
M-N
EP52-12
M-N
GP52A-42
M-N
Domestic well, Bartow
USGS monitor well
USGS monitor well
Private well, restaurant,
Lake Wales
Private well, trailer park,
Mulberry
Mulberry well 1
Mulberry well 2
Private well, trailer park
Mulberry
Mulberry Heights well
Domestic well
751201422
Sand mine tailings area
Industrial well
SWFWMD observation
well-B(3]
USGS monitor well
Hwy 640 E. of Hwy 555
Bartow well 1, Power Plant
Bartow well 2, Power Plant
Bartow well 10, Water Plant
Bartow well 3
Monitor well (4)
29.25.23dc
29.26.14dc
29.26.19da
29.27.27cd
30. 23. led
30.23.12bc
30.23.12bc
30.23.14aa
30.23.14aa
30.23.21ca
30.23.35cc
30.23.36bb
30.24.12bb
30.24.35ca
30.24.35ab
30.25.5cc
30.25.5cc
30.25.5cc
30.25.5cc
30.25.5dd
12/11/74
3/18/75
3/19/75
8/22/73
8/21/73
9/10/74
9/10/74
8/21/73
7/30/74
12/10/74
12/04/74
9/10/74
1/08/75
2/26/75
9/10/74
7/30/74
7/30/74
7/30/74
7/30/74
12/04/74
-/-
22/17
39/34
405/-
156/-
776/78
833/80
258/-
-/-
80/60
24/21
-/-
-/-
22/17
-/-
600/100
765/125
683/59
663/565
1348/1270
LF 0.72
WT 2.2
WT 7.7
LF 0.23 ' .018
LF 3.5 .07
LF ,2.0
LF <2.0
LF 1.9 t .019
? 1.8 i .054
UF 1.5
WT 0.26
? 2.0
IIP r n
Ur 6.0
UF 0.64
? r r '
5.5 - .11
LF 2.24 .045
LF 2.3 - .046
LF 1-4 i .042
LF 2.6 _ .052
LF 1.2
-------
EP53-15
M-N
EP54-14
M-N
EP55-13
M-N
GP56-54
M-N
EP57-8
M-N
GP57A-43
M-N
GP58-67
M-M
GP59-4
M-M
GP60-2'4'
M-M
6P61-34
M-M
GP62-72
M-M
GP63-73
M-M
GP63A-14
N
GP65-75
M-M
GP66-76
M-M
EP67-13
N
EP68-2
M
GP68A-4
N
EP69-18
N
GP69A-35
M-N
Bartow well 4, Commerce Park
Bartow well 7, Floral Avenue
Bartow well 5, Jordan Pk.
Jordan Pk.
Lake Garfield
Domestic well
Industrial well
Nursery well, Bartow
Private well
Private well
Industrial well 30A
Industrial well 46A
Domestic well
Monitor well
Monitor well
Motel and restaurant,
Lake Wales
Camp ground., Lake Wales
USGS monitor well
Trailer park, Babson
USGS monitor well
30.25.7cb
30.25.7dc
30.25.8cb
30.25.8dc
30.25.12ad
30.25.14
30.25.15bc
30,25.17ab
30.25.18ca
30.25.18ca
30.25.08dd
30.25.19aa
30.27.4cc
30.25.33ca
30.25.33db
30.27.14bc
30.27.23dc
30.27.29ca
30.28.29cd
31.22.13ba
7/30/74
7/30/74
7/30/74
12/12/74
7/30/74
12/11/74
12/05/74
1/08/75
12/13/74
12/13/74
12/12/74
12/04/74
12/T3/76
12/04/74
12/04/74
8/22/73
8/22/73
2/26/75
8/22/73
3/11/75
315/110 LF
555/65 LF
525/93 LF
-/- UF
360/- LF
173/84 LF
772/196 LF
-/- LF/UF
-/- UF
-/- UF
700/272 LF
-/- UF
200/99 UF
-/- WT
39/37 WT
700/- LF
450/- LF
21/16 UF
800/- LF
15/10 WT
1.75 .052
1.1 .033
1.5 .045
0.96
0.54 * .027
.72
0.79
14.
0.58
1.5
0.37
1.6
7.3
0.26
0.23
3.4 t .068
3.7 ± .074
0.20
14.7 i .147
2.0
-------
GP70-13
M-M
GP71-46A
M-M
GP71A-30
M-M
EP72-27
M-N
GP73-47A
M-M
EP74-29
M-N
GP74A-40
M-M
GP74B-39
M-M
EP76-81
M-M
EP77-35
M-M
EP78-34
M-M
EP79-36
M-H
EP81-33
M-N
EP82-32
M-N
EP83-31
M-N
GP83A-29
M-N
EP84-30
M-M
GP84A-34
M-N
GP84B-32
N
Industrial well 21
Bradley Jet.
USGS monitor well
Domestic well at
grove W. of Agrico
Bradley Jet.
Church, Hwy 555 and 630
USGS monitor wel 1
USGS monitor well
Private well
Hwy. 17, 3/4 mile N. of Ft. Meade
Hwy. 17, h mile N. of Ft. Meade
Well in rest area on Hwy. 98 E. of
Ft. Meade
Ft. Meade well 2
Ft. Meade well 3
Ft. Meade well 1
USGS monitor well
2 miles W. of Ft. Meade
USGS monitor well
USGS monitor well
31.23.4ba
31.23.lldb
31.23,27da
31.24.7ab
31.24.15da
31.24.26cc
31.25.laa
31.24.4ba
31.25.20bc
31.25.22bb
31.25.22bd
31.25.25ca
31.25.27ba
31.25.27ba
31.25.27ca
31.25.28dc
31.25.30dd
31.26.17ba
31.26.23bc
12/13/74
12/12/74
2/25/75
9/10/74
12/02/74
9/10/74
2/25/75
2/26/75
12/05/75
9/10/74
9/10/74
9/07/74
9/10/74
9/10/74
9/10/74
2/25/75
9/10/74
3/19/75
3/19/75
289/20-52 WPb'
200/87 UF
16/11 WT
^
803/284 LF
-/-
31/21 WT
31/21 WT
-/- WT
175/126 UF'
42/21 WT
-'-
900/- LF
900/450 LF
850/- LF
23/18 UF
27/14 WT
39/34 UF
27/22 WT
1.2
4.5
0.28
< 2.0
1.4
< 2.0
1 .8
0.46
5.3
6.2 _± .124
< 2.0
< 2.0
< 2.0
< 2.0
< 2.0
0.32
* 2.0
7.7
0.20
-------
GP84C-27
M-M
EP85-12
M-N
GP85A-23
M-M
GP86-50
M-M
GP87-51
M-M
GP88-93
M-M
GP88A-19
M-M
GP88B-18
M-M
GP88C-26
M-N
GP88D-22
M-N
GP88E-20
M-N
GP88F-1
M-N
EP89-14
EP91-17
EP92-37
EM3-10
M-N
GM3A-5
M-N
EM4-7
M-N
EM5-8
M-N
EM8-1
M-N
USGS monitor well
Camp ground, Lake Wales
USGS monitor well
741156121-232 ,
Domestic well (*•'
USGS monitor well
Bowling Green
USGS monitor well
USGS monitor wel 1
USGS monitor well
USGS monitor well
USGS monitor well
USGS monitor well
Trailer park, Frostproof
Trailer park
Kelley Rd. at Hwy 27A
(1 m. So. of Frostproof)
Domestic well
USGS monitor well
Domestic well
Private well, school
Private well, fire tower
31.26,31ad
31.27.10dd
32.23.13da
32.24.18aa
32.24.18bb
32.24.18bb
32.24.26cb
32.25.28cd
32.26.2bb
32.26.23bb
32.27.28ba
32.27.31cc
32.28.2dc
32.28.6bd
32.28.9bb
33.21.22cd
33.22.laa
33.22.15ba
33.22.20dd
34.21.32bc
12/10/75
8/22/73
2/25/75
12/05/74
12/13/74
12/03/74
2/26/75
2/26/75
3/19/75
2/27/75
3/12/75
3/12/75
8/22/73
8/22/73
9/10/74
6/24/75
2/25/75
6/24/75
6/24/75
6/24/75
122/84 UF
705/- LF
27/22 WT
-/60 WT
-/280 LF
23/20 WT
23/18 WT
22/17 WT
20/15 WT
27/22 WT
32/27 WT
51/46 UF
220/- UF
200/- UF
900/- LF
-/- ?
18/8 WT
93/63 WT
-/- ?
-/- ?
0.26
7.3 ' .073
0.20
3.4
0.14
4.4
0.54
0.20
0.20
22.
1.4
0.20
5.2 t .052
10.6 * .106
2.2 .066
0.40 ' .024
0.20
3.3 - .066
1.2 - .036
5.1 * .051
-------
GM10A-1
M-N
GM10B-2
M-N
GM10C-3
M-N
GM10D-4
M-N
EMI 3-1 5
N
EMI 4- 2
M-N
EMI 6- 16
EMI 8- 18
M-N
EMI 9-1 9
M-N
EM20-20
EM21-21
M-iJ
GHB-1
M-N
GHB-2
M-N
GHB-3
M-N
GHB-4
M-N
GHB-5
M-N
GHB-6
M-N
GHB-7
M-N
GH8-8
M-M
GH-1
M-N
Future phosphate mining area
Future phosphate mining area
Future phosphate mining area
Future phosphate mining area
Domestic well
Domestic wel 1
Domestic well
Domestic well
Domestic well
Domestic well
Domestic well
275006/821442
USGS monitor well
275110/820255
USGS monitor well
274033/820536
Domestic well
274216/820847
USGS monitor well
274544/821442
USGS monitor well
275514/820732
USGS monitor well'6'
275711/820329
old irrigation well
275918/820719
USGS monitor well
272954/814930
future phosphorus mine
34.22.19dd 1/29/75
34.22.19dd 1/29/75
34,22.19dd 1/29/75
34.22.19dd 1/29/75
35.19.22ba 6/24/75
35. 21. Sac 6/24/75
36.21.4cc 6/24/75
36.22.28da 6/24/75
37.22.7cb 6/24/75
37.22.12cb 6/24/75
37.22.17 6/24/75
3/17/75
3/11/75
12/10/74
2/26/75
2/25/75
3/11/75
2/25/75
2/25/75
12/03/74
1225/750
195/130
30/20
650/500
-/-
400/90
-/-
260/220
285/200
141/127
-'-
31/26
17/12
11/11
22/17
22/17
23/18
22/17
22/17
220/84
LF
LF
WT
LF
?
LF
t
LF
LF
LF
7
WT
WT
WT
WT
WT
WT
WT
WT
UF
4.7
0.54
..20
1 .4
0.11
12.1
1.1
5.0 1
4.0
3.9 i
3.7 f
.20
4.5
1.5
.20
.32
.20
.94
.94
1.5
.014
.121
.044
.10
.08
.078
.074
-------
to
--J
GH-2
M-N
6H-3
M-N
GH-4
M-N
GH-5
M-N
GH-6
M-N
GH-7
M-N
GH-8
M-N
GD-1
N-N
272954/814930
future phosphorus mine
273516/814930
USGS monitor well
273528/813448
USGS monitor well
273532/814024
USGS monitor well
273540/815216
USGS monitor well
273541/820203
273659/815639
271303/815037
near future phosphate
mining area
12/03/74
3/12/75
3/12/75
2/27/75
2/25/75
2/25/75
2/26/75
12/03/74
32/21
17/12
17/12
26/21
17/12
18/13
23/18
320/141
WT
WT
WT
WT
WT
WT
WT
UF
.20
.24
.20
.05
.20
1.9
.20
7.9
(1) Domestic well, water level 97 feet below land surface
(2) Domestic well along edge of phosphate mining district in a populated area
(3) Near slime pit
(4) Chemical waste injection site
(5) Uncased segment of well also open to UF and LF; drains water from surface aquifer to Floridan aquifer at phosphate mine
(6) Well surrounded by an extensive area of mining and tailings disposal
-------
APPENDIX 2
ANALYTICAL RESULTS FROM THE 1966 FWPCA SURVEY
OF RADIUM-226 IN CENTRAL FLORIDA GROUND WATER
(tables III and IV from Shearer et al., 1966)
98
-------
Municipal Well Si.ippli.es - Central i''lQrjda
Gross Gross
Sample Depth of Rn-222 Ra-226 U a Th Fb-210 Po-210 Alpha Beta
Number Municipal Supply Well (ft) (pc/l)^ (pc/l) (p.g/1) .(PJiZii (pc/l)^ (pc/l) (pc/l) (pc/1^
83 Eartow Well No. 3
(RAW) 650 260 1.6
Bartow (Treated) -
Aeration, Filtration,
Clilorination 80 l.k
Qk- Winter Haven Wells 1 and
2 (RAW) 1-593 20 0.67
2 - 816
85 Winter Haven Wells 3 and
k (RAW) 3 - 6^8 95 0.58 1.2 0.12 o.O 0.0 2.7 2.7
86 Lake Wales Well 1,
Market St. Plant (RAW) 1022 60 0.76 - - 0.2 0.1 3.3 8.2
87 Lake Wales Well 1,
Grove Ave.Plant (RAW) 1063 35 0.47 -
88 Avon Park (RAW) 560 0.98 0.7 0.11 0.5 0.1 3.7
3-7
-89"- Souring Franklin St.
Well (RAW) lll-80 140 - 0.7 0.11 0.2 0.1 O.h 3.8
90 Arcadia Well 1 (RAW) U95 ^80 3-3 0.9 0.52 0.6 0.1 k.$ 2k
Arcadia (Treated)
Aeration, Chlorination 210 2.5
(continued)
-------
Table III (continued)
o
o
Sample
Wuinbc r
91
92
93
<*
95
96
97
98
99
100
101
102
Munic
Depth of
Municipal Supply Well (ft)
J3o'..'!l ing Green at
"Water Tank
Plant City Well 3
Plant City Well 2
Zcphyrhills
Dcidc City Well 1
Cler::iont South Well
Clermont Highland Well
L.OKC Alfred Well k (RAW)
Lake Alfred (Treated)
Ac; ration, Ciiiorination
Dundee Well at City Park
!I;iin-cG City Well 7
IlaineG City Well B
750
368
te5
150
525
550
560
Soo
565
i£al_Well
En- 222
260
150
360
360
305
600
720
220
50
560
125
115
Supplies - Central Florida
Ra-226 U a T'n Pb-210
(pc/i) (nr;/i) (p'V'1) (PC/!)
2.7 0,o 0.15 0.3
0.77
0.00 1.2 0.06 0.1
0.31 1.2 0.00 0.3
0,00 -
0.39 -
0,29 1.8 o.ok o.l
k.l
1.8 -
0.0 - - 0.0
0.71!- -
0.73 -
Po-210
(pc/1)
0.3
-
0.0
0.0
-
-
0.0
-
-
0..0
-
-
Gross Croc
Alpha Beta
ir_£/ll IPS/
11
0..8
0.7
•616
3JLO
0.50
103 Aviburndale; Water
PLv,nt 'Well
(continued)
-------
o
Table_ m (continued)
Municipal Well 'Supplies - Central Florida
Sample
Uiiniber
H*
105
106
107
iiui : i c. ipal 3u pply
y u . florlda Ave.
Lake-land Well 22
luilLcrry Well 1
• 4edula Eecreation
Depth of
YJcll (ft)
865
891
778
En- 222
160
250
165
Ra-226 U
O..GO
0.84
0.45 1.4
O. Th
-
0.06
Center - North oi'
1-tu.rborry
320
0.23
Fb-210 Po-210
0.1
0.1
0.1
GJ-O
Grose
j je"ca
-------
o
ro
Cample
Number
3
k
Table IV
Privately Owned Wells - Central i>'lorida
Location
Depth of
Well (ft)
H.W. 60 one mile west
of Liii'Lov, Polk Cciuity 1100
H.W. 60 two miles west
of Bartow, Polk County 551
Ridge Wood Rod and Gun
Club, 3 miles cast of
Mulberry, Polk County 170
5 H.W. 5^8 five miles
N.E. of Mulberry,, Polk
County 300
6 H.W. 5'MDA one-half
mile couth of High-
land c City> Polk
County 155
7 1-1/2 miles north of
Bartov, polk County 200
8 II.W. 60 t'/o miles
eact of r/irto1..', Polk
County
9 H.W. 17 one and one-
half ):i:iles so\ith of
Aii'liO.oe^ rolk County
76
68o
2880
210
95
970
3060
1030
26 U
0.08
1.86
1.71 0.7
.18
0.69
1.5
Pb-230 Po-2].0
0.01
0.0
0.0
12
23
0.0
0.1
(continued)
-------
Table IV (continued)
Privately Owned Wells - Central Florida
Sample
Number
10
11
12
13
Ik
15
16
17
Depth of Rn-222
Location Well (ft) (pc/l)
H.U. 17 one mile
ooui/li of Airba.se,
Polk County 35 3550
H.W. 60 tvo miles
east of Mulberry,
Polk County - 32^0
Intei-national Min-
erals Chemical Co.
(Bonnie Plant),
Polk County 900 860
0.1 miles east of
Bonnie Mine Rd. on
Pebbledale Ed.., Polk
County 60 3890
1/2 mile vest of CCA
on Pebblcdale Rd.,
Polk County - 990
H.W. 6hO at Jet. S.R.
555, Polk County - 18,200
Bartov, 860 Herner St.
Poll: County 80 1010
Ra-226
(PC/1)
o.6l
0.12
2.35
0.21
1.19
u
0.9
o.i
a Th
0.38
0.06
Gross Gross
Fb-210 Po-210 Alpha Beta
.L££/ii (
0.6
1.2
0.1
0.4
13
0-9 15
(continued)
-------
Table IV (continued)
Privately Owned
Sample
Mumber
Depth of
Location Well (ft)
Rn-222
(pc/D
Wells -
Ka-226
(pc/D
Central
U
(nc/D
Florida
a Th
(PC/D
Fb-210
(rc/D
Po- 210
(pc/l).
Gross
Alpha
(pc/D
Gros
Beta
(Pc/
s
18 International Min-
er; u.u Chemical Co.,
iloralyn Plant, Polk
County 92 8370
19 Homeland Rd. 1-1/2
miles north of Home-
land, Polk County 98 '^730
20 One mile south of
Bar'cow on Homeland
Ed., Polk County 200 4l60
21 H.W. 17. 1-1/2 miles
south of Bartow, Polk
County 89 4190
22 N.U.17. 3 miles
couth of Bartow, Polk
County 160 5600
ILW. 5^5. One mile
south of intersection
of il.W. yfi and II.W.
6-':0, Polk County
70 Ji 6, ooo
ijv;ift Co. East Deep
V.'oll, A^ricol'a, Florida,
Itolk County oOO 1750
2.70
1.87
1.0'f-
1.99
1.89
2.2
0.03
0.00
0.03
0.8
0.5
0.2
0.2
10
10
16
(continued)
-------
o
01
Table IV (continued)
Sample
Number
25
26
27
28
29
30
31
32
33
Privately Owned
Depth of to- 222
Location Well (ft) (pc/l)
Svift Co. veil B-3, one
mile T-'oot of !>rift Co.,
Polk County 1100 2000
Minute Maid Co . , tvo
miles vest of Agricola - 28,800
Off Be vis Rd. two miles
vest of Ft. Meade, Polk
County 105 5780
Be vis Rd,, 2-1/2 miles
vest, of Ft. Meade, Polk
County - 22,700
K.W. 17, 1/2 mile north of
Ft. Meade,, Polk County 97 9750
II. U. 17, 1/2 mile north of
Ft. Meade, Polk County 187 10,850
Homeland, Polk County 105 ^80
Homeland, Polk County - ^130
Wells -
Ra-226
1.78
^9
5.22
76
0.21
5.13
2.68
2.1»1
Central Florida
U a Th
(|t(f/l) (pc/l)
l.'l- O.O^l-
11 0.37
IK 2 0.58
Homeland, Polk County 88 2310 2.15
Gross Gross
Pb-210 Po-210 Alpha Beta
o.:
7-6
3.7
o.o
1.5 75
0.8 97
16
(continued)
-------
Table IV (continued)
Privately Ovned Wells - Central Florida
Q
0>
3ample
Number
35
36
37
39
Depth of
Location Well (ft)
Durant Section -
Durtuii, Rd. ab Tuikey
Creek, Ilillsborough
County 201
Intersection of H.W.
640 and Runyon Rd.,
Iljllsborough County 90
II. W. 60, 6 miles vest
of Mulberry, Hills-
borough County 150
S.R. 60 and Coronet
Rd.,
-------
Table JV (continued)
Privately Owned Wells - Central Florida
Gross Gross
Sample Depth of Rn-222 Ra-226 U a Th Fo-210 Po-210 Alpha Be la
number Location Well (ft) (pc/l) (pc/l) ilio/l). (pc/l) (pc/l) jpc/l) j_pe/l)_ (pc/l)
II.W. 2'7A, 2 miles
oGuuii uf i-'rOot Proof,
Polk County 37 25 I-2
H.W. 27A, 3 miles
north of Gebring,
Highlands County 300 3330 2.3 - - 0.2 0.2 5 13
II.W. 27, 7 mile£
south of Sebring,
Hic!i3.ands County 69 23 lA2
Grassy I,
-------
Table IV (continued)
o
00
Privately Owned Wells -
Sample
Numbe r
Location
4 9 11. W. 17, one mile
north of Wanchula,
Hardee County
50*
51*
H.W
Spi-
ll. W
. 39, at Crystal
ings, Pasco County
. 50, 2 miles east
Depth of
Well (ft)
85
36
Rn-222 Ra-226
(pc/l) (pc/l)
kOhO 0 . kk
U70 0.82
Central Florida
U a Th
(uS/l) (pc/l).
0.8 0.09
Gross
Pb-210 Po-210 Alpha
(pc/l) (pc/l) (pc/l)
0.1 0.1 2
Gross
Beta
(pc/D
13
of I/inter Garden, Orange
County - 365 0.^9
52* H.W. 1*393, l.k miles ^
south of II.W. 50, Orange
County - 2780 2.33
53* H.W. 17, 2 Miles south
of Kicsiiumee, Osceola
County 60 - 70 153
5^* H.W. 17, 5 miles south
of Kissiixiee, Osceola
County 90 - 115 870 2.17
55* II.W. 17, south of
Kinsii.mee, Osceola
County 7^ 1350 0.1*3
56 1.5 miles vest of Haines
City off H.W. 92,Polk Co. 200 23^0 3.9
0.1
0.04
0.0
0.1
(continued)
-------
Table.IV (continued)
Privately Owned Wells - Central Florida
Gross Gross
Sample Depth of Rn-222 Ra-226 U a Th Fo-210 Po-210 Alpha Beta
Number Location Well (ft) (pc/l) (pc/l) (iWl) (pc/l) (pc/l) (pc/l) (pc/l) (pc/l)
57 3.5 niles east of Lake-
land off II. W. 92, Polk
County 162 i860 2.7
58 3-5 miles east of Lake-
land off H.W. 92, Polk
County 53 2610 0-75
59 2.5 miles east of Lake-
land off H.W. 92, Polk
County 150 6210
60 H.W. 542, one mile east
of Lakeland, Polk County 27 10,000 1.20 1.4 0.04 0.7 0.4 1.9 0
6l Intersection of Fields Rd.
and H.W. 542 east of Lake-
land, Polk County 287 3810 1.89
62 Old Auburndale Rd. 4 miles
east of Lakeland, Polk
County 120 2410 3.58
63* H.W. 92, 1.75 miles vest
of Plant City, Hills-
borough County 134 550 0.03
64* H.W. 92, 2 miles east of
H.W. 579, Hillsborough
County ' 150 240 0.02
-------
Table IV (continued)
Privately Owned Wells - Central Florida
Sample
Humbe r
65*
66*
67*
68*
69
70
71
Location
Depth of
Well (ft)
Intersection of H.W. 301
cu id Puliil Rlvt: i' Rd . ,
uplllsborough County 80
River View, Hillsborough
County
II.W. 301, .75 mile south
of Alafia River Bridge,
Hillsborough County
H.W. 301, 2.5 miles
south of Riverviev,
Hillsborough County 96
II.W. 672, 2 miles vest
of Picnic, Hillsborough
County 187
Intersection of H.W.
672 and II.W. 39 at
Picnic, Hillsborough
County
Intersection of H.W.
and H.W. 39 at Ft. Lone
some, Hillsborough Co.
Rn-222
(PC/1)
2020
750
630
3^00
7650
2650
Ra-226
(pc/D
0.30
1-93
1.73
0.3
0.28
0.33
U
0.8
a. Th
0.04
0.01
Gi'oss
0.2
0.6
0.1
0.1
o.k
Gross
Pb-210 Po-210 Alpha Leta
(pc/l) (pc/1) (pc/1) (pc/1)
0.3 0
(continued)
-------
Table TV (continued)
Privately Owned
Sample
Number Location
Depth of
Well (ft)
Bn-222
(pc/1)
Wells - Central Florida
Ra-226
(pc/1)
(MS/I)
a ih
(pc/1)
Pb-210
(pc/1)
Po-210
(pc/1)
Gross
Alpha
(pc/1)
Gross
Beta
(pc/1
72 H.¥. 630, 1.7 miles
west of Armour Ft.
Meade, Polk County
73 k miles south of
.Mulberry on "E.W. 37,
Po.lk County
74 Bradley Jet., 6.5
miles soutli of Mul-
berry, Polk County
75 Bradley Jet., 7 miles
sou tli of Mulberry,
Polk County
76 One mile couth of
Bradley Jet., Polk
County
77 Intersection of new
and old H.W, 37,
Polk County
78 Old H.W. 37 south into
Bradley Jet., Polk Co.
143 11,300 7.41 1.6
200 5120 2.14 1.4
183 7330 3-90
2070 o.ll 1.4
So - 90 9460 4.32
3300 1.31
15 7600 19
o.ok
0.11
1.3
0.3
0.03 0.4
0.3
0.2
0.1
12
(continued)
-------
Table IV (continued)
Privately Owned Wells - Central Florida
Gross Gross
Sample Depth of Rn-222 Ra-226 U a Th Fb-210 Po-210 Alpha Beta
Number Location Well (ft) (pc/l) (pc/l) (nc/l) (pe/l) (pc/l) (pc/l) (pc/l) (pc/l)
79 Bradley Jet., Polk Co. - 1480 1.31
80 Rolling Mills, Polk
CouuLy - 6550 2.3 1.4 0.04 0.7 0.3
8.1 Oak Terrace, Polk Co. - 8680 4.6
82 Oak Terrace, Polk Co. - 8430 0.0
* Considered outside of area.
-------
APPENDIX 3. WELL NUMBERING SYSTEM
The well numbering system used in this report is based on the Federal
system of land divisions. Data from the U. S. Geological Survey are located
by latitude and longitude and are additionally located herein by township and
range.
Under the Federal system of land divisions, a location is specified in
terms of three principal parts: township (T), range (R), and section (S). In
the study area, townships are measured northward at six-mile intervals from
the meridian. An area of 36 square miles is defined by a given township and
range and, for example, is stated as T.27S., R.25E. Within a given township
there are 36 sections of one square mile each (640 acres) and numbered from 1
to 36 as shown below. The letters a, b, c, and d designate, respectively, the
northeast, northwest, southwest, and southeast quadrants. The first letter
designates the quarter section (160 acre tract) and the second letter indicates
the quarter quarter section (40 acres). A well located in the NW3* SW3*, sec.24
T.27S., R.25E. would be numbered 27.25.24cb. This is illustrated follows:
R25E
27 25 24 cb
113
-------
Site No
SI
S2
S3
S4
S5
S6
S7
S8
S9
S10
Sll
S12
S13
S14
S15
S16
SI 7
S18
S19
S20
S21
S22
S23
S24
S25
S26
S27
APPENDIX 4. DISSOLVED
Name of Well
TRAILER PARK
TRAILER PARK
TRAILER COURT
BUSINESS ESTABLISHMENT
MOBILE HOME PARK
APARTMENTS
MOBILE HOME PARK
FRUITVILLE ELEMENTARY SCHOOL
VENICE CAMPGROUNDS
ORGANIZATION OFFICE
NOKOMIS ELEMENTARY SCHOOL
TRAILER PARK
TRAILER PARK
CLUB
ASHTON-BLISS SCHOOL
PARK PICNIC AREA
PARK PICNIC AREA
PARK PICNIC AREA
PARK PICNIC AREA
PARK PICNIC AREA
PARK CONCESSION STAND
PARK CONCESSION STAND
PARK CONCESSION STAND
PARK CONCESSION STAND
PARK CONCESSION STAND
PARK
MOBILE HOME PARK
RADIUM
Sample
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
- 226 CONCENTRATION (pCi/1) IN GROUND
Code T.R.S Date Sampled
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
1793
1796
40
37
38
36
38
37
28
36
39
37
39
37
37
37
37
38
38
38
38
38
37
37
37
37
37
36
37
.19.
.18.
.18.
.18.
.18.
.18.
14
20
3
24
10
29
.18.25
.18.
.20.
.18.
.19.
.18.
.18.
.17.
.18.
.18.
.18.
.18.
.18.
.18.
.20.
.20.
.20.
.20.
.20.
.18.
.18.
24
21
18
6
7
13
13
11
14
14
14
14
14
15
15
15
15
15
15
12
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
2/25/76
9/9/75
9/9/75
9/9/75
9/9/75
9/9/75
<)/9/7r>
9/9/75
9/9/75
9/9/75
9/9/75
8/7/75
8/7/75
WATER IN SARASOTA COUNTY '
Dissolved
Aquifer Radium - 226 (pC1/l)
F2
F
F
F
F
F
F
F
F
F
F
F
F
F
F
WT3
WT
WT
WT
WT
WT
WT
WT
WT
WT
F
F
19.
17.
14.
21.
3
8
9
,7
5.4
13.
15.
12.
12.
8.
12.
10.
10,
4.
3.
23.
22
24
23
24
11
12
14
,0
0
4
3
,9
,1
.3
•1
.8
.5
.0
.3
.0
.5
.4
.8
.1
.1
14.5
14,
5,
6.
.1
.1
5
± .015
+ .014
+ .012
± .015
± .054
± .12
± .14
± .12
± .12
± .089
± .12
± .10
± .10
± .096
± .070
± .28
± .27
± .29
± .28
± .29
± .11
± .11
± .13
± .12
±.13
± .051
±.52
-------
S28
S29
S30
S31
S32
S33
S34
S35
S36
S37
S38
S39
S40
S41
S42
S43
S44
S45
S46
S47
548
S49
CITY OF VENICE
UTILITY
PARK CONCESSION STAND
UTILITY CO.
MOBILE HOME PARK
ENGLEWOOD WATER DISTRICT
MOBILE HOME PARK
MOBILE HOME PARK
MOBILE HOME PARK
TRAILER PARK
STATE PARK
CITY OF SARASOTA
MOBILE HOME PARK
MOBILE HOME PARK
OSPREY SCHOOL
WATER COMPANY
MOBILE HOME PARK
UTILITY
UTILITY
UTILITY
PHILLIPI SHORES SCHOOL
UTILITY
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
PW -
1997
1799
1800
1801
1802
1803
1805
1806
1807
1808
1809
1810
1812
1814
1815
1816
1817
181S
1819
1821
1822
1823
39.
36.
37.
39.
36.
40.
37.
39.
36.
36,
17.
7
18.6
20.
19,
18,
,20.
19,
19,
.19
.18
38.18
36,
38
38.
38.
36.
40.
36.
38.
36.
37.
37.
.20
.18
18.
18,
18.
19.
18.
,15
,26
,27
.31
.5
,23
.19
.28
.14
.4
.3
.25
,3
,32
4
11
18.23
18.
18.
17.
35
5
12
8/?/75
8/7/75
8/7/75
8/?/75
8/8/75
8/7/75
8/7/75
8/7/75
8/7/75
8/7/75
8/?/75
8/7/75
8/7/75
8/?/75
8/7/75
8/7/75
8/7/75
8/7/75
8/8/75
8/7/75
8/7/75
8/8/75
F
f
WT
F
F
T
F
f
F
F
WT
F
F
F
--
F
F
F
F
F
F
F
8.5
2.1
14.7
10.8
8.1
1.5
3.6
4.0
8.7
6.0
7.3
4.2
27
15. 3
7.3
6.8
1.6
3.6
4.6
3.5
11.7
4.9
± .085
±.042
-t
+ ,
+ _
T<
+ f
,13
12
081
045
072
±.080
+t
+,
087
060
±.073
±.084
+_
+
t
t
«.
+
-).
-
+
-
15
.14
.073
.068
.048
.072
.046
.070
.12
.098
1 Data supplied by Sarasota Cou.ity Health Department
2 Floridan aquifer
3 Water table aquifer
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA/520-6-77-010
3. RECIPIENT'S ACCESSI ON-NO.
4. TITLE AND SUBTITLE
Effects of Phosphate Mineralization
and the Phosphate Industry on Radium-226 in Ground
Water of Central Florida
5. REPORT DATE
October 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robert F. Kaufmann
James D. Bl iss
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Radiation Programs - Las Vegas Facility
U.S. Environmental Protection Agency
P. 0. Box 15027
Las Vegas, Nevada 89114
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Final
Same as above
14. SPONSORING AGENCY CODE
200/03
15. SUPPLEMENTARY NOTES
characterize
is. ABSTRACT
Dissolved radium-226 data were statistically analyzed to
water quality in the water table, Upper Floridan and Lower Floridan aquifers in six
counties of west central Florida where major strip mining and beneficiation of phos-
phate deposits is underway. Mineralization and mining have not significantly
increased the dissolved radium content of native ground water, although locally
elevated levels of radium in both mineralized and nonmineralized areas are naturally
present. In Hillsborough, Polk, Hardee, and De Soto Counties, mean radium content of
ground water beneath mined and unmined lands is 5 pCi/1 or less, with maximum values
on the order of 15 to 20 pCi/1 occurring in unmined areas. For Sarasota and Manatee
Counties, average radium content is 4 to 15 pCi/1 in the water table and Floridan
aquifers. Although portions of these latter two areas are mineralized, there has been
no mining activity to date. Other hydrogeologic and hydrogeochemical factors such as
position near the discharge portion of the ground-water flow system and increased
radium solubility in-water enriched in IDS are believed responsible for the elevated
concentrations, particularly in the Floridan aquifer in Sarasota County. The radium-
226 data base collected in the period 1966 to present is marginal for determining
environmental quality trends and spatial or temporal variations because too few
samples have been collected and diverse sample handling procedures affect the values
produced. Recommendations for additional monitoring and technical studies are
Outlined to -inipr""g v/ator and land managpmpnt __ _____
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Ground water
Hydrogeology
Water pollution
Water quality
Radium
Phosphate deposits
Statistical inference
b.IDENTIFIERS/OPEN ENDED TERMS
Florida hydrology study
Phosphate mining
Radiation surveys
Environmental surveys
c. cos AT l Field/Group
1808
0807
0808
3. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
125
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
*U. S. GOVERNMENT PRINTING OFFICE: 1977-785-037
------- |