This item has the following downloads:

PDF ( 1 MBs ) ( PDF )

STATE OF FLORIDA
DEPARTMENT OF ENVIRONMENTAL PROTECTION
David B. Struhs, Secretary




DIVISION OF RESOURCE ASSESSMENT AND MANAGEMENT
Edwin J. Conklin, Division Director




FLORIDA GEOLOGICAL SURVEY
Walter Schmidt, State Geologist and Chief




OPEN FILE REPORT 83




FLORIDA AQUIFER STORAGE AND RECOVERY GEOCHEMICAL STUDY:
YEAR THREE PROGRESS REPORT


By

Jonathan D. Arthur, P.G. #1149, James B. Cowart and Adel A. Dabous


FLORIDA GEOLOGICAL SURVEY
Tallahassee
A report submitted to the Underground Injection Control Program, Bureau of Water
Facilities Regulation, Division of Water Resource Management, Florida Department of
Environmental Protection
2001


ISSN 1058-1391










CONTENTS

INTRODUCTION ................................................ ........ 1
PURPOSE OF INVESTIGATION .............................................. 2
METHODS
Whole rock chemistry. .................. .................................. 4
W after chem istry ............................................... ......... 6
ARSENIC GEOCHEMISTRY................................................. 6
Arsenic in the hydrogeologic environment ..................................... 11
RESULTS AND DISCUSSION
WHOLE ROCK CHEMISTRY ............................................. 12
Rare-earth elements ................... .............................. 15
WATER CHEMISTRY
U ranium ...... ............................................ ..... .... 18
Rom e Avenue.................................. ........ .. ...... 23
Punta G orda .............................................. .......... 24
Other trace metals
Rom e Avenue.................................. ........ .. ...... 27
Punta G orda .............................................. .......... 32
Origin of the m obilized m etals ............................................ 32
PLANNED RESEARCH ......................................... ........ 38
R EFER EN C E S ................................................ ........ 40


LIST OF TABLES

1. Major-element concentrations; Floridan aquifer system carbonates;
Hillsborough County, Florida. ................ ........................... 5
2. Trace-element concentrations; Floridan aquifer system carbonates;
Hillsborough County, Florida. ................ ........................... 7
3. Rare earth element concentrations; Floridan aquifer system carbonates;
Hillsborough County, Florida ............................ ...... ......... 10
4. Uranium analyses for Cycle Test samples from Punta Gorda ASR .................. 21
5. Trace metal analyses for Rome Avenue ASR samples............................ 29
6. Trace metal analyses of water samples from Cycles 1 and 2 (Upper Zone) and
Cycles 1A and 2A (Lower Zone), Punta Gorda ASR ........................... 35


LIST OF FIGURES

1. Location of study areas ................................................. 3
2. Iron oxide (Fe 203) As variation in Suwannee Limestone, Hillsborough County, Florida. 14
3. Zinc As variation in Suwannee Limestone, Hillsborough County, Florida ............ 14
4. Normalized rare earth element distribution in Suwannee Limestone samples from
W-16618, Hillsborough County, Florida. ................. .................. 16
5. Normalized rare earth element distribution in Suwannee Limestone samples from
W-17687, Hillsborough County, Florida. ................. .................. 16
6. Normalized rare earth element distribution in Suwannee Limestone samples from
W-16574, Hillsborough County, Florida. ................. .................. 17










7. Normalized rare earth element ratios showing Suwannee Limestone samples relative
to fields of enrichment and depletion (LREE light REE, MREE middle REE, and
HREE -heavy REE) .................................................... 17
8. Total rare earth element (estimated, see text) distribution in Suwannee Limestone
samples versus loss on ignition (LOI) ....................................... 19
9. Uranium mobilization models for single carbonate grains (coated and uncoated)
exposed to flowing and non-flowing groundwater ............................ 22
10. Uranium activity ratios through time for Cycles 2 and 3, Rome Avenue ASR,
Hillsborough County, Florida ........................................... 25
11. Uranium concentration through time for Cycles 2 and 3, Rome Avenue ASR,
Hillsborough County, Florida ........................................... 25
12. Uranium activity ratio through time for Cycle 1 (Upper Zone), Punta Gorda ASR,
Charlotte County, Florida.................... .. .........................26
13. Uranium activity ratio through time for Cycles 1A and 2A, Punta Gorda ASR,
Charlotte County, Florida ............... ..............................26
14. Uranium concentration through time for Cycles 1A and 2A, Punta Gorda ASR,
Charlotte County, Florida...................................... ........28
15. Arsenic concentration through time for Cycles 2 and 3, Rome Avenue ASR,
Hillsborough County, Florida ............................................ 30
16. Arsenic concentration through time for Rome Avenue ASR pH-adjusted cycle test,
May-June, 1999 ........................................................ 30
17. Arsenic concentration through time for Cycles 1 and 2 (Upper Zone),
Punta Gorda ASR, Charlotte County, Florida. ................................. 33
18. Arsenic concentration through time for Cycles 1A and 2A (Lower Zone),
Punta Gorda ASR, Charlotte County, Florida. ................................ 33
19. Iron concentration through time for Cycles 1A and 2A (Lower Zone),
Punta Gorda ASR, Charlotte County, Florida. ................................. 34
20. Calcium concentration through time for Cycles 1A and 2A (Lower Zone),
Punta Gorda ASR, Charlotte County, Florida. ................................. 34
21. Manganese concentration through time for Cycles 1A and 2A (Lower Zone),
Punta Gorda ASR, Charlotte County, Florida. ................................ 36
22. Sodium concentration through time for Cycles 1A and 2A (Lower Zone),
Punta Gorda ASR, Charlotte County, Florida. ................................. 36
23. Strontium concentration through time for Cycles 1A and 2A (Lower Zone),
Punta Gorda ASR, Charlotte County, Florida. .............................. 37
24. Sulfate concentration though time for Rome Avenue ASR pH-adjusted cycle test,
May-June, 1999 ......................................... ............ 37
25. Iron and Mn concentrations through time for Rome Avenue ASR pH-adjusted
cycle test, May-June, 1999 ................ ................... ......... 38


APPENDICES

1. Lithologic descriptions of samples analyzed for whole rock chemistry. Wells include
SWFWMD ROMP TR9-2 (W-16618), NW Hillsborough Reclaimed ASR TPW-1
(W-17687), and SWFWMD WRAP 2D (W-16574) ............................ 43
2. Cycle test and drilling activities at the Punta Gorda ASR, Charlotte County, Florida .... 46









EXECUTIVE SUMMARY


Aquifer storage and recovery (ASR) is a cost-effective, viable solution to address
drinking-water shortages in Florida. ASR wells are Class 5 injection wells regulated by the
Underground Injection Control Program of the Florida Department of Environmental Protection.
Six ASR facilities are in operation in Florida and more than 25 additional sites are under
development. Some of the sites include reclaimed water ASR facilities, which are also cost-
effective solutions to local water shortages. ASR is a proposed major component of the
Everglades restoration plan, which calls for the installation of approximately 300 ASR wells in
the Lake Okeechobee region within the next 20 years.


The Florida Aquifer Storage and Recovery Geochemical Study is an ongoing
investigation by the Florida Geological Survey, in cooperation with the Florida State University
Department of Geological Sciences, to examine water-rock geochemical interactions that take
place during ASR cycles. This report includes results from Year Three of the study. Results
from Years One and Two are presented in Williams and others (in review). Year Four of the
project is underway. Water-quality variations and aquifer system characteristics (including three
injection zones) at two ASR facilities, the Rome Avenue ASR (Hillsborough County) and the
Punta Gorda ASR (Charlotte County), are the focus of the current study.


Research presented herein confirms that understanding water-rock geochemical
interactions is important to the continued success of ASR in Florida. Results of this
investigation indicate the following: 1) chemical (including isotopic) variability exists within
groundwaters and carbonates of the Floridan aquifer system; 2) this variability may result in
site-specific geochemical processes affecting ASR well performance (e.g., plugging) and water
quality; and 3) as oxygen-rich surface waters are injected into the Floridan aquifer system, trace
metals such as arsenic (As), iron (Fe), manganese (Mn) and uranium (U) are mobilized
(chemically leached) from the carbonate rocks and withdrawn during recovery. With regard to
the third item, some of the periods of higher metals concentrations in recovered waters are short-
lived, depending on the duration of the injection-storage-recovery cycle. It is significant that
mobilization of U and As into recovered ASR waters has occurred within all three of the
aquifer-storage zones investigated in this study. On the other hand, it is important to emphasize
that only Fe and Mn concentrations (for relatively few samples) have exceeded secondary
drinking water standards (i.e., maximum contaminant levels MCL). With the exception of one
sample, As is not observed to have exceeded the MCL.









The current MCL for As is 50 ug/1. The U.S. Environmental Protection Agency (EPA)
has proposed to lower the MCL for As to 5 ug/1 (Federal Register, 2000). If the As MCL is
lowered*, mobilization of metals into injected and recovered waters may become even more of an
issue from both a regulatory and human health perspective. For example, more than 50% of the
recovery water samples analyzed in this study would exceed the proposed MCL. The EPA
proposal is lower than the 10 ug/1 MCL established by the World Health Organization.


These results underscore the need for further research on the geochemistry of ASR in
Florida. Ongoing research at the Florida Geological Survey will continue to evaluate the
geochemical effects of continued ASR cycling and further characterize the lithology and
geochemistry of the Floridan aquifer system. Although the observations reported herein
concerning water-quality changes during ASR are significant, proper design of ASR facilities
(including installation of monitor wells), as well as proper design and monitoring of ASR cycles
should be able to overcome any human health concerns.


* Note: During preparation of this document for publication, the EPA published a final rule lowering
the MCL for As to 10 ug/1 (Federal Register, 2001). This rule is being re-evaluted by the EPA.









Acknowledgements


This study presents results of Year Three of the Florida Department of Environmental
Protection (FDEP)/Florida Geological Survey (FGS) Aquifer Storage and Recovery Geochemical
study. This research is funded through a cooperative agreement between the FDEP/FGS and the
Underground Injection Control (UIC) Program, Bureau of Water Facilities Regulation, Division
of Water Resource Management, FDEP. The UIC program administers funds from the U.S.
Environmental Protection Agency for this project. The authors thank George Heuler and Richard
Deuerling (UIC Program) for their continued support of this research. We are also thankful for
administrative support from Joe May and Judy Richtar of the FDEP Southwest District Office in
Tampa. Other individuals who have provided valuable administrative support include Mark
McNeal (CH2M HILL), Malachi (Mike) C. Bennett and Marjorie F. Guillory (City of Tampa
Water Department), A. Ed Fox (Hillsborough County), and Mark Abbott and Rob Dennis
(Montgomery Watson Harza). We thank Ed Fox and CH2M HILL for core collected from
the NW Hillsborough reclaimed water ASR in support of this project.


Collection of ASR cycle-test samples by cooperators has allowed fund reallocation in support
of additional laboratory analyses. We are especially grateful to the following people who collected
samples for this project: Brian Fuller (City of Punta Gorda Water Treatment Plant), Pete Larkin
and Marty Clasen (CH2M HILL).


Laboratory analyses for this study were completed at Activation Analytical Labs (Ontario,
Canada), the Florida State University Isotope Geochemistry Laboratory, and the FDEP Bureau of
Laboratories. Conversion of this manuscript to a digital (.pdf file) format was difficult and time-
consuming, and we are indebted to Holly Tulpin and Charlie Steed for their work in bringing this
aspect of the project to completion. We are also grateful for review of this report by the
following FGS staff: Carol Armstrong, Paulette Bond, Jackie Lloyd, Guy Means, Deborah
Mekeel, Frank Rupert, Dr. Walt Schmidt, and Dr. Tom Scott.









INTRODUCTION


This annual report summarizes activities and results of Year Three of the Florida
Aquifer Storage and Recovery (ASR) Geochemical Study. The reader is referred to the Year
One and Two progress report by Williams and others (in review) for background and initial
results of the project, including an introduction to ASR, the geological setting of the study area,
an introduction to relevant geochemical systems, initial research activities, data and interpretations.


Although the first two years of this investigation significantly improved our
understanding of ASR geochemical processes and has resulted in modification of ASR wellfield
design and monitoring, many questions remain unresolved. Summarized below are major
highlights of the initial study (Williams and others, in review) as well as more recent
observations based on their data.


Uranium isotopes are useful tools in modeling the geochemical history and mixing of
the injected surface waters and aquifer system matrix waters (i.e. native water). Ground-
water U-isotopic ratios are the result of a dual porosity aquifer system, which may consist of
three "end-member" waters: flowing intergranular, relatively non-flowing pore and conduit
flow groundwaters. The type and degree of mixing of these waters during ASR cycling,
however, is site specific and varies as a function of the hydrogeologic characteristics of the
area (Cowart and others, 1998).


In the Hillsborough County study area, the use of Sr isotopic ratios to determine the
degree of limestone dissolution in the aquifer system during ASR is inconclusive (Williams
and others, in review). 87Sr/86Sr ratios for the injected waters (from the Hillsborough River)
contain approximately the same ratios as that of the native groundwater, and both waters
indicate equilibrium with Oligocene rocks (-37 to 24 million years old). According to data
from the U.S. Geological Survey National Water Information System, the Hillsborough River
was flowing at base flow levels during March, 1997, when the sample was collected and
analyzed for 87Sr/86Sr. Base flow in the upper reaches of the Hillsborough River in
Hillsborough County comes from the Oligocene Suwannee Limestone (Campbell and Arthur,
1993). This suggests that the observed Oligocene 87Sr/86Sr ratio for the injection water
sample (pre-treated river water, collected 3/14/97) represents waters that have equilibrated
with aquifer system carbonates exposed in the upper river basin. Had there been more
surface water runoff in March, 1997, perhaps the 87Sr/86Sr ratio of input waters would have
been more useful in modeling dissolution in the aquifer system due to cycle testing.











Trace metals are leached from the aquifer system matrix during the ASR cycle testing.
Of the metals analyzed during the first two years of this study (As, Ba, Cd, Ca, Pb, Mn, Na,
Sr, U and Zn), results indicate that As and U are leached by injected, oxygen-rich surface
waters. Other metals may be mobilized during ASR cycle testing, however, the data
presented in Williams and others (in review) are inconclusive. Although short-term As and
U increases are observed, U and As concentrations did not exceed maximum contaminant
levels (MCL) (Williams and others, in review). Additional data reported herein, however,
suggests that leaching of metals during ASR yield concentrations higher than previously
observed and thus warrants further study, monitoring and site specific analysis.


PURPOSE OF INVESTIGATION


The purpose of this investigation is to evaluate the hydrochemical changes that occur
during the ASR process within injected and native waters of the Floridan aquifer system at
these selected sites. Specific goals for Year Three of the study include:


Continue to investigate the nature of water-rock interaction as it affects ground-water
chemistry.
Identify the source and mechanism for mobility of trace metals into the injected waters.
Determine the effect of repeated cycle testing on dissolved metal concentrations.
Continue to examine uranium concentrations and isotopic ratios (234U/238U) in native
aquifer water (when available), recharge water, recovered water, and water in surrounding
wells; explore use of U geochemistry to further understand mixing of pore, native and
injected waters.
Determine the relation between gamma-ray log response and trace metal concentrations
in the aquifer system carbonates.
Provide the Florida Department of Environmental Protection (FDEP) with additional
scientific knowledge on which to base protective criteria for future ASR projects.


The two ASR facilities in southwestern Florida that were initially included in this multi-
year study remain the focus of this investigation: the Rome Avenue ASR facility in Hillsborough
County, and the Punta Gorda ASR facility in Charlotte County (Figure 1). Additional ASR
facilities did not perform cycle testing during Year Three (1999). Analyses of rock samples from
the NW Hillsborough Reclaimed Water ASR facility are included in this report. Cycle testing is
anticipated to begin in the near future at the NW Hillsborough facility.













Rome Ave. ASR site, Hillsborough County


700 0 700 1400 Feet





Punta Gorda ASR site, Charlotte County


Figure 1. Location of study areas.









In a scientific investigation under optimum experimental conditions, data collection is
designed to support the goals of the study. This investigation, however, is appended to existing
activities at relatively new ASR facilities. As such, data collection is not driven by this
research, but rather by the need to place the facilities in full production while meeting all water
quality standards. Limiting factors in this study include gaps in time series data, lack of control
over cycle test parameters (e.g., input/output rates and volumes) and types of borehole samples
collected. Despite these limitations, significant progress has been made toward addressing the
goals of this multi-year investigation.


METHODS


Whole rock chemistry


Whole rock analyses of 67 major, trace and rare-earth elements (REE's) were completed
on 21 samples. These samples were collected from three wells at or near the ASR sites in the
present study. Samples from the Rome Avenue injection well were not available for analysis.
However, a nearby monitor well, WRAP 2D, (W-16574) provided cuttings at five-foot intervals.
At the NW Hillsborough site, samples from the injection well TPW-1 (W-17687) were available
as core segments and cuttings. Samples from another well in Hillsborough County, W-16618,
were also included for regional comparison. W-16618 is located approximately 17 miles south-
southeast of the study area. Uranium leaching experiments were completed on rock samples
from this well (Green and others, 1995). Four of these samples are included in the present
investigation to compare high-U samples (and high leaching potential) with variations in other
trace metals in the carbonate matrix.


Rock samples were collected and analyzed based on the following criteria: lithology,
gamma-ray activity and stratigraphic unit. Representative samples of limestones and two
dolostone samples were collected for chemical comparison. The presence/absence of organic
material in the samples was also noted. Using gamma-ray logs, samples were selected "on peak"
and "off peak." This qualitative data is denoted on Table 1 as "HI" and "LO," respectively, in the
column labeled "Gamma." Although higher concentrations of U, Th, and K are expected with
increasing gamma-ray activity, the relation between other metals (and organic content) and
gamma-ray response is not well understood for these rocks. All samples were collected from the
Suwannee Limestone, which is utilized as the ASR storage zone in the region.


XRAL Laboratories (Toronto, Canada) completed all whole rock chemical analyses,























OX o
LJL















a





OX
-
2.























O LL


0a w








o w
g X











0 LJ



w-e
0 x


G)c

0 "-'










C C




x






S z u





















N X


LcO W 0 0 (N CO CO (N I N (N CM L CM W fl C (O N ( CN Cl C)




7 CO m N- C- (N ( N 0- N (N 7 7 7 m
dddddddddddddddddddd





















dd dddddddddddo a
0 00000000000000000000
d 6dddddddddd66dddddddd





d d d d d dddd dddddd d ddddd
0O0 0000000 .0000000000





Oo 0000000000000000000
0ocoo .000006o00 000000000










CO zl O CO CO CO I C C 0) N
0000 10- -1- c o 0` ':O











ddgdddddddddQQ^^^^^^^^





CO D CD C D- C0 00
uq zl 00 zl LO w CO ;, m M0 W M O u u N I- M
0000000c00100000000000

66 660 66 6666666


EEE


oo
s-e



*"n I
2 2
Ell






Io 2 .0





LL 6a



w LU LU LU WLU LU




ID
a)


ZZ>-Z)>>>>z>>>>zZZZZZz


LO 0 LO LO LO OOLOOLO


CO N C CN CO 'C C CoN C CO CO Co I-




CO CO CO CO CO CO CO CO CO CO CO CO
CO O CO d V LV CO~)) ^- - ^.^.^.^- - -.' t' t t' t


CD CD CD CD [^ ^~[^ [~ ^ [~ ^~ [~ ^ [~ D D D CD CD CD CD C


C) C
0 N



a _o






x 11 D 15
LU C LL
-
W U









utilizing a high precision, multi-method technique. Analytical methods for each element are
listed in Tables 1 through 3. Due to limited sample availability, borehole cuttings were analyzed
in addition to core chips. Cuttings were handpicked under binocular microscope to minimize
contaminants and maximize lithologic homogeneity. All drill-bit contact surfaces were removed
from core samples taken from W-16574 and W-17687.


Water chemistry


Uranium analyses were performed at the Florida State University Geochemistry
Laboratory in Tallahassee, Florida. This laboratory, under Comprehensive Quality Assurance
Plan #960078-5, is approved by the FDEP. Thirty-four 5-gallon (18.9 liters) water samples were
collected during cycle testing and delivered to the laboratory, where samples were then split for
uranium analyses. Detailed sample preparation is described in Williams and others (in review). The
234U/238U activity ratio and uranium concentration in each sample was determined with alpha

pulse-height spectrometry.


Seventy-five trace metal samples were collected in 250 ml amber-colored glass jars,
preserved with 5% HN03 per FDEP groundwater sampling protocol. Unless otherwise noted,
all water samples were unfiltered. The samples were first transported to the Florida Geological
Survey, and then to the FDEP Central Laboratory. All samples were analyzed for trace metals
by inductively coupled plasma emission spectroscopy at the FDEP Central Laboratory. Some
samples did not meet FDEP sample preparation protocol because they were splits taken from
the five gallon uranium samples. Interpretation limitations for these particular samples are
addressed in a later section.


ARSENIC GEOCHEMISTRY


This section on As geochemistry is included herein based on the results of the Williams
and others' (in review) report to the FDEP Underground Injection Control Program, as well as
subsequent events such as maximum contaminant level (MCL) exceedences and changes in
ASR facility design and testing. This information is provided as an overview to describe the
natural occurrence of As in the hydrogeochemical environment.


Arsenic contamination in groundwater and As-associated illness in humans has been
observed around the globe since the late 1950's (Das and others, 1996). Perhaps the most recent
and notable occurrence of As contamination during the last decade has been the health crisis in
















CL
oft





2- Ev -
0u




CL

0 .






Eo





we a.


U-
2








O E


LU-0





















2
m E



















EO M .
U
2


















w

~X ENc
OE
U0
E















,Oo
LL
























C,X(
Z


LL
E c,4




LLE
m~m


LL





X E
m 0. 0

LU 0



En,







EO



a m








E 0


LO m m md d dc c dd d





-CO co





0 ccO r- c z C) ) Lo Lo z cc cc cc cc cc

















li CO C i COCO


cc cc cc c O Nil') r- cc cc r- CL j N Ni U Nl~ co Ni co) LO O










co~~~~~~C Lo Ccc 0)c DZ-Uc z- OD 0 ) CD'o) occc oC




0)~~ ~ ~ ~ NoL cc a Z C') c Lo co Cj co C) 0 D N







O~OO--)-OO----OO--NNNO

Lo D cc o u zi co
r- co 0) 2 c~~~~





















C') LL C') LL 00L L LL L L L 000 L L LL LL 0 L LL 0 L LL









C' O 'NCD) O U CDNCD CO CD LO



co OD mO Nil C CD CO CD CD CD CD Nil Nil Nil LO U CD U CD Nl~ U Nl


CO 'CO CO CO r- r- r- r- r- r- r- r- r- r- col col col col col col col col
coc oc oc DO DO r-r-r-r-r-r r
cc c c c c c c c c c c c c c U U U L L L L L
R R R R : : : : : : :


E

C)


0







c c
N a
II

LU C LL


A 0


EEE


oo o



983
0,,
L22 a




o)o)- 8o
E 0
O OO



0 20
o> r
oe

-a 0
-j E E t


I) y a-< LL LL u
x 0-j3



0

E
a)a






-C
Ci)
















ZE ,











w L D d d d d d d d d d d d d d d d d d d C"D







ZE

z 0 d d dddddddddddddddd




























. .
XQ. oLoo oo --ooo D oooo D



















-0 EC I -O 0 -O 0 -O 0 -" -00 -0 -0 -0 -0 0-0" LO-CD >








SE C - a --
SoCIo oIh oCI Ih-h-h-h-hI-- O-O o 0)o(


(U 0. 0
E








0 0 1
Wa) .O
E









00
0










5o CD co'sf D DLo D D D D'sf'sf's DLO LO DO LO LO LOL
0) t- CNJ CN O'r zl CO CN 0) cl co o rC -




OC D O CD CD CD CD CD CD CO CD .- r- r- r- LO LO LO Co I
IU

LuCL















LL




' LS c T-c c c L c







LU


D 5 mS N mo Lo -ood c o
w

I-
w






-o E o --oo-ooooD ooo o-




SEE ci coN coooo-1oo-ooooc o
00



























.z E
(LU C CCC C C







0 0 0O DN C

Qcj, 0






















--e-
w.2
0) cw
00 E 0 00000000000000000 ^^

(B) O O) 5


ES 0



0, 0



o co














z - l-I-'- -a -0 CO
S.w i



















-J | o ( d d o 3COcd ocod d 0 C


S UJ c d d. .66 ddddd ddddddddd d d .
E cLN C o C ~ -0 r




Q c CQ CQ 040OOOOOCOOOCOOOOOO0O 000

-wo d d oocoo a a a a a oo o a 0a






x L0o cc0c 00.0000. 0000000000
a









C 0
o E m d
w Er dddd OdddC4COC CCdddddddd dd0d o





















z E
S, .... . ...... ... .... ..













x
- o d 0000 0000 za00C00 m0












SL L00000 U00
00 0

X0 CW
CC




( IE LU :S :5




























.) ol ?
^-E 040-0-0r0r0 0 0 (D




U n5E 2L I 1
0 0 a0 a a00000000 000
























o 2

o 0U_
C)U




E3-0 0 00000--

CO CO 1 C CO 0 1 N CO C O w r
" aa a 0 0 a 0 0 0 0 0 0











00
C o w0.4 E 0O 004M0) 0)000 00 c








0 V V) I I
Fl a 51' 90





o l~
Q- oO E








0- CO 0C N
COOOOOO 4OOOOwi
(-fl


0d COOC ONN -N --N -N-Ntt mtt ~t ~ 4
0 E CCCCCCCCCCCCCCCCCCCCC1010010010110
CCCCCC--NN--NNN--NCCCCCCCCCC
z----------------------------------------------inI
cW aLL









West Bengal, India, and in adjacent Bangladesh. Approximately one-third of the 50 million
people in Bangladesh are feared to be at risk from chronic exposure to As (Bhattacharya and
others, 1999a). Along with other incidents, the India crisis has led to an increased awareness
and response in the scientific community. Since 1998, organizations in the United States (e.g.,
Geological Society of America and the American Geophysical Union) and other countries have
held special sessions on the subject of As in groundwater. In 1999, the National Research
Council released their report, Arsenic in Drinking Water.


As part of the Safe Drinking Water Act, the U.S. Environmental Protection Agency has
set an MCL of 50 ug/1 (ppb) for As. A proposal to revise this standard to 5 ug/1, however, is
under review (Federal Register, 2000). For comparison, the World Health Organization MCL
for As is 10 ug/1.


Arsenic in the hydrogeologic environment


In groundwater, two oxidation states of arsenic are arsenate (As5+) and arsenite (As3+;
Hemm, 1985). In most natural waters, As3+ exists as arsenite hydroxide (As(OH)3) and arsenous
acid (H3AsO3), whereas the less mobile As5+ occurs as arsenic acid (H3AsO4). Several
methylated acid species exist as well (National Research Council, 1999). Arsenic is present in
a variety of solid phases in the hydrogeological environment. Summarized below are examples
of these As-mobilization mechanisms and associated mineral phases:


Oxidation ofsulfide minerals such as pyrite and arsenopyrite (e.g., White and Scully,
1999; Nickson and others, 2000). Pyrite (FeS2) contains the following trace elements as
impurities or in solid solution: Cu, Pb, As, Zn, and Mn (Deer and others, 1992). Nickel and
Co commonly substitute for Fe in the pyrite lattice. Thomas and Saunders (1998) report As
concentrations in authigenic pyrite up to 1000 ppm. A variant of this category is oxidation
of framboid masses (Kolker and others, 1999), which are microscopic, spheroidal clusters of
sulfide minerals associated with organic material.


Fe andMn hydroxide grain coating desorption/dissolution (e.g., Schlottmann and Breit,
1992; Welch, A.H., 1999). These coatings are reported to contain up to 30 ppm As (Thomas
and Saunders, 1998). Nickson and others (2000) report As concentrations in grain coatings
(recalculated as pure FeOOH) in excess of 500 ppm.


Oxidation-reduction of organic material (e.g., Welch and Lico, 1998; Bhattacharya and









others, 1999b; Stollenwerk and Colman, 1999). Redox reactions can mobilize organically
completed As.

Arsenic occurrences in groundwater are also related to biological transformations
(Maddox and others, 1992; Ahmann, D., 1999; National Research Council, 1999) and
anthropogenic activities such as contamination from pesticides (Maddox and others, 1992),
tannery waste (Brandon and others, 1999) and landfill leachate (Colman and Lyford, 1999).


Recent studies document As mobility during artificial recharge of aquifers (Stuyfzand,
1998; Ruiter and Stuyfzand, 1998, Brun and others, 1998). In a study of 11 deep well recharge
experiments, oxidation of pyrite is reported, resulting in mobilization of As, Co, and Ni (Stuyfzand,
1998). Stuyfzand found that As remains relatively mobile while Ni and Co are less mobile and
likely coprecipitate or adsorb on to Fe-hydroxides further away from the injection well. Although
Zn is associated with pyrite, it did not mobilize, possibly due to scavenging by Fe-hydroxide
(Stuyfzand, 1998).


In a similar study utilizing separate recharge and recovery wells, Ruiter and Stuyfzand
(1998) also report oxidation of pyrite leading to mobilization of As, Co and Ni. In their study,
Zn is also reported to be mobile. Of the mobilized metals, only As reached the most distant
recovery well (relative to the recharge well) due to its higher mobility (Ruiter and Stuyfzand,
1998). An increase in SO42- and diminishing 02 and NO3- is also attributed to pyrite oxidation.
Of the several important geochemical reactions that may occur during artificial recharge listed in
Stuyfzand (1998), the following reaction represents oxidation of pyrite leading to precipitation
of Fe-hydroxides and increases in sulfate in the injected water:


3.75 02 + FeS2 + 4HCO3- -> Fe(OH) 3 + 2S042- + 4CO2 + 0.5H20.


This reaction is one of many listed in Stuyfzand (1998) that may be relevant to
geochemical processes taking place in ASR wells utilizing the Floridan aquifer system.


RESULTS AND DISCUSSION


WHOLE ROCK CHEMISTRY


Multi-method, multi-element chemical analyses of Suwannee Limestone samples from
Hillsborough County are listed in Tables 1, 2 and 3. The tables also show minimum, maximum
and average concentrations for the samples. These samples provide reconnaissance chemistry of
a variety of carbonate lithologies (Appendix 1). A statistical evaluation of these and additional
data is planned.











Two of the samples (FGS-12 and FGS-13) are dolostones, with MgO concentrations up
to 19 weight percent. These two samples have slightly higher loss on ignition (LOI) values than
the limestones. Moreover, the samples are depleted in Ca, Sr and REE's and enriched in Zr
relative to the limestones. Both samples contain trace amounts of organic material and represent
zones of higher gamma-ray activity.


Of the other carbonates analyzed, one sample, FGS-3 contains trace metal
concentrations exceeding all other samples in this study. This sample contains the highest
reported values for Al, Si, K, Ti, As, Sr, Zr, B, Ni, Cu, Zn, Cr, Co, Se, Mo, Sb, Hf, Th and U
and contains the lowest LOI and LREE concentrations. High Si, Al and K concentrations
suggest that this sample has a siliciclastic component. Possible heavy mineral content (likely in
a fine-sand to silt-sized fraction) in the siliciclastic component of this sample may account for
maximum observed concentrations of Ti and Zr, among other metals. The As concentration in
this sample is 10 ppm. Of the U content in this sample (FGS-3; U = 28 ppm), approximately
one-third of the U is leachable under oxidizing (laboratory) conditions (Green and others, 1995).
Sulfur was below detection in this sample. Samples FGS 1-4 were analyzed using an analytical
package that could not resolve concentrations of S, Cl, Hg and Cd in these carbonates. The
choice of analytical packages will be re-evaluated for subsequent chemical analyses to maximize
detection of S, which may reflect, in part, pyrite content. Mineralogical investigation of these
samples is planned for Phase 5 of the project.


Trends observed in most bivariate plots tend to be anchored by the high-concentrations in
sample FGS-3. Although many of these "trends" are not discussed herein, a few plots indicate a
possibly important trend that is not entirely biased by FGS-3. Figure 2 shows the relation
between Fe203 and As, suggesting a possible presence of Fe-sulfides and/or Fe-hydroxides in the
samples. Arsenic is present in both of these phases. Zinc is also associated with sulfides. A
positive correlation exists between Zn and As as well (Figure 3).


Available data suggest no relation between As and U concentrations, organic content or
high gamma-ray activity in these samples. There exists, however, a relation between high
gamma-ray activity and organic content within the samples. In general, those samples with high
gamma-ray activity contain organic material and those samples with low gamma-ray activity do
not contain observable organic material. Of 21 samples, six do not fit this generalization.
Additional data and a statistical evaluation may further substantiate this observation.













0.45



0.40



0.35



0.30



0.25



0.20 -
uS


0.15 -.,-



0.10



0.05



0.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00
As (ppm)


Figure 2. Iron oxide (Fe203) As variation in Suwannee Limestone, Hillsborough County, Florida



14.00




12.00




10.00




8.00

a. *


6.00




4.00




2.00




0.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00
As (ppm)


Figure 3. Zinc As variation in Suwannee Limestone, Hillsborough County, Florida









The range of chemical compositions in these rocks provides some insight into the
chemical heterogeneity in the aquifer system in this region (Tables 1, 2 and 3). Based on this
variability, one may conclude that water-rock interaction modeling based on laboratory
experiments may not completely represent the hydrogeochemical processes taking place over
large areas during ASR activities. Such laboratory ground work, however, is important and
geochemical analyses of core samples are required to provide control in these models and to
help understand observed water quality trends during ASR.


Rare-earth elements


To allow comparison of REE distributions between samples, all REE data in this report
have been normalized to the North American Shale Composite (Grommet and others, 1984).
The REE patterns for samples from the three boreholes in this study (Figures 4, 5 and 6) all
cluster around an NASC normalized value of 0.1, which is typical of carbonate compositions
(Bellanca and others, 1997). Moreover, the nearly ubiquitous negative Ce anomaly is consistent
with a marine depositional origin (Bellanca and others, 1997), and indicates that post-
depositional chemical changes may not have greatly influenced the REE distributions.
Variations in REE patterns do, however, exist relative to trace metal concentrations and depth.
The somewhat fragmented REE pattern (i.e., elements not analyzed or detected) in Figure 4,
when compared to the samples from other wells (Figures 5 and 6), is due to use of a different
analytical package for these samples.


A comparison of all patterns indicates that the high-As sample, FGS-3, as well as samples
FGS-5 and FGS-6 plot at the high end of the REE-normalized range. Upon analysis of all
available data, these three samples share two common characteristics. First, the samples contain
organic material. This characteristic, however, is not unique to these samples. Seven other
samples also contain organic (possibly as grain coatings), thus the enrichment may not be related.
The second characteristic is the Cr content of these samples, which is higher than any other sample
in this data set. The relation between Cr and total REE is the subject of further study.


Slight differences in the slope of each REE pattern indicates enrichment or depletion in
light or heavy REE's. Ratios of Sm to Yb and to Pr allow an alternative to comparing REE
patterns for heavy (HREE) and light (LREE) pattern enrichment or depletion. This distribution
of REE ratios (Figure 7), indicates that Suwannee Limestone samples exhibit a variety of
enrichment from LREE to HREE. At present, no discernable relationships are observed
between LREE and HREE and other chemical or physical characteristics of the rocks. None of












La Ce Pr Nd Sm Eu Gd Dy Er Yb Lu


10











--- FGS-1
--W-- FGS-2
-A-- FGS-3





^ ^ // XS^^^--------------- -------- ^o
S,------ FGS-4







I I



001

Figure 4. Normalized rare earth element distribution in Suwannee Limestone samples from
W-16618, Hillsborough County, Florida.






La Ce Pr Nd Sm Eu Gd Dy Er Yb Lu

10







--- FGS-5
1 FGS-6

-- FGS-7

-A- FGS-8

FGS-9

X FGS-10
m
0-.1 -------- FG S-11

0I-- FGS-12

IX-- FGS-13

--- FGS-14



0.01



Figure 5. Normalized rare earth element distribution in Suwannee Limestone samples from W-17687,
Hillsborough County, Florida.












La Ce Pr Nd Sm Eu Gd Dy Er Yb Lu


-4-
-,--

-4-o-
m
m - -
0.1 z
o
0)


FGS-15
FGS-16
FGS-17

FGS-18
FGS-19
FGS-20
FGS-21
FGS-22


I 1I 0.01

Figure 6. Normalized rare earth element distribution in Suwannee Limestone samples from W-16574,
Hillsborough County, Florida.



10


LREE Enriched


MREE Enriched


MREE depleted


HREE Enriched


SmnPrn
SmnIPrn


Figure 7. Normalized rare earth elements ratios showing Suwannee Limestone samples relative to fields
of enrichment and depletion (LREE light REE, MREE middle REE, and HREE heavy REE).


J-
1
E
v


I:









the samples in this study are depleted in middle REE's (MREE; Figure 7).


Total REE concentrations in the samples are estimated from the sum of all REE data
plus the detection limits for those elements reported as "n.d." in Tables 1, 2 and 3. Any error or
uncertainty associated with this estimation is minimal compared to observed trends. In fact, the
same trends are observed when replacing "n.d." with null values.


Figure 8 is a plot of total estimated REE concentrations versus loss on ignition (LOI).
LOI is a measure of the volatile content of a lithologic sample. Among other minor
constituents, LOI includes structural water from phyllosilicates, adsorbed/absorbed water, CO2
from carbonate material, and reflects weight changes due to oxidation of organic and metals
(e.g., Fe). Most carbonates in this study have LOI values centered around 43 percent. Five
samples, however, show a negative correlation with REE concentrations. This variation is the
result of two overprinting lithologic characteristics: 1) the high LOI samples are the only two
dolomitic samples in the data set, and 2) low LOI (<42%) samples increase in Si and Al with
increasing REE content. The latter association is also observed in a plot of SiO2 plus A1203
versus total REE (not shown).


Samples from W-17687 also indicate lithologic/depositional control of REE
concentrations. With one exception, total REE content decreases with depth from 310 to 430 feet
below land surface in that borehole (Table 3). Measured trace metals do not exhibit such variation
with depth. Further sampling may or may not confirm this trend. Significance of the possible
relation between high Cr, (and As in one sample) and total REE content may become more
apparent during chemical investigation of mineral phases in the rocks during future FGS studies.


WATER CHEMISTRY


Uranium


In an aquifer system, the behavior and relationship between uranium concentration and
uranium isotope ratios released by the action of aggressive waters is thought to be a function of
the geochemical history of the rocks and the permeating waters. An idealized schematic model
of the behavior of uranium under different conditions is presented along with a comparison
between actual measurements and those predicted by the model.


In the following discussion, it will be assumed that recoil of the nucleus resulting from














25



20

C \ Limestones

u15

Siliciclastics increase? *

10

S' Dolostones

5



0*
39 40 41 42 43 44 45 46 47
Loss on ignition (weight %)

Figure 8. Total rare earth element (estimated, see text) distribution in Suwannee Limestone samples versus
loss on ignition (LOI.)


the alpha particle decay of an atom is the most prominent mechanism affecting uranium isotope

fractionation. Selective leaching of 234U atoms can also play an important role in fractionating

uranium isotopes in locations where oxic waters have continuously permeated a rock; however,

in the present study area, it is probable that reducing waters have been in contact with the

aquifer system matrix until the introduction of oxic waters by injection.


The redox environment is important because the oxidized uranium ion, U6+, is highly

mobile whereas the reduced uranium ion, U4+, is quite immobile. In oxygenated water uranium

is easily transported but if that water becomes reduced, the uranium contained in it will

precipitate. This being the case, selective leaching of any isotope of uranium is suppressed. On

the other hand, ejection of the nucleus resulting from the alpha decay of 238U, which is in effect
234U, is not inhibited because it is a physical mechanism, not a chemical one. If the 234U is

ejected from a rock into moving, albeit reduced water, the rock containing the uranium will

become depleted in 234U relative to 238U because the 238U can only become mobile through

chemical means and some portion of the 234U has been transported away. Therefore, if

aggressive oxygenated waters invade a reduced aquifer, the initial uranium released from the

rock will usually be deficient in 234U.









With the above in mind, four scenarios are presented which provide insight into
behavior of the uranium isotopes as shown by the measurements made of the waters recovered
from several ASR sites (Table 4). In all the cases, the water in the aquifer is assumed to have
been reducing before the injection of oxygen-rich surface waters. In two of the cases (1 and 2)
the water bathing the rock is moving and in the other two cases (3 and 4) the water is assumed
to have no significant motion (Figure 9). Each of the diagrams is meant to convey the uranium
concentration and isotope ratio (234U/238U) for a single grain as oxic water mobilizes the
uranium at deeper and deeper levels within that grain. In a manner of speaking, we are peeling
off onion-like layers that contain the variously affected uranium. It should be noted that the
concentration of uranium depends almost totally on the amount of 238U present and is not
measurably affected by the deficiency or excess of 234U.


Case 1. A grain containing uranium homogeneously distributed within it (Figure 9).
The nucleus recoiled from a 238U atom originally sited at the grain-water boundary has a
50% chance of recoiling into the water. The nucleus recoiling from a 238U atom located
a small distance below the grain-water boundary would have a smaller probability of
being ejected into the water and so on down to a distance equal to the recoil distance at
which point no recoiling nucleus would be able to be ejected. In this case 0.50 would
be the lowest uranium isotope ratio and it would be seen first (as the uranium in the
outermost layer of the grain is mobilized), followed by ratios tending towards 1.00, the
ratio of the interior of the grain.


Case 2. A grain having a very thin coating or layer of relatively high concentration
uranium resting on top of a grain having uranium homogeneously distributed within it
(Figure 9). The coating might consist of organic material or might result from
precipitation of uranium due to changing redox character of the aquifer in the past a
mechanism responsible for the formation of many sedimentary uranium ore bodies. It is
assumed that the layer is significantly thinner than the recoil distance of recoiling
uranium nuclei. In this case the lowest 234U/238U ratio can be less than 0.50 because both
above and below the coating the uranium concentration is less than that of the coating.
Those nuclei recoiling into the water will be carried off as in Case 1. Those recoiling
downward into the grain engender a zone in which the 234U is in excess relative to 238U
and below the recoil distance the 234U/238U ratio is 1.00. Invading oxic waters first
mobilize the coating so that the concentration rises rapidly but then decreases as the
coating is dissipated. The concentration is then dictated by mobilization of uranium from
the underlying unenriched rock material. The zone directly beneath the high









Table 4. Uranium analyses for Cycle Test samples from Punta Gorda ASR. Uranium detection
limit is 0.01 ug/1


Sample No.
(FLW)


NATIVE
5173


INJECTION
5174
5176
5195
5196
5197
5198


RECOVERY
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5193
5192
5194
5200
5201
5202
5203
5204
5205
5206
5207
5208


Date


6/28/99 6:00 PM


7/1/99 12:00 PM
7/19/99 10:00 AM
8/17/99 2:30 PM
8/18/99 11:00 AM
8/19/99 12:00 AM
8/30/99 1:00 PM



7/22/99 9:15 AM
7/22/99 10:00 PM
7/23/99 10:00 AM
7/23/99 10:00 PM
7/24/99 10:00 AM
7/24/99 10:00 PM
7/25/99 10:00 AM
7/25/99 10:00 PM
7/26/99 10:00 AM
7/26/99 10:00 PM
7/27/99 10:00 AM
7/27/99 10:00 PM
7/28/99 10:00 AM
7/28/99 10:00 PM
8/5/99 12:00 AM
8/5/99 12:00 AM
8/6/99 12:00 AM
8/6/99 12:00 AM
9/1/99 2:30 PM
9/1/99 10:00 PM
9/2/99 2:00 PM
9/2/99 10:00 PM
9/3/99 10:00 AM
9/4/99 10:00 AM
9/5/99 10:00 AM
9/6/99 10:00 AM
9/8/99 9:30 AM


U
(ug/l)



2.99


0.03
0.025
0.064
0.056
0.708
0.038



0.027
2.03
3.79
4.75
5.79
5.88
7.93
7.92
7.23
7.08
7.85
7.42
7.2
7.37
3.6
4.08
6.23
7.1
1.56
2.07
3.76
3.98
4.58
3.9
2.92
2.36
2.14


Activity Ratio
234U/238U


0.85


1.21
2.4
1.37
1.02
0.93
2.4



1.08
0.74
0.76
0.81
0.81
0.84
0.83
0.83
0.86
0.87
0.85
0.84
0.84
0.84
0.82
0.85
0.87
0.85
0.85
0.86
0.87
0.91
0.9
0.88
0.9
0.88
0.96

















//


.jfj ;~


Maximum recoil
distance through CASE 1
the mineral


Thin
S Layer


CASE 2


NO WATER MOVEMENT


AR0
>1.0


NO WATER MOVEMENT

1ThinLaoyer
. ............. ....... .... . .. ......... .... .. .... . L ye


<1.0
iii~i 7/~i


CASE 4


LEGEND

0*4f+ Groundwater
flow
High U
thin
coating
AR 234U/238U


Recoil
vector


Figure 9. Uranium mobilization models for single carbonate grains (coated and uncoated) exposed to
flowing and non-flowing groundwater. Cases 1-4 reflect conditions that may occur during ASR.


>1.0

1.0O


CASE


h _n


~ ; ~~...,:; ^~*\i
...,. ....,.-
i - r.
`~":


OW* /
I'lllt









concentration layer has a 234U/238U ratio greater than 1.0 because the number of nuclei
that recoiled down into this zone far exceed those moving upward.


Case 3. In this case and in Case 4, the water surrounding the grain is both reducing and
essentially "stagnant" (Figure 9). In Case 3, as in Case 1, the rock material is
homogeneous with regard to uranium distribution. However, in this situation, for those
recoiling nuclei that reach the water there is no transport away from the grain but rather
an eventual precipitation of the 234U on the surface of the grain. The effect is that all the
nuclei that reach the water end up on the grain boundary so that invading oxic water will
first mobilize uranium which is enriched in 234U so that the 234U/238U ratio will be greater
than 1.00. After this 234U-enriched zone has been mobilized, the uranium will have a
ratio of less than 1.00, even approaching 0.5. Finally the ratio will approach 1.00 as the
limit of recoil within the rock material is reached. Because 234U does not contribute
significantly to total uranium concentration, the initial concentration will be high.


Case 4. In this case, as in Case 2, an initial coating with a relatively high concentration
of uranium is hypothesized (Figure 9). In such a situation, those nuclei recoiled into the
water will eventually be precipitated onto the grain boundary and those recoiled
downward will generate a zone enriched in 234U. Invading oxic water mobilizes the
high uranium concentration coating first so that the 234U/238U ratio will initially be less
than 1.00 (although probably not approaching 0.5) followed by the mobilization of the
zone just below the high uranium concentration coating, which has a ratio greater than
1.00. Finally the ratio will approach 1.00. Because there is a coating, the initial
concentration will tend to be relatively high, followed by a lower concentration,
reflecting the difference in uranium concentration between the coating and the
underlying grain.


The scenarios described above are based on a single grain being acted upon by waters
that promote the mobilization of the uranium present. In an actual aquifer the mobilizing water
affects many grains, not all of which are identical. Further, the injected water can mix with
native waters or with waters that affected the same grains, but at a different time. Nevertheless,
the wide range of isotopic values and changing uranium concentrations appear to reflect the
behavior suggested by some of these scenarios.


Rome Avenue


The Rome Avenue, Tampa (Hillsborough County) ASR Test Production Well (TPW-1)









provided samples that appear similar to those of Case 2. No samples were taken during the first
cycle of testing. The initial isotope ratios measured during Cycle 2 were somewhat below 0.5 and
as more of the injected water was recovered the ratio increased such that the final three samples
had ratios of about 0.85 (Figure 10). The uranium concentration increased in the first few samples
taken, peaked, and then decreased (Figure 11). It is likely that the samples collected during Cycle 2
were somewhat modified by the previous exposure of the aquifer rocks to injected oxic waters.


Cycle 3 consisted of a long period of injection followed by storage. The only discharge
samples that were available from TPW-1 were taken during a short period of recovery. Each of
the five samples taken during this time had a uranium concentration of about 1.0 ug/1 (Figure
11) and an isotope ratio which was very low (approximately 0.38; Figure 10). These values
are very similar to the initial samples taken during Cycle 2 recovery. The interpretation of
these low-isotope ratio samples is that they are found in injected oxic waters that have had
time to penetrate into the micropores found in all aquifer rocks and have mobilized the uranium
located there. With increased flow back to the well (as in Cycle 2) the isotope ratio increases
and the concentration increases as the uranium is mobilized from the previously somewhat
leached grains along the major conduits or flow passages.


Punta Gorda


Two injection zones were tested at the Punta Gorda (Charlotte County) ASR facility.
Appendix 2 summarizes relevant drilling, cycle testing and chemical modeling activities at this
site. The upper zone tested at Punta Gorda (Cycle 1) seems to fit Case 3 in which the recovered
waters first have an isotope ratio greater than 1.00 (Figure 12). Relatively small amounts of
water were input during Cycle 1 injection, thus only the first part of Case 3 was observed,
suggesting that native water in that zone migrated little, if at all. This is consistent with the
assumptions of Case 3.


The lower injection zone at Punta Gorda (Cycles 1A and 2A) is more difficult to assign
to a particular scenario. The initial isotope ratios are less than 1.00 but not in the vicinity of 0.5
and, in fact, they remain more-or-less constant through time (Figure 13). A possibility is that
some water movement has occurred in this zone (which has a greater acceptance of injected
water than does the upper zone) so that some portion of the recoiled 234U was transported away
in the past. This would result in an initial mobilization ratio between 0.5 and 1.0. A
combination of environments combining Cases 1 and 4 approximate this suite of measurements.
























OSZOW-1
STPW-1
OSMW-1
A2-D
XTRW
XPTRW


12/1/96 3/11/97 6/19197 9/27/97 1/5/98 4/15/98 7/24/98 11/1/98



Figure 10. Uranium activity ratios through time for Cycles 2 and 3, Rome Avenue ASR,
Hillsborough County, Florida.


SSZOW-1
A2-D
DSMW-1
eTPW-1
XTRW
XPTRW


0 !l A- r "-I X-.X--x I
12/1/96 3/11/97 6/19/97 9/27/97 1/5/98 4/15/98 7/24/98 11/1/98




Figure 11. Uranium concentration through time for Cycles 2 and 3, Rome Avenue ASR, Hillsborough
County, Florida (data source: Williams and others, in review). All concentrations are below the
proposed 20 ug/1 MCL for U.






























* ASRwell
-Native
X PTRW
X TRW


7/22/1998 7/24/1998 7/26/1998 7/28/1998 7/30/1998 8/1/1998 8/3/1998 8/5/1998 8/7/1998 8/9/1998


Figure 12. Uranium activity ratio through time for Cycle 1 (Upper Zone), Punta Gorda ASR, Charlotte
County, Florida.


* Recovery AR
* Injection AR


6/29/99 7/9/99 7/19/99 7/29/99


8/8/99 8/18/99 8/28/99 9/7/99


Figure 13. Uranium activity ratio through time for Cycles 1A and 2A, Punta Gorda ASR, Charlotte
County, Florida.


16



14



12


08



06


04 -
7/20/1998


NATIVE WATER VALUE

















0
0
*








x


2.5





2





S 1.5

0
CO



> 1




0.5


Cycle 1A Cycle 1A Cycle 2A Cycle A
Injection Recovery Injection Recovery
















I


061
6/19/


9/17/99


I


99











Thus far in this multi-year investigation, cycle testing at Punta Gorda (Lower Zone)
provides the only opportunity to compare chemical trends in two successive cycles. In Cycle
1A, U concentrations approach 8 ug/1, whereas in 2A, the maximum U concentration observed is
4.6 ug/1 (Figure 14). The observed trends indicate leaching of total U from the aquifer system
matrix. Significance of the observed decrease in maximum U values between Cycles 1A and
2A is difficult to ascertain. Had cycle testing conditions been consistent (and the results directly
comparable), the data would suggest that with continued cycling U is being successively
depleted from the source and may ultimately decline to native groundwater values. Conditions,
however, were not consistent; differences exist in injection duration, injection and recovery
rates, and probable FeOH precipitation in the borehole (see Appendix 2). Further study under
more controlled conditions is required to determine if U concentrations (and other metals) in
recovered waters decrease with increasing ASR cycling.


Other trace metals


Rome Avenue


Williams and others (in review) report As values up to 36 ug/1 during the Cycle 2
recovery phase for TPW-1. Although the observed values for that cycle were below the current
As MCL of 50 ug/1, the increase in As warranted further study at this site. Subsequent to the
first progress report (Williams and others, in review), additional Cycle 2 samples were analyzed.
These additional Cycle 2 samples were initially collected as 5 gallons samples for U analysis.
These splits did not meet FDEP sample preservation protocol; however, they were the only
samples available to improve resolution of the observed Cycle 2 trend. It was hoped that
analysis of these additional samples would better delineate the behavior of As, at least to the
degree that changes in water quality monitoring could be suggested, if necessary. Given that the
Cycle 2 samples did not meet FDEP protocol, the results are only considered an indication of
the geochemical processes that occur during this ASR cycle.


The As distribution through time for Cycle 2 recovery and all of Cycle 3 is shown in
Figure 15. This data set includes 24 new analyses (Table 5). Because Cycle 3 took place over a
period of only three days and likely represents only a partial recovery trend, a quantitative
comparison of results from the two cycles is not possible. The pre-treated river water sample
(PTRW) represents an As concentration for injected water (< 1 ug/1 As). During the recovery
period for both cycles, a distinctive increase in As is observed, with one sample exceeding the











































6/19/99 6/29/99 7/9/99 7/19/99 7/29/99 8/8/99 8/18/99


8/28/99 9/7/99


9 ------------------------------------------------------------
Cycle 1A
Recovery
8 _
Cycle 1A Cycle 2A Cycle 2A
Injection eS Injection Recovery
7


6-


5-
5 --------- ----------- ----- -
S

4-


3 -


20
2 ------------*-------------
3 --------------------------------------------- -- ----

1 _________________ _____________________________ *
2 -------------------* ----------------------< ----


9/17/99


Figure 14. Uranium concentration through time for Cycles 1A and 2A, Punta Gorda ASR,
Charlotte County, Florida


As MCL. (This sample, although significant, did not meet all FDEP sampling protocols.)

Although the cycles did not continue to the extent that native groundwater As levels (< 2 ug/l;

Maddox and others, 1992) were reached, both trends show a peak in As, followed by a decline.

The decline represents mixing of the injected waters with low-As native groundwater.



Collection of samples from SZOW-1, the storage zone monitor well, began in March,

1997 at values likely representing the outer edge of the injection bubble (Figure 15). During

Cycle 2 recovery, the injected "bubble" is withdrawn from the SZOW-1 well and the

concentration decreases toward background groundwater levels for As. An increase is again

observed in SZOW-1 as Cycle 3 recharge begins, reflecting migration of reactive injected waters

across the location of that well, which is 192.5 feet south of TPW-1.



To address the issue of mobilized metals in recovered waters at Rome Avenue, the

project consultants, CH2M Hill, conducted a cycle test during May and June, 1999. Abatement

of As mobility was evaluated through a pH-adjusted cycle test. The pH of injected waters was

elevated in an attempt to reduce As concentrations in recovered waters. Changes in As

concentration during the test are shown in Figure 16. During the initial pump-out (May 10,

1999 to May 24, 1999) for this cycle, As concentrations vary similarly to those found during


* Recovery U
*Injection U










Table 5. Trace metal analyses for Rome Ave. ASR samples. Shaded data are from Williams and others
(1999). Non-shaded data do not fully meet FDEP sampling protocols. See text for further explanation.
"BD"- Below detection.


Na Sr
1 mg/I ug/1


0.1 0.4 1.5


62.8 6.20 0.733 7.9
62.7 6.82 0.743 8.4
64.0 5.44 0.778 9.4


BD 77.6 7.29 2.45
BD 70.7 13.8 2.29


72.1 5.54


117 335 2.50 34.5
91.9 59.2 1.97 24.2
84.9 9.40 2.08 25.6
80.4 12.0 2.15 26.4


SZOW-1
3/7/97
3/14/97
4/1/97
6/9/97
9/24/97
12/8/97
12/15/97
12/22/97
12/29/97
1/12/98
1/26/98
2/9/98
2/23/98
3/9/98
3/21/98
5/11/98
8/3/98

TPW-1
3/10/97
3/11/97
3/14/97
3/20/97
3/21/97
4/3/97
4/17/97
5/1/97
5/22/97
6/9/97

2-D
3/7/97
3/14/97
1/5/98
3/16/98

PTRW
3/20/97

TRW
3/14/97


BD 134 13.2 6.88 219
BD 133 10.8 7.00 218


BD 12.5 BD 65.9 9.25 1.64


2400
2350


553 29


As Ba Cd Ca Mn K
ug/1 ug/1 ug/1 mg/1 ug/1 mg/


4.0 0.7 0.4 0.1 0.25 0.01


DETECTION
LIMIT

SMW-1
3/7/97
3/14/97
6/9/97
1/5/98
3/16/98


11.9 BD 70.5 6.04 1.64

















40



B 30



20



10



0
12/1/96


X PTRW
---SZOW-1
-*--TPW-1


3/11/97 6/19/97 9/27/97 1/5/98 4/15/98 7/24/98 11/1/98 2/9/99


Figure 15. Arsenic concentration through time for Cycles 2 and 3, Rome Avenue ASR, Hillsborough
County, Florida. (Note: some high-As samples do not meet FDEP protocol for regulatory purposes.)


50


0


S ... b ": Piz : Pz "z Pz P ": P ": P ": P 1: b ": P ": P ": P
4&\ & PC C PC C P'

Figure 16. Arsenic concentrations through time for Rome Avenue ASR pH-adjusted cycle test,
May-June, 1999 (data source: CH2M Hill, December, 1999).


Maximum contaminant level












Recovery Recharge Storage Recovery




*





* *I ** *









Cycles 1 and 2. These recovered waters represent the unrecovered injected waters from Cycle 3
recharge during Spring/Summer, 1998.


Cycle 3 storage began in August 1998 and aside from a brief (3 day) recovery in
October (Figure 14), the injected waters were not withdrawn until mid-May, 1999 -
approximately a 10-month storage period. Initial pump-out waters collected before the pH-
adjusted cycle test did not reach the As MCL (i.e., <50 ug/l; Figure 16). This suggests that with
increasing storage time, As did not continue to leach from the aquifer system carbonates to yield
higher concentrations. Although As did not reach levels observed during earlier cycles, it is
likely that the As "front" increases in size because of the increase in storage time and thus has
more of an opportunity for leaching to reach its full extent within the injected "bubble." It is
also possible that As reached higher levels, but injected waters equilibrated and metals had time
to precipitate or resorb in the aquifer-system matrix. For reference, Stuyfzand (1998) reports
that 10-100 days were required in Netherlands ASR sites for pyrite oxidation to consume 02 and
NO3- in the injected waters.


For the pH-adjusted cycle test, surface waters with an adjusted pH of approximately 7.8
were injected from May 25, 1999 to June 3, 1999. Reported pH of previously injected waters
(PTRW) is less than 7.2 (CH2M Hill, 1999). Native ground-water pH, based on an earlier
aquifer performance test (February, 1996) is approximately 7.5. Input waters were stored in the
aquifer from May 10, 1999 to May 24, 1999. Recovery extended from June 7, 1999 to June 22,
1999. Although the pH-adjusted cycle test was successful in that the maximum observed As
concentrations were significantly reduced (Figure 16), As concentrations still increase with time
during recovery.


Despite the positive results of the pH-adjusted cycle test in reducing observed As,
mechanical complications occurred. Upon raising the pH to a more basic level, excessive scaling
occurred in the physical plant (pumps, valves, etc.). Moreover, Fe and Mn in some samples
(CH2M Hill, 1999) exceeded secondary drinking water standards. The pH-adjusted solution
was not cost-effective.


The Rome Avenue facility currently has seven additional ASR wells being drilled as
part of a major expansion project. Once online, these wells will be connected with TPW-1 via a
pipeline that will allow blending of the recovered water. Carefully monitored, it is expected that
the blended water will result in reduction of dissolved metals concentrations.









Punta Gorda


Relevant drilling, chemical modeling and cycle-testing activities at the Punta Gorda
ASR facility are summarized in Appendix 2. The distributions of As concentrations during
Cycle tests 1 and 2, and 1A and 2A, are shown in Figures 17 and 18, respectively. Although no
samples exceed the current MCL for As, the trend observed in Cycle 1 (Williams and others, in
review) is clearly repeated in Cycles 2 (Upper Zone) and 1A and 2A (Lower Zone) indicating
that a similar geochemical process occurs in both zones during ASR cycle testing. Similar to the
U concentrations for these cycles (Figure 14), As maximum concentrations decrease from Cycle
1A to Cycle 2A (Figure 18). Iron exhibits a similar pattern (Figure 19) and some samples exceed
the MCL for Fe. The same observations presented above (e.g., the need for improved
experimentation control) apply here for As, and to a lesser extent, Fe. Although depletion (i.e.,
leaching) of the As (and perhaps Fe) source in the aquifer system may be taking place during
successive cycle tests, other chemical processes may be involved. We are fortunate that native
concentrations and injection concentrations for Fe and As are low, thus enabling the patterns to be
observed in time-transgressive plots without being masked or overprinted by trends representing
mixing of the two waters.


Other metals analyzed in Punta Gorda samples are listed in Table 6. Distribution of these
metals (e.g., Ca, Mn, Na and Sr; Figures 20-23) reveals mixing of injected and native waters. For
these parameters, the mixing curves mask any geochemical effects due to water-rock interaction.


Origin of the mobilized metals


The association of As and Fe in recovered waters from the Rome Avenue and Punta
Gorda wells suggests that oxidation of an As and Fe-bearing phase (or organic material) is taking
place during the cycle tests. With the available data, it is not possible to determine if the source
is pyrite, Fe-hydroxide grain coatings or organic. If oxidation of pyrite is occurring, a release of
S042-, and perhaps Co, Ni and Zn would be observed. Pyrite has been observed (FGS lithologic
data) in carbonates of the Floridan aquifer system in southwest Florida. Input waters at Rome
Avenue and native groundwater at Punta Gorda have S042- concentrations that, due to mixing,
mask any trends that could otherwise be attributed to pyrite oxidation (Figure 24). The metals
Co, Ni and Zn have not been analyzed in the recovered water, however, these metals will be
investigated in the Year 2000 study.


The relation between Fe and Mn during the Rome Avenue pH-adjusted cycle test (Figure 25)










































08/03/98 08/13/98 08/23/98 09/02/98 09/12/98 09/22/98 10/02/98 10/12/98


Figure 17. Arsenic concentration through time for Cycles
Charlotte County, Florida


60



50



40



. 30



20



10


0


1 and 2 (Upper Zone), Punta Gorda ASR,


*RECOVERY
INJECTION
NATIVE


19-Jun-99 29-Jun-99 09-Jul-99 19-Jul-99 29-Jul-99 08-Aug-99 18-Aug-99 28-Aug-99 07-Sep-99 17-Sep-99




Figure 18. Arsenic concentration through time for Cycles 1A and 2A (Lower Zone), Punta Gorda ASR,
Charlotte County, Florida


15


10


5


0
07/2


RECOVERY
m INJECTION (TREATED)
INJECTION (UNTREATED)
As MCL = 50 ug/1












* 0




,"


4/98


Cycle 1A Cycle 1A Cycle 2A C.2A
Injection Recovery Injection Rec.
Maximurr contanimant level

OS
0




*












.. ,. Native
g Concentration
0 Iml
. . . . . . . . . . . .ativ e.








































0


19-Jun-99 29-Jun-99 09-Jul-99 19-Jul-99 29-Jul-99 08-Aug-99 18-Aug-99 28-Aug-99 07-Sep-99 17-Sep-99



Figure 19. Iron concentration through time for Cycles 1A and 2A (Lower Zone), Punta Gorda ASR, Charlotte
County, Florida


160
160 I-----------------------------------------------------------------i_
Cycle 1A Cycle 1A Cycle 2A C.2A
Injection Recovery Injection Rec.
140 ......_ ..
Native concentration *

120 0 ------


100


*0*
80 _

60*



40 2 J-----u 9A-9---S


20

10 9Jun99 -29Jun99 09Jul99 19Jul99 29Jul9 08Aug99 18Aug99 28A 9 0 9 1

19-Jun-99 29-Jun-99 09-Jul-99 19-Jul-99 29-Jul-99 08-Aug-99 18-Aug-99 28-Aug-99 07-Sep-99 17-Sep-99


Figure 20. Calcium concentration through time for Cycles 1A and 2A (Lower Zone), Punta Gorda ASR,
Charlotte County, Florida


Cycle 1A Cycle 1A Cycle 2A Cycle 2A
Injection Recovery Injection Recovery















a
Fe MCL = 300 ug/l


Native concentration -
.----------- ------------------------------


. . .____ > _.. .__..__.. .__.. .__..__. .__.. __. .


*RECOVERY
*INJECTION
NATIVE


RECOVERY
*INJECTION
NATIVE












Table 6. Trace metal analyses of water samples from Cycles 1 and 2 (Upper Zone) and Cycles 1A and 2A

(Lower Zone), Punta Gorda ASR. Shaded data are from Williams and others (in review).


As Ba Ca Mn
ug/1 ug/1 mg/1 ug/1


DETECTION
LIMIT


Na Sr Fe Pb Cd
mg/1 ug/1 ug/1 ug/1 ug/1


4.0 0.7 0.1 0.25 0.1 0.4


NATIVE
6/28/99 6:00 PM 7.96

INJECTION
7/14/99 10:00 AM BD
8/17/99 2:30 PM BD
8/18/99 11:00 AM BD
8/19/99 12:00 AM BD
8/30/99 1:00 PM BD

RECOVERY
7/22/99 9:15 AM BD
7/22/99 10:00 PM BD
7/23/99 10:00 AM 4.69
7/23/99 10:00 PM 6.33
7/24/99 10:00 AM 8.87
7/24/99 10:00 PM 12.7
7/25/99 10:00 AM 16.8
7/25/99 10:00 PM 22.4
7/26/99 10:00 AM 30.1
7/26/99 10:00 PM 35.6
7/27/99 10:00 PM 45.9
7/28/99 10:00 PM 48.9
7/29/99 10:00 AM 49.7
7/29/99 10:00 PM 46.5
7/30/99 10:00 PM 37.5
8/1/99 10:00 PM 21.2
8/2/99 10:00 PM 8.57
8/5/99 10:00 PM 15.8
9/1/99 9:00 AM BD
9/1/99 2:30 PM BD
9/1/99 10:00 P BD
9/2/99 2:00 PM 14.1
9/2/99 10:00 PM 18.9
9/3/99 10:00 AM 19.7
9/4/99 10:00 AM 15.4
9/5/99 10:00 AM 10.5
9/6/99 10:00 AM 7.15
9/8/99 9:30 AM 4.38


INJECTION (UNTREATED)
7/29/98 12:00 AM 10
9/30/98 12:00 AM BD

INJECTION (TREATED)
7/29/98 12:00 AM BD
9/24/98 12:00 AM BD
9/30/98 12:00 AM BD

RECOVERY
7/29/98 12:00 AM BD
7/30/98 12:00 AM 10
7/31/98 12:00 AM 32
8/3/98 12:00 AM 37
8/4/98 12:00 AM 28
8/5/98 12:00 AM 18
8/6/98 12:00 AM 13
9/29/98 2:00 PM BD
9/30/98 2:00 AM BD
9/30/98 2:00 PM BD
10/1/98 2:00 AM BD
10/1/98 2:00 PM BD
10/2/98 2:00 AM 4.6
10/2/98 2:00 PM 6.7
10/3/98 2:00 AM 9.3
10/3/98 2:00 PM 12.8


45 139 5.9 403 34500


41.5 65.8
18.2 31.7


40.7 63.1
26.6 35.9
16.3 32.7


2540
3140
2980
2900
1250


2100
3190
4390
5350
7000
8540
10900
12800
15100
17000
21900
25700
28000
28900
30900
31800
32900
32400
1500
3200
4380
11500
19200
23900
28400
29500
31600
32300



4590
1470


22.8
10.2
6


29.9
14.3
9.52
3.73
3.93
3.85
2.87
18.7
20.4
19.7
18.1
16.7
14.3
13.7
12.6
11.7


4570
10200
19100
30900
33000
32700
32700
2100
3500
4820
5990
7440
8970
10900
12400
13900


10 10 0.4

















50



40



. 30



20



10



0


Cycle 1A Cycle 1A Cycle 2A C.2A
Injection Recovery Injection Rec.


















4.
0
0* *


------- ------------------ -------- -----------------
S Native concentration


19-Jun-99 29-Jun-99 09-Jul-99 19-Jul-99 29-Jul-99 08-Aug-99 18-Aug-99 28-Aug-99 07-Sep-99 17-Sep-99


Figure 21. Manganese concentration through time for Cycles 1A and 2A (Lower Zone), Punta Gorda ASR,
Charlotte County, Florida.


450
Cycle 1A Cycle 1A Cycle 2A C.2A
Injection Recovery Injection Rec.
400 --__ _- .
Native concentration

350


300

250
S250 *RECOVERY
E m INJECTION
S200 __ NATIVE
*

150
*

100
5

0

0 ....-----------------------------------------------------------
19-Jun-99 29-Jun-99 09-Jul-99 19-Jul-99 29-Jul-99 08-Aug-99 18-Aug-99 28-Aug-99 07-Sep-99 17-Sep-99



Figure 22. Sodium concentration through time for Cycles 1A and 2A (Lower Zone), Punta Gorda ASR,
Charlotte County, Florida


RECOVERY
*INJECTION
NATIVE












40000
Cycle Cycle 1A Cycle C.2A
1A Injection Recovery 2A Injection Rec.
35000-------------------------------------------------------
Native concentration

30000 --------


25000
*
*RECOVERY
S20000 INJECTION
NATIVE

15000 *


10000


5000



19-Jun-99 29-Jun-99 09-Jul-99 19-Jul-99 29-Jul-99 08-Aug-99 18-Aug-99 28-Aug-99 07-Sep-99 17-Sep-99


Figure 23. Strontium concentration through time for Cycles 1A and 2A (Lower Zone), Punta Gorda ASR,
Charlotte County, Florida.


160



140



120



100



S80



0 60



40



20


Figure 24. Sulfate concentration though time for Rome Avenue ASR pH-adjusted cycle test, May-June, 1999
(data source: CH2M Hill, December, 1999).


t







Recovery Recharge Storage Recovery


* .
4











Fe MCL = 300 ug/1
Mn MCL = 50 ug/1





[ e a Storage

*Recovery Recharge
100

SRecovery


In* *


* Mn(ug/l)
--- Fe (ug/l)


K0 4' e 0 4


Figure 25. Iron and Manganese concentrations through time for Rome Avenue ASR pH-adjusted
cycle test, May-June, 1999 (source: CH2M Hill, December, 1999).


is similar to As in that the metals show an increase with time during recovery. Association

of As-Fe-Zn (and Mn) in rocks (Figures 2 and 3) suggests pyrite involvement, however, the

trends are not well developed and require additional data.



PLANNED RESEARCH


Sampling of waters from ASR cycle tests at Rome Avenue and Punta Gorda are

continuing. Given the association of Co, Ni and Zn to pyrite and the possible oxidation

of pyrite in the aquifer system during cycle testing, these metals will be added to those being

monitored.


Cores from the Rome Avenue expansion have already been received and are being

described by Florida Geological Survey (FGS) staff. Moreover, the FGS is performing

hydraulic conductivity analyses on the core samples. In cooperation with the FGS, CH2M Hill

consultants have agreed to core targeted intervals in the expansion wells. This core and the

accompanying gamma-ray log for the well will allow comparison of whole rock chemistry,

organic content and gamma-ray activity in a given borehole. If time and funds permit,

chemistry of phases within the rock samples (e.g., grain coatings, interstitial minerals) will be









determined from microprobe analysis. Moreover, total organic carbon will be measured in
selected samples.


As other ASR facilities come online and begin cycle testing, suites of samples from
those sites will likely be included in the ongoing study. These sites include the NW
Hillsborough County Reclaimed Water ASR, Englewood Reclaimed Water ASR and Bonita
Springs ASR.


The results of this continuing investigation are extremely relevant to the ASR pilot
studies proposed under the Comprehensive Everglades Restoration Plan. The site-specific
nature of geochemical processes during ASR cycles warrants appropriate sampling of core
material from the aquifer system as well as appropriate water sampling schedules during ASR
cycle testing. During the Restudy pilot ASR programs, we hope that cycle tests can be designed
to accommodate the need for evaluating water-rock geochemical interactions taking place in the
Floridan aquifer system and their affect on recovered water quality through time. This would
include coring of selected injection-zone horizons, native ground-water sampling, collection of a
comprehensive suite of geophysical logs, and trace metal analyses of test cycle waters. As
demonstrated herein and in our Year One Year Two study (Williams and others, in review),
chemical variability exists within Floridan aquifer system carbonates. As such, geochemical
processes taking place during ASR operation can be site specific.










REFERENCES


Ahmann, D., 1999, Microbial arsenic reduction: A new, ubiquitous, yet still mysterious factor in
aquatic arsenic mobility: Eos v. 80, no. 17 Supplement, p. S148.

Ahmed, K.M., Imam, M.B., Akhter, S.H., Hasan, M.A., and Khan, A.A., 1999, Sedimentology
and mineralogy of the arsenic contaminated aquifers in the Bengal Delta of Bangladesh,
Kth-Dhaka University Seminar on ground-water arsenic contamination in the Bengal
Delta Plains of Bangladesh, Feb. 7-8, 1999 (see http://phys4.harvard.edu/-wilson
/arsenic_proj ect Dhakaconferenceabstracts.html).

Bellanca, A., Masetti, D., and Neri, R., 1997, Rare earth elements in limestone/marlstone
couplets from the Albanian-Cenomanian Cismon section (Venetian region, northern
Italy): Assessing REE sensitivity to environmental changes: Chemical Geology, v. 141,
p. 141-152.

Bhattacharya, P., Jacks, G., Gustafsson, J.P., Sracek, A., Olofsson, B., Aaltonen, J., Khan, A.A.,
Akhter, S.H., and Ahmed, K.M., 1999a, High arsenic in the alluvial aquifers of Bengal
Delta Plains in Bangladesh: Genesis and low-cost remediation, 1999 [abstracts] Kth-
Dhaka University Seminar on groundwater arsenic contamination in the Bengal Delta
Plains of Bangladesh, Feb. 7-8, 1999 (see http://phys4.harvard.edu/-wilson
/arsenic_proj ect Dhakaconferenceabstracts.html).

Bhattachaya, P., Jacks, J., Sracek, A., Gustafsson, J.P., and Chatterjee, D., 1999b, Geochemistry
of the Holocene alluvial sediments in the Bengal Delta Plains and their implications in
ground-water arsenic contamination [abstracts] Kth-Dhaka University Seminar on
ground-water arsenic contamination in the Bengal Delta Plains of Bangladesh, Feb. 7-8,
1999 (see http://phys4.harvard.edu/~wilson/arsenicprojectDhakaconference
abstracts.html).

Brandon, W., Byrne, J., McTigue, D., and Stein, C., 1999, Arsenic contaminated sediment and
groundwater: A case study in central Massachusetts: Eos v. 80, no. 17 Supplement, p.
S146.

Brun, A., Christensen, F.D., Christiansen, J.S., Stuyfzand, P.J., and Timmer, H., 1998, Water
quality modeling at the Langerak deep-well recharge site: in Peters, J.H., et al., eds.,
Artificial Recharge of Groundwater: Rotterdam, Netherlands, A.A. Balkema, 474 p.

Campbell, K.M., and Arthur, J.D., 1993, Geologic map of Hillsborough County: Florida
Geological Survey Open File Map Series No. 45.

CH2M Hill, 1999, Rome Avenue Park cycle test unpublished analytical data, May/June, 1999
(provided by M.Clasen, CH2M Hill, December, 1999).

Colman, J.A. and Lyford, F.P., 1999, Sources and geochemical associations of arsenic in
leachate plumes from a landfill in Saco, Maine: Eos v. 80, no. 17 Supplement, p. S148.

Cowart, J.B., Williams, H.K., and Arthur, J.D., 1998, Mobilization of U isotopes by the
introduction of surface waters into a carbonate aquifer: Geological Society of America
Abstracts with Programs, v. 30, no. 7, p. A-86.










REFERENCES (continued)


Das, D., Samanta, G., Mandal, B.K., Chowdhury, T.R., Chanda, C.R., Chowdhury, P.P., Basu,
G.K., and Chakraborti, D., 1996, Arsenic in groundwater in six districts of West Bengal,
India: Environmental Geochemistry and Health, v. 18, p. 5-15.

Deer, W.A., Howie, R.A., and Zussman, J., 1992, An introduction to rock forming minerals -
Second edition, Longman Scientific and Technical, Hong Kong, 696 p.

Federal Register, 2000, 40 CFR Parts 141 and 142, National Primary Drinking Water
Regulations; Arsenic and clarifications to compliance and new source contaminants
monitoring; Proposed rule: v. 65, no. 121, p. 38888-38893.

Federal Register, 2001, 40 CFR Parts 9, 141 and 142, National Primary Drinking Water
Regulations; Arsenic and clarifications to compliance and new source contaminants
monitoring; Final rule: v. 66, no. 14, p. 6975-7066.

Green, R., Arthur, J.D., and DeWitt, D., 1995, Lithostratigraphic and hydrostratigraphic cross
sections through Pinellas and Hillsborough Counties, Southwest Florida: Florida
Geological Survey Open File Report 61, 26 p.

Grommet, L.P., Dymek, R.F., Haskin, L.A., and Kortev, R.L., 1984, The "North American shale
composite": Its compilation, major and trace element characteristics: Geochimica et
Cosmochimica Acta, v. 48, p. 2469-2482.

Hemm, J.D., 1985, Study and interpretation of chemical characteristics of natural water: U.S.
Geological Survey Water-Supply Paper 2254, 263 p.

Jacks, G., Bhattacharya, P., and Chaterjee, D., 1999, Artificial recharge as a remedial measure,
[abstracts] Kth-Dhaka University Seminar on groundwater arsenic contamination in the
Bengal Delta Plains of Bangladesh, Feb. 7-8, 1999 (see http://phys4.harvard.edu
/-wilson/arsenicproject Dhakaconferenceabstracts.html).

Kolker, A., Cannon, W.F., Woodruff, L.G., Westjohn, D.B., Haack, S.K. and Kim, M., 1999,
Arsenic in southeastern Michigan groundwater: Results of USGS test drilling: Eos v. 80,
no. 17 Supplement, p. S146.

Maddox, G.L., Lloyd, J.M., Scott, T.M., Upchurch, S.B., and Copeland, R., 1992, Florida's
groundwater quality monitoring program: Background hydrogeochemistry: Florida
Geological Survey Special Publication 34, 364 p.

Montgomery-Watson, 1999, Punta Gorda ASR Exploratory Well, Cycle Test la and 2a Results,
Memorandum dated November 15, 1999, 32 p.

National Research Council, 1999, Arsenic in Drinking Water: Washington D.C., National
Academy Press, 310 p.

Nickson, R.T., McArthur, J.M., Ravenscroft, P., Burgess, W.G., and Ahmed, K.M., 2000,
Mechanism of arsenic release to groundwater, Bangladesh and West Bengal: Applied
Geochemistry, v. 15, p. 403-413.

Ruiter, H. and Stuyfzand, P.J., 1998, An experiment on well recharge of oxic water into an
anoxic aquifer: in Peters, J.H., et al., eds., Artificial Recharge of Groundwater:
Rotterdam, Netherlands, A.A. Balkema, 474 p.










REFERENCES (continued)


Schlottmann, J.L., and Briet, G.N., 1992, Mobilization of As and U in the central Oklahoma
aquifer, USA: in Kharaka, Y.K. and Maest, A.S., eds., Water-Rock Interaction:
Rotterdam, Netherlands, A.A. Balkema, p. 835-838.

Sracek, A., Bhattacharya, P., and Jacks, G., 1999, Basic principles of groundwater flow and
considerations for sampling of arsenic, [abstracts] Kth-Dhaka University Seminar on
groundwater arsenic contamination in the Bengal Delta Plains of Bangladesh, Feb. 7-8,
1999 (see http://phys4.harvard.edu/-wilson/arsenicprojectDhakaconference
abstracts.html).

Sracek, A., Bhattacharya, P., Jacks, G., and Gustafsson, J.P., 1999, Mobility of arsenic and
geochemical modeling applications [abstracts] Kth-Dhaka University Seminar on
groundwater arsenic contamination in the Bengal Delta Plains of Bangladesh, Feb. 7-8,
1999 (see http://phys4.harvard.edu/-wilson/arsenicprojectDhakaconference
abstracts.html).

Stollenwerk, K.G. and Colman, J.A., 1999, Natural remediation of arsenic contaminated
groundwater: Solute-transport model predictions: Eos v. 80, no. 17 Supplement, p. S148.

Stuyfzand, P.J., 1998, Quality changes upon injection into anoxic aquifers in the Netherlands:
Evaluation of 11 experiments, in Peters, J.H., et al., eds., Artificial Recharge of
Groundwater: Rotterdam, Netherlands, A.A. Balkema, 474 p.

Thomas, R.C. and Saunders, J.A., 1998, Arsenic co-precipitation in low-temperature pyrites:
implications for bioremediation via sulfate reducing bacteria: Geological Society of
America Abstracts with Program, v. 30, no. 7, p. A-58.

Welch, A.H. and Lico, M.S., 1998, Factors controlling As and U in shallow ground water,
southern Carson Desert, Nevada: Applied Geochemistry, v. 13, no. 4, p. 521-539.

Welch, A.H., 1999, Arsenic in groundwater of the United States: Processes leading to
widespread high concentrations: Eos v. 80, no. 17 Supplement, p. S147.

Wendell, D. and Glanzman, D., 1998, Well recharge in California: Water quantity and quality
considerations: in Peters, J.H., et al., eds., Artificial Recharge of Groundwater:
Rotterdam, Netherlands, A.A. Balkema, 474 p.

White, C.A. and Scully, M.V., 1999, An evaluation of potential arsenic sources to the
groundwater at a landfill site in eastern Maine: Eos v. 80, no. 17 Supplement, p. S148.

Williams, H., Cowart, J.B. and Arthur, J.D., in review, Florida aquifer storage and recovery
(ASR) geochemical project: Year One and Year Two Progress Report Submitted to
Bureau of Water Facilities Regulation, Florida Department of Environmental Protection,
January, 1999, Florida Geological Survey Open File Report 82, 132 p.










Appendix 1. Lithologic descriptions of samples analyzed for whole rock chemistry. Wells include
SWFWMD ROMP TR9-2 (W-16618), NW Hillsborough Recalimed ASR TPW-1 (W-17687) and
SWFWMD WRAP 2D (W16574).

Well Depth
Number (ft. below land surface)

16618 368 FGS-1 PACKSTONE; YELLOWISH GRAY
POROSITY: INTERGRANULAR, LOW PERMEABILITY
GRAIN TYPE: BIOGENIC, SKELETAL, CALCILUTITE
GRAIN SIZE: MEDIUM; WELL-SORTED
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
SEDIMENTARY STRUCTURES: INTERBEDDED, MASSIVE
OTHER FEATURES: WEATHERED, CHALKY, GRANULAR
FOSSILS: FOSSIL FRAGMENTS
16618 382.5 FGS-2 MUDSTONE-WACKESTONE; YELLOWISH GRAY
POROSITY: INTERGRANULAR, LOW PERMEABILITY
GRAIN TYPE: BIOGENIC, SKELETAL, CALCILUTITE
GRAIN SIZE: MEDIUM; WELL-SORTED
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
SEDIMENTARY STRUCTURES: INTERBEDDED, MASSIVE
OTHER FEATURES: WEATHERED, CHALKY
FOSSILS: FOSSIL FRAGMENTS
16618 392 FGS-3 MUDSTONE; LIGHT OLIVE
POROSITY: INTERGRANULAR, LOW PERMEABILITY
GRAIN TYPE: BIOGENIC, SKELETAL, CALCILUTITE
GRAIN SIZE: MEDIUM; WELL-SORTED
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
SEDIMENTARY STRUCTURES: INTERBEDDED, MASSIVE
OTHER FEATURES: WEATHERED, CHALKY
FOSSILS: ORGANIC? (BLACK LAMINATIONS COMPOSITION
UNKNOWN)
16618 425 FGS-4 WACKESTONE; YELLOWISH GRAY
POROSITY: INTERGRANULAR, LOW PERMEABILITY
GRAIN TYPE: BIOGENIC, SKELETAL, CALCILUTITE
GRAIN SIZE: MEDIUM; WELL-SORTED
MODERATE TO GOOD INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
SEDIMENTARY STRUCTURES: INTERBEDDED, MASSIVE
OTHER FEATURES: WEATHERED, CHALKY
FOSSILS: FOSSIL MOLDS
17687 310-315 FGS-5 PACKSTONE-GRAINSTONE; YELLOWISH GRAY
20% POROSITY: INTERGRANULAR
GRAIN TYPE: BIOGENIC, CALCILUTITE, SKELETAL
GRAIN SIZE: MEDIUM; RANGE: CRYPTOCRYSTALLINE TO MEDIUM
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
OTHER FEATURES: LOW RECRYSTALLIZATION
FOSSILS: MOLLUSKS
TRACE ORGANIC PRESENT.










Appendix 1. (continued)


Well Depth
Number (ft. below land surface)



17687 320-325 FGS-6 WACKESTONE; PALE YELLOWISH BROWN TO
YELLOWISH GRAY
20% POROSITY: INTERGRANULAR
GRAIN TYPE: BIOGENIC, CALCILUTITE, SKELETAL
45% ALLOCHEMICAL CONSTITUENTS
GRAIN SIZE: MEDIUM; RANGE: CRYPTOCRYSTALLINE TO MEDIUM
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
OTHER FEATURES: LOW RECRYSTALLIZATION, DOLOMITIC?
FOSSILS: MOLLUSKS
SOME ORGANIC (OR GRAIN COATINGS; DARK BLEBS) PRESENT.
17687 325-330 FGS-7 AS ABOVE
17687 330-335 FGS-8 AS ABOVE; TWO LITHOLOGIES: MUDSTONE AND
WACKESTONE, TRACE SILT? AND ORGANIC
17687 340-345 FGS-9 GRAINSTONE; YELLOWISH GRAY
GRAIN TYPE: BIOGENIC, CALCILUTITE, SKELETAL
GRAIN SIZE: MEDIUM; RANGE: MEDIUM COARSE
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
OTHER FEATURES: LOW RECRYSTALLIZATION
FOSSILS: MOLLUSKS
17687 360-365 FGS-10 PACKSTONE; YELLOWISH GRAY
25% POROSITY: INTERGRANULAR, MOLDIC
GRAIN TYPE: BIOGENIC, CALCILUTITE, SKELETAL
55% ALLOCHEMICAL CONSTITUENTS
GRAIN SIZE: MEDIUM; RANGE: CRYPTOCRYSTALLINE TO COARSE
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
OTHER FEATURES: LOW RECRYSTALLIZATION
FOSSILS: MOLLUSKS, FOSSIL MOLDS
SOME ORGANIC PRESENT AS FINELY INTERSPERSED SPECS.
17687 420-421 FGS-11 WACKESTONE; YELLOWISH GRAY
20% POROSITY: INTERGRANULAR
GRAIN TYPE: BIOGENIC, CALCILUTITE, SKELETAL
40% ALLOCHEMICAL CONSTITUENTS
GRAIN SIZE: FINE; RANGE: CRYPTOCRYSTALLINE TO MEDIUM
MODERATE INDURATION
CEMENT TYPE(S): CALCILUTITE MATRIX
CORED INTERVAL 420'-421' IS A
WACKESTONE WITH MEDIUM INDURATION AND MICRITE CEMENT.
MOLLUSK MOLDS AND SOME ORGANIC ARE PRESENT AS TRACE
FINELY DISPERSED SPECS.










Appendix 1. (continued)


Well Depth
Number (ft. below land surface)

17687 429 FGS-12 WACKESTONE; GRAYISH BROWN
20% POROSITY: INTERGRANULAR
GRAIN TYPE: BIOGENIC, CALCILUTITE, SKELETAL
15% ALLOCHEMICAL CONSTITUENTS
GRAIN SIZE: VERY FINE; RANGE: CRYPTOCRYSTALLINE TO FINE
MODERATE INDURATION
CEMENT TYPE(S): DOLOMITE AND CALCILUTITE MATRIX
OTHER FEATURES: MEDIUM RECRYSTALLIZATION
FOSSILS: MOLLUSKS, FOSSIL MOLDS
FINE GRAINED DOLOMITE MAKES UP ABOUT 50% OF THIS
SAMPLE. SOME ORGANIC PRESENT. CORED INTERVAL 429'-430'
IS A WACKESTONE WITH 15% POROSITY AND GOOD
INDURATION WITH MICRITE CEMENT. FINE BLACK LAMINATIONS
(PROBABLY ORGANIC LAYERS) WITHIN THE LIMESTONE ARE
PRESENT. LAMINATIONS ARE UP TO 1/4 INCH THICK, WITH
VARYING DISTANCES BETWEEN
LAMINATIONS. POSSIBLE BURROW PERPENDICULAR TO CORE,
FILLED WITH CALCILUTITE, ORGANIC MATERIAL AND
FORAMINIFERA. FRACTURE OBSERVED IN THIS INTERVAL LIKELY
FORMED DURING OR AFTER DRILLING. THIS OBSERVATION IS
MADE DUE TO THE LACK OF SLICKENSIDES, MINERALIZATION, OR
DISSOLUTION ON THE FRACTURE SURFACE.

17687 431 FGS-13 DOLOSTONE, YELLOWISH GRAY TO GRAYISH BROWN
MOLDIC, GRANULAR, RECRYSTALIZED, ORGANIC COATINGS?
17687 432 FGS-14 PACKSTONE, YELLOWISH GRAY, CHALKY, "OCALA"
SPARRY
ALLOCHEMS, TRACE BLACK SPECS ORGANICCS)
16574 295-300 FGS-15 GRAINSTONE; YELLOWISH GRAY TO TAN
POOR INDURATION
ACCESSORY MINERALS: TRACE SPAR
FOSSILS: BENTHIC FORAMINIFERA, ECHINOID
POROUS.
16574 325-330 FGS-16 GRAINSTONE; YELLOWISH GRAY TO TAN
POOR INDURATION; WELL SORTED
ACCESSORY MINERALS: TRACE SPAR
OTHER FEATURES: FOSSILIFEROUS
16574 330-335 FGS-17 AS ABOVE; VARIABLY RECRYSTALLIZED PACKSTONE TO
GRAINSTONE; TRACE ORGANICS() INTERSPERSED
16574 335-340 FGS-18 AS ABOVE
16574 370-375 FGS-19 PACKSTONE-GRAINSTONE; LIGHT YELLOWISH GRAY
POOR INDURATION
ACCESSORY MINERALS: TRACE QUARTZ SAND/SILT
OTHER FEATURES: FOSSILIFEROUS, CHALKY
16574 459 FGS-20 VERY FINE-GRAINED PACKSTONE (OR MUDSTONE); VERY
PALE ORANGE TO YELLOWISH GRAY; GRAIN SIZE: FINE; RANGE: VERY FINE
TO FINE CHALKY, FOSSILS: BENTHIC FORAMS
16574 459.5 FGS-21 AS ABOVE
16574 459 FGS-22 SPLIT SAMPLE OF FGS-20









Appendix 2. Cycle test and drilling activities at the Punta Gorda ASR (modified from
Montgomery-Watson (1999).


The Punta Gorda ASR well has undergone cycle testing in two permeable storage zones
within the Suwannee Limestone, which comprises the upper part of the Floridan aquifer system.
Cycle tests 1 and 2 utilize the Upper Zone, which includes an open borehole interval from 700'
to 764' below land surface (bls). Cycle tests 1A and 2A utilize a Lower Zone, which extends
from 764' to 933' bls.


Injection flow rates during cycle tests in the Upper Zone revealed significant decreases
in average flow. During the injection period for cycle 2 (8/24/98 to 9/29/98), for example, daily
average flows decreased from 200 gallons per minute (gpm) to 10 gpm. The decrease in
injection flow rates for cycle tests 1 and 2 were attributed to formation plugging (Montgomery-
Watson, 1999). Consultants attributed the plugging of pore space in the aquifer system matrix
(or along borehole walls) to either particle rearrangement or precipitation of solids. Rather than
acidizing the Upper Zone or making chemical adjustments to the recharge water, consultants
opted to utilize a lower flow zone in the well for further cycle tests and possibly full ASR
implementation. Similar injection-rate declines were observed in the Lower Zone during cycle
tests 1A and 2A.


During Cycle 1A, several water quality parameters were monitored and later used to
compare to geochemical model results from PHREEQC (Parkhurst, 1995). As demonstrated in
the MW (1999) report, it is clear that the PHREEQC model is a very useful predictive tool
concerning the movement of cations and anions from the aquifer system matrix to the injected
"bubble" during an ASR cycle test. The model also predicts that Fe-hydroxide precipitation may
be occurring in the aquifer. Moreover, the model was used to predict the ability to reduce the
observed pore plugging by the acid addition during injection.