xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001914892
007 cr mnu|||uuuuu
008 071029s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001998
Murphy, Kathryn E.
Hydrological connectivity between clay settling areas and surrounding hydrological landscapes in the phosphate mining district, Peninsular Florida, USA
h [electronic resource] /
by Kathryn E. Murphy.
[Tampa, Fla.] :
b University of South Florida,
ABSTRACT: The objective of this study was to use applied and naturally-occurring geochemical tracers to study the hydrology of clay settling areas (CSAs) and the hydrological connectivity between CSAs and surrounding hydrological landscapes. The study site is located on the Fort Meade Mine in Polk County, Florida. The surface of the CSA is covered in desiccation cracks which swell and shrink in response to wetting and drying. Bromide was used as an applied tracer to study hydrological processes in the upper part of the CSA. Bromide infiltrated rapidly and perched on an uncracked massive sublayer. Bromide concentrations attenuated in the upper part of the profile without being translated vertically down through the lower part of the profile suggesting that bromide was lost to lateral rather than to vertical downward transport. Infiltration and lateral flow were rapid suggesting that preferential flow through desiccation cracks and other macropores likely dominates flow in the upper part of the profile. Naturally-occurring dissolved constituents and stable isotopes of hydrogen and oxygen were used as naturally-occurring tracers to study the hydrological connectivity between the CSA and the surrounding hydrological landscape. The relative contributions of source waters were determined using a two-end, mass-balance mixing model with sodium as a conservative natural tracer. On average, water samples downgradient from the CSA were ~80% rainfall/ambient water and ~20% CSA water. Discharge from the CSA to the surrounding surface water bodies and surficial aquifer occurs laterally over, through, and/or under the berms and/or vertically through the thick uncracked massive sublayer. However, the precise flowpaths from the CSA to the surrounding hydrological landscape are unclear and the fluxes remain unquantified, so the effects of CSAs on the hydrology of the surrounding and underlying hydrological landscape also remain unquantified.
Thesis (M.S.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 39 pages.
Advisor: Mark C. Rains, Ph.D.
Clay settling area.
t USF Electronic Theses and Dissertations.
Hydrological Connectivity Between Clay Se ttling Areas and Surrounding Hydrological Landscapes in the Phosphate Mining District, Peninsular Florida, USA by Kathryn E. Murphy A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology College of Arts and Sciences University of South Florida Major Professor: Mark C. Rains, Ph.D. Mark Stewart, Ph.D. Mark A. Ross, Ph.D. Date of Approval: April 6, 2007 Keywords: clay settling area, geochemical tr acer, preferential tr ansport, desiccation crack, perched aquifer Copyright 2007 Kathryn E. Murphy
ACKNOWLEDGMENTS This project was funded by the Florida In stitute of Phosphate Research project 0303-150s. A large thanks to Mosaic for letting us conduct our research at the Fort Meade Mine. I would like to extend my sincerest appr eciation to Mark Rains, my adviser for his advising and guidance. A large thanks to my committee, Mark Stewart and Mark Ross. To my fellow labmates, Mike Kittridge fo r countless days in the field and hours doing GIS, and Christina Stringer for your editing and keeping me sane, this project would not have been completed without you. Jon Spencer and Ken Nilsson, thank you for the help with fieldwork and technical questions. Las tly, to my family and friends, thank you so much for all of the support and encourag ement and laughs, it was greatly needed.
i TABLE OF CONTENTS LIST OF TABLES iii LIST OF FIGURES iv ABSTRACT vi INTRODUCTION 1 SITE DESCRIPTION 4 Location and Setting 4 Climate 6 Geology and Soils 6 Vegetation 7 METHODS 8 Physical Hydrological Meas urements and Analyses 8 Tracer Application and Data Collection 9 Natural Geochemical Tracer Data Collection 11 Laboratory Analyses 12 Mass-Balance Mixing Modeling 13 RESULTS 14
ii Applied Tracer 14 Pond Water Budget 14 Natural Tracers 18 Mass-Balance Mixing Modeling 22 DISCUSSION 27 CONCLUSIONS 30 REFERENCES 31 APPENDICES 34 Appendix I. List of all bromide concen trations for all control and treatment samples in g/L. 35 Appendix IIa. Dry season values for all constituents for all samples. 36 Appendix IIb. Wet season values for al l constituents for all samples. 37 Appendix IIIa. Dry season results of mass-balance mixing modeling using sodium. End members are CSA water and rainfall/ambient surficial and surface water. 38 Appendix IIIb. Wet season results of mass-balance mixing modeling using sodium. End members are CSA water and rainfall/ambient surficial and surface water. 39
iii LIST OF TABLES Table 1. Physical and chemical properties of rainfall/ambient water, CSA water, and downgradient waters. 21 Table 2. Proportion of rainfall/ambient water and CSA water in downgradient water as determined by mass-balance mixing analyses using sodium as a conservative, natural tracer. 26
iv LIST OF FIGURES Figure 1. Topography and piezometer locations. 5 Figure 2. Rainfall in cm/d at the CSA imme diately before, during, and after tracer application. 15 Figure 3. Average bromide concentrations in g/L v. depth in th e 5 control and 5 treatment infiltration rings. 16 Figure 4. Precipitation in cm/d and stage of the north pond in cm immediately before and during the entire period of inundation. 17 Figure 5. Net groundwater inflow to the nor th pond in cm/d immediately before and during the entire period of inundation. 19 Figure 6. Stiff diagrams of the 4 types of water. 20 Figure 7. Scatterplot of sodium v. manganese con centration in mg/L. 23
v Figure 8a. Postplot of the proportional cont ribution of CSA water to downgradient water in dry season. 24 Figure 8b. Postplot of the proportional cont ribution of CSA water to downgradient water in wet season. 25
vi Hydrological Connectivity Between Clay Se ttling Areas and Surrounding Hydrological Landscapes in the Phosphate Mining District, Peninsular Florida, USA Kathryn E. Murphy ABSTRACT The objective of this study was to use applied and na turally-occurring geochemical tracers to study the hydrology of clay settling areas (CSAs) and the hydrological connectivity between CSAs a nd surrounding hydrological landscapes. The study site is located on the Fort Meade Mine in Polk County, Florida. The surface of the CSA is covered in desiccation cracks which swell and shrink in response to wetting and drying. Bromide was used as an applied tr acer to study hydrologica l processes in the upper part of the CSA. Bromide infiltrated rapidly and perched on an uncracked massive sublayer. Bromide concentrations attenuate d in the upper part of the profile without being translated vertically down through the lower part of the profile suggesting that bromide was lost to lateral rather than to vertical downward trans port. Infiltration and lateral flow were rapid sugges ting that preferenti al flow through desiccation cracks and other macropores likely dominates flow in the upper part of the profile. Naturallyoccurring dissolved constituents and stable isotopes of hydrogen and oxygen were used as naturally-occurring tracers to study th e hydrological connectivity between the CSA and the surrounding hydrological landscape. The relative contributions of source waters
vii were determined using a two-end, mass-ba lance mixing model with sodium as a conservative natural tracer. On average, water samples downgradient from the CSA were ~80% rainfall/ambient water and ~20% CSA water. Discharge from the CSA to the surrounding surface water bodies and surficial aquifer occu rs laterally over, through, and/or under the berms and/or vertically th rough the thick uncracked massive sublayer. However, the precise flowpaths from the CS A to the surrounding hydr ological landscape are unclear and the fluxes re main unquantified, so the effects of CSAs on the hydrology of the surrounding and underlying hydrologi cal landscape also remain unquantified.
1 INTRODUCTION Clay-rich deposits have low perm eabilities (Manning, 1997; Fetter, 2001; Deming, 2002). Because of their low permeab ilities, clay-rich depos its often confine aquifers (Fetter, 2001) and perch surface wa ter and/or groundwater (Rains et al., 2006; Rains et al., In Review), and clay liners are often used to restrict flow down well annuli (Sanders, 1998) and into and out of solid waste disposal sites (Johnson et al., 1989). Therefore, clay-rich deposits are typically treated as homogeneous and isotropic deposits in which hydraulic properties are identical at all locations and in all directions (Dekker and Bouma, 1984). However, these assumptions are not true in cl ay-rich deposits with desiccation cracks (Bouma et al., 1981). Due to interlayer water ad sorption, clay-rich deposits swell when wet and shrink when dry (Tuller and Or, 2003). Where swelling and shrink ing are pronounced, desiccation cracks can form in the surface and shallow subsurface. The depth of desiccation cracking depends on clay mineralo gy, and is increased by factors such as wetting and drying (Vogel et al., 2005) and compaction due to grazing (Reid and Parkinson, 1984) and decreased by factors such as disper sion due to high sodium concentrations (Qui rk and Schofield, 1955) Desiccation cracks, along w ith other macropores associated with bioturbation such as burrows and root channels (Dekke r and Ritsema, 1996), create preferential
2 flowpaths through which water can rapidly flow (Thomas an d Phillips, 1979). This phenomenon is also known as Â“short-circuitingÂ”, because water flows through the desiccation cracks and macropores rather than through the low-permeability matrix (Bouma and Dekker, 1978). Though desicc ation cracks and macropores may only represent a small proportion of the total porosity of a de posit, desiccation cracks and macropores can dominate flow and transport in the shallow subsurface of many clay-rich deposits (Heppell et al., 2000; Rains et al., In Review). Clay settling areas (CSAs) are waste pr oducts of phosphate mining in peninsular Florida. Phosphate-rich deposits cover much of peninsular Florida. Where mined, the phosphate-rich layer averages 3-4 m in thic kness and is buried beneath an overburden layer that is typically 4-8 m in thickness. The phosphate-rich layer is ~50-60 % clay, ~30 40 % quartz sand, and ~2-5 % heavy minera ls and other miscellaneous materials (Hawkins, 1973). Beneficiation yields phosphoric acid, gypsum, and a slurry of clay and sand that is predominantly clay with a d50 typically less than 0.2 microns (Hawkins, 1973). The slurry of clay and sand is ~3% solids and is disposed of in CSAs, large, above-grade reservoirs contai ned by rectangular, earthen be rms that are ~6-15 m above grade and ~120-320 hectares in area (Lewel ling and Wylie, 1993; Zang and Albarelli, 1995). Rapid consolidation and drainage en sues, with the slurry of clay and sand reaching ~18-22% solids within a few months (Zang and Albarelli, 1995). After ~5 years, a surface crust that is ~50-60% solid s forms, though the deeper subsurface remains ~25% solids for many years thereafter (Ervin et al., 1997). As of 1998, ~120,000 hectares of land had b een mined in peninsular Florida, with ~40% or ~50,000 hectares covered with CSAs (Stricker, 2000). Mining continues and,
3 depending on economical and environmental f actors, as much as ~175,000 hectares of land could be mined and as much as ~70,000 hectares of CSAs could remain when mining is complete. The hydrology of CSAs and the subsequent effects of CSAs on water resources are poorly understood. To be sure, annual ra infall is captured by the CSAs. However, the fate of this rainfall and the original pr ocessing water still cont ained within the CSAs is unclear. Does it recharge the underlying surficial aquife r and flow to nearby wetlands and streams? Does it recharge the underl ying Floridan aquifer? Or does it simply evaporate or enter into storage within the cl ay matrix and not become available for any beneficial use? This effort is part of a four-year project designed to address these and other fundamental questions of hydrological and ecological importance in Florida. Specifically, the objective of this study was to use applied and natural-occurring geochemical tracers to study the hydrology of CSAs and the hydr ological connectivity between CSAs and surrounding hydrological landscapes.
4 SITE DESCRIPTION Location and Setting The study site is located on the Fort Meade Mine in Polk County, Florida (Figure 1). The CSA is ~6 m above grade a nd ~75 hectares in size. The CSA is ~20 years old, and the surface crus t is sufficiently solid to s upport vehicles during the dry season. The CSA is located on un-mined land, so bottom boundary conditions are reasonably well known. The CSA is nearly level to undulating with a slight topographic gradient from north to south (Figure 1). The southwest corner of the berm has slumped, so the topographic gradient steepens toward the sout hwest corner and the adjacent field. There are several closed-basin depressions that pond water seasonally. There are no surface water inflows, but there are interconnected ditches that discharge through a culvert through the east berm. The su rface of the CSA is covered in desiccation cracks which swell and shrink due to wetting and drying. The surrounding land has been mined. Th e mined land to the north, east, and southwest has been reclaimed, while the land to the northwest and south has not been reclaimed and remains in open pits filled w ith water. There was no active mining in the immediate vicinity of the CSA during the course of the study.
5 Figure 1. Topography and piezometer locatio ns. Darker lines represent higher elevations. The CSA is the re ctangular feature, and the t opographic gradient of the CSA is generally from N to S.
6 Climate The climate at the study area is subtropical with warm, relatively dry winters and hot, relatively wet summers. Summer rain fall is due to frequent, local convective thunderstorms, while winter rainfall is due to infrequent cold fronts (Lewelling and Wylie, 1993). Mean ( standard deviation) annual precipit ation is 135.3 cm ( 24.5 cm), with ~65% falling during the months of Ma y-September (Southeastern Regional Climate Center data for Bartow, Florida for calenda r years 1931-2006). Annua l precipitation was 154.1 cm and 100.4 cm for water years 2005 and 2006, so rainfall was normal in water year 2005 and slightly lower than normal in water year 2006 (Southeastern Regional Climate Center data for Bartow Florida for water years 2005-2006). Geology and Soils The central Florida phosphate mining district where the study s ite is located, has three principal hydrostratigraphic units: the su rficial aquifer, the intermediate aquifer system, and the Floridan aquifer system (Ervin et al., 1997). Phosphate is mined from the Bone Valley Member, a part of the Miocen e-age Hawthorne Formation, located within the upper part of the intermed iate aquifer system. The CSA berms are comprised of overburden from the surficial aquifer, and the CSA deposits are comprised of clay and sand from the Bone Valley Member mixed w ith native groundwater and processing water (Reigner and Winkler, 2001).
7 Vegetation The predominant vegetation is the invasive Cogon grass ( Imperata cylindrical (L.) Raeuschel). Towards the southern end, and in topographically low areas, the predominant vegetation is the Florida willow ( Salix floridana Chapm.).
8 METHODS Physical Hydrological Measurements and Analyses Precipitation was measured continuously and solar radiation, net radiation, soil density heat flux, temperature, humidity, and wind speed and direction were measured hourly (Campbell Scientific, Inc.). Daily reference evapotranspiration was computed using the ASCE 2005 reference ET equa tion (Allen et al., 2005). Actual evapotranspiration was computed by multiplyi ng reference evapotranspiration by a dailyvarying crop coefficient. The crop coe fficient was computed using eddy-flux and meteorological data concurrently collected at a nearby study s ite with different soils but similar vegetation (D. Sumner, unpublished data ). The crop coefficient for the nearby site was computed by dividing daily actual ev apotranspiration computed using the data from the eddy-flux tower by daily reference ev apotranspiration computed using the data from the meteorological station. A leas t-squares polynomial was fit to the crop coefficient v. annual water year day data a nd used to compute a generic daily-varying crop coefficient for use in this study. Stage in a closed-basin depression wa s measured hourly with a pressure transducer and datalogger (Sol inst). Piezometers were inst alled in the CSA and in the surficial, intermediate, and Floridan aquifers surrounding and underneath the CSA
9 (Figure 1). Piezometers were installed using either the dire ct rotary or auger methods. Piezometers had 5 cm inside diameter by 1.5 m length PVC screens and 5 cm inside diameter PVC standpipes. Time-lag errors can arise in piezometers screened in lowconductivity formations such as the CSA (H anschke and Baird, 2001). The potential for time-lag errors was minimized during data anal ysis by interpreting time-series data over the course of days and weeks, which eliminat ed time-lags that occurred over the course of hours. A water budget was computed for a select ed closed-basin depression, hereafter called the north pond, where GW SW ET P S and where S is the change in storage (i.e., the stage of north pond), P is precipitation, ET is evapotranspiration, SW is net surface water inflow, and GW is net groundwater inflow. There were no su rface water inflows or outflows so SW was assumed to be negligible, and the simplif ied water budget was resolved in terms ofGW Therefore, the simplified water budget was ET P S GW and was resolved on daily time steps during a portion of the wet season to determine the relative contributions of groundw ater inflow and outflow to observed changes in stage. Tracer Application and Data Collection Bromide was used as an applied tracer to study hydrological processes in the upper part of the CSA. Ten infiltration rings 60 cm in diameter and 10 cm in height were
10 placed 5 m apart perpendicular to the surface slope on the norther n part of the CSA. Five infiltration rings were randomly assigned to be controls while five in filtration rings were randomly assigned to be treatments. Five L of 60 g/L LiBr soluti on was applied with a shower-head watering can to the surface of the treatment infiltration rings. All water immediately infiltrated and there was no ponding in any of the infiltration rings. Porewaters were collected from core sa mples. Cores were obtained by augering with an open-barrel auger to predetermined de pths, extracting the core material from the open-barrel auger, and placi ng the core material in wide-mouthed 300 mL HDPE sample bottles. Core samples ranged from approximately 100-300 g. Core samples were collected at 0-5 cm, 5-15 cm, 15-25 cm 25-35 cm, 35-45 cm, 45-55 cm, and 95-105 cm depths. Core samples from control infiltration rings were obtained on day 0 (June 28) right after application of the deionized water. Core samples from treatment rings were obtained on day 1 (June 29), week 1 (July 5), month 1 (July 26), and month 3 (September 29). After core samples were collected, bore holes were backfilled with native materials to avoid down-borehole contamination. Porewaters were extracted from the core samples using a hydraulic sediment squeezer (Manheim et al., 1994). Approximate ly 30-50 g well-mixed subsamples were placed in capsules and subject ed to pressures of approximately 500-1500 psi, depending on clay content. Porewaters were filter ed through 2.7 m hardened low ash filters (Whatman). Porewaters were collected in 10 mL syringes and transferred to 30 mL HDPE sample bottles. Porewater samples ranged from 1-10 mL.
11 Natural Geochemical Tracer Data Collection Naturally-occurring dissolved constituen ts and stable isotopes of hydrogen and oxygen were used as natural tracers to st udy the hydrological conn ectivity between the CSA and the surrounding hydrological landscap e. Rainfall samples were collected periodically, and surface water and groundwater samples were collected in the dry season (i.e., April 7, 2006) and wet season (i.e., Se ptember 4, 2007). Samples were collected from ponds and ditches on and around the CSA, from piezometers in the CSA, and from piezometers in the surficial, intermediate, and Floridan aquifers around and underneath the CSA (Figure 1). Some samples were co llected far from and/or upgradient of the CSA. These samples were considered to represent ambient conditions. Since the topographic gradient is generally north to south, these samples were largely collected north of the CSA. Other samples were coll ected close to and/or downgradient of the CSA. These samples were considered to potentially represen t mixes of ambient conditions and CSA conditions. Again, since th e topographic gradient is generally north to south, these samples were largely collected west, south, or east of the CSA. In both the dry season and the wet season, two ra infall samples and 23 surface water and groundwater samples were collected and anal yzed for dissolved constituents and 23 surface water and groundwater samples were collect ed and analyzed for stable isotopes. Samples were pumped through 0.45 m in-line filters (Whatman) directly into either 60 mL HDPE sample bottles for disso lved constituent analyses or 30 mL HDPE sample bottles for stable isotope analyses When possible, piezometers were purged
12 approximately ~3 volumes before samples were collected. Stable isotope sample bottles were filled completely with negligible head space and sealed with parafilm to prevent air contamination. Anion sample bottles were pr e-acidified with 1 mL of nitric acid. Laboratory Analyses Temperature, pH, and electrical conductiv ity were measured in the field (YSI). Major cation and anion analyses were conduc ted at the University of South Florida Center for Water Analysis. Major cation c oncentrations were determined by inductively coupled plasma-emission spectroscopy (ICP-OES) following the EPA 200 method (Clesceri et al. 1998), and major anion c oncentrations were determined by ion chromatography following the EPA 300 met hod (Clesceri et al. 1998). Bicarbonate concentrations were back-calculated by assu ming all other cation and anions had been measured and that bicarbonate accounted for the entire missing charge in charge balance error analysis. Stable isotope analyses were conducted at the UC Davis Department of Geology Stable Isotope Laborat ory. Deuterium analyses were performed using the hydrogen equilibration technique (Coplen et al. 1991), while oxygen-18 analyses were performed using the carbon dioxide equilibrati on technique (Epstein and Mayeda 1953). Deuterium and oxygen-18 are reported in the conventional, delta notation ( ) where 1000 1tan dard s sampleR R and where R is the ratio D/H or 18O/16O for deuterium and oxygen-18, respectively (Craig 1961). The result ing sample values of D and 18O are reported in per mil (Â‰)
13 deviation relative to Vienna Standard M ean Ocean Water (VSMOW) and, by convention the D and 18O of VSMOW are set at 0 Â‰ VSMOW (Gonfiantini 1978). Mass-Balance Mixing Modeling The relative contributions of rainfa ll, ambient surface wa ter and surficial groundwater (hereafter referred to as ambi ent water), and CSA gr oundwater (hereafter referred to as CSA water) to surface water a nd surficial groundwater downgradient of the CSA (hereafter referred to as downgradient water) were determined using a two-end, mass-balance mixing model with sodium as a conservative natural tracer. The massbalance mixing model was run in both dry and wet seasons. Rainfall and ambient water were similar, so rainfall and ambient wate r were considered a si ngle end member. End members were assigned mean sodium concentr ations of 0.83 mg/L for rainfall/ambient water, and 36.27 mg/L for CSA water. The mass-balance mixing model was 1 CSA RA CSA CSA RA RA DGf f Na f Na f Na where Na are sodium concentration, f are proportions that sum to one, and the subscripts Â‘DGÂ’, Â‘RAÂ’, and Â‘CSAÂ’ refer to downgradient water, rainfall/ambient water, and CSA water respectively. The primary assumption of the two-end, mass-balance mixing model was that a given sample was an instantaneous mix of the two end members.
14 RESULTS Applied Tracer Rain fell between the initia l tracer application and the first sampling effort, and continued to fall periodically between each a dditional sampling effort (Figure 2). Mean ( standard deviation) bromide concentrati ons in the control infiltration rings were 0.0019 ( 0.0023) g/L, while bromide concentra tions in the treatment infiltration rings were 2.10 ( 3.29) g/L (Appendix I). Maximum and minimum bromide concentrations in the treatment infiltration rings were found on day 1 and month 3, respectively. Vertical profiles show transport to ~0.5 m depth was s ubstantial by day 1, but th at transport to ~1 m depth was negligible even by month 3 (Fi gure 3). Bromide concentrations attenuated in the upper part of the profile without bei ng translated vertically down through the lower part of the profile suggesting that bromide wa s lost to lateral rather than to vertical downward flow. Pond Water Budget The north pond responded slowly to early we t season rainfall (Figure 4). Rain fell throughout the dry season and into the wet season without any surface water accumulation within the north pond indicati ng that rainfall was infiltrating and
15 Figure 2. Rainfall in cm/d at the CSA imme diately before, during, and after the tracer application. Arrows indicate core sampling efforts on day 0, day 1, week 1, month 1, and month 3, respectively.
16 Figure 3. Average bromide concentrations in g/L v. depth in the 5 control and 5 treatment infiltration rings. The 5 control infiltration rings were sampled on day 0, and the 5 treatment sites were sampled on day 1, week1, month 1, and month 3.
17 Figure 4. Precipitation in cm/d and stage of the north pond in cm immediately before and during the entire pe riod of inundation.
18 augmenting soil moisture. On July 15, stag e in the north pond rose 0.27 cm even though no rain fell on July 15th, and only 0.025 cm fell July 14th. Thereafter, stage in the north pond typically rose higher than could be exp ected from direct precipitation alone. In total, surface water was present in the north pond from July 15th-November 7th. During this time, water levels in the piezometer s within the CSA remained well-below the ground surface. Therefore, this water was perched above free groundwater in the lower portion of the CSA. There were no surface water inflows or outflows, and overland flow was never observed even during the rapid water applicatio ns to the control and treatment infiltration rings. Therefore, surface water inflows beyond some localized overland flow were believed negligible. If surface water inflow is neglected, then net groundwater inflow ranged from Â–3.9-12.0 cm/day (Figure 5). In Oc tober, when rainfall was negligible, net groundwater inflow was ~0.9 cm/day. Eva potranspiration was ~0.5 cm/day, so net groundwater inflow even during this period of steady drawdown was much larger than could be explained by error in th e evapotranspiration estimate alone. Natural Tracers Rainfall, ambient water, downgradient water, and CSA water had different geochemical signatures (Table 1; Figure 6; Appendix II). Rainfall and ambient water had relatively low concentrations of all dissolved constituents and were therefore combined in Table 1 and in the subsequent mass-balan ce mixing analyses. Dissolved constituent concentrations and isotopic ratios of the dow ngradient water were intermediate between
19 Figure 5. Net groundwater inflow to the nor th pond in cm/d immediately before and during the entire pe riod of inundation.
20 Figure 6. Stiff diagrams of the four types of water: (a) rainfall; (b) ambient water; (c) downgradient water; and (d) CSA water On ly the most common dissolved constituents are used, and concentrations are in meq/L.
21 Table 1. Physical and chemical properties of rainfall/ambient water, CSA water, and downgradient water. Values are means ( standard deviations).
22 dissolved constituent concentrations and isotopic ratios of rainfall/ambient water and CSA water. This suggests that downgradient water is a mix of rainfall/ambient water and CSA water. Mass-Balance Mixing Modeling Sodium best delineates rainfall/ambient water, downgradient water, and CSA water, with downgradient water having sodium concentrations that were intermediate between rainfall/ambient water and CSA wate r (Figure 7). The mean ( standard deviation) contribution of rainfall/ambient water to downgradient water was 0.81 ( 0.13), while the mean ( standard deviation) contribution of CSA water to downgradient water was 0.19 ( 0.13) (Table 2; Figure 8; Appendix III). Evidence of CSA water was found in surface water and in the surficial aq uifer around the perimeter of the CSA and in the surficial aquife r underneath the CSA.
23 Figure 7. Scatterplot of sodi um v. manganese concentrati on in mg/L i ndicating that downgradient water is a mix of rain fall/ambient water and CSA water.
24 Figure 8a. Postplot of the proportional cont ribution of CSA water to downgradient water in dry season. The size of th e circles is proportional to th e contribution of CSA water to downgradient water.
25 Figure 8b. Postplot of the pr oportional contribution of CSA water to downgradient water in wet season. The size of the circles is pr oportional to the contri bution of CSA water to downgradient water.
26 Table 2. Proportion of rainfall/ambient wate r and CSA water in downgradient water as determined by mass-balance mixing analyses using sodium as a conservative, natural tracer.
27 DISCUSSION Bromide infiltrated rapidly through the upper ~0.5 m but either did not infiltrate or infiltrated much more slowly below ~ 0.5 m. With time, bromide concentrations attenuated without being measurably translated down the profile. Uptake by plants might have accounted for some but not all of the observed bromide loss (Whitmer et al., 2000). The remainder was likely lost to lateral transport. Lateral flow was evident in the water budget for the north pond, which had substant ial amounts of groundwat er inflow and/or outlflow throughout the period of inundation. The hydraulic conductivity of clay typically ranges from 10-5-10-7 m/d (Morris and Johnson, 1967; Davis, 1969). These values were confirmed by slug tests (A. Cirra, unpublished data). Therefore, flow through the clay matrix alone cannot explain the apparent flow rates. Instead, preferen tial flow through desiccation cracks and other macropores likely dominates flow and transpor t in the upper ~0.5 m of the CSA. Rainfall infiltrates rapidly th rough the cracked clay surface la yer, and the subsurface water perches on the uncracked massive clay subl ayer below and flows laterally through the cracked clay surface layer across the CSA. The geochemical signature of CSA water was evident in the downgradient water both around the perimeter and underneath the CSA throughout the year. The relative proportions of CSA water in th e downgradient water were hi ghest in the dry season and
28 lowest in the wet season, perhaps due to diluti on by the additional rainfall. However, the precise flowpaths and fluxes from the CSA to the surrounding hydrological landscape are unclear. CSA water may discharge laterally over, through, or under the berms. The CSA has a slight topographic gradie nt from north to south, though smaller areas have slight topographic gradients to the west and east. Therefore, some of the perched subsurface water flows laterally toward the west, sout h, and east berms. Some of this water discharges to a system of ditches that di scharge from the CSA to the perimeter ditch through a culvert through the eas t berm. However, much of this water remains in the shallow subsurface and may discharge from the CSA to the surrounding surface water bodies and surficial aquifer over, through, and/or under the berms. This may be particularly true in the sout hwest corner where the berm has slumped and the topographic gradient steepens toward the southw est corner and the adjacent field. CSA water also may discharge vertically through the bottom of the CSA. CSAs are the highest topographic featur es in the region. Therefore, water stored in the CSAs is stored well-above the water in the surficial a quifer. Therefore, downward driving forces are high and may be high enough to drive water through the thick uncracked massive clay sublayer. However, it may also be true that the CSA water in the surficial aquifer below the CSA flows laterally from where it accu mulates around the perimeter of the CSA to the surficial underneath the CS A, and/or down the annuli of the surficial, intermediate, and Floridan piezometers drilled through the CSA. Closed-basin depressions form on CSAs because the mixed clay and sand deposits settle differentially a nd have low hydraulic conductivities. Lewelling and Wylie
29 (1993) suggested that surface water storage in these closed-basin depressions reduces runoff from CSAs. Our results indicate th at the hydrology of CSAs is more complex, with rapid infiltration, storage and flow in a perched subsurface flow system, and subsurface flow flowing through the closed-bas in depressions. Therefore, the closedbasin depressions and cracked clay surface la yer are assumed to be part of the same integrated surface water and perched subsurf ace water flow systems. Regardless, surface and subsurface water are temporarily stored above the uncracked massive sublayer and must either flow laterally over, through, or under the berms or verti cally through the thick uncracked massive sublayer before discharg ing to the surrounding surface water bodies or to the surrounding and underl ying surficial aquifer.
30 CONCLUSIONS In this study, we used applied and na turally-occurring geoche mical tracers to study the hydrology of CSAs and the hydrologi cal connectivity between CSAs and the surrounding hydrological landscapes. The results of this study indicate that older CSAs with well-developed cracked surface layers may have integrated surface water and perched subsurface water flow systems, and that water flows from older CSAs with welldeveloped cracked surface layers into the surrounding and underlying hydrological landscapes laterally over, through, or under the berms and/or vertically through the thick uncracked massive sublayer. However, the fluxes remain unquantified, so the effects of CSAs on the hydrology of the surrounding and underlying hydrological landscape also remain unquantified.
31 REFERENCES Allen, R.G. et al. (Editors), 2005. The ASCE Standardized Reference Evapotranspiration Equation. American Society of Ci vil Engineers, Washington DC. Bouma, J., Decker, L.W. and Muilwijk, C. J., 1981. A field method for measuring shortcircuiting in clay soils. J ournal of Hydrology, 52: 347-354. Bouma, J. and Dekker, L.W., 1978. A case study on infiltration into dry clay soil, I. Morphological observations. Geoderma, 20: 27-40. Davis, S.N., 1969. Porosity and permeability of natural materials. In: R.J.M.D. Wiest (Editor), Flow Through Porous Media. Academic Press, New York, pp. 54-89. Dekker, L.W. and Bouma, J., 1984. Nitrogen leaching during sprinkl er irrigation of a Dutch clay soil. Agricultural Water Management, 9: 37-45. Dekker, L.W. and Ritsema, C.J., 1996. Preferen tial flow paths in a water repellent clay soil with grass cover. Water Resources Research, 32: 1239-1249. Deming, D., 2002. Introduction to Hydrogeology. McGraw Hill, Boston. Ervin, K.L., Doherty, S.J. and Brown, M.T. (Editors), 1997. Evaluation of constructed wetlands on phosphate mined lands in Flor ida. Volume II. Final Report FIPR Project 92-03-103. Florida Institute of Phosphate Research, Bartow, FL. Fetter, C.W., 2001. Applied Hydrogeology, 4th edn. Prentice Hall, U pper Saddle River, NJ. Hanschke, T. and Baird, A.J., 2001. Time-lag er rors associated with the use of simple standpipe piezometers in wetla nd soils. Wetlands, 21: 412-421. Hawkins, W.H., 1973. Physical, chemical, a nd mineralogical properties of phosphatic clay slimes from the Bone Valley Formati on, University of Florida, Gainesville, FL. Heppell, C.M., Burt, T.P. and Williams, R.K ., 2000. Variations in the hydrology of an underdrained clay hillslope. Journal of Hydrology, 227: 236-256.
32 Johnson, R.L., Cherry, J.A. and Pankow, J.F., 1989. Diffusive contaminant transport in natural clay: A field example and implicati ons for clay-lined wa ste disposal sites. Environmental Science Technology, 23: 340-349. Lewelling, B.R. and Wylie, R.W., 1993. Hydr ology and water quality of unmined and reclaimed basins in phosphate-mining areas, West-Central Florida. U.S. Geological Survey Water-Resources Investigations Report 93-4002, US Government Printing Office: Washington, DC. Manheim, F.T., Brooks, E.G. and Winters, W.J., 1994. Description of a hydraulic sediment squeezer. 94-584, United States Geological Survey. Manning, J.C., 1997. Applied Principles of Hydrology. Prentice Hall, Upper Saddle River, NJ. Morris, D.A. and Johnson, A.I., 1967. Summary of Hydrologic and Phys ical Properties of Rock and Soil Materials, as Analyzed by the Hydrologic Laboratories of the U.S. Geological Survey 1948-1960, US Governme nt Printing Office, Washington, DC. Quirk, J.P. and Schofield, R.K., 1955. The e ffect on electrolyte concentration on soil permeability. Soil Science, 6: 165-178. Rains, M.C., Dahlgren, R.A., Williamson, R.J ., Fogg, G.E. and Harter, T., In Review. Geological control of physical and chem ical hydrology in vernal pools, Central Valley, California. Water Resources Research. Rains, M.C., Fogg, G.R., Harter, T., Dahlgr en, R.A. and Williamson, R.J., 2006. The role of perched aquifers in hydrological conn ectivity and biogeochemical processes in vernal pool landscapes, Central Valley, Ca lifornia. Hydrological Processes, 20: 1157-1175. Reid, I. and Parkinson, R.J., 1984. Nature of the tile-drain outfall hydrograph in heavy clay soils. Journal of Hydrology, 72: 289-305. Reigner, W.R. and Winkler, C., 2001. R eclaimed phosphate clay settling area investigation: Hydrologic model cal ibration and ultimate clay elevation prediction, Florida Ins titute of Phosphate Research, Bartow, FL. Sanders, L.L., 1998. A Manual of Field Hydrog eology. Prentice Hall, Inc., Upper Saddle River, New Jersey. Stricker, J.A., 2000. High value crop potential of reclaimed phosphatic clay soil. In: W.L. Daniels and S.G. Richardson (Editors), Annual Meeting of the American Society for Surface Mining and Reclamation, Tampa, FL, pp. 644-654.
33 Thomas, G.W. and Phillips, R.E., 1979. Consequences of water movement in macropores. Journal of Envi ronmental Quality, 8: 149-156. Tuller, M. and Or, D., 2003. Hydraulic f unctions for swelling soils: pore scale considerations. Journal of Hydrology, 272: 50-71. Vogel, H.J., Hoffman, H. and K.Roth, 2005. Studi es of crack dynamics in clay soil I. Experimental methods, results, and mor phological quantification. Geoderma, 125: 203-211. Whitmer, S., Baker, L. and Wass, R., 2000. Loss of Bromide in a Wetland Tracer Experiment. Journal of Envi ronmental Quality, 29: 2043-2045. Zang, P. and Albarelli, G.R., 1995. Phosphatic Clay Bibliography. In: FIPR (Editor), pp. 322.
34 LIST OF APPENDICES
35 Appendix I. List of all bromide concentrati ons for all control and treatment samples in g/L.
36 Appendix IIa. Dry season values for all constituents for all samples.
37 Appendix IIb. Wet season values for all constituents for all samples.
38 Appendix IIIa. Dry season results of mass-balance mixi ng modeling using sodium. End members are CSA water and rainfall/ambient surficial and surface water.
39 Appendix IIIb. Wet season results of mass-balance mixing modeling using sodium. End members are CSA water and rainfall/ambient surficial and surface water.