Final report on the effects of on-site sewage disposal systems on nutrient relations of groundwater and nearshore waters of the Florida Keys

[From the U.S. Government Printing Office, www.gpo.gov]

NOV 1987

FINAL REPORT ON THE EFFECTS OF

ON-S.ITE SEWAGE D1ISPOSAL SYSTEMS ON NUTRIENT RELATIONS

OF GROUNDWATER AND NEARSHORE WATERS OF THE FLORIDA KEYS

Prepared for:

Lr\
COASTAL ZONE

INFORMATION CENTER

Prepared by:

Brian E. Lapointe, Ph.D.
International Marine Research, Inc.
Rt. 3, Box 297A, Big Pine Key, FL 33043
305-872-2247

Funds for this project were provided by the Department of Enviornmental
Regulation, Office of Coastal Management, through funds made available
by the National Oceanic and Atmospheric Administration under the
Coastal Zone Management Act of 1972, as amended.

TD
771.5
.L37
1987

TABLE OF CONTENTS

LIST OF TABLES ----------------------------------------------------------- ii

LIST OF FIGURES --------------------------------------------------------- iii

PROJECT SUMMARY ---------------------------------------------------------- iv

INTRODUCTION -------------------------------------------------------------- I

BACKGROUND ---------------------------------------------------------------- I

MATERIALS AND METHODS ----------------------------------------------------- 6

RESULTS ------------------------------------------------------------------ 14

-----------
DISCUSSION --------------------------------------------------- --19

CONCLUSIONS AND RECOMMENDATIONS ---------------------------------- --------- 30

REFERENCES --------------------------------------------------------------- 33@

APPENDIX TABLE I ------------------------------------------------------------

APPENDIX TABLE II ------------------------------------------------------------

LIST OF TABLES

Table Page

1. List of groundwater monitoring stations in the Florida Keys .......... 33

2. List of hydrographic monitoring stations in nearshore waters of
Big Pine Key, FL ...................................................... 34

3. Mean values and ranges for nutrient concentrations, temperature
and salinity of groundwaters and adjacent surface waters in winter,
1986/1987 and summer, 1987 ........................................... 35

4. Comparison of nitrate concentrations of groundwater and adjacent
surface waters in winter, 1986/1987 versus summer, 1987 .............. 36

5. Comparison of ammonium concentrations of groundwater and adjacent
surface waters in winter, 1986/1987 versus summer, 1987 .............. 37

6. Comparison of soluble reactive phosphate concentrations of
groundwater and adjacent surface waters in winter, 1986/1987
versus summer, 1987 .................................................. 38

7. Comparison of nutrient concentrations in groundwaters from
residential stations versus the control station at the Key Deer
National Wildlife Refuge ............................................. 39

8. Mean values for nutrients and salinity at Halcyon Trailer Park
and Key Deer National Wildlife'Refuge .......................... ...... 40

9. Rate and direction of groundwater flow at Port Pine Heights,
Big Pine Key, during ebbing and flooding tides ....................... 41

10. Rate and direction of groundwater flow at Halcyon Trailer Park
and Key Deer National Wildlife Refuge.during high and low tides ...... 42

11. Mean values and ranges for concentrations of nutrients and
chlorophyll in seawater sampled at hydrographic stations along an
orishore-offshore transect in nearshore waters of Big Pine Key, FL.... 43

12, Correlation matrices of chlorophyll and nutrient concentration
data collected at hydrographic stations along an onshore-offshore
transect in nearshore waters of Big Pine'Key, FL ...................... 44

LIST OF FIGURES

Figure Page

1. Location of groundwater monitor stations on Key Largo, FL ............ 50

2. Location of groundwater monitor stations on Plantation Key,.FL ....... 51

3. 'Location of groundwater monitor stations on Marathon, FL .............. 52

4. Location of groundwater monitor stations on Big Pine Key, FL ......... 53

5. Calibration curve for GeoFlo groundwater flow meter .................. 54

6. Location of hydrographic stations in nearshore waters of Big Pine
Key, FL ......................... .................................... 55

7. GroundwaLer flow rate in Port Pine Heights, Big Pine Key; ebbing
tide .................................................................. 56

8. Groundwater flow rate in Port Pine Heigh ts, Big Pine Key; flooding
tide ................................................................. 57

9., Seasonal trends in hydraulic head based on monthly average
'groundwater table height and mean sea level .......................... 58

PROJECT SUMMARY

The effects of on-site sewage disposal systems (OSDS; septic tanks and

aerobic treatment units) on nutrient concentrations of upland groundwaters and

adjacent inshore waters of the Florid a Keys, Monroe County, was studied between

December, 1986 and September, 1987. Monitor wells designed to sample

groundwater at 8-10 ft below grade were installed at four residential stations

in the Upper Keys (Key Largo Limestone substrate) and four residential stations

in the Lower Keys (Miami Oolite substrate) to determine the effects of septic

tank systems,on groundwater quality. A deeper monitor well clu ster, co"sisting

of three wells of different depths (15-, 30-, 60' below grad e), was installed

adjacent to the injection well of an aerobic treatment unit. Control monitor

wells (15' and 30- below grade) were located in the pristene environs of the Key

Deer National Wildlife Refuge (KDNWR) and remote from potential sources of

contamination. All the monitor wells, as well as the most adjacent inshore

surface water (i.e. canal), were sampled monthly for determination of ammonium,

nitrate, soluble reactive phosphate (SRP), salinity, and temperature.

The results indicated that septic tank/drainfield systems cause extreme

nutrient enrichment of groundwaters, which apparently seep into adjacent surface

waters through natural groundwater flow.patterns. The highest concentrations of

nutrients were found in groundwaters adjacent to drainfields and the effluent of

the aerobic treatment units where concentrations as high as 2.5 mM for

ammonium, 2.3 mM for nitrate, and 120 pM for SRP occurred. Annual mean

concentrations of ammonium and nitrate in residential groundwaters were

approximately 350-fold higher than the control, whereas concentrations of SRP

were approximately 60-fold higher. The reduced level of SRP enrichment of

groundwaters in the Keys appears to be due to mineral formation associated with

carbonate geologies (i.e. fluoroapatite) and scavenging by oxides of iron and

alumiaum. A significant seasonality was evident-for the concentrations of all

nutrients in the groundwaters and surface waters; maximum concentrations of

groundwater nutrients occurred during winter (minimum during summer), whereas

maximum concentrations in surface waters occurred during summer (and minimum in

winter). This suggests that maximum discharge of groundwater nutrient loads into

surface waters occurrs during summer, which may be due to increased groundwater

recharge, hydraulic head (and groundwater flow), and mixing processes during the

higher sea levels that occur in summer compared to winter. Approximately

three-fold higher concentrations of chlorophyll also occurred in inshore surface

waters during summer, apparently in respone to groundwater nutrient enrichment.

Direct determiaations of groundwater flow rate at several locations on Big

Pine Key during summer 1987 indicated an average flow rate of 2.8 f t/day.

However, increased flo w was observed in response to ebbing tides (up to 5

ft/day) and rain events (up to 12 ft/day). Direction of groundwater flow could

not be pred icted as based solely on the decreasing grade elevations to the most

adjacent surface water, and explains previous failures of rhodamine dye to trace

septic effluents in to adjacent surface waters. Based on an average travel path

to the down stream receiving waters of 350' for one canal residence on Big Pine

Key, an average of 6 months would be required for discharge of nutrients into

surface waters. Considering the expanded tourist and resident population of the

Keys during winter, the resulting nutrient load might be expected to reach

surface waters during the following summ er. Such a "delayed discharge" would be

facilitated by increased sea level, groundwater recharge, hydraulic head, and

groundwater flow during late spring and summer.

INTRODUCTION

The marine environment of the Florida Keys is one of the most important

the economy of Monroe County. The clear, oligotrophic waters of the
assets to

Keys support extensive gr owth of corals and seagrasses that are linked directly

or indirectly, to tourism, commercial and sport fisheries, and the distinctive

"Key s" way of life. To determine potential impacts on the marine environment by

increasing use of on-site sewage disposal systems (OSDS), a one year study was

conducted to quantify effects of OSDS on nutrient concentrations of upland

groundwaters and adjacent inshore waters. Twenty groundwater monitor wells were

installed and, together with adjacent inshore waters, sampled for nutrient

concentrations (ammonium, nitrate, soluble reactive phosphate) as wellas

salinity and temperature. Rate and direction of groundwater flow was determined

at several sites on Big Pine Key to determine subsurface flow patterns and

quantify potential dispersion of septic leachate associated with OSDS in Miami

Oolite limestone substrate. Concentrations of nutrients and chlorophyll-a were

also measured at hydrographic stations in inshore and nearshore waters along a

transect extending from canal waters on Big Pine Key to offshore waters at Looe

Key National Marine Sanctuary-

BACKGROUND

The Florida Keys are truly unique in North America in Inaving the most

biologically diverse and productive shallow water (<10m) tropical marine

ecosystem. The archipelago island chain of the Keys separates the marine

environments of Florida Bay and the Gulf of Mexico from that of the Florida

Straits and Atlantic Ocean, with numerous tidal passes between the Keys

providing water exchange between these water bodies. The most unique feature of

the Keys marine environment is the extensive coral reef f ormations that extend

for 220 miles between Soldier Key and Dry Tortugas that include patch and bank

reef systems ranging from 25m to 1.3 km offshore and collectively referred to as

t
he Florida Reef Tract (Vaughn, 1914; Jaap, 1984). Coral reef ecosytems are the

most biologically diverse and productive ecosystems on earth (Goreau, t979) and

make tourism and recreation a major factor in the local and regional economy.

Several species of tropical seagrasses also form extensivemeadows in the

shallow Keys marine environment and are important as sources of food and habitat

for marine organisms, stabalization of sediments, and recycling and storage of

nutrients (Zieman, 1982).

The importance of maintaining water quality with increasing upland

development was a major element that spurred legislation t o protect the unique

Keys marine environme nt. In 1974, the Florida Keys were designated as an "Area

of Critical State Concern" under Chapter 380 of the Florida Statutes. Section

380.0552, "The Principles for Guiding Development in the Florida Keys Area of

Critical State Concern" was adopted by the administrative Commission ten years

later, in 1984, to "insure a water management system that will reverse the

deterioration of water quality and provide optimum utilization of our limited

aquatic resources, facilitate orderly and well planned development, and protect

the health, welfare, safety, and quality of life of the residents of this

state." Subsequently, in 1,985, the waters of the Florida Keys were designted as

"Outstanding Florida Waters" under Chapter 403 of the Florida Statutes.

in accordance with the guidelines of Section 380.0552 cited above, the use

of OSDS for wastewater management in the Florida Keys remains controversial. The

bulk (70%) of liquid domestic waste disposal in Monroe County is met by OSDS,

either septic tank/drainfield systems or aerobic treatment units, such as the

11multi-Flo" unit, coupled with injection wells. Considering that only 20% of the

53,000 platted, subdivided.lots in the Florida Keys are developed at present,

cumulative effects of increasing use of OSDS could have potentialy dramatic

effects on inshore water quality. While studies of septic tank systems in Dade

County, FL have concluded that septic leachate enters surface waters in canals

and causes water quality degradation (Barada, 1972) there still exists no

general agreement regarding the effects of OSDS on water quality in the Florida

Keys.

This lack of general agreement regarding impacts of canalization and septic

tank use on water quality in the Keys results from the disparate conclusions of

previous studies that have addressed this issue. For example, the studies of

Chesher (1973) reported satisfactory water quality in virtua-11y all of the

canals studied and concluded that "there were no adverse environmental

conditions attributable to septic tanks." However, the.studies of Hicks et al.

(1974) concluded that canal systems in the Keys have poor flushing

characteristics that result in frequent violations of both State and Federal

water quality criteria. Furthermore, although dye studies failed to demonstrate

septic leachate directly entering canal waters, elevated dissolved nutrients and

total organic carbon in developed canals appeared to be responsible for lower

oxygen levels in developed compared to undeveloped canals (Hicks et al., 1974).

An assessment of the use of OSDS on inshore water quality of the Keys will

necessarily have to address groundwater quality also. Groundwaters are known to

be important nutrient sources to lakes (Keeney et alw, 1971; Brock et al., 1982;

Loeb and Goldman, 1979) but only recently have.,they been demonstrated as

significant@nutrient sources to inshore marine waters. For example, groundwaters

are an important nutrient source in salt marsh systems on Cape Cod,

Massachusetts (Valiela et al., 1978), in nearshore waters of Long Island Sound,

New York (Capone and Bautista, 1985), and in back-reef habitats on coral reefs

along the north shore of Jamaica D'Elia et al., 1981). While a historical

importance of groundwater nutrient inputs on nutrient budgets of coastal waters

is recognized (e.g. Manheim, 1967) increasing development and agriculture in

upland, coastal areas are dramatically increasing the role of groundwaters in

nutrient budgets of nearshore marine waters (Capone and Bautista, 1985; Pye and

Patrick, 1983). The potential for nutrient enrichment of nearshore waters by

groundwaters enriched with septic leachate is exacerbated in the Florida Keys

because of the high porosity and permeability of its coral-derived substrate and

the-ubiquitous close proximity of OSDS to oligotrophic marine ecosystems; there

also exists the possibility that elevated tides and cross island heads

accelerate flow of enriched groundwaters toward oligotrophic marine habitats.

Studies of groundwater flow alo ng Florida-s east coast have clearly demonstrated

brackish groundwater fluxes into nearshore marine waters on the order of 45

3
m /day from a strip-one meter wide (Kohout, 1966), suggesting that enriched

groundwaters could become a major source of nutrients to nearshore environments

in South Florida.

Special concern regarding increased nutrient availability on the ecology of

nearshore waters of the Florida Keys is based on the known high degree of

nutrient limitation that is key to maintaining outstanding water quality in

these oligotrophic waters. Nutrient-limitation bioassays with several species of

dominant macroalgae in Pine Channel and Florida Bay demonstrated severe

limitation of productivity by phosphorus and nitrogen (Lapointe, 1987; Lapointe,

1988), supporting the the contention.that limited nutrient availability

regulates, to a large extent, marine plant growth in these waters. Consequently,

increased.nitrogen and phosphorus flux to nearshore waters of the Keys marine

environment could lead to increased marine biomass (phytoplankton and

macrophytes) resulting in increased microbial decomposition, decreased submarine

irradiance, and cumulative water quality degradation. This process of water

quality degradation, referred to as cultural eutrophication, is receiving

increas 'ed attention by marine scientists who are in general agreement that the

process is regulated primarily by nutrient loading (e.g. Ryther and Dunstan,

1971; Lee and Jones, 1981). Cumulative effects of eutrophication include reduced

water transparency, reduced dissolved oxygen Concentrations, odo rs (hydrogen

sulfide), fish kills, and a reduction of biological diversity. Because inorganic

forms of dissolved nitrogen (nitrate and ammonium) and phosphorus (soluble

reactive phosphate, SRP) are the most important forms associated with domestic

wastewater that support algal growth (Parsons et al., 1977), a knowledge of the

contribution of OSDS to concentrations of these nutrients in groundwaters and

surface waters of the Florida Keys is needed.

SCOPE OF THE PRESENT STUDY

The objectives of the present study were to:

1) determine if use of OSDS affects nutrient concentrations of

upland groundwaters and/or adjacent inshore waters.

2) determine groundwater flo w rates in typical geologies of the Keys to

quantify interaction of groundwaters with inshore marine waters.

3) De termine the relationship, if one exists, between dissolved

in organic nutrients and phytoplankton chlorophyll in nearshore waters

to predict potential effects of increased nutrient availabili.ty.

MATERIALS AND METHODS

Study Area

This study took place in Monroe County, FL, and included a variety of

residential canal locations that extended from Key Largo in the Upper Keys to

Big Pine Key in the Lower Keys. The surafce geology of the Upper Keys is

composed primarily of coral reef rock known as Key La rgo Limestone whereas that

of the Lower Keys is formed of small spherical grains of calcium carbonate

cemented together and known as Miami Oolite (Hoffmeister, 1974; Multer, 1971).

Key Largo Limestone is ver.y porous and riddled with numerous solution features

and voids that allow rapid vertical and horizontal groundwater flow;

consequently, this formations retains little fresh water because of its high

permeability and therefore, no freshwater lenses occur in the Upper Keys (Parker

et al., 1955; Hoffmeister and Multer, 1968). Although Miami Oolite in the Lower

Keys is also quite porous, it has fewer horizontal voids than the Key Largo

Limestone so retention of fresh water is enhanced and results in several fresh

water Ghyben-Herzberg lenses, most notably those of Big P ine Key (Hanson, 1980).

Monitor Wells: Design and Installation

To determine potential effects of septic tank leachate on groundwater

nutrient concentrations, nutrient concentrations were determined in groundwaters

of-16 monitor wells on eight upland residential lots with septic tank/drainfield
sy@tems currently in use'and.com pared to values for pristene groundwaters of the

Key Deer National Wildlife Refuge (KDNWR) that were considered the "control"(see

list of stations in Table 1 and location of the stations in Figs,1-4). Four

stations were selected in the Upper Keys and four in the Lower Keys to address

the different geologies of these areas and its possible effects on nutrient

relations of groundwaters and inshore waters.

Two monitor wells were installed on each of the eight residential lots. One

well was installed in the vicinity of the septic tank drainfield and the other

well was installed on the waterfront side of the lot, approximately halfway

towards the surface water closest to the septic drainfield. A portable hand-held

I" auger was used to bore a 10- deep borehole; initial development utilized a

portable well point sampler designed for groundwater plume tracking (Kerfoot,

1984). Following development, the boreholes were cased with 1/2 inch PVC pipe

that had a 2' long section of continuous slotted well screen (0.010 of an inch

slot width) adjusted to a horizon 8-10 ft below grade at each location.

Satura.ted groundwaters were commonly reached at 3-4 ft below grade at most

locations, but we believed that by sampling somewhat deeper groundwaters a more

representative and consistent sample of gr oundwaters would be realized.

To determine potential effects of injection well wastewater on groundwater

nutrient concentrations both at depth and in near-surface groundwat ers, a site

in the lower Keys (Halcyon Trailer Park; See Table 1) with a "Multi-flow"

aerobic treatment unit coupled to a 60- injection well was monitored. At this

site, a cluster of three monitor wells, each sampling a different and discrete

depth, was installed according to guidelines outlined in Driscoll -0986). From

grade, three boreholes (60', 30-, and 15') were augered along a transect

towards the closestadjacent surface waters from the injection well. The se

monitor wells were cased with 2" PVC pipe (to also allow groundwater flow

determinations, see below) a nd each casing had a 5' section of .010 inch

continuous slotted well screen at the bottom; annular space in these wells was

packed with with1coarse.carbonate substrate.

These monitor wells meet or exceed published guidelines for monitor well

installation for monitoring of dissolved inorganic nutrients as outlined by

Driscoll (1986) and for groundwater flow determinations as suggested by Kerfoot

(1986); well logs that contain details of the wells (e.g. grade and exact

locations of wells) are available from the Monroe County Planning Department.

Sampling of Groundwaters

The monitor wells were sampled monthly following installation, which began

in December 1986 and ended in September 1987. Initial groundwater samples were

collected with a portable well point sampler (Kerfoot-, 1984@; after casing of

the wells, either a submersible pump fitted with Tygon plastic tubing (for 2"

I.D. monitor wells) or a peristaltic pump fitted with teflon'tubing was used for

sample extraction. Sampling protocal consisted of initially removing

approximately 3-5 casing volumes prior to sample collection (Driscoll, 1986).

All wellpoint sampler parts, pump tubing an d fittings were rinsed with tap

water between well samplings to prevent cross contaminationof different

groundwater samples.

Collection and preservation of the groundwater,samples followed the

methodologies suggested in Standard Methods (Greenbaum, 1975). Specifically,

protoca.1 consisted of sample collection into acid-washed I liter Nalg6ne

polyethylene containers, immediate determination of temperature (using a mercury

thermometer) and salinity (using a Rausch and Lomb hand-held refractometer) and

subsequent preservation with a biocide (10mg/l HgCl). The samples were held on

ice in the dark until return'to the laboratory where they were immediately

filtered through a 0.45,p Gelman glass fiber filter and either analyzed

immediately for nutrient analysis, or frozen for subsequent analysis (within 2

weeks).

Determination of Ground-water Flow Rate and Direction

To understand interactions of groundwaters with inshore surface waters, a

knowledge of the direction and rate oE horizontal subsurface flow is needed.

Direct determinations of the rate and direction of groundwater flow were made at

several lo cations on Big Pine Key using a Model 30 GeoFlo groundwater flowmeter

(K-V Associates, Inc; Falmouth, MA). This flowmeter is a portable,

self -contained system that allows direct measurement of the rate and direction

of lateral flow of groundwater through permeable saturated geologies. The Model

30 GeoFlo flowmeter uses a submersible sensor consisting of a circular array of

thermistors arranged around a central heat source. A five v6ctor response is

displayed on an LCD readout which, in combination with a calibration curve and

vector worksheet, can allow det ermination of groundwater flow rate in the range

of .03-500 ft/day (+ 15%) and direction of groundwater flow (+ 10%).

The accuracy and precision of the GeoFlo flowmeter is greatly affected by

monitor well design and calibration of the flowmeter; therefore, procedures

suggested by the manufacturer were closely followed (See Kerfoot, 1986). The

GeoFlo flowmeter was deployed in 2" I.D. PVC monitor wells that had sections of

slotted well screen inserted at desired depths below grade. This involved use

of a high quality, continuous slotted well screen (Timco, sch 40, .010 inch slit

width, 59 slots per foot), centralization of the well, and careful annular

packing with washed, coarse carbonate substrate. Groundwater flow monitor wells

were installed at several locations on Big Pir@e Key and include three wells at

Halcyon Trailer Park (HTP, see above description of.monitor well installation)

as well as two wells (15' and 30') in the Key Deer Nati onal Wildlife Refuge

(KDNWR) and one well (15') in Port Pine Heights subdivision (PPH, See Table 1).

Calibration of the GeoFlo flowmeter. involved use of a flow chamber packed

with carbonate substrate similar to that surrounding th e monitor wells (Miami

Oolite) and a metering pump to provide controlled and variable flow rates. A

typical calibratio n curve for the GeoFlo flowmeter in Miami Oolite substrate is

illustrated in Fig 5; regression of the readout of the GeoFlo versus known rates

of lateral flow allows rapid, on-site determination of groundwater flow rates.

The five vector response is used with a worksheet to Lest for uniform cosine

flow and then plotted on polar gra@ph paper to determine direction of flow. Field

logs for all groundwater flow determinations are available from Monroe County

Planning Department.

Sampling of Surface Waters

During the monthly sampling at the eight residential monitor stations and

the "Multi-Flo" station on Big Pine Key, samples of adjacent surface waters were

also collected for determination of nutrient concentrations. Samples were

collected either by hand using a 3 liter Nalgene container or using a 3.7 liter

Niskin bottle sampler-Surface water samples were also collected at monthly

intervals along an onshore-offshore transect extending from an inshcre canal

system (PPH) on Big Pine Key to Looe Key National Marine Sanctuary (LKNMS),

located 5 miles south.of Big Pine Key; the hydrographic stations along th .is

transect are listed in Table 2 and locations are illustrated in Fig 6. Surface,

water samples coll ected'from canals adjacent to the residential stations were

handled in identical fashion to that described above for groundwater samples;

however,-the transect 'samples, which were also analyzed for chlorophyll-a, were

not spiked with a biocide because of interference with the chlorophyll

analysis.

Nutrient and Chlorophyll Analysis: Methodologies and Quality Assurance

Surface and groundwater samples were analyzed for dissolved inorganic

nitrogen in the form of nitrate, nitrite and ammonium using a Technicon

Autoanalyzer II system Preliminary analyses indicated that nitrite was

negligible compared to nitrate in groundwater and surface water samples at all

stations; comparable results have also been reported for similar carbonate

groundwaters and inshore waters of Bermuda (Simmons et al., 1984) and Jamaica

(D-Elia et al., 1979). Therefore, we determined and report herein total nitrate

and nitrite (referred to as nitrate, NO ) using the copper-cadmium reduction
3
method according to standard Technicon Industrial methodology (Technicon,1.972).

The ammonium determinations were made using a modified phenol-hypochlorite

method described bY Slawyk and MacIssac (1972). Nutrient concentrations are

reported in units of pM (= pg-at/1) to conform with the marine chemistry

literature. The detection limit during these analyses was 0.10 pM for nitrate

and 0.20,uM for ammonium.

Because phos phate appears to. be the primary limiting nutrient in the

nearshore Keys marine environment (Lapointe, 1987; Lapointe, 1988) and is

present at very low concentrations (e.g. frequently < 30-50 rLM), concentrations

of soluble reactive phosphate (SRP) were determined using the highly sensitive

manual method described by Strickland and Parsons (1977). This method is a

modification of the Murphey.and Riley (1962) molybdenum blue method and utilized

a Bausch and Lomb spectrophotometer 88 fitted with a 10 cm. cell for maximum

sensitivity.

Quality assurance of our nutri ent determinations is based on known internal

standards that we re analyzed regularly with all unknown samples. A continuous

record of analyses of standards.and recoveries was maintained and assures that.

our determinations were accurate and within acceptable upper and lower limits

(EPA, 1972). Interlaboratory comparisons of unknown samples.(available from EPA)

indicated that mean recoveries for our nutrient determinations are excellent and

range from 95-103% of stated EPA values.

Chlorophyll-a concentrations of seawater were determined using a Turner

Designs Model 10 fluorometer that was calibrated using known concentrations of

chlorophyll. A 800 ml seawater sample was filtered through a 0.45,u Gelman

glass fiber filter and analyzed for chlorophyll-a using a modified dimethyl

sulfoxide (DMSO)-acetone method (Burnison, 1979). Immediately following

filtration, the filters were placed''in 10 mls DMSO in a cool, dark place to

extract for one hour; following this, 15 mls of acetone was added. After two

hours of further ex.traction, the samples were analyzed for fluorescence.

Subsequent acidification with 10% HCL was also performed to correct for

phaeophytin.

Statistical Analyses

Several hypotheses were tested in this study. First, to consider potential

effects of OSDS on nutrient concentrations (ammonium, nitrate, and,SRP) of

groundwaters in the Florida Keys, groundwater nutrient concentrations of the

eight residential monitor stations were compared to nutrient concentrations of

the "control" station at the KDNWR. Specifically, the following null and

alternative hypotheses were tested:

H : nutrient concentrations of.groundwaters adjacent to OSDS are
0

equal to those of groun.dwaters of the KDNWR.

HA nutri ent concentrations of groundwaters adjacent to OSDS systems

are greater than groundwaters of the KDNWR.

Second, to consider potential seasonal seepage of nutrients from

groundwaters into the nearshore marine environment due to climatological

forcing, nutrient concentrations (ammonium, nitrate, SRP) of groundwaters and

nearshore marine waters (canal waters adjacent to sites) at the eight

residential OSDS sites during winter (December-April) were compared to those of

summer (May-September). Specifically, the follow ing null and alternative

hypotheses were tested:

H0: nutrient concentrations of groundwaters during winter are equal to

those of summer.

HA nutrient concentrations of groundwaters during summer are lower

than those of winter.

H nutrient concentrations of inshore marine waters are the same in

summer and winter

HA nutrient concentrations of inshore marine waters are higher

in summer than winter.

These a priori hypotheses were teste.d.using the Kruskall-Wallis test, a

conservative nonparametric test statistic. The experimental design involved

comparisons of data within individual stationsto reduce station-to-station

variability and increase the.power of these statistical tests. Because of the

sensitivity of our nutrient analyses and the randomized block experimental

design, we used a conserv ative alpha level of P=0.05 to represent the

probability of makin g a Type I error; thus, significance re ported in the results

below indicates the probability of making an incorrect inference is <0-05 or

less than 1 chance in 20.

RESULTS

Groundwater Nutrient Concentrations

Concentrations of ammonium, nitrate, and SRP as well as salinity and

temperature of groundwaters sampled at monthly intervals between December 1986

and September 1987 from groundwater monitor wells at the eight septic sites and

KDNWR are presented in Appendix Table 1. The highest groundwater nutrient

concentrations generally occurred during winter at the septic locations; lower

concentrations occurred at the midpoint locations and the lowest concentrations

typically occurred in groundwaters of the KDNWR. Over the entire study, ammonium

and nitrate concentrations ranged from 0.77 pM to 2.75, mM and 0.03,UM to 2.89

mM, respectively; SRP concentrations ranged from 0.06 pM to 107.4 PI (Table 3).

Salinity of the groundwater samples ranged from 0% (fresh) to 27% (saline) and

temperature ranged from a low of 21.01C in January to a high of 32.OOC in August

(Table 3).

Nutrient concentrations in groundwaters did not vary significantly between

the Upper and Lower Keys,'but an overall seasonal trend was evident.

Concentrations of ammonium, nitrate, and SRP in groundwaters from a majority of

the monitor wells decreased significantly from winter to summer, 1987 (See Table

3). Overall, nutrient concentrations decreased from winter to summer in 8 out

of 13 moni tor wells for nitrate (Table 4), 7 out of 13 wells for ammonium (Table

5), and 8 out of 13 wells for SRP (Table 6). The average groundwater nutrient

concentration (average of both septic and midpoint locations) during winter was

541 pM.for ammonium,.494)uM for nitrate, and 10.3 pM for SRP;. the average

concentration during summer w as 145 )uM for ammonium, 125 pM for nitrate,3nd 4 10

pm for SRP (Table 3).

Nutrient concentrations of groundwaters at the KDNWR "control" station did

not vary significantly from winter to summer and were consistently lower than

nutrient concentrations characteristic of the residential stations. During

winter, nutrient concentrations averaged 1.91 pM for ammonium, 0.76 pM for

nitrate, and 0.11 pM for SRP; during summer, nutrient co:lcentrations averaged
1.40 pM for ammonium, 0.20 pM for nitrate, and 0.14,pM for SRP (Tables 4-6).

A comparison of groundwater nutrient concentrations fro-m the residential

stations and the KDNWR station indicates significantly elevated nutrient

concentrations on developed, residential lots with septic tank/drainfield

systems. For example, concentrations of ammonium, nitrate, and SRP during winter

were all significantly higher in residential groundwaters compared to

groundwaters in the KDNWR in all 24 cases analysed; during summer, 17 out of 24

cases indicated significantly elevated nutrient concentrations in residential

groundwaters compared to groundwaters of the KDNWR (Table 7). The various

nutrients were.enriched in residential groundwaters, relative to groundwaters of

the KDNWR, some 625 to 650-fold for nitrat e, 246 to 283-fold for ammonium, and

29 to 94-fold for SRP.

Nutrient concentrations of groundwaters adjacent to the "Multi-Flo"

injection well were also significantly elevated compared to those of the KDNWR.

During summe r, concentrations of ammonium and SRP averaged 19.1,uM and 0.67 pl,

respectively, about 5-fold higher than concentrations of 3.35 PM and 0.12 PM*for
ammonium and SRP, respectively (Table 8). Concentrations of ammonium and SRP

were approximately the same at different depths.at the "Multi-Flo" site (15' to

60') whereas ammonium concentrations appeared'to increase with depth at the

KDNWR. Salinity increased with depth at both sites and waters at '00' at the

"Multi-Flo" site were always hypersaline (Table 8).

Groundwater Flow

Direct measurement of groundwater flow indicates that rainfall and tides

affect the instantaneous lateral velocity of subsurface groundwater movements.

At Port Pine Heights (PPH) on Big Pine Key, gr oundwater flow rate ranged between

0 and 5.0 ft/day during an ebbing tide on July 24 1987 with the lowest rates

occuring during peak high tide and highest rates occuring during the ebbing tide

(Fig 7); over the full 12 hrs of the ebbing tide, the flow rate averaged 2.3

ft/day (Table 10). On October 2 1987, groundwater flow rate at the same"site

during a flood ing tide ranged between 2.8 and 12.1 ft/day with.anomoulously high

flow rates occurring between 2130 and 2300 hrs when a major rain event occurred

(>1.0 inchds of rain fell in 6 hrs; Fig 8); the average flow rate during this

flooding tide was 5.5 ft/day (Table 9). During both groundwater flow studies,
the direction of groundwater flow ranged between 1.63 0 and 2220 and averaged 184 0

or southward (Table 9).

Groundwater flow rates in the KDNWR monitor wells (15' and 30' wells)

ranged from 0 to 4.10 ft/day and averaged 2.1 ft/day; the direction of flow at

0
both depths was consistently between 71 and 104 or eastward (Table 10).

However, flow rates were greater in the deeper (30') Key Largo Limestone

formation, which averaged 3.7 ft/day and ranged from 3.38 to 4.10 ft/day; flow

rates in the shallowe r Miami Oolite (15-) averaged 0.38.ft/day and ranged from

0.0 to 0.75 ft/day..

Groundwater flow rates at the Halcyon Trailer Park monitor wells (15', 30',

and 60'),.@veraged 2.2 ft/day and ranged from 2.17 to 3.38 f t/day; no differences

in flow rate-at different depths were apparent. However, at this station,

direction of groundwater flow was different belweoii the different depths flow
at 15' was 1970 or southwestward whereas flow at 60' was 3050 or northwestward

(Table 10).

Surface Water Nutrient Concentrations and Chlorophyll

Concentrations of ammonium, nitrate, and SRP as well as salinity and

temperature of surface waters sampled at monthly intervals between December 1986

and September 1987 from inshore waters adjacent to the groundwater monitor wells

at the eight residential sites and KDNWR are presented in Ap pendix Table 1. A

significant seasonal trend opposite that of groundwaters was observed for- the

inshore waters; consistently, the lowest nutrient concentrations occurred during

winter and the highest occurred during summer (Table 4-6). For example,

concentrations of nitrate ranged from 0.27/iM to 4.05/iM during winter and from

0.28 pM to 49.02 pM during summer; ammonium ranged from 0.15 pM to 2.39,UM

during winter and from 0.33 pM to 6.92 pM during summer; SRP ranged from 0.03 PM

to 0.35 pM during winter and from 0.12 uM to 1.60,uM during summer (Table 3).

In all 12 surface water samplings along the hydrographic transect, nutrient

concentrations and chlo rophyll consistently decreased towards offshore waters.

Over all these samplings, concentrations of nitrate ranged from 0.08,uM (LKNMS)

to 3.02,uM (PPH canal), ammonium ranged from 0.02 (LKNMS) to 0.94,uM (PPH

canal), SRP ranged from 0.03 pM (LKNMS) to 0.29 pM (PPH canal) and chlorophyll

ranged from 0.04 pg/1 (LKNMS) to 0.78,ug/l,(PPH canal; Table 11;,Appendix Table

2). Out of the twelve samplings, chlorophyll wds'significantly and positively (r

> 0.70) correlated with SRP six times, with ammonium twice, and with nitrate

three times (Table 12).

Chlorophyll concentrations were significantly greater during summer

compared to winter in the inshore waters during-these studies. Mean chlorophyll

concentration in PPH canal waters during winter were 0.22 jig/l (+ 0.08, N = 16),
about three-fold lower than the mean value of M2 jjg1l (+ 0.19, N =10) during

summer*

DISCUSSION

OSDS as a Nutri ent Source to Groundwaters and Nearshore Waters

Elevated nutrient concentrations in residential groundwaters compared to

pristene groundwaters in the Florida Keys indicates that OSDS represent a

significant source of nutrients to groundwaters of the Florida Keys. It is

unlikely that the high concentrations of ammonium (2.5 mM), nitrate (2.5 mM) and

SRP (30 jjM) observed in residential groundwaters would result from a natural

source such as the decomposition of leguminous matter, nitrogen fixation, or

rainwater (D'Elia et al., 1979). That nutrient concentrations were generally

higher at the septic locations compared to midpoint locations during the study

clearly point to a septic drainfield source for these elevated nutrient

concentrations. Furthermore, the highest nutrient concentrations observed in

groundwaters during this study are typical of secondarily-treated wastewater,

characterized by concentrations of 2-3 mM of inorganic nitrogen (either in the

form of ammonia or nitrate, depending on the degree of oxidation of effluent)

and 200-300,pM inorganic phosphate (Ryther et al., 1975; Goldman and Ryther,

1975), and are probably represent groundwater septic plumes. These nutrient

concentrations are also typical of effluents sampled from several aerobic

treatment units ("Multi-Flo") during this study.

Clearly, and not surprisingly, use of OSDS is resulting in increased

nutrient concentrations of associated groundwaters. The mean SRP concentration

of residential groundwaters (both midpoint and septic locations) was 10. 3,uM

during winter and 4.0 pM during summer, considerably higher than background SRP

concentrations of 0.11,,juM to 0.14,uM in the KDNWR during winter and summer,
respectively. Thus, SRP concentrations of enriched groundwaters on residential

sites of the Keys were on the order of 29- to 94-fold higher than background

concentrations with an annual average of 60-fold'higher SRP concentrations. For

nitrogen, even higher enrichment occurred. The average total nitrogen

concentration ( IN = ammonium and nitrate) on residential stations during winter

was 1036,uM, compared to a lower value of 471 pM during summer. Relative to

background concentrations in the KDNWR, this represents a 370 to 346-fold

increase during winter and summer, respectively,, with an annual average of

358- fold increase above background concentrations.

Elevated N:P molar ratios of enriched residential groundwaters during this

study suggests that SRP is, to some extent, scavenged during flow through

carbonate geologies. The N:P ratio of groundwaters (both the septic and midpoint

locations) ranged from 94:1 to 147:1, much higher than the 10:1 molar ratio

typical of domestic wastewater (Ryther and Dunstan,1971). In carbonate

geologies, SRP is well known to react and co -precipitate with calcium carbonate

to form calcium carbonate-phosphate surface complexes and/or the mineral apatite

(Berner, 1981); additionally, SRP can be scavenged by adsorption onto oxides of

iron (II), iron (M), and aluminum (III).- In both.oxic and anoxic sulfidic

sediments, apatite formation is considered the dominant mineral sink of
phosphate. The two primary minerals are fluroapatite (Ca5(PO 4)3 F) and
hydroxyapatite (Ca5(PO 4)3 OH), with the fluoride-rich.form being the most

predominant form in marine sediments. Consequently, concentrations of SRP are

often low in natuarl carbonate-rich waters because of equilibrium with carbonate

fluoroapatite (Gulbrandsen and Robertson, 1973). However, while removal of SRP

by calcium carbonate in upland groundwaters appears to be efficient wh6n based

on N:P molar ratios, SRP is not complet ely removed from enriched groundwaters as

described above by comparison of residential groundwaters to those of the KDNWR.

Accordingly, upland watersheds of the Keys enriched by OSDS appear to be

geochemically phosphorus-limited regarding their stoichiometric impact on

nearshore primary production. The most widely used approach to identify the

relative importance of nitrogen versus phosphorus limitation is through

determination of N:P ratios of seawater and comparison of these ratios with

those of marine plant populations. This approach is based on extensive analyses

of marine plankton and dissolved inorganic.nutrients, both of which indicate an

average N:P ratio of 16:1 by atoms in oceanic waters (Redfield, 1958).

Deviations of either marine plant composition or seawater nutrients from this

ratio is then used as an indirect method to infer which nutrient element limits

productivity; N:P ratios < 10:1 indicate N-limitation wheras ratios> 30:1

indicate P-limitation. While enriched groundwaters are clearly P-limited in that

N:P ratios are typically > 94, surface waters in canals during this study had

average ratios of ranging from 11 to 17, suggesting a trend towards N limitation

in enriched inshore receiving waters. However, frequent correlation of SRP and

chlorophyll-a in our nearshore transect studies suggest that elevated SRP is the

primary limiting nutrient to phytoplankton production, a conclusion also

reported for phytoplankton on the northwest Florida continental shelf (Myers and

Iverson, 1981). These results also concur with macrophyte bioassays in nearshore

waters of the Keys and Florida Bay where productivity was limited primarily by

phosphorous and secondarily by nitrogen (Lapointe, 1987; Lapointe, 1988). A

similar predominance of P-limitation occurs in carbonate-rich waters of Shark

Bay, Australia, (Smith and Atkinson, 1984) and contasts precepts of nitrogen

rather than phosphorus limitation of marine productivity in clastic environments

of temperate climes (Rythe r and Dunstan, 1971).

Our conclusion that elevated nutrient concentrations of groundwaters in

residential areas of the Florida Keys are anthropogenic in origin are in

agree .. ont with similar conclusions for Bermuda (Simmons et al., 1984) that were

based on elevated concentrations of nitrate in groundwaters and use of cesspits

as the dominant form of domestic wastew ater disposal. However, while the major

form of inorganic nitrogen in Bermuda's groundwaters is nitrate, groundwaters of

the Keys have higher concentrations of ammonium compared to nitrate. For

example, the mean concentration of nitrate in Bermuda's groundwaters is 749 PM

(ammonium is negligible) compared to the Keys where the annual mean

concentration of total inorganic nitrogen is 753 pM and consists of 310 PM

nitrate and 443 pM of ammonium.

We believe this difference in nitrogeno us composition of groundwaters

between Bermuda and the Florida Keys is significant and suggests generally low

oxygen availability in groundwaters of the Keys. Ammonium represents the most

reduced species of inorganic nitrogen and results from decomposition of organic

matter in oxygen-depleted waters; in the presence of oxygen, ammonium is rapidly

oxidized to nitrite and subsequently to nitrate by nitrifying bacteria. The

apparent lack of adequate oxic conditions needed to support oxidation of OSDS

wastewater (e.g. ammonium) in shallow grouridwaters may be related to the limited

vadose zone,underlying drainfields associated with OSDS in the Keys; an average

of 12" of vadose zone (above mean high water) separates drainfield wastewater

from the piezometric (hydrated groundwater) surface in the Keys. Considering

that groundwaters typically have low oxygen tensions even without impacts of

organic wastes, those impacted by organic wastewater loads will become suboxic

or anoxic. Because microbial mineralization of wastewater is much more rapid

under aerobic. conditions,,septic tank/drainfield systems with such limited

vadose zones may not provide complete oxidation of their wastewater, as
suggested by the elevate@ ammonium concentrations of residential groundwaters

found in this study. As-a consequence of impacting low oxygen groundwaters with

organic wastewater and associated oxygen demands (both BOD and COD),

biogeochemical zones (e.g. sulfide reduction zone that produces hydrogen

sulfide) usually restricted to deeper anoxic groundwaters will shift upward

towards to near-surface levels, increasing the occurrence of hydrogen sulfide

odors to the atmosphere. In contrast to ammonium that was characteristic of

sept ic tank enriched groundwaters, effluents of the aerobic t,reatment units were

composed primarily of nitrate, indicating a better-oxidized effluent.

The significant decrease of groundwater nutrient concentrations and

parallel increase in nutrient concentrations of inshore surface waters during

summer suggests that nutrients derived from enriched groundwaters are discharged

to adjacent inshore waters, to a large extent, in a seasonal manner. This

finding contrasts the view that groundwater nutrient concentrations are

relatively constant compared to surface waters (Freeze and Cherry, 1979).

Considering that all the residential stations used in our study were located on

canal sytems, which effectively increase shoreline development and decrease

groundwater dilution potential, then mixing and seepage of enriched groundwaters

with inshore surface waters is dramatically enhanced. Based on elevated ammonium

and total organic carbon concentrations in developed canals compared to

undeveloped canals, Hicks et al. (1975) also concluded that septic leachate

enters surface waters and be partially responsible for observed ecological

imbalances in adjacent waters. Considering the dramatic seasonality in extent of

groundwater intrusion to inshore waters, we sought to explain this phenomena by

considering the various mechanisms that regulate groundwater seepage to inshore

receiving waters.

Mechanisms of Groundwater Seepage to Inshore Waters

In general, nutrients associated with fresh groundwaters enter adjacent

surface waters primarily through horizontal groundwater flow, although inclined

and vertical flow can also occur (Visher and Mink, 1964; Cooper, 1959; Kohout,

1960). The general direction of flow is offshore because of decreasing hydraulic

gradients between the piezometric surface on land and sea level at adjacent

inshore surface waters where seepage through canal walls or bottom sediments

occurs.. The instantaneous rate of groundwater flow is a function of porosity

and permeability of the substrate and hydraulic head. Direct determination of

groundwater flow rates in the Lower Keys (PPH) during summer 1987 indicated

average instantaneous groundwater flow rates in Miami Oolite sub strate of 2.8

ft/day (omitting higher flow rates during rain events), a value consistent with

the known high porosity of this geology (bulk porosity of 40-60%; Evans, 1983).

Distinctly lower flow rates were associated with Miami Oolite in the KDNWR on

Big Pine Key where flow rates ranged from 0 to 0.75 ft/day, a finding consistent

with the ability of this geology to support a Ghyben-Herzberg fresh water lens

(Hanson, 1980).

Our observed decrease of dissolved nutrients in groundwater and

simultaneous increase in adjacent surface waters during summer suggests that a

seasonal climatic or astronomical forcing mechanism enhances flow rates of

nutrient-rich groundwaters into surface waters. Such a dramatic discharge would

require enhanced lateral groundwater flow during summer compared to winter, and

would be best explained by increased hydraulic head of groundwater during summer

compared to winter. Although freshwater recharge to groundwaters of the Keys is

ordinarily maximum during the "wet" summer.-and early fall in the Keys and could

explain seasonal differences in hydraulic head, such a typical rainfall pattern

lid no, occur during this study (NOAA weather, Key West) and no significant

seasonal correlation between rainfall and nutrient concentrations of

groundwaters and surface waters was apparent.

However, seasonal patterns in tide height did covary significantly with

our observed patterns for dissolved nutrients. Sea level is general ly

acknowledged as the dominant component of fluctuations of groundwater surfaces

on carbonate oceanic islands (Hanson, 1980; Rowe, 1984), although rainfall

becomes the more important factor during periods of heavy rain (Hanson,1980).

Annual variations of 0.8' in sea level occur in the Keys in response to

astronomical, isostatic, and mass transfer effects (Marmer, 1954; Pattullo,

1963) and results in maximum sea levels and groundwater tables during summer and

fall at which time the minimum seasonal potential evapotranspiration from

groundwaters occurs (the maximum evapotranspiration occurs during the "dry"

winter months). Consequently, a seasonal maximum in hydraulic head could occur

during summer and fall, during which the mixing rate of fresh groundwaters and

marine surface waters would result due to increased inclined flow (Visher and

Mink, 1964) and subsurface dispersion (Cooper, 1959; Kohout, 1960). This

possibility is directly supported by monthly average sea level and groundwater

table height data reported by Hanson (1980) for Big Pine Key and ill ustrated in

Fig 9; the increased hydraulic head during summer and early fall would support

increased lateral flow compared to reduced hydraulic head and flow during winter

and early spring. Such a seasonal mechanism is also supported by the elevated

salinities commonly observed in groundwaters of the Keys during summer as

compared to.winter (Table 3).

While groundwater,flow rates may vary seasonally as a function of hydraulic

head and regulate seasonal patterns in discharge of nutrients to inshore waters,

the most dramatic increases in groundwater flow occurs on shorter time scales

(i.e. hours-days) during rain events. Lateral groundwater velocities up to 12

ft/day occurred during rain events, causing rapid increases in groundwater

discharge to inshore waters that were some 5- to 7- fold higher than background

flow rates. The rapid flow response and lack of long lag periods of increased

flow is consistent with the high porosity and permeability of carbonate

geologies of the Keys. This finding.also explains the algal blooms that often

follow major rain events in inshore waters of the Keys (personal communication).

The direction of lateral, shallow groundwater flow appears to be related

primarily to the direction of the hydraulic gradient as affected by large scale

natural elevation grades and not necessarily by smaller scale man made changes.

For example, groundwater flow at the PPH residential station on Big Pine Key was

consistently (14 determinations over 3 months) southerly, the direction of

decreases in natural grade, even though the most adjacent surface water body

(canal) was 50- to the west of the flow monitor well. Previous studies using

Rhodamine dye injections into septic tanks on Big Pine Key failed to detect

septic leachate entering adjacent canal waters (Hicks et al., 1975), quite

possibly because the surface waters sampled for dye intrusion were not

downstream of the groundwater flow from the septic tank drainfield. Future dye

studies in the Keys need to obtain preliminary information regarding natural

subsurface flow patterns to support the assumptions generally made in such

tracer dye studies.

Patterns and Effects of Nutrient Flux to Nearshore Waters of the Keys

By assuming an average groundwater flow rate and path length to probable

receiving waters (not n,ecessarily the closest surface water), the travel time of

nutrients associated with groundwaters towards surface waters can be estimated.

For example, assuming an average flow rate of 2.0 ft/day for nutrients

associated with groundwaters and an average travel path of 350' to inshore

waters typical of the OSDS in our study, a travel time of approximately 0.5 yrs

would be required for discharge of new septic-derived nutrients into surface

waters. This suggests that seasonally increased nutrient input during the winter

tourist season would require about 6 months before discharge to inshore waters,

resulting in "delayed discharge" that would exacerbate increased flow of

grou ndwaters into surface waters during summer months. Considering that

irradiance and temperature are also maximum during summer, ideal conditions

favoring phytoplankton blooms result. A similar delayed discharge of nutrients

occurs on Cape Cod, MA, where nutrients introduced to groundwaters during the

summer tourist season are discharged,to surface waters during winter (Kerfoot,

personal communication).

The effects of increased nutrient seepage to nearshore waters of the Keys

will be cumulative enhancement of natural eutrophication processes that will

result in further water-quality degradation. Because the productivity and

nutrient dynamics of inshore waters of the Keys are controlled to a large extent

by benthic macrophytes that are themselves nutrient limited (Lapointe, 1987;
'Lapointe, 1988), increases in biomass of marine macrophytes, both seagrasses and

macroalgae, will undoubtedly occurr. Quite possibly, previous nutrient

enrichment has enhanced historic seagrass and macroalgal growth rates, resulting

in the extensive seagrass biomass harvests that are increasingly affecting

canals and shorelines when their harvested biomass decomposes, causing local

reduction of dissolved oxygen and strong hydrogen sulfide odors.

Several examples serve to illustrate the initially subtle but often

devastating ecological'imbalances that cultural eutrophication can have on

benthic marine ecosystems. Reef corals in Kaneohe Bay, Hawaii slowly became

overgrown with the green "bubble alga" Dictyospheria that bloomed in response to

nutrient enrichment from a secondary sewage,outfall; diversion of the sewage

outfall to an offshore location has since partially restored water quality

(Smith, 1981). In inshore waters of Bermuda, the green alga Cladophora prolifera

has increased dramatically over the past ten years in response to nutrient

enrichment by groundwater seepage resulting from widespread use of unlined

cesspits (Lapointe and O'Connell, 1988); this has resulted in water quality

degradation of Harrington Sound, including loss of the commercially-valuable

Calico clam, as well as cave systems that border Harrington Sound - some of

Bermuda's major tourist attractions (Ileffe et al., 1984). Phytoplankton blooms

resulting from sewage enrichment,of Hillsborough Bay, FL (adjacent to Tampa)

caused reduction of seagrass biomass and replacement by the red alga Gracilaria

(Taylor et al., 1973), a phenomenon that has also occurred in many other inshore

areas of Florida*.

However, effects of eutrophication will not be limited solely to enhanced

growth of marine macrophytes. The significant correlation of phytoplankton

chlorophyll and dissolved nutrients, particularly SRP, during our study suggests

that cumulative nutrient enrichment of inshore waters of thE Keys will also

result in a trend toward increased phytoplankton biomass that could cause the

most dramatic water quality degradation. Increased phytoplankton chlorophyll

concentrations are well known to degrade seagrass and coral reef ecosystems by

decreasing water transparency, submarine irradiance, dissolved oxygen, biotic

diversity and secondary production (e.g. see Zieman, 1982; Johannes, 1975).

Furthermore, blooms of the toxic red tide dinoflagellate species are increasing

world-wide in waters adjacent to developing coastlines, sugessting that run-off

derived from mans activities favor initiation of these devastating blooms. Red

tides of Gymnodinium breve on the east coast of Florida in 1972 originated from

se.ed that was carried from the west coast of Florida through the Keys to the

CONCLUSIONS and RECOMMENDATIONS

1) Based on the demonstrated extreme nutrient (ammonium, nitrate, and SRP)

enrichment of residential upland groundwaters by OSDS as well as

observed discharge of these nutrients into inshore waters, most notably

during summer and early fall and during rain events, we concur with

conclusions of Hicks et al. (1975) that water quality of inshore waters

of the Keys is measurably impacted by use of OSDS. As-described by Hicks

et al (1975), such impacts result in frequent and gross violations of

established Class III water quality criteria regarding dissolved oxygen

and biological integrity. Our results are also in agreement with those

of Pitt et al (1975) that also reported septic leaching into

groundwaters and found that hydraulic conductivity was the primary

factor controlling the extent of groundwater contamination. We further

suggest that increased growth of marine plants, both phytoplankton and

macrophytes, results from this enrichment and that this biomass

production exacerbates background BOD arriving in inshore waters with

nutrient enriched groundwaters. Our results do not support the

conclusion of Chesher (1974) that " there are no adverse environmental

conditions (in canal waters) attributable to septic tanks "; in

assessing that study, we believe that inadequate methodology and bias

confounded interpretation of data and precluded valid inferences.

Cumulative impacts of nutrients associated with OSDS as well as

ancillary but relat ed sources, should be quantitatively assess6d on an

areal basis prior to permitting of development of upland recharge areas

of any inshore or nearshore waters that are to be given the highest

degree of environmental protection. The following standards, adapted

from the Planning Board of the Town of Falmouth, MA, are given as an

example:

a) Loading per person: 5 lbs nitrogen/person/year; .25lbs

phosphorus/person/year for OSDS within 300 ft of shoreline.

Persons per dwelling unit = 3.0

b) Loading from lawn fertilizers: 3 lbs nitrogen/1000 ft 2 /year

c) Loading from road runoff: .19 lbs nirrogen/curb mile/day;

.15 lbs phosphorus /curb mile/day

d) Critical marine eutrophic levels: 16 lbs N/40,000ft 2/yr

While the above loading rates represent a long-term national,average, the

critical eutrophic levels cited above are specific to Cape Cod waters (i.e.

nitrogen-limited marine waters compared to phosphorus limited waters in the

Keys). We suggest that a study be performed to provide a quantitative data base

(i.e. flushing rates, concentrations of inorganic and organic nutrients,

dissolved oxygen, chlorophyll, etc.) that would allow development of accurate

critical eutrophic levels specific to inshore marine waters of the Keys.

3) Increased use of waste treatment should be considered for high

density, platted subdivisions that clearly are in excess of critical

eutrophic levels as cited above. A study should also be conducted to

consider cost/benefit relationships of various types of waste

treatment suitable for the Keys, which needs to include assessment of

improved OSDS (i.e. better drainfield designs), alternate sewers
(Godfrey, 1986; coupled to aerobic and/or tertiery treatment

facilities (assessing potential use of mangrove systems to strip

nutrients), ocean outfalls (Officer and Ryther, 1977), and deep

(below an aquiclude) well injection. In advanced waste treatment

designs, priority should be placed on phosphorus stripping as this

nutrient appears to limit eutrophication and, compared to nitrogen,

is more easily removed from wastewater effluents.

4) Considering the unique marine resources of the Florida Keys and the

importance of maintaining adequate water quality to the economic

well-being of Monroe County, a water quality monitoring program

specifically designed to address nutrient enrichment via groundwater

seepage, wastewater outfalls, and stormwater runoff and related

eutrophication of adjacent surface waters needs to be initiated. A

majority of experts in the water quality field believe that

eutrophication is potentially the most severe source of degradation of

natural waters (Clark et al., 1977). Such a monitoring program should

address the current status of water quality in Monroe County and also

use whatever existing historical data bases are available to document

any significant degradation. In addition to nutrients, groundwaters are

often contaminated by a spectrum of synthetic organic and metal

pollutants. Thus, the identification in the present study of nutrient

enrichment from groundwaters suggests that this is an additional route

that should be monitored for transfer of persistent metal and organic

pollutants to inshore waters of,the Keys.

REFERENCES

Berner, R.A. 1981. Authigenic mineral formation resulting from organic matter
decomposition in modern sediments. Fortschr. Miner. 59:117-135.

Brock, T.D., D.R. Lee, D. Janes and D. Winek. 1982. Groundwater seepage as a
nutrient source to a drainage lake; Lake Mendota, Wisconsin. Water Res.
16:1255-1263.

Burnison, B.K. 1980. Modified dimethyl sulfoxide (DMSO) extraction for
chlorophyll analysis of phytoplankton. Can. J. Fish Aquat. Sci. 37:729-733.

Capone, D.G. and M.F. Bautista. 1985. A groundwater source of nitrate in
nearshore marine sediments. Nature. 313:214-216.

Chesher, R.H. 1974. Canal survey, Florida Keys. Marine Research Foundation,
Inc., P.O. Box 2328, Key West, Florida.

Clark, J.W., W. Viessman, Jr. and M.J. Hammer. 1977. Water supply and pollution
control, 3rd. ed., IEP-A Dun-Donnelley Publisher

Cooper, H.H. 1959. A hypothesis concerning the dynamic balance of fresh water
and salt water in a coastal aquifer. J. Geophys. Res. 64:461-467.

D'Elia, C.F., K.L. Webb and J.W. Porter. 1981. Nitrate-rich groundwater inputs
to Discovery Bay, Jamaica: A significant source of N to local coral reefs?
Bull. of Mar. Sc i. 31:903-910.

Driscoll, F.G. 1986. Groundwater and wells. 2nd ed., Johnson Division, St. Paul,
MN. 1089pp.

Dugdale, R.C. and J.J. Goering. 1967. Uptake of new and regenerated forms of
nitrogen in primary productivity. Limnol. Oceanogr. 12:196-206.

EPA. 1972. Handbook for analytical quality control in water and wastewater
laboratories. U.S. Environmental Protection Agency, Analytical Quality
Control Laboratory, Cincinnati, Ohio. 103pp.

Hicks, D.B., T.R. Cavinder, B.S. Carroll, R.L. Raschke and P.M. Murphy. 1975.
Finger-fill canal studies, Florida and North Carolina. EPA-904/0-76/017.
U.S. Environmental Protection Agency, Surveillance and Analysis Division,
Athens, GA. 427pp.

Evans, C.C. 1983. Aspects of the depositional and diagenetic history of the
Miami limestone: Control of primary sedimentary fabric over early *
cementation and porosity development. Compariative Sedimentary Laboratoryl
U. of Miami, Rosenstiel School of Marine and,Atmospheric Sciences. 136pp.

Freeze, R.A. and Cherry, J.A. 1979. Grouridwater. Prentice-Hall, Englewood
Cliffs, N.J. 604pp.

Godfrey,. Jr, K.A. 1986. Alternate sewers-no longer alternate@. Civil Eng.
885:66-69.

Goldman, J.C. and J.H. Ryther. 1976. Nutrient transformations in mass cultures
of marine algae. J. Environ. Eng. Div. 101:351-364.

Goreau, T.F., N.I. Goreau and T.J. Goreau. 1979. Corals and coral reefs. Sci.
Amer. 123-136.

Greenberg, A.E., R.R. Trussell and L.S. Clesceri. 1985. Standard methods for the
examination of water and wastewater. American Public Health Association,
Washington, D.C., USA. 1268pp.

Gulbrandsen, R.A. and C.E. Roberson. 1973. Inorganic phosphorus in seawater. In:
Environmental Phosphorus Handbook. (Ed. by E.J. Griffith, A. Beeton, J.M.
Spencer and D.T. Mitchell), A Wiley-Interscience Publication, N.Y. 117-140.

Hanson, C.E. 1980. Freshwater resources of Big Pine Key, Florida. U.S. Dept. of
the Interior, Geological Survey. 35pp.

Hirayama, K. and S. Iizuka. 1975. Effect of anaerobic conditions in bottom
waters on the occurrence of red tides at Omura Bay. Environ. Let.
9:167-178.

Hoffmeister, J.E. and H.G. Multe r. 1968. Geology and origin of the Florida Keys.
Geological Society of America Bulletin. 79:1487-1502.

Hoffmeister, J.E. 1974. Land from the sea. University of Miami Press. 138pp.

Iliffe, T.M. and T.D. Jickells. 1984. Organic pollution of an inland marine cave
from Bermuda. Mar. Environ. Res. 12:173-189.

Jaap, W.C. 1984. The ecology of the south Florida coral reefs: a community
profile. U.S. Fish and Wildlife Service report Number, FWS/OBS-82/08.

Johannes, R.E. 1980. Mar. Ecol. Prog. Ser. 3:365-373.

Keeney, D.R., R.L. Chen and D.A. Graetz. 1971. Nature. 233:66-67.

Kerfoot, W.B. 1984. A Portable well point sampler for plume tracking.
Groundwater Monitoring Review. 4:38-43.

Kerfoot,, W.B. 1986. Monitoring well construction and recommended procedures for
direct groundwater flow measurements. Presented at: Symposium on field
methods for groundwater contamination studies and their standardization.
Cocoa Beach, Florida. 21pp.

Kimmmerer, W.J., T.W. Walsh, and J. Hirota. 1981. The effects of sewage
discharge into a wind-induced plume front. In: Estuaries and Nutrients
(eds. B.J. Neilson and L.E. Cronin) Humana Press, Clifton, N.J. 531-548.

Kohout, F.A. 1960. Cyclic flow of salt water in the Biscayne aquifer of
Southeastern Florida. J. of Geophys. Res. 65:2133-2141.

Lapointe, B.E. 1987. Macroalgal production and nutrient relations in
oligotrophic areas of Florida Bay. Bull. Mar. Sci. In press.

Lapointe, 8*E. 1181* Phosphorus and nitrogen-limited photosynthe 9is and growth
of Gracilaria tikvahiae (Rhodophyceae) in the Florida Keys: an experimental
field study. Mar. Biol. 93:561-568.

Lapointe, B.E. and J.D. O-Connell. 1988. Growth of Cladophora prolifera in
Harrington Sound, Bermuda: evidence for euLrophication of a P-limited
marine ecosystem.

Lee, G.F. and R.A. Jones. 1981. Application of the OECD eutrophication modeling
approach to estuaries. In: Estuaries and Nutrients (eds Nielson, B.J. and
Cronin, L.E.) 111-138. Humana, New Jersey.

Loeb, S.L. and C.R. Goldman. 1979. Water and nutrient transport via groundwater
from Ward Valley into Lake Tahoe. Limnol.' Oceanogr. 24:1146-1154.

Murphy, E.B., K.A. Steidinger, B.S. Roberts, J. Williams and J.W. Jolley, Jr.
1975. An explanation for the Florida east coast Gymnodinium breve red tide
of November 1972. Limnol. Oceanogr. 20:481-486.

Myers, V.B. and R.I. Iverson. 1981. Phosphorus and nitrogen limited
phytoplankton productivity in northeastern Gulf of Mexico coastal
estuaries. In: Estuaries and Nutrients (eds. B.J. Neilson and L.E. Cronin)
Humana Press, Clifton, N.J. 531-548.

Nixon, S.W. 1981. Remineralization and nutrient cycling in coastal marine
ecosystems. In: Estuaries and Nutrients (eds Nielson, B.J. and Cronin,
L.E.) 111-138. Humana, New-Jersey.

Officer, C.B. and J.H. Ryther. 1977. Secondary sewage treatment versus ocean
outfalls:,an assessment. Science. 197:1056-1060.

Parker, G.G., et al. 1955. Water resources of southeastern Florida with special
reference@`Tt7the geology and groundwater of the Miami area. U.S.G.S. Water
Supply Paper 1255

Parsons, T.R., M. Takahashi and B. Hargrave. 1977. Biological oceanographic
processes. Pergamon Press, New York. 332pp.

Pattullo, J.C. 1963. Seasonal changes in sea level. In: M.N. Hill (editor), The
sea. Wiley, New York, NY..485-496.

Pitt, Jr., W.A.J., H.C. Mattraw, Jr. and H. Klein. 1975. Ground-water quality in
selected areas serviced by septic tanks, Dade County, Florida. Open File
Report 75-607. U.S. Dept. of the Interior Geological Survey, Tallahassee,
Florida.

Pye, V.I. and R. Patrick. 1983. Ground water contamination in the United States.
Science 221:713-718.

Redfield, A.C. 1958. The biological control of chemical factors in the
environment. Am. Sci. 46:205-222.

Rowe, M. 1984. The fres.hwater "central lens" of Bermuda. J. of Hydrol.
73:165-176.

Ryther, J.H. and W.M. Dunstan. 1971. Nitrogen, phosphorus and eutrophication in
the coastal marine environment. Science 171:1008-1012.

Ryther, J.H., J.C. Goldman, C.E. Gifford, J*H* Huguenin, A*1. Wing, J *P
Clarner, L.D. Williams and B.E. Lapointe. 1975. Physical models of
integrated waste-recycling aquaculture systems. Aquaculture. 5:163-177.

Simmons, J.A.K., T. Jickells, A. Knap and W.B. Lyons. 1984. Nutrient
concentrations in ground waters from Bermuda: Anthropogenic effects. In:
383-398.

Slawyk G. and J.J. MacIsaac. 1972. Comparison of two automated ammonium methods
in a region of coastal upwelling. Deep-Sea Res. 19:521-524.

Smith, S.V. and M.J. Atkinson. 1984. Phosphorus limitation of net production in
a confined aquatic ecosystem. Nature. 307:626-627.

@.Strickland, J.D.H. and T.R..Parsons. 1977. A practical handbook of seawater
analysis. Alger press, Ltd., Ottawa, Canada. 310pp.

Technicon. 1973. Technicon AutoAnalyzer II, Industrial methods. Technicon
Industrial Systems, Tarrytown, N.Y.

Vacher, H.L. 1974. Groundwater hydrology of Bermuda. Bermuda Public Works
Department, Hamilton, 85pp.

Valiela, I., J.M. Teal, S. Volkmann, D. Shafer and E.J. Carpenter. 1978.
Nutrient and particulate fluxes in a salt marsh ecosystem: tidal exchanges
and inputs by precipitation and groundwater. Limnol. Oceanogr. 23:798-812.

Visher, F.N. and J.F. Mink. 1964. Ground-water resources in Southern Oahu,
Hawaii. U.S. Geological Survey. Water-Supply Pap. 1778.

von Bodungen, B., T.D. Jickells, S.R. Smith, J.A.D. Ward and G.B. Hillier. 1982.
The Bermuda Marine Environment, Vol. III., The final report of the Bermuda
Inshore Waters Investigations 1975-1980. Bermuda Biological Station Special
Publication No. 18. 123pp.

Zieman, J.C. 1982. The ecology of the seagrasses of South Florida: a community
profile. U.S. Fish and Wildlife Services, Office of Biological Services,
Washington, D.C. FWS/OBS-82/25. 158pp.

TABLE 1. Key to abbreviations for groundwater monitoring stations.

ABBREVIATION STATION

KDNWR ------------ -------------------- Key Deer National Wildlife Refuge

PPH ----------------------------------- Port Pine Heights, Big Pine Key, FL

EP ------------------------------------ Eden Pines, Big Pine Key, FL

.DA ------------------------------------ Doctor's Arm, Big Pine Key, FL

WP ------------------------------------ Whispering Pines, Big Pine Key, FL

DL ------------------------------------ Dodge Lake, Marathon, FL

YT ------------------------------------ Yello w Tail, Marathon, FL

TH ------------------------------------ Treas ure Harbor, Plantation Key, FL

OS ------------------------------------ Ocean Shores, Key Largo', FL

TABLE 2. Hydrographic monitoring stations and locations in nearshore waters
of Big Pine Key, FL.

STATION LOCATION LAT/LON
I-------------------- Port Pine Heights finger canal 24 042.88'N
81 023.87'W

2-------------------- Port Pine Hei ghts main canal 24 042.86'N
81 023.88'W

3-------------------- South Pine Channel 24 041.58'N
81 024.50'W

4-------------------- Munson Island 24 038.30'N
81 023.64'W

5-------------------- Hawks Channel 24 034'13'N
81 023:62-W
6-------------------- Looe Key National Marine 24 032.90'N
Sanctuary Back Reef 81 023.46'W

7-------------------- Looe Key National Marine 24 031-96'N
Sanctuary Fore Reef 81 024.22'W

M vim
TABLE 3. Mean values and ranges for nutrient concentrations (PM), temperature ('C) and salinity (o/oo) at
canal, midpoint and septic locations in winter vs.. summer. Groundwater represents the average of the
the midpoint and septic values. Values represent meanst standard deviation.

WINTER

LOCATION N N03 NH4 SRP N:P TEMPERATURE SALINITY

CANAL 32 MEAN 1.61 0.55 0.88.t 0.29 0.15 + 0.05 16.6 25.5 � 3.1 37.1 + 2.1
RANGE 0.27-----4.05 0.15 ----- 2.39 0.03 ---- 0.35 19.0--30.1 33.0--42.0

MIDPOINT 28 MEAN 118.25 + 179.89 256.94 + 325.88 2.54 :L 2.83 147.7 25.2 + 1.9 4.7 :L3.4
RANGE 0.03 --- 545.45 3.61--2750.34 0.12 --- 13.79 21.0=30.0 0.5--12.0

SEPTIC 32 MEAN 817.36.tIO52.80 784.89 + 808.27 17.00 26.85 94.3 26.0 + 1.9 3.3-�-3.9
RANGE 6.69--2896.70 32.41--2417.79 0.14--107.39 21.0--29.3 0.0--19.0

GROUNDWATER MEAN 467.81 + 494.34 520.92 + 373.32 9.77 + 10.22 25.6 �.0.6 4.0 -:L 1.0
RANGE 0.03--2896.70 3.61--2417.79 0.12-107,39 21.0--30.0 0.0--19.0

SUMMER

LOCATION N N03 NH4 SRP N:P TEMPERATURE SALINITY
CANAL 30 MEAN 3.22 + 8.38 1.69 1.48 0.43 + 0.38 11.4 30.7_� 1.3 40.9 zt 2.1
RANGE 0.28 ---- 49.02 0.33 ----- 6.92 0.12 ---- 1.60 .28.5--33.9 37.0--45.0

MIDPOINT 26 MEAN 30.76 + 89.91 188.45 � 306.32 1.63-t 2.78 134.5 27.9 -t 1.9 15.2 �12.4
RANGE 0.04 --- 409.02 0.77--1046.95 0.09 --- 14.14 25.5--32.0 1.0--45.0
SEPTiC 30 MEAN 220.12.L 482.13 502.97 + 784.07 6.37 + 16.28 113-5. 28.1 � 1.7 11.6 -+11.9
RANGE 0.10--1969.71 3.60--2579.00 0.06 --- 85.34 25.5--31.5 1.0--44.5

GROUNDWATER MEAN 125.44 � 133-89 345.71 -t 222.40 4.00 -t 3.35 117.8 28.0 O'l 13.4@� 2.5
RANGE 0.04--1969.71 0.77--2579.00 0.06 --- 85.34 25.5--32:0 1.0--45.0

TABLE 4. Statistical analysis (Kruskal-Wallis test) of summer vs. winter
nitrate concentrations (yM) at residential sites and a control site
at Key Deer National Wildlife Refuge. Values represent means I
standard deviation.

STATION LOCATION Y SUMMER N WINTER N P

PPH CANAL 2.00.t 1.41 10 1.71 ;L 1.35 10 N.S.

MIDPOINT 164.73.� [email protected] 6 258.46 t- 187.79 10 N.S.

SEPTIC 1.17 + 1.09 8 18.77-t 10.04 10 <.001

EP CANAL 1.32.� 0.36 8 1.02.� 0.32 10 N.S.

MIDPOINT 0. 57 :L 0.36 8 4.55 -t- 2.62 10 <.001

SEPTIC 196.49.� 385.52 8 66.77 71.77 10 N.S.

DA CANAL 14.65 23.20 8 1.31 0.73 10 N.S.

MIDPOINT 0.85 0.83 6 277.25 :L 303.57 10 .001

SEPTIC 393.91 783.84 8 1989.61.� 909.59 10 .004

DL CANAL 2.21 1.99 8 0.47 0.55 10- .047

MIDPOINT 2.03 + 3.13 8 9.02 11.92 10 N.S.

SEPTIC 2.54.� 2.42 8 350.28 527.79 10 <.001

WP CANAL 0.98 + 0.49 8 2.18� 1.63 8 N.S.

MIDPOINT 2.93 + 2.10 8 8.58 4.89 8 .019

SEPTIC 340.60 + 244.70 8 197.07 249.98 8 N.S.

YT CANAL 4.32 + 4.27 8 2.43 1.04 8 N.S.

SEPTIC 496.33.� 982.26 8 2742.03 150.83 8 .001

TH CANAL 1.05 + 0.63 8 1.65 1.00 8 N.S.

MIDPOINT 53.92 + 105.97 8 6.93 10.88 8 N.S.
SEPTIC 194.35 + 385.22 8 506.59 842.36 8 .034

KDR CONTROL 0.20 + 0.23 8 0.76 1.20 12 N.S.

N.S. P>0.05

TABLE 5. Statistical analysis (Kruskal-Wallis test) of summer vs. winter
ammonium concentrations (pM) at residential sites and a control
site at Key Deer National Wildlife Refuge. Values represent means
+ I standard deviation.

STATION LOCATION SUMMER N X WINTER N P

PPH CANAL 1.88 1.33 10 0.62.� 0.21 10 .003

MIDPOINT 11.18 9.95 6 16.38-�- 12.41 10 N.S.

SEPTIC 37.73 3.46 8 83.89 � 59.62 10 .001

EP CANAL, 1.74 + 0.45 8 0.84 + 0.13 10 <.001

MIDPOINT 16.49 � 11.13 8 23.63 9.99 10 N.S.

SEPTIC 2311.19 + 252.22 8 2146.90 473.46 10 N.S.

DA CANAL 2.69 � 2.95 8 0.81 0.51 10 .035

MIDPOINT 200.71 + 132.34 6 62.86 43.89 10 .016

SEPTIC 743.08 +1175.76 8 1415.57 778.03 10 N.S.

DL CANAL 0.83 0.83, 8 0.50 :L 0.23 10 N.S.

MIDPOINT 19.21 + 13.22 8 27.71 + 10.29 10 N.S.

SEPTIC 28.99 + 7.09 8 201.57 + 224.91 9 .007

WP CANAL 2.24 + 1.72 8 1.03 + 1.03 8 N.S.

MIDPOINT 188.10 + 59.39 8 137.80 51.25 8 .018

SEPTIC 507.51 + 191.99 8 259.79@ 176.84 8 N.S.

YT CANAL 1*35 + 1*32 8 0.96 0.75 8 N.S.

SEPTIC 156.0 274.36 8 254.50 + 169.63 8 N.S.

TH CANAL 1.55 0.75 8 1.40'� 0.67 8 N.S.

MIDPOINT 939.86 109.74 8 146.12 + 9.00 8 N.S.

SEPTIC 603.54 + 198.55 8 1588.40 � 400.84 .8 .001

KDR 1.40 + 0*48 8 1.91 + 1.39 12 N.S.

N.S. P>0.05

TABLE 1* Statistical analysis (Kruskal-Wallis test) of summer vs. winter
soluble reactive phosphate (,uM) concentrations at residential
sites and a control site at the Key Deer National Wildlife
Refuge. Values represent means -t I standard deviation.

STATION LOCATION 7 SUMMER N 7 WINTER N P

PPH CANAL 0.32 0.21 8 0.08 :L 0.02 10 <.00i

MIDPOINT 0.43 0.17 8 0.97 -t 0.78 10 N.S.

SEPTIC 0.15 0.07 8 0.92 0.81 10 <.001

EP CANAL 0.30 0.18 8 0.13 0.03 10 .008

MIDPOINT 0.23 0.17 8 0.86 0.97 10 .016

SEPTIC 18.16 + 7.62 8 65.82 38.07 10 .029

DA CANAL 0.94 + 0.66 8 0.19 0.09 10 .008

MIDPOINT 2.14 + 0.65 8 4.92 3.05 10 .031

SEPTIC 2.26 � 0.86 8 5.46 2.32 10 .001

DL CANAL 0.25 + 0.15 8 0.15 + 0.06 10 N.S.

MIDPOINT 6.05 5.40 8 7.15 + 4.08 10 .029

SEPTIC 0.84 0.72 8 0.77 t 0.37 10 .001

WP CANAL 0.68 0.64 8 0.12 � 0.02 8 .002

MIDPOINT 0.70 0.18 8 0.50 + 0.18 8 N.S.

SEPTIC 0.23,� 0.14 8 0.51 0.21 8 .006

YT CANAL 0.35 1- 0.17 8 0.15 0.07 8 .011

SEPTIC 24.73 + 40.54 8 44.53 32.94 8 N.S.

TH CANAL 0.37 + 0.11 8 + 0.06 8 .015

MIDPOINT 0.44 � 0.18 8 0.76.� 0.43 8 N.S.

SEPTIC 0.67 + 0.88 8 :1.08 1.02 8 N.S.

KDR CONTROL 0.14 �0.11 8 0.11 0.02 12 N.S.

N.S.= P>0.05

TABLE 7. Statistical analysis (Kruskal-Wallis test) to determine differences in
nutrient concentrations at residential stations and the control station
at Key Deer National Wildlife Refuge. Values represent the probability
that no difference exists between residential sites and the control
station; N.S.= not significant (P >.05).

LOWER KEY@S

WINTER SUMMER

LOCATION N03 NH4 SRP LOCATION N03 NH4 SRP

PPHM <.001 .001 <.001 PPHM .002 .117 .002

PPHS <.001 <.001 <.001 PPHS .012 .001 N.S.

EPM <.001 <.001 <.001 EPM N.S. .056 N.S.

EPS <.001 <.001 <.001 EPS N.S. .001 .001

DAM <.001 <.001 <.001 DAM .037 .002 .001

DAS <.001 <.001 <.001 DAS .001 .001 .001

WPM .001 <.001 <.001 WPM .001 .001 .001

WPS <.001 <.001 <.001 WPS .001 N.S.

UPPER KEYS.

WINTER SUMMER

LOCATION N03 NH4 SRP LOCATION N03 NH4 SRP

DLM .045 <.001 <.001 DLM N.S. .001 .001

DLS <.001 <.001 .008 DLS .012 .001 N.S.

THM .008 <.001 <-001 THM .055 .001 .004

THS <.001 <.001 <.001 THS .003 .001 N.S.

YTS <.001 <.001 <.001 YTS .003 .001 .001

0SM --- --- --- OSM .002 .006 .002

OSS --- OSS .019 .002 .002

TABLE 8. Mean values for nutrients (pM) and salinity (parts per thousand)
for Halcyon Trailer Park and Key Deer National Wildlife Refuge.
Values represent + I standard deviati6lie

LOCATION NO3 NH4 SRP SALINITY

KDNWR-30- 0.07 + 0.02 5.67 + 7.00 0.12 + 0.08 2.0 + 0.0

KDNWR-60' 0.33 + 0 .21 1.03 + 0.46 0.13 + 0.12 0.3 + 0.6

HTP-MF 2691.29 + 518.15 203.29 + 342.33 117.06 + 6.46 2.9 + 2.7

HTP-60' 0.13 + 0.04 17.96 + 4.37 0.74 + 0.38 41.8 + 2.9

HTP-30' 0.29 + 0.24 19.05 + 7.34 0.93 + 0.50 39.5 + 1.1

HTP-15' 2.63 + 4.11 20.29 + 8.49 0.34 + 0.18 212.2 + 7.1

HTP-Open 0.44 + 0.19 0.91-+ 0.94 0.35 + 0.17 42.6 + 3.0
Water

TABLE 9. Rate and direction of groundwater flow during a flooding and
ebbing tide at Port Pine Heights, Big,Pine Key, FL. Denotes
occurrence of a heavy rain event.

LOCATION DATE TIME TIDE RATE(ft/d) DIRECTION
PPH 7-24-87 1215 Flooding 2.58 1990

1345 High 0.00 ----
1515 Ebbing 2.21 193 0
1800 Ebbing 5.00 188 0
1930 Ebbing 2.64 163 0
2130 Ebbing 1.52 176 0

2300 Low 2.08 181 0

0
PPH Average Ebbing 2.29 183

0
PPH 10-02-87 1830 Low 2.83 190

2130 Flooding 12.10* 222

0
2300 Flooding 9.68* 160
10-03-87 0320 Flooding 4.19* 172 0
1130 High 3.32 187 0
1500 Ebbing 3.73 174 0
2230 Low 2.83 190 0

PPH Average Flooding 6.40 185 0

Heavy rain event

TABLE 10. Rate and direction of groundwater flow at Halcyon Trailer Park
(HTP) and the Key Deer National Wildlife Refuge (KDNWR) at
different depths and tides.

LOCATION DATE TIME TIDE RATE(ft/d) DIRECTION

0
HTP 15', 8-09-87 1500 High 2.17 305
HTP 15' 10-20-87 1830 Low 3.32 2770
HTP 60' 8-09-87 1330 High 2.17 1970

0
HTP 30' 10-20-87 1730 Low 1.21 188

KDNWR 15' 8-08-87 1345 High 0.00

0
KDNWR 15' 10-15-87 1710 Low 0.75 71

KDNW`R 30' 8-08-87 1515 High 3.38 1040

0
KDNWR 30' 10-15-87 1750 Low 4.10 87

TABLE It. Mean values and ranges of nutrient concentrations (juM) and chlorphyll-a (,ug/1)
along an onshore-offshore transect from inshore canal waters on Big Pine
Key (PPH) to offshore waters at Looe Key National Marine Sanctuary (LK).
Values represent means + I standard deviation.

STATION N03 NH4 SRP EN:P Chl-a

1--- PPH-BEL MEAN 1.60 + 0.92 0.57 + 0.23 0.09 + 0.07 24.1 0.34 + 0.22
RANGE 0.39 --- 3.02 0.22 --- 0.94 0.03 --- 0.29 0.09 --- 0.78

2--- PPH-MAIN MEAN 1.67 + 0.57 0.49 + 0.18 0.07 + 0.05 30.8 0.19 + 0.09
RANGE 0.38 --- 2.36 0.18---0.79 0.03 --- 0.22 0.07 --- 0.42

3 --- PINE CHANNEL MEAN 0.84 + 0.32 0.32 � 0.12 0.05 � 0.02 0.14 � 0.06
RANGE 0.26 --- 1.29 0.12 --- 0.48 0.03 --- 0.10 0.06 --- 0.21

4--- LITTLE MEAN 0.78 � 0.42 0.27 + 0.11 0.05 + 0.02 21.0 0.22 :t 0.11
MUNSON IS. RANGE 0.27---l.75 0.09 --- 0.49 0.03 --- 0.08 0.08 --- 0.50

5--- HAWKS MEAN 0.40 � 0.34 0.16 + 0.07 0.05 + 0.03 11.2 0.24 + 0.13
CHANNEL RANGE 0.13 --- 1.05 0.05 --- 0.31 0.03 --- 0.11 0.09 --- 0.50

6 --- LK-BACK REEF MEAN 0.40 + 0.32 0.21 � 0.16 0.05 + 0.02 12.2 0.18 � 0.09
RANGE 0.14 --- 0.94 0.02 --- 0.06 0.03 --- 0.10 0.10 --- 0.43

7S--LK-FORE REEF MEAN 0.38 + 0.33 0.23 + 0.14 0.04 � 0.02 15.3 0.13 t 0.07
SURFACE RANGE 0.08---0.96 0.02 --- 0.53 0.03 --- 0.08 0.05 --- 0.32

7B--LK-FORE REEF MEAN 0.52 � 0.38 0.22 + 0.09 0.06 � 0.04 12.3 0.19 � 0.10
BOTTOM RANGE 0.14 --- 1.35 0.02 --- 0.33 0.03 --- 0.16 0.07 --- 0.37

TABLE 11, Correlation matrices ol nutrlenL concentrations and c 410 rophyll-a
data along and onshore-offshore transect from inshore canal waters
on Big Pine Key to offshore waters of Looe Key National Marine
Sanctuary (LK). Data were collected At samplings from 10/17/86 to
6/19/87. Values represent correlation coefficients; *r >0.70 is
considered significant.

10/17/86 LK #1 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .82 .67 .52

AMMONIA .66 .82 *

PHOSPHATE .80 *

11/15/86 LK #2 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .68 .52 .86

AMMONIA .47 .49

PHOSPHATE .76

11/23/86 LK #3 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .96 .35 .78

AMMONIA .41 .68

PHOSPHATE .13

12/06/86 LK #4 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .89 .66

AMMONIA .82 .05

PHOSPHATE .39

12/21/86 LK #5 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .96 .94 * .81 *

AMMONIA .93 * .78 *

PHOSPHATE .88 *

1/08/87 LK #6 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .91 .95 * .27

AMMONIA .79 * .40

PHOSPHATE .24

.-TABLE 12. CONTINUED

1/28/87 LK #7 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .91 .72 .59

AMMONIA .81.* .47

PHOSPHATE .70

2/18/87 LK #8 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .84 .27 .34

AMMONIA .37 .43

PHOSPHATE .12

3/20/87 LK #9 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .81 .29

AMMONIA .21 .39

PHOSPHATE .31

4/09/87 LK #10 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .85 18 .04

AMMONIA .11 .30

PHOSPHATE .49

4/26/87 LK #11 AMMONIA PHOSPHATE CHLOROPHYLL

NITRATE .68 .02 .31

AMMONIA .45 .36

PHOSPHATE .86

6/19/87 LK #12 AMMONIA* PHOSPHATE CHLOROPHYLL

NITRATE .63 .55 .37

AMMONIA .93 .65

PHOSPHATE .79

LIME

BLACIMArER
&A RWS SOUNAD
SOUND

PT.

0 M

SCALE IN FEET

CROSS Us
LOCATION MAP SHELL KEY
KEY

INAKE Pl-
8LALWWA7*.WR 504VIV

T---T 11

BOGGY KEY

C:.

Burromwoo
$0110 5d*5 40

rAll0N 8AS.

&V770N;JlO0D
SOUND *S

10* --l-
Xe rl-
OCEAN
SHORES

FiEure 1. Location of groundwater monitor stations on Key Largo.

P t CIARLES

S.CALI it, FEE-.

TREASURE
HARBOR

LOCATION MAP

(9 RWRKWU KEY

A 41

CURRIE
COVE

Figure 2. Location of groundwater monitor stations on Plantation Key.

'T
rI
j

(ND
noo 7.-

SCALE IN FEET

LOCATION MAP

YELLOW
TAIL STIPP

DODGE
LAKE

KNIGHT
US KE

PIGEON
KEY

BOOT
KEY

@DLASSES ET
W SISTER 'ROCK E, SWER RO.CK
tiArKS CjiAVX0

Figure 3. Location of groundwater monitor stations ra".
m m m m = Oak

PORT PINE
HEIGHTS

f, 8:G TORCH
PINE K; f
KEY DEER
REFUGE
DOCTORS
ARM

Tcp I@EE NO NAME. KEY
KA@

TORCH
WDLE
EDEN
PINES

0
C, HALCYON
TRAILER
PARK
;SUMMERLAND
KEY WHISPERING
PINES

us@-j
SPANISH

NiWFOUND HARSOR
LITTLE I C R CH

RAMROD KEY
T@p -

Figure 4. Location of g-roundwater monitor stations on Big Pine Key.

Figure 5. Cal-ibiatJon Curve. for GeoFlow groundwater flow meter.!,'

Uj 10

0 r2=0.92
0
8 b=O-28

z 6
Ln
Uj

4 0
0

U-

25 50 75 100 125

INSTRUMENT READOUT IN COLUMN "F" (Relative Units)

PORT PINE HEIGHTS CA
r FINGER CANAL
@FIIG TORCH
q@j@jjl I BIG PIPIF kFy
P
@i 11119rud
T" PINE %."
3 sou PORT PINE HEIG Tq
CHANNEL 2MAIN CANAL

tIn IIAMf.
MIDDLE TORCH

"'."ERLAND
p KEY

US.%,

$PA)V;SH jj4A-30.,

LITTLE TORCH AWWFOuNa HAR80.9

RAMROD KEY

4 MAVER
_-----@-@@@MLINSON ISLAND

SCALE IN FEET

AHAWKS CHANNEL

6LOOE KEY NATIONAL
jrMARINE SANCTUARY
13ACK REEF
luld

LOOE KEY NATIONAL
c@@71MARINE SANCTUARY
FORE REEF

F19'ure 6. Location of hydrographic.stations in nearshore waters of Big Pine Key.
55

TIDE HEIGHT(MSL)
(inches)

0 0 N -P. 0) Go
1800

ago
Ln
2200

0200

CD
0600

0
-h
0 1000

1400
16

1800

CD
Cr

r+
2200

0 Co

LATERAL VEL06TY of GROUNDWATER
ft-d-1

TIDE HEIGHT (MSI.
(inches)

+ +
N) 41
1200 1

1300

0
a 1400'

,D
010i,
1500

1600

00S

so
0
r+ 1700 so

is
1800

1900

CD 2000

.%#
2100 oil
0
0

2200

2300

2400

LATERAL VELOCITY of GROUNDWATER
f t - d-'

Well 2
(1980)
18.0 Hanson

4.5 increased hydraulic head
increased conductivity a f ow

40'
15.0

4f.

> 40,'

Mean sea level
Ln
12.0

CU
J@

OF,

9.0
41f

6.0 A

CD Cr_ z CL
LLJ < CL
D
TIME
Figure. 9. Seasonal trends in- hydraulic head based on monthly average groundwat
ter.
table hei and"n Ef&f evjjM
"M

OAA COASTAL SERVICES CTR LIBRARY

3 6668 14110278 2