ML12338A684

Official Exhibit - ENT000328 -00-BD01 - USGS, Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, Fairfield County, Connecticut and Westchester County, New York, 2000-2002 (2004)

ML12338A684
Person / Time
Site:	Indian Point
Issue date:	12/31/2004
From:	US Dept of Interior, Geological Survey (USGS)
To:	Atomic Safety and Licensing Board Panel
SECY RAS
References
RAS 22135, 50-247-LR, 50-286-LR, ASLBP 07-858-03-LR-BD01
Download: ML12338A684 (71)

Text

Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, Fairfield County, Connecticut and Westchester County, New York, 2000-2002 In cooperation with the town of Greenwich, Connecticut Water-Resources Investigations Report 03-4300 U.S. Department of the Interior U.S. Geological Survey ENT000328 Submitted: March 29, 2012 United States Nuclear Regulatory Commission Official Hearing Exhibit In the Matter of:

Entergy Nuclear Operations, Inc.

(Indian Point Nuclear Generating Units 2 and 3)

ASLBP #: 07-858-03-LR-BD01 Docket #: 05000247 l 05000286 Exhibit #:

Identified:

Admitted:

Withdrawn:

Rejected:

Stricken:

Other:

ENT000328-00-BD01 10/15/2012 10/15/2012

U.S. Department of the Interior

U.S. Geological Survey Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, Fairfield County, Connecticut and Westchester County, 

New York, 2000-2002 In cooperation with the town of Greenwich, Connecticut By John R. Mullaney East Hartford, Connecticut 2004 Water-Resources Investigations Report 03-4300

District Chief

U.S. Geological Survey

101 Pitkin Street

East Hartford, CT 06108

http://ct.water.usgs.gov U.S. DEPARTMENT OF THE INTERIOR GALE A. NORTON, Secretary U.S. GEOLOGICAL SURVEY Charles G. Groat, Director The use of firm, trade, and brand names in this report is for identification purposes only and does not constitute endorsement by the U.S. Geological Survey.

For additional information write to:

Copies of this report can be purchased from:

U.S. Geological Survey

Branch of Information Services

Box 25286, Federal Center

Denver, CO 80225

Contents iii CONTENTS Abstract......................................................................................................................................................................

1 Introduction................................................................................................................................................................

1 Purpose and Scope............................................................................................................................................

3 Previous Investigations.....................................................................................................................................

3 Acknowledgments.............................................................................................................................................

3 Description of the Study Area...................................................................................................................................

3 Geohydrology....................................................................................................................................................

5 Surficial Deposits.....................................................................................................................................

5 Bedrock.....................................................................................................................................................

5 Precipitation and Runoff...................................................................................................................................

9 Methods of Data Collection and Analysis................................................................................................................. 13 Streamflow Measurements................................................................................................................................ 13 Water-Level Measurements.............................................................................................................................. 15 Water-Quality Samples..................................................................................................................................... 15 Water-Use Data................................................................................................................................................. 15 Ground-Water Recharge in the Greenwich Area....................................................................................................... 17 Factors Affecting Ground-Water Recharge...................................................................................................... 17 Ground-Water Recharge and Discharge, 2001-02........................................................................................... 18 Water Use in the Greenwich Area............................................................................................................................. 20 Water Use at Residences with Public Water Supply......................................................................................... 20 Log-Linear Regression Models of Residential Water Use................................................................................ 21 Variable Selection..................................................................................................................................... 21 Prediction of Residential Water Use in Areas with Domestic Wells....................................................... 22 Estimation of Consumptive Water Use.................................................................................................... 27 Water Use at Nonresidential Properties.................................................................................................... 27 Simulation of Ground-Water Flow in the Greenwich Area....................................................................................... 28 Description of Flow Model and Model Assumptions....................................................................................... 28 Boundary Conditions......................................................................................................................................... 33 Ground-Water Recharge................................................................................................................................... 33 Aquifer Properties............................................................................................................................................. 37 Internal Sources and Sinks of Water................................................................................................................. 38 Streamflow................................................................................................................................................ 38 Reservoirs................................................................................................................................................. 38 Ground-Water Withdrawals and Return Flow from Septic Systems....................................................... 38 Model Calibration............................................................................................................................................. 39 Hydrologic Budget............................................................................................................................................ 41 Ground-Water Availability in the Greenwich Area................................................................................................... 47 Water Quality in the Greenwich Area....................................................................................................................... 50 Summary and Conclusions........................................................................................................................................ 53 References Cited........................................................................................................................................................ 54 Appendixes 1-4........................................................................................................................................................... 57

iv Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 FIGURES

1.

Map showing location of public water and sewer service areas, Greenwich area, Connecticut.................

2

2.

Map showing location of subregional basins and data-collection sites for streamflow and 

ground-water-level measurements, Greenwich area, Connecticut..............................................................

4

3.

Map showing generalized surficial geology of the Greenwich area, Connecticut......................................

6

4.

Map showing generalized bedrock geology of the Greenwich area, Connecticut......................................

8

5.

Graph showing precipitation at Putnam Lake, Greenwich, Connecticut, 1967-2001................................ 10

6.

Map showing additional data-collection sites in Fairfield County, Connecticut........................................ 11

7.

Graph showing annual mean runoff at two long-term streamflow-gaging stations near Greenwich, 

Connecticut, 1967-01.................................................................................................................................. 12

8.

Graph showing precipitation at Bridgeport, Connecticut, and runoff at Sasco Brook near Southport, 

Connecticut (USGS station number 01208950), calendar years 1966-96.................................................. 12

9.

Cross section through wells installed for water-level measurements.......................................................... 16

10.

Graph showing water levels in bedrock wells in the East Branch of the Byram River Basin and 

precipitation at Putnam Lake, Greenwich area, Connecticut, September 2001-October 2002................... 19

11.

Graph showing quarterly base flow at U.S. Geological Survey streamflow-gaging station 

number 01211699, East Branch Byram River below Lake Mead at Round Hill, Connecticut................... 20

12.

Graph showing frequency distribution for residuals and the best-fit normal distribution 

approximation for (A) average daily water use, (B) average daily summer water use, and 

(C) average daily winter water use, Greenwich area, Connecticut............................................................. 23

13.

Graph showing estimates of average daily residential water use per square mile for small basins, 

Greenwich area, Connecticut....................................................................................................................... 25

14.

Graph showing estimates of average daily return flow per square mile from residential septic systems 

supplied by public water supply for small basins, Greenwich area, Connecticut....................................... 26

15.

Map showing extent of finite-difference ground-water-flow model grid, Greenwich area, Connecticut... 29

16.

Cross section through row 14 finite-difference ground-water-flow model, Greenwich area, 

Connecticut.................................................................................................................................................. 30

17.

Map showing location of observations for ground-water levels and streamflow, Greenwich area, 

Connecticut.................................................................................................................................................. 31

18.

Graph showing ground-water levels and streamflow, March to May 2002, Greenwich area, 

Connecticut.................................................................................................................................................. 32

19.

Map showing zone array based on geology used for (A) application of recharge to the model and 

(B) application of hydraulic properties to layers 1and 2, steady-state ground-water simulation, 

Greenwich area, Connecticut....................................................................................................................... 34

20.

Graph showing instantaneous streamflows per square mile at two continuous streamflow-gaging stations, 

Greenwich area, Connecticut, March to April 2001.................................................................................... 36

21.

Graph showing measured and predicted impervious areas for a random sample of model cells, 

Greenwich area, Connecticut....................................................................................................................... 37

22.

Graph showing weighted simulated equivalent plotted against weighted observations of hydraulic head 

and streamflow for model simulation of the Greenwich area, Connecticut, April 16-24, 2002................. 41

23.

Map showing simulated long-term, average annual ground-water recharge from precipitation 

aggregated by zones, Greenwich area, Connecticut.................................................................................... 44 24.

Map showing simulated long-term average net residential consumptive water use for zones, 

Greenwich area, Connecticut....................................................................................................................... 46

25.

Graph showing simulated ground-water outflow to streams, calculated base flow, and low flow for 

basins in the Greenwich area, Connecticut.................................................................................................. 48

26.

Graph showing percentage of urban land use and concentrations of selected water-quality constituents, 

Greenwich area, Connecticut....................................................................................................................... 51

Contents v TABLES 1.

Streamflow measurements, Greenwich area, Connecticut, 2000-02........................................................... 13 2.

Wells in the East Branch Byram River Basin, Greenwich, Connecticut...................................................... 15 3.

Median daily water use, and median daily seasonal water use by property size, in calendar year 2000 

at selected residences with public water supply.......................................................................................... 21 4.

Parameter estimates, standard errors, t-statistics, and p-values for multiple linear regression model of 

average daily water use at residences with public water supply, calendar year 2000, Greenwich, area, 

Connecticut................................................................................................................................................. 24 5.

Parameter estimates, standard errors, t-statistics and p-values for multiple linear regression model of 

average daily summer water use at residences with public water supply, calendar year 2000, 

Greenwich area, Connecticut...................................................................................................................... 24 6.

Parameter estimates, standard errors, t-statistics and p-values for multiple linear regression model of 

average daily winter water use at residences with public water supply, calendar year 2000, 

Greenwich area, Connecticut...................................................................................................................... 24 7.

Water levels in April 2002 and average water levels at selected U.S. Geological Survey 

network wells in Fairfield County, Connecticut.......................................................................................... 33 8.

Recharge estimates from historical values and nonlinear regression for April 2002 calibration................. 35 9.

Parameter estimates, standard errors, t-statistics and p-values for multiple linear regression model of 

percentage of impervious area for randomly selected cells of the finite-difference ground-water-flow 

model, Greenwich area, Connecticut.......................................................................................................... 35 10.

Simulated values of hydraulic conductivity in finite-difference ground-water-flow model, 

Greenwich area, Connecticut...................................................................................................................... 37 11.

Parameter estimates of hydraulic conductivity and recharge from nonlinear regression, finite-

difference ground-water flow model, Greenwich area Connecticut........................................................... 39 12.

Summary of error statistics and comparison of observed and simulated ground-water levels, 

Greenwich area, Connecticut, April 16-20, 2002....................................................................................... 40 13.

Observed and simulated streamflows, Greenwich area, Connecticut, April 24-25, 2002............................ 40 14.

Annual ground-water budget for modeled zones, based on calibration data from April 18-25, 2002, 

Greenwich area, Connecticut...................................................................................................................... 42 15.

Annual ground-water budget for modeled zones, based on calibration data from April 18-25, 2002 and 

adjusted to average residential water withdrawals, Greenwich area, Connecticut..................................... 43 16.

Estimated net annual residential consumptive water use and the difference between estimated long-term 

average recharge and the 30-day 2-year low flow....................................................................................... 49 17.

Percentage of urban land use and concentration of selected water-quality constituents in surface-water 

base-flow samples collected October 2000, Greenwich area, Connecticut................................................ 51 18.

Wastewater compounds detected on October 25, 2000 at station number 01211110, Greenwich area, 

Connecticut................................................................................................................................................. 52 APPENDIXES 1.

Records from two streamflow-gaging stations, March 2001-September 2002............................................ 58 2.

Estimated average daily ground-water use, Greenwich area, Connecticut.................................................. 62 3.

Estimated average daily summer ground-water use, Greenwich area, Connecticut..................................... 63 4.

Estimated average daily winter ground-water use, Greenwich area, Connecticut....................................... 64

vi Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 CONVERSION FACTORS AND VERTICAL DATUM Temperature in degrees Fahrenheit (F) may be converted to degrees Celsius (C) 

as follows:

C = (F - 32) / 1.8 Vertical coordinate information is referenced to the North American Vertical Datum of 1988 (NAVD 88).

Horizontal coordinate information is referenced to the North American Datum of 1927 (NAD 27).

Altitude, as used in this report, refers to distance above or below sea level.

Concentrations of chemical constituents in water are given in milligrams per liter 

(mg/L) and micrograms per liter (g/L).

Concentrations of indicator bacteria are given in colonies per 100 milliliters (mL) 

of sample.

Multiply By To obtain inch (in.)

25.4 millimeter inch per year (in/yr) 25.4 millimeter per year foot (ft) 0.3048 meter foot per day (ft/d) 0.3048 meter per day mile (mi) 1.609 kilometer square mile (mi2) 2.590 square kilometer cubic foot per second (ft3/s) 0.02832 cubic meter per second gallon (gal) 3.785 liter gallon per day (gal/d) 0.003785 cubic meter per day million gallons per day (Mgal/d) 0.04381 cubic meter per second million gallons per day per square mile

[(Mgal/d)/mi2]

1,461 cubic meter per day per square kilometer

Abstract 1 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 

Fairfield County, Connecticut and Westchester County, New York, 2000-2002 by John R. Mullaney ABSTRACT Ground-water budgets were developed for 32 small basin-based zones in the Greenwich area of southwestern Connecticut, where crystalline-bedrock aquifers supply private wells, to determine the status of residential ground-water consumption relative to rates of ground-water recharge and discharge. Estimated residential ground-water withdrawals for small basins (averaging 1.7 square miles (mi2)) ranged from 0 to 0.16 million gallons per day per square mile (Mgal/d/mi2). To develop these budgets, residential ground-water withdrawals were estimated using multiple-linear regression models that relate water use from public water supply to data on residential property characteristics. Average daily water use of households with public water supply ranged from 219 to 1,082 gallons per day (gal/d).

A steady-state finite-difference ground-water-flow model was developed to track water budgets, and to estimate optimal values for hydraulic conductivity of the bedrock (0.05 feet per day) and recharge to the overlying till deposits (6.9 inches) using nonlinear regression. Estimated recharge rates to the small basins ranged from 3.6 to 7.5 inches per year (in/yr) and relate to the percentage of the basin underlain by coarse-grained glacial stratified deposits. Recharge was not applied to impervious areas to account for the effects of urbanization. Net residential ground-water consump-tion was estimated as ground-water withdrawals increased during the growing season, and ranged from 0 to 0.9 in/yr.

Long-term average stream base flows simulated by the ground-water-flow model were compared to calculated values of average base flow and low flow to determine if base flow was substantially reduced in any of the basins studied. Three of the 32 basins studied had simulated base flows less than 3 in/yr, as a result of either ground-water withdrawals or reduced recharge due to urbanization. A water-availability criteria of the difference between the 30-day 2-year low flow and the recharge rate for each basin was explored as a method to rate the status of water consumption in each basin.

Water consumption ranged from 0 to 14.3 percent of available water based on this criteria for the 32 basins studied.

Base-flow water quality was related to the amount of urbanized area in each basin sampled.

Concentrations of total nitrogen and phosphorus, chlo-ride, indicator bacteria, and the number of pesticide detections increased with basin urbanization, which ranged from 18 to 63 percent of basin area.

INTRODUCTION New residential development in Connecticut is taking place in rural upland areas that are outside public water-supply areas. Residents in these areas rely on individual private wells drilled in the fractured crys-talline-bedrock aquifer for their water supply. Very little information is available about recharge rates to this aquifer or about the water withdrawals and consumptive use from private domestic wells.

The majority of residents in Greenwich, a community in southwestern Connecticut (fig. 1), rely on public water supply from local and regional surface-water reservoirs. Aquifers in the glacial stratified deposits in this part of Connecticut are limited in extent; therefore, any future development must rely on ground water from the crystalline-bedrock aquifer for water supply. Ground water is the primary source of water supply for approximately 12 percent of the towns households that are outside the public water-supply area (fig. 1) (U.S. Department of Commerce, 1991). Ground-water recharge is the ultimate source of

2 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD GREENWICH Westchester County, New York Westchester County, New York EXPLANATION Area with private wells and septic systems Greenwich Connecticut Massachusetts New York Rhode Island Area with public water supply and septic systems Area with public water supply and sewers Long Island Sound Connecticut New York Putnam Lake Rockwood Lake Converse Lake Bargh Reservoir Figure 1. Location of public water and sewer service areas, Greenwich area, Connecticut.

Description of the Study Area 3 water to the crystalline-bedrock aquifers, but informa-tion is sparse on recharge rates to these bedrock aqui-fers in New England.

Increasing development and the lack of large glacial stratified aquifers for public water supply have led to a need to study ground-water availability and water use in basins in the Greenwich area. The U.S.

Geological Survey (USGS) and the town of Greenwich began a cooperative study in 2000 on water use, ground-water recharge and availability, and quality of water in several coastal basins in southwestern Connecticut. The information presented in this report may be applicable to other parts of New England and New York where the issue of sustainable water use from bedrock aquifers is of increasing concern. Water-use estimates are relevant especially to other parts of Fairfield County, Connecticut, and Westchester County, New York.

Purpose and Scope The purpose of this report is to characterize resi-dential ground-water use, ground-water recharge and availability, and quality of water in selected coastal basins in Fairfield County, Connecticut and Westchester County, New York. The report focuses primarily on the Greenwich, Connecticut area. It presents data collected during 2000 through 2002 on streamflow, ground-water levels, and water quality of base flow. Three multiple-linear regression models were used to estimate self-supplied water use from ground-water sources for average daily, winter, and summer conditions in each of 32 small basin-based zones. A steady-state ground-water-flow model was used as a tool to assist in estimating recharge rates, hydraulic conductivity of crystalline bedrock, and in the analysis of ground-water budgets and ground-water availability for each of the 32 zones.

Previous Investigations The ground-water resources of the study area were described previously in several reports. Gregory and Ellis (1916) described the hydrogeology, water supply, and water use in the Greenwich-Stamford area.

Ryder and others (1970) conducted a water-resources inventory of the southwestern Connecticut coastal basins, including information on the hydrogeology, surface water, water quality, and water use of the region. Wolcott and Snow (1995) estimated ground-water recharge for northern Westchester County, including some areas in the Byram and Mianus River Basins that drain through Greenwich.

Acknowledgments The following town of Greenwich officials and departments are thanked for their assistance with this investigation: Aleksandra Moch, Denise Savageau, and Steven Danzer of the Greenwich Conservation Depart-ment; the Greenwich Health Department; and the Greenwich Department of Public Works. The National Audubon Society and the Boy Scouts of America are thanked for allowing ground-water level and stream-flow monitoring on their properties; the numerous resi-dents of Greenwich are thanked who allowed access to their wells for water-level monitoring; and David Medd of the Aquarion Water Company of Connecticut is thanked for providing water-use and precipitation information. Thanks also are due to Anna Maria Marrone, assessor of the town of North Castle, New York, and Cindy Barber, GIS analyst for the city of Stamford, Connecticut. The following USGS employees are thanked for their contributions to this study: Gregory Schwarz, Bruce Davies 3rd, J. Jeffrey Starn, Remo Mondazzi, Staunton Williams Jr.,

Timothy Frick, Joseph Martin, and Barbara Korzen-dorfer.

DESCRIPTION OF THE STUDY AREA The study area (fig. 2), referred to in this report as the Greenwich area, contains 52.8 mi2 in Fairfield County in the southwestern coastal part of Connecticut and Westchester County, New York. The town of Greenwich, which makes up most of the study area, had a population of 61,101 in 2000 (data accessed on August 26, 2003, on the World Wide Web at URL http://www.census.gov). Development in the Green-wich area ranges from homes on small lots (less than 0.5 acres) and commercial and business districts with public water supply and sewer in the southern part of the town, to large lots (more than 4 acres) and no commercial activity in the more rural northern areas.

Residential water use in some parts of the town likely is higher than in other parts of Connecticut due to the large lot size and lawn area and the large number of swimming pools. In 1999, median annual household

4 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 01211699 01211110 01211010 01211040 CT-GW-21 CT-GW-23 CT-GW-22 01211450 01211600 01212100 01211210 01212550 01212600 Greenwich Creek Byram River Horseneck Brook Putnam Lake Mianus River Blind Brook Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD GREENWICH Westchester County, New York Westchester County, New York EXPLANATION A

A' Long Island Sound Rockwood Lake Converse Lake Bargh Reservoir Kensico Reservoir A

A' Line of section (fig. 8)

Subregional basin boundary Study area boundary 01211600 Water-level observation well, with identifier Continuous streamflow-gaging station and water-quality site, with identifier Water-quality site with streamflow measurement(s) and identifier Streamflow measurement site, with identifier CT-GW-21 01211699 01211450 01211140 MILES KILOMETERS Figure 2, Location of subregional basins and data-collection sites for streamflow and ground-water-level measurements, Greenwich area, Connecticut.

Description of the Study Area 5 income in Greenwich was $99,086 as compared to the median annual household income of $53,935 in the State of Connecticut (data accessed on August 26, 2003, on the World Wide Web at URL http://www.cen-sus.gov).

The town of Greenwich is in the Byram River and Mianus River drainage basins. Several smaller subregional basins that also drain to Long Island Sound are within the town (fig. 2). Altitude of the Greenwich area ranges from sea level on Long Island Sound to almost 600 ft near the northern border with Westchester County, New York. The water table is generally shallow and at a consistent depth below land surface at most places in the study area. Water levels in upland areas, such as Greenwich, typically are a subdued reflection of the topography.

Geohydrology Two types of aquifers are present in the Green-wich area: (1) aquifers in surficial deposits including till, glacial stratified deposits, post-glacial alluvium, and swamp deposits; and (2) aquifers in the fractured crystalline bedrock.

Surficial Deposits The surficial geology of Connecticut has been most recently described by Stone and others (1992).

The major surficial deposits in the Greenwich area are till and glacial stratified deposits. Till is an ice-laid deposit containing a nonsorted mixture of gravel, sand, silt, and clay. Till overlies the bedrock in most places in the Greenwich area and is the primary unconsolidated material.

Tills of two separate glaciations are present in Connecticut. The characteristics of these two tills are summarized in Melvin and others (1992). The upper till (surface till) was deposited during the last glaciation from about 23,000 to 16,000 years before present.

Surface till is generally sandy with many boulders and is thin in areas with numerous bedrock outcrops.

Surface till has been noted up to 33 ft thick in Connect-icut (Melvin and others, 1992). The average thickness of the upper till is 12 ft based on well-completion reports for 462 selected wells in Greenwich. The composition of the till is related to nearby bedrock types and to other surficial deposits that were present before the last glaciation. Drumlin till (lower till) is interpreted to have been deposited during an earlier glaciation (Melvin and others, 1992). Drumlin till is inferred to be present in the cores of drumlins and in thick till deposits (drumlin till possibly covered by surface till) in the Greenwich area (fig. 3), and may range in thickness from 30 to 100 ft. Based on the selected records of 121 wells in Greenwich, the average thickness of till deposits in areas with thick till is 35 ft.

Thick till deposits containing drumlin till are present primarily in the northern half of Greenwich (fig. 3).

Glacial stratified deposits include sand and gravel, silt, and clay deposited in glacial meltwater streams or lakes by the retreating glacier (fig. 3). In other parts of Connecticut, these deposits are the most productive aquifers for public ground-water supplies.

They are important areas for storing ground water and providing base flow to streams because ground-water recharge rates to these deposits are greater than recharge rates to deposits of till. Glacial stratified deposits are limited in the Greenwich area, as compared to other parts of Connecticut, and consist mostly of small and thinly saturated areas of sand and gravel. The largest area of glacial stratified deposits underlies the Tamarack swamp (fig. 3), in the north-western part of Greenwich, where swamp deposits overlie fine-grained material and sand. According to Ryder and others (1970), this deposit may contain as much as 120 ft of saturated thickness with a transmis-sivity of up to 6,700 ft2/d. Several wells completed in another stratified deposit in the Banksville section (fig.

3) of Greenwich (northeastern part of town) were reported to penetrate up to 53 ft of fine-grained sand.

Some residents in Banksville obtain their water from shallow dug wells completed in this material (Aleksandra Moch, Greenwich Conservation Depart-ment, oral commun., 2001).

Bedrock The Greenwich area is underlain by metamor-phic bedrock of two different geologic terranes (Rodgers, 1985). The northwestern part of Greenwich is underlain by Proto-North American terrane, which is thought to be the edge of the Proto-North American continent. The remainder of Greenwich is in the Iapetos terrane of the western uplands of Connecticut, which is thought to be oceanic sediments that were offshore from Proto-North America. Bedrock to the northwest of Camerons line (a fault, see fig. 4) 

includes granitic gneiss, gneiss, schistose marble, schist, and amphibolite (fig. 4). The oldest bedrock

6 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD GREENWICH Geology modified from Stone and others, 1992 Tamarack Swamp Banksville Figure 3, Generalized surficial geology of the Greenwich area, Connecticut.

Description of the Study Area 7 FINE DEPOSITS--Composed of fine sand, silt, and clay particles generally in well sorted, thin layers of alternating silt and clay and (or) very fine sand; locally may contain lenses of coarser material. Fines, if present, overlie sand and gravel described below All sorted and stratified sediments composed of gravel, sand, silt and clay laid down by flowing meltwater during retreat of the last ice sheet; includes minor lenses of flowtill and other diamict sediments EXPLANATION POSTGLACIAL DEPOSITS WATER BODIES SAND AND GRAVEL--Composed of mixtures of sand and gravel within individual layers and as alternating layers. Sand and gravel layers generally range from 50-75 percent sand particles and from 25-50 percent gravel particles. Unit locally contains zones that are entirely sand GLACIAL ICE-LAID DEPOSITS ALLUVIUM OVERLYING SAND AND GRAVEL--Sand, gravel, silt, and some organic material on the flood plains of modern streams; overlie "sand and gravel" described below SWAMP--Muck and peat that contain minor amounts of sand and silt accumulated in poorly drained areas. Generally less than 10 feet thick GLACIAL MELTWATER DEPOSITS TILL--Poorly sorted, generally nonstratified mixture of grain sizes ranging from clay to large boulders; the matrix of most tills is composed dominantly of sand and silt. Darker green areas indicate till 15 feet or greater in thickness

8 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' Geology from Rodgers, 1985 Otf+Og Otfc Otfc Otfs Otfs Otfs Otfc Otfg l

Otfg Otfg Cm Oh Cma Ow Owm Yg Oh Yg Cma OCs Cd Yg Oh Ohn Ogh?

Otfg Ohn Otfg Ohn Ogh Oh Cd Cm Cameron's Line Figure 4. Generalized bedrock geology of the Greenwich area, Connecticut.

Description of the Study Area 9 Yg Gneiss of Highlands masifs Cd Dalton Formation OCs Stockbridge Marble Ow Walloomsac Schist Owm Basal marble member of Walloomsac Schist Cm Manhattan Schist Cma Amphibolite-bearing unit of Manhattan Schist Ogh Golden Hill Schist Oh Harrison Gneiss Ohn Nodular member of Harrison Gneiss Otfg Schist and granulite member of Trap Falls Formation Otfs Shelton (white gneiss) Member of Trap Falls Formation Otfc Carringtons Pond Member of Trap Falls Formation EXPLANATION Faults Otf+Og Trap Falls Formation and granitic gneiss units are the gneisses of Highland Massifs and are Pro-terozoic in age. The shelf-sequence rocks (marble and schistose marble) are Cambrian and Ordovician in age.

Bedrock in the area southeast of Camerons line prima-rily includes schist, gneiss, and granitic gneiss. These rocks are lower to middle Ordovician in age.

Foliation of bedrock in the Greenwich area typi-cally strikes N-NE and dips west at 34 to 80 degrees (Rodgers, 1985). The stream-drainage network appar-ently is strongly controlled by the underlying rock structure, and many streams are parallel to the strike of the foliation. Many high-angle fractures in the bedrock likely coincide with the direction of foliation. Another pattern in the stream-drainage network suggests struc-tural features that strike N-NW; this could be another major direction of fractures in the underlying bedrock.

Water moving through the crystalline bedrock, prima-rily through networks of interconnected fractures, supplies most private wells in Greenwich.

Precipitation and Runoff The climate of the Greenwich area is humid, and the average annual precipitation during the 30-year period 1967-96 was 49.9 in/yr as measured at Putnam Lake (fig. 2) (David Medd, Aquarion Water Company

10 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 of Connecticut, written commun., 2002) (fig. 5). 

During this period, precipitation ranged from 30.5 in.

(in 1995) to 68.9 in. (in 1983). Median annual precipi-tation in the Greenwich area during 1951-80 ranged from 46 to 50 in. (Hunter and Meade, 1983). Smaller amounts of precipitation occur near the coast; the larger amounts occur at higher elevations in northern Green-wich and are the results of orographic uplift.

Mean annual runoff from streams in the Green-wich area during 1930-60 ranged from about 21 to 23 in. (Ryder and others, 1970). Annual mean runoff from two nearby USGS streamflow-gaging stations (01208950, 01208990, fig. 6) during 1967-01 ranged from 13 to 46 in. (fig. 7). The difference between precipitation and runoff represents water lost through evapotranspiration. The average difference between precipitation at Bridgeport, Connecticut, and annual mean runoff from Sasco Brook near Southport (station 01208950, fig. 6), Connecticut, during 1966-96 was 17.9 in. (fig. 8). This value is probably very similar to the evapotranspiration rate for the Greenwich area.

Annual variations in the relation between precipitation and runoff can be attributed primarily to variations in ground-water storage and climatic factors such as temperature, humidity, wind speed, and cloud cover.

Natural long-term differences in annual mean runoff are related primarily to long-term local differences in precipitation.

1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 YEAR 20 30 40 50 60 PRECIPITATION, IN INCHES 70 Figure 5. Precipitation at Putnam Lake, Greenwich, Connecticut, 1967-2001.

Description of the Study Area 11 GREENWICH BRIDGEPORT FAIRFIELD 01208990 01208950 33 32 31 30 23 GREENWICH STUDY AREA 21 23 22 BROOKFIELD 8

FAIRFIELD COUNTY, CONNECTICUT NEW YORK NEW YORK LONG ISLAND SOUND 01208990 Continuous streamflow-gaging station and station-identification number Water-level observation well and town well number 21 EXPLANATION Rain gage at Bridgeport climatological station Figure 6. Additional data-collection sites in Fairfield County, Connecticut.

12 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 10 20 30 40 50 01208990, Saugatuck River near Redding, Conn.

01208950, Sasco Brook near Southport, Conn.

RUNOFF, IN INCHES PER YEAR 30 40 50 60 70 10 20 30 40 Least squares fitted line PRECIPITATION, IN INCHES ANNUAL MEAN RUNOFF, IN INCHES 50 80 Figure 7. Annual mean runoff at two long-term streamflow-gaging stations near Greenwich, Connecticut, 1967-01.

Figure 8. Precipitation at Bridgeport, Connecticut, and runoff at Sasco Brook near Southport, Connecticut (USGS station number 01208950), calendar years 1966-96.

Description of the Study Area 13 METHODS OF DATA COLLECTION AND ANALYSIS Data collected for this study include (1) stream-flow measurements at two continuous record stream-flow-gaging stations (2) streamflow measurements at miscellaneous stations, (3) ground-water-level measurements, (4) water-quality samples, and (5) water-use data.

Streamflow Measurements Continuous streamflow was measured at two locations (USGS stations 01211699 and 01211110) from February 2001 to September 2002 (fig. 2; table 1) to estimate ground-water recharge, to examine the difference in peak flows among primarily urbanized and primarily forested basins, and to assess the status of streamflow conditions during the study period (2001-02). Streamflow measurements at selected miscella-neous stations (table 1) were made on October 24-25, 2000 in conjunction with collection of water-quality samples and on April 24-25, 2002 to calibrate a finite-difference ground-water-flow model. Records of streamflow were computed using methods described by Rantz (1982a, b). Streamflow records from February 2001 to September 2002 for stations 01211699 and 01211110 (fig. 2) are in appendix 1.

Table 1. Streamflow measurements, Greenwich area, Connecticut, 2000-2002.

[USGS, U.S. Geological Survey; mi2, square miles; ft3/s, cubic feet per second; mi, miles; ft, feet; lat., latitude; long, longitude; Type of station: C, continu-ous; M, miscellaneous.

Purpose:

QW, water-quality samples collected; C, measurement to calibrate ground-water-flow model; R, measurement to cre-ate/maintain rating curve at continuous measurement site]

USGS station identification number and stream Tributary to Location Drainage area (mi2)

Type of station Purpose Date of measure-ment Dis-charge (ft3/s) 01211010

Brothers Brook Mianus River Lat. 410326, long 733558, Fair-field County, at Montgomery Pine-tum Park (known in Greenwich as Strickland Brook) 1.39 M

QW C

10/25/00 04/24/02 0.173 1.01 01211040

Greenwich Creek Long Island Sound Lat. 410356, long 733626, Fair-field County, at bridge on Hill Street near Cos Cob (known in Greenwich as Beaver Brook) 0.69 M

C 04/24/02

.538 01211110

Unnamed tributary Greenwich Creek Lat. 410234, long 733659, Fair-field County, on Old Church Road, at Cos Cob (known in Greenwich as West Brothers Brook) 2.19 C

QW R

R R

R R

R R

R R

R R

R R

R R

R R

R R

R R

R 10/25/00 02/22/01 03/13/01 03/22/01 03/31/01 04/19/01 05/09/01 05/18/01 06/04/01 06/11/01 09/07/01 10/09/01 10/11/01 11/06/01 11/30/01 04/10/02 04/20/02 06/07/02 06/13/02 08/06/02 09/24/02

.389 3.68 37.0 14.0 23.1 3.84

.977

.472 5.27 1.07

.314

.729

.686

.495 1.15 2.98 6.21 36.6 1.85

.040

.842

14 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 01211140

Horseneck Brook Long Island Sound Lat. 410650, long 733800, Fair-field County, at Lower Cross Road near Stanwich 0.52 M

QW C

10/25/00 04/24/02 0.091

.220 01211210

Horseneck Brook Long Island Sound Lat. 410541, long 733558, Fair-field County, at Valley Drive near Rock Ridge 4.81 M

QW 10/25/00

.877 01211450

Byram River Long Island Sound Lat. 410326, long 734218, Fair-field County, at bridge on Bedford Road near North Greenwich 9.10 M

C 04/25/02 4.55 01211600

Byram River Long Island Sound Lat. 410338, long 734041, Fair-field County, at bridge on Sher-wood Avenue, at Riversville 11.7 M

C 10/24/00 1.87 01211699

East Branch Byram River Byram River Lat. 410558, long 734102, Fair-field County, below Lake Mead, 200 ft upstream from John Street, at Round Hill (known in Green-wich as Middle Branch Byram River) 1.65 C

QW R

R R

R R

R R

R R

R R

R R

R R, C R

R R

R 10/24/00 02/22/01 03/13/01 03/20/01 04/19/01 05/18/01 06/04/01 06/11/01 09/07/01 09/28/01 11/06/01 11/30/01 01/03/02 02/05/02 04/10/02 04/24/02 05/20/02 06/07/02 08/06/02 09/24/02

.393 3.83 13.8 4.64 3.50

.654 2.89

.839

.029 1.60

.19

.620

.289

.687 1.98 1.09 5.53 25.6

.16

.069 01212100

East Branch Byram River Long Island Sound Lat. 410339, long 734031, Fair-field County, at bridge on 

Riversville Road just downstream from Merritt Parkway, 0.2 mi upstream from mouth 17.4 M

QW 10/24/00 1.86 01212550

Unnamed tributary Byram River Lat. 410139, long 733941, Fair-field County, upstream of bridge on Pemberwick Road (known in Greenwich as Pemberwick Brook) 1.4 M

C 04/25/02

.839 01212600

Byram River Long Island Sound Lat. 410138, long 733941, Fair-field County, downstream of unnamed tributary at Pemberwick 123.6 M

C 04/25/02 13.6 1Drainage area does not include basins diverted to Putnam Lake (fig. 2).

Table 1. Streamflow measurements, Greenwich area, Connecticut, 2000-2002.Continued

[USGS, U.S. Geological Survey; mi2, square miles; ft3/s, cubic feet per second; mi, miles; ft, feet; lat., latitude; long, longitude; Type of station: C, continu-ous; M, miscellaneous.

Purpose:

QW, water-quality samples collected; C, measurement to calibrate ground-water-flow model; R, measurement to cre-ate/maintain rating curve at continuous measurement site]

USGS station identification number and stream Tributary to Location Drainage area (mi2)

Type of station Purpose Date of measure-ment Dis-charge (ft3/s)

Description of the Study Area 15 Water-Level Measurements Three wells (table 2) were installed in the East Branch of the Byram River Basin, using air-rotary drilling techniques, to monitor water levels and 

ground-water storage in the bedrock aquifer. The wells were installed at different topographic positions in the watershed to evaluate the magnitude of water-level fluctuations in different parts of the ground-water-flow system (GW-21 on the hilltop on the drainage divide, GW-23 on the hillside, and GW-22 in the bottom of the valley, figs. 2, 9). This approach of installing wells in different parts of the ground-water-flow system was discussed by Melvin (1986) and has been implemented in many basins in Connecticut with USGS streamflow-gaging stations. Water-level fluctuations generally are greater on hilltops and hillsides, where ground-water-flow gradients are low. Water-level fluctuations in the valley bottom typically are small, because these tend to be areas of ground-water discharge, and ground-water-flow gradients are high.

Water levels were measured sporadically begin-ning in September 2001; water levels were monitored continuously using submersible pressure transducers with data loggers from November 20, 2001, to October 2002. Data on daily water levels were published by Morrison and others (2003).

Water levels also were measured in 36 domestic wells on April 16-20, 2002 to calibrate the ground-water-flow model. Wells were selected from a list of voluntary homeowners who were interested in partici-pating in the study. Wells were located using a global positioning system (GPS), and the altitude of the land surface was determined from 2-ft contour maps provided by the town of Greenwich. In addition, water levels measured by well drillers were obtained from 402 well-completion reports as part of calibrating the ground-water-flow model.

Water-Quality Samples Water-quality samples were collected from seven sites (fig. 2; table 1) during base-flow conditions on October 24-25, 2000. Samples were collected during a base-flow period to ensure that streams contained primarily ground-water discharge. These samples represented an integrated sample of ground-water quality in basins with different land-use charac-teristics. Samples were collected using methods described by Wilde and Radtke (1998), Wilde and others (1998a-c), and Wilde and others (1999a, b) and were analyzed at the USGS laboratory in Denver, Colo-rado, for major ions, dissolved trace elements, nutri-ents, volatile organic compounds (VOCs), pesticides, and indicator bacteria. Water-quality analyses were reported in Morrison and others (2002). An additional sample was collected at site 01211110 (fig. 2) for 

experimental analysis of pharmaceuticals, hormones, and other organic wastewater contaminants, and analyzed using methods described in Kolpin and others (2002). This sample was collected to determine if compounds associated with septic-system waste were entering streams from ground water.

Water-Use Data Water-use data were compiled from public water suppliers for calendar year 2000 (David Medd, Aquarion Water Company of Connecticut, written commun., 2002). These data were used to create a statistical model of water use at residences with private wells and to estimate return flow from private septic systems.

Table 2. Wells in the East Branch Byram River Basin, Greenwich, Connecticut.

[USGS, U.S. Geological Survey]

USGS local well identifier USGS station identification number Latitude (NAD83)

Longitude (NAD83)

Altitude (feet)

Depth of well (feet)

Depth to bedrock (feet)

Depth to bottom of casing CT-GW 21 410628073413301 41o 06 28.4 73o 41 32.6 465 350 11 18.7 CT-GW 22 410443073414101 41o 04 43.3 73o 41 40.7 222 250 14 18.6 CT-GW 23 410515073415901 41o 05 15 73o 41 59.5 365 250 9

17.8

16 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 0

100 200 300 400 500 ALTITUDE, IN FEET ABOVE SEA LEVEL John Street North Porchuck Road Porchuck Road A

A' Thick Glacial Till Walloomsac Schist Manhattan Schist Cameron's Line Carrington's Pond Member of the Trap Falls Formation (interlayered schist and gneiss)

CT-GW-21 CT-GW-23 CT-GW-22 0

0 1

2 KILOMETERS 1

2 MILES 444 341 216 EXPLANATION Inferred contact between bedrock and thick glacial till Inferred contact between bedrock types Water-level altitude (feet) in well on April 16, 2002 444 Bend in section Figure 9. Cross section through wells installed for water-level measurements. (Geology based on Rodgers, 1985.)

Ground-Water Recharge 17 GROUND-WATER RECHARGE IN THE 

GREENWICH AREA In the Greenwich area, natural ground-water recharge to bedrock aquifers is derived primarily from precipitation that infiltrates from the land surface to the water table. Other natural sources of recharge include loss of water from lakes, wetlands, and streams.

Previous studies in Connecticut have shown that there are differences in ground-water recharge in areas covered with till compared to areas covered with glacial stratified deposits. These differences have been observed in the base-flow component of streamflow (ground-water runoff). Mazzaferro and others (1979) determined the following mathematical relation between geology and ground-water runoff based on work by Thomas (1966):

(1) where Y= Ground-water runoff, as a ratio to total basin runoff, and X= Percentage of the basin underlain by coarse-grained stratified drift (glacial strati-fied deposits).

Recharge rates calculated with this equation represent the total recharge in the basin, but do not describe how much recharge actually infiltrates to the bedrock aquifer from the glacial material. For instance, a large percentage of recharge to till areas discharges directly from the till into small nearby streams and intermittent watercourses during wet periods. Under natural conditions, most recharge to glacial stratified deposits does not enter bedrock, because glacial strati-fied deposits are mostly in valley-bottom locations.

These areas are associated with ground-water discharge; therefore, ground water is more likely to flow upward from the bedrock aquifer to the glacial stratified deposits than downward from the surficial deposits to the bedrock aquifer. Under stressed condi-tions, such as those that take place when wells in the bedrock aquifer are pumped, water in the till and glacial stratified deposits, which normally would have discharged to a local surface-water body, may be drawn into the bedrock aquifer and into wells.

Factors Affecting Ground-Water Recharge Factors other than geologic properties may affect recharge rates locally and temporally. Annual and long-term variability and timing of precipitation and subse-quent runoff from a basin are very important in deter-mining recharge rates. In Connecticut, recharge takes place primarily during the nongrowing season from October to May (Melvin, 1986) although precipitation is generally distributed evenly throughout the year. In a typical year, ground-water levels begin to rise in October, because there is more recharge to the aquifer than discharge to streams. During the growing season, water levels typically follow a downward trend 

because of evapotranspiration and depletion of soil moisture, despite temporary peaks caused by large rainfall events. A lack of precipitation during the nongrowing season can cause lower than normal water levels in the aquifer even in a year with above-normal precipitation during the growing season. In addition, water-level fluctuations are likely to be larger in upland areas than in valley bottoms; therefore, the effects of drought conditions may be more pronounced in the upland areas.

In areas where the water table is at land surface, recharge is rejected as surface runoff (Kontis, 2001).

Channelized or unchanneled surface runoff from till deposits may be another important source of recharge to glacial stratified deposits (Kontis, 2001). The natural drainage patterns, the degree to which channelized flow takes place, and the position of the water table during the nongrowing season may affect local recharge rates in any basin.

Other factors that may affect natural recharge include slope, vegetation cover, and local variations in geology and soil moisture. For example, Bauer and Mastin (1997) showed that areas with coniferous forest may have less ground-water recharge than areas with deciduous forest, due to retention and evaporation of rain and snow from the evergreen canopy.

In a developed area such as Greenwich, human activities can affect recharge rates and can change the ground-water budget of a particular basin. Develop-ment adds impervious surfaces that can impede natural recharge. Ground-water withdrawals can reduce the base flow of streams or increase the recharge from surface-water bodies and wetlands. Reduction in ground-water levels in areas where the water table is at land surface also may increase recharge from precipita-tion. The discharge of wastewater through septic Y

0.6X 35

+

=

18 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 systems can affect ground-water-flow patterns, and, in areas like Greenwich, can be an additional source of recharge if the water is imported from public water-supply reservoirs inside or outside the basin.

Additional recharge may be obtained from leaking water mains, storm sewers, and sanitary sewers, and ground-water withdrawals may occur from infiltration into sanitary sewer lines. The amount of water entering and leaving the ground-water system from these sources is difficult to quantify, but due to the aging water and sewer infrastructure in Greenwich, it is likely that some amount of water is gained or lost from these sources. A study of these issues would lead to enhanced understanding of the ground-water budget in the Greenwich area.

Ground-Water Recharge and Discharge, 2001-02 Ground-water recharge and discharge can be observed in the hydrographs of ground-water levels in wells GW-21, 22, and 23 (fig. 10), and in records of streamflow (appendix 1). Ground-water recharge during the winter of 2001-02 was delayed in wells GW 21 and 23 by a lack of precipitation. Ground-water levels began to rise in well GW-23 during November 2001, but did not rise in well GW-21 until February 2002 (fig. 10). The delay probably was caused by below-normal precipitation. Precipitation from October 2001 to February 2002 at nearby Putnam Lake was 8.11 in., compared to the 20-year average of 18.58 in. for these months during 1982-01 (David Medd, Aquarion Water Company of Connecticut, written commun., 2002). The highest water levels in the three wells installed for this study were during May 2002.

Water levels generally declined in all three wells during the summer of 2002; however, only GW-23 responded to summer precipitation. It may be that some additional source of recharge in this part of the basin allows water into the aquifer. One hypothesis is that an intermittent stream about 50 ft from this well affects ground-water rechargeduring the spring, when this stream is flowing, it may be a conduit for ground-water discharge; when this stream dries up, it may become a source for ground-water recharge from upland surface-water runoff collected during storms. Water-level fluc-tuations in well GW-22 (in the valley bottom of the basin) were much smaller than fluctuations in GW-21 or GW-23, as expected, due to ground-water inflow from upgradient areas.

Stream base flow can be equal to ground-water recharge during periods when there is no net change in ground-water storage. Ground-water discharge to a stream can be determined by the separation of base flow from a streamflow hydrograph. Streamflow was continuously monitored (appendix 1) at two locations (fig. 2) from March 2001 to September 2002. A base-flow separation was conducted at station 01211699 to determine the ground-water discharge from this basin during April 2001-September 2002. The computer hydrograph separation program PART developed by Rutledge (1997) was used to estimate the quarterly ground-water runoff. (fig. 11).

During April-December 2001, the total runoff calculated for station 01211699 was 8.48 in. This compares with 8.60 in. of runoff at nearby Sasco Brook (station 01208950; fig. 6) during the same period. The mean annual runoff for April-December at Sasco Brook from 1967-2001 was 16.2 in., indicating that most of 2001 was drier than average.

At station 01211699 in Greenwich, the results of hydrograph separation from PART (fig. 11) indicate that most ground-water discharge from this basin took place from April to June (4.5 in.) with a very small amount (1.5 in.) from July to December. Ground-water runoff (6.0 in.) represented 70 percent of the total runoff from April to December 2001.

Data from USGS streamflow-gaging stations commonly are reported by water year,1 which repre-sents the annual water cycle. If the baseflow-separa-tion data from 01211699 are calculated on the basis of water year 2002 (October 2001 to September 2002),

the total base flow for this period was 7.81 in. and the total runoff was 12.21 in (fig 10); therefore, base flow was about 64 percent of the total annual runoff. In comparison, the mean annual runoff at Sasco Brook (station 01208950; fig. 6) was 25.7 in. from 1967-2001.

There probably was more base flow than recharge for water year 2002 because ground water that was in storage before the water year began was discharged and because 2002 was drier than normal.

1A water year is defined as the 12-month period October 1 through September 30. The water year is designated by the calen-dar year in which it ends and which includes 9 of the 12 months.

Thus, the year ending September 30, 2002 is called the 2002 water year.

Ground-Water Recharge 19 Sep Oct Nov Dec 2002 Jan Feb Mar Apr May Jun Jul Aug Sep Oct 0

5 10 15 20 25 30 35 40 45 50 0

0.5 1

1.5 2

2.5 3

3.5 4

4.5 5

5.5 6

GW21 hilltop GW22 valley bottom GW23 hillside Precipitation at Putnam Lake, Greenwich, Connecticut EXPLANATION PRECIPITATION, IN INCHES GROUND-WATER LEVEL, IN FEET BELOW LAND SURFACE 2001 manual measurements Gaps indicate missing record or equipment failure continuous measurements Figure 10. Water levels in bedrock wells in the East Branch of the Byram River Basin and precipitation at Putnam Lake, Greenwich area, Connecticut, September 2001-October 2002.

20 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Apr-Jun Jul-Sep Oct-Dec Jan-Mar Apr-Jun Jul-Sep 0

8 0

2 4

6 2001 2002 RUNOFF, IN INCHES STATION 01211699 OVERLAND RUNOFF BASEFLOW Water Year 2002 WATER USE IN THE GREENWICH AREA Residential water use was estimated for basins in Greenwich to (1) determine water-use patterns, (2) provide data for input into a finite-difference ground-water-flow model, as either water withdrawals from wells or as return flow in areas with septic systems and public water supply, and (3) estimate consumptive water use, with respect to average rates of ground-water recharge.

Ground-water withdrawals for residences with private wells in Greenwich were estimated by evalu-ating water-use data and geographic information system (GIS) characteristics of residences with public water supply for calendar year 2000 and creating log-linear ordinary least-squares (OLS) regression models to estimate average daily water use, average summer water use, and average daily winter water use. Predic-tions of water use and confidence intervals were aggre-gated by drainage basin (appendix 2).

Additional estimates of water use for golf courses, commercial buildings, schools, churches and other government or institutional buildings, which would be necessary to determine total water use for each basin, would be useful to understand the full effect of water use on downstream resources, but were beyond the scope of this study.

Water Use at Residences with Public Water Supply The water-use data (meter readings) for indi-vidual domestic users obtained from Aquarion Water Company of Connecticut for calendar year 2000 were matched to addresses in the town of Greenwich prop-erty GIS database. Average daily water use was calcu-lated for each property with public water supply by summing the quarterly (or more frequent) meter read-ings and dividing by 365 days. Average daily summer water use was calculated by summing meter readings from April to September, and values for average daily winter water use were calculated by summing readings from January to March and October to December.

These values were divided by 182.5 days, or one-half of a year. A subset of these properties had data that could be extracted only seasonally, because of water billing cycles, or meter readings that were not recorded at least quarterly; therefore, a smaller number of observations were used to predict winter and summer water use than were used for average daily water use. Data were summarized initially by property size (table 3).

Figure 11. Quarterly base flow at U.S. Geological Survey streamflow-gaging station number 01211699, East Branch Byram River below Lake Mead at Round Hill, Connecticut.

Ground-Water Recharge 21 Table 3. Median daily water use, and median daily seasonal water use by property size, in calendar year 2000 at selected residences with public water supply.

[gal/d, gallons per day; <, less than; >, greater than]

Residential property size (acres)

Median daily water use calendar year 2000 (based on average daily use at individual properties)

Median daily water use, April-September 2000 (based on average daily summer use at individual properties)

Median daily water use, January-March 2000 and October-December 2000 (based on average daily winter use at individual properties)

(gal/d)

Number of households (gal/d)

Number of households (gal/d)

Number of households

<0.5 219 309 230 240 204 240

>.5-1 295 342 324 251 225 251

>1-2 357 1,297 410 1,011 275 1011

>2-4 535 828 704 723 385 723

>4 1,082 112 1389 89 848 89 Typical water use in Connecticut is estimated to be about 76 gal/person/d (Solley and others, 1998).

Based on data compiled for this report (table 3), it is apparent that water use in the Greenwich area may be larger than what is typical, depending on the type of development. In areas with less than 1-acre lot size, per capita use is about 113 gal/person/d or less. In areas with larger lots, per capita use could be as high as 416 gal/person/d (based on table 3 values and assuming an average of 2.6 persons per household (U.S. Department of Commerce, 1991)).

Log-Linear Regression Models of Residential Water Use The regression models that were used to predict water use for residences with private wells were of the form (2) where WU is water use, in gallons per day,

0 is an intercept,

1, 2, 3 are coefficients, X is an independent variable, and E is a random error.

Information from the Greenwich GIS database that was compiled for each property included the pres-ence and size of an outdoor swimming pool, the lot size, the number and total footprint of buildings, and the area of the lot that was forested. The area of forest cover was subtracted from the total lot area to create a variable related to lawn size. These variables were chosen because of their availability and the hypothesis that properties with large acreage, swimming pools, large or numerous buildings, and large lawns use more water. A study done by the American Water Works Association Research Foundation in several metropol-itan areas in the United States determined that these are important factors in determining residential water use (American Water Works Association Research Foun-dation, 1999). A similar approach was undertaken in this study.

Variable Selection The GIS information that was compiled was tested in different combinations as variables in a multiple linear-regression model to predict the natural log of average daily water use for each property with public water-supply information. A log-transformation of the response variable (water-use data) was necessary because the data appear to be log-normally distributed.

Variables (tables 4, 5, and 6) were selected on the basis of physical plausibility, the evaluation of R2 (coeffi-cient of determination), the distribution of residuals, WU





ln

0

1 X1





2 X2





3 X3



...

n Xn





E

+

+

+

+

=

22 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 and whether or not the variable was statistically signif-icant at the 95-percent confidence interval. Regression models were created for average daily water use, average summer water use, and average winter water use.

Selected variables (tables 4, 5, and 6) were similar for all three regression models. The variables as a group have some predictive ability and are highly significant, but the coefficients of determination (R2) for each model are low and indicate that additional variables would be useful. For example, a variable for the number of residents per household would help explain additional error; however, many of the vari-ables for predicting water use probably are socioeco-nomic and therefore are difficult to quantify or are unavailable. Model residuals are generally normally distributed (figs. 12A-C). This indicates that it is reasonable to apply confidence intervals to water-use predictions made with these models.

Prediction of Residential Water Use in Areas with Domestic Wells A major assumption in using the regression method to predict water use for residences with private wells is that residences with wells use water at similar rates to similar residences with public water supply. It is possible that water use in homes with private wells may differ from those with public water supply. For instance, homeowners with wells may use less water indoors because of the lower water pressure that is sometimes associated with private domestic water systems, or may be more or less inclined to fill or top off swimming pools, to water lawns, or to use water for other landscaping purposes.

To use the OLS regression model described above for making water-use predictions, an algorithm was applied to correct for retransformation bias in converting the predictions from natural log space. Data that are retransformed from log space will not be normally distributed; therefore, predictions and the associated confidence intervals will be biased (Helsel and Hirsch, 1992). The method used to account for retransformation bias is the Minimum Variance Unbi-ased Estimator (MVUE) (Cohn and others, 1989).

A computer program was written (Gregory Schwarz, U.S. Geological Survey, written commun.

2002) to run the OLS regression model and make predictions using the MVUE. The program sorts predictions by water-use type (private well and septic system, or public water supply and septic system),

aggregates data by basin (fig. 13), and computes the 90-percent confidence interval of the aggregated values (

2). The program uses actual water-use data in areas with public water supply and private septic systems, or predicts water use for properties in the public water-supply area that were not matched to the GIS data.

Return flow from septic systems (fig. 14) was assumed to be equivalent to the winter rate of water use. Return flow from septic systems in areas with public water supply were separated from return flow in areas with private wells, because this is generally an outside source of additional recharge to the aquifer.

Predictions for areas outside the Greenwich town boundaries were made using different methods because of the availability and comparability of GIS datasets.

Water-use predictions were made for the Stamford part of the study area using the same water-use model as for Greenwich, because the City of Stamford has compa-rable GIS datasets to Greenwich, except for forest cover. The unforested part of each property in the study area in Stamford was digitized from high-resolution, digital aerial photographs. Water-use estimates for resi-dential properties in the Westchester County part of the study area were made based on property size only, because GIS data layers were unavailable. The number of residential properties of each size in each basin were summarized from town of North Castle, New York assessors records. Values from table 3 were applied to these properties. These estimates are subject to an unknown amount of error. The error will be less in basins where the estimate of water use for the Westchester part of the basin is within the 90-percent confidence interval for water use in the rest of the basin (appendix 2).

Average annual residential ground-water with-drawals from the study area are estimated to total 2.1 Mgal/year (appendix 2). Return flow from septic systems at residences served by public water supply was estimated using the winter water use model to total 1.4 Mgal/year.

Ground-Water Recharge 23

-5

-4

-3

-2

-1 0.0 0.1 0.2 0.3 0.4 0.5 DENSITY

-4

-3

-2

-1 0.0 0.1 0.2 0.3 0.4

-3

-2

-1 0.0 0.1 0.2 0.3 0.4 0.5 RESIDUAL A.

B.

C.

DENSITY DENSITY 0

1 2

3 0

1 2

3 0

1 2

Figure 12. Frequency distribution for residuals and the best-fit normal distribution approximation for (A) average daily 

water use, (B) average daily summer water use, and (C) average daily winter water use, Greenwich area, Connecticut.

24 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Table 4. Parameter estimates, standard errors, t-statistics, and p-values for multiple linear regression model of average daily water use at residences with public water supply, calendar year 2000, Greenwich, area, Connecticut.

[Coefficient of determination (R2) = 0.25]

Variable Units Parameter estimate Standard error t-statistic p-value Intercept dimensionless 8.75 0.167 52.3

<0.0001 Log (unforested area) acres

.219

.041 5.38

<.0001 Outdoor swimming pool size acres 19.6 1.62 12.1

<.0001 Log (total footprint of buildings) acres 3.62

.211 17.1

<.0001 Log (total footprint of buildings) squared acres squared

.915

.060 15.4

<.0001 Table 5. Parameter estimates, standard errors, t-statistics and p-values for multiple linear regression model of average daily summer water use at residences with public water supply, calendar year 2000, Greenwich area, Connecticut.

[Coefficient of determination (R2) = 0.26]

Variable Units Parameter estimate Standard error t-statistic p-value Intercept dimensionless 9.23 0.210 44.0

<0.0001 Log (unforested area) acres

.258

.050 5.20

<.0001 Outdoor swimming pool size acres 22.4 1.99 11.3

<.0001 Log (total footprint of buildings) acres 4.06

.268 15.2

<.0001 Log (total footprint of buildings) squared acres squared 1.05

.078 13.6

<.0001 Table 6. Parameter estimates, standard errors, t-statistics and p-values for multiple linear regression model of average daily winter water use at residences with public water supply, calendar year 2000, Greenwich area, Connecticut.

[Coefficient of determination (R2) = 0.21]

Variable Units Parameter estimate Standard error t-statistic p-value Intercept dimensionless 7.73 0.200 38.7

<0.0001 Log (property size) acres

.345

.069 5.00

<.0001 Log (unforested area) acres

.107

.046 2.34

.019 Outdoor swimming pool size acres 12.5 1.73 7.25

<.0001 Log (total footprint of buildings) acres 2.80

.246 11.4

<.0001 Log (total footprint of buildings) squared acres squared

.757

.069 11.0

<.0001

Ground-Water Recharge 25 Estimated average daily withdrawals from residential wells, in million gallons per day per square mile of basin drainage area 0 - 0.02

>0.02 - 0.04

>0.04 - 0.07

>0.07 - 0.1

>0.1 - 0.16 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' GREENWICH STAMFORD Westchester County, New York Westchester County, New York Long Island Sound Basins with potential of 50 percent or more increase in average daily water use due to commerical, public, or institutional properties Basin number EXPLANATION Figure 13. Estimates of average daily residential water use per square mile for small basins, Greenwich area, Connecticut.

26 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Estimate average daily return flow from residential septic systems served by public water supply, in million gallons per day per square mile of basin drainage area 0

>0 - 0.02

>0.02 - 0.04

>0.04 - 0.05

>0.05 - 0.08 GREENWICH Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD Westchester County, New York Westchester County, New York Long Island Sound Basin number EXPLANATION Figure 14. Estimates of average daily return flow per square mile from residential septic systems supplied by public water supply for small basins, Greenwich area, Connecticut.

Simulation of Ground-Water Flow 27 Estimation of Consumptive Water Use Consumptive water use is the part of water with-drawn from the aquifer that is evaporated, transpired, incorporated into products or crops, consumed by humans or livestock, or otherwise removed from the immediate water environment. Estimates of consump-tive water use are necessary to determine the amount of water withdrawn from the bedrock-aquifer system that is not returned, for comparison with available ground-water recharge and stream base flow. Estimates of consumptive use also are necessary to determine the amount of water imported from public water supply in Greenwich (surface-water sources) that is exfiltrated from septic systems and becomes an additional source of recharge.

The majority of consumptive water use is seasonal outdoor water use, including lawn or land-scape watering, filling swimming pools, and washing vehicles. These uses are generally consumptive because water evaporates directly from surfaces or is taken up and transpired by plants to the atmosphere (U.S. Geological Survey, 1995).

Consumptive water use in Connecticut averages about 20 percent of water delivered by public water suppliers (Solley and others, 1998). In this report, consumptive use is estimated to be equal to outdoor water use. Indoor water use is probably returned to the aquifer or added as additional recharge to areas with public water supply and septic systems. Indoor water use is estimated to equal the winter water use.

Consumptive use in the study area was estimated by subtracting the winter water-use data from the average daily water use at properties that had seasonal use information. The estimated average consumptive use for 2,315 properties (with public water supply) in the Greenwich area with seasonal-use information for calendar year 2000 is 20 percent, the median is 19, and the interquartile range is from 3 to 39 percent. This is consistent with published values for percentages of consumptive use (U.S. Geological Survey, 1995).

Consumptive use estimated by the water-use models (tables 4 and 6, appendix 2) and aggregated by basin is higher than published values; consumptive use is esti-mated to average 29 percent.

Water Use at Nonresidential Properties In addition to water use by residences, water also is withdrawn from wells for use at golf courses, country clubs, schools, churches, nurseries, and properties with commercial or institutional buildings. The determina-tion of use for these properties is beyond the scope of this project and would require additional monitoring and willing participation from owners of these proper-ties; however, if data were collected and analyzed, esti-mates of water use could be updated for each basin area shown in figure 13. Five of the 32 basins might have a large amount (50 percent or more increase) of addi-tional average daily water use due to nonresidential water uses (fig. 13); however, some of this water use (at least in the case of golf-course irrigation) may not be from ground-water sources. More information is needed to update water-use estimates for these basins.

Water use at golf courses probably represents the largest individual category of water use in Greenwich.

Water generally is withdrawn from on-site ponds and wells at golf courses, and information is limited on the ratio of ground-water to surface-water supply. Golf courses can use water stored in ponds from earlier spring runoff, overland runoff of surface water from storms during the irrigation season, and by pumping of ground water. Because of these different sources, the total ground-water use and consumption would be difficult to estimate without metering wells and total system uses simultaneously.

Water use is variable among golf courses and from year to year, depending on annual or seasonal precipitation, soil type, or other localized conditions.

Estimates of typical golf course irrigation rates were provided by several golf course superintendents in Greenwich (Jeffrey Scott, Tamarack Country Club; Gary Glazier, Burning Tree Country Club; Scott Niven, Stanwich Club, oral commun., 2002) and range from 60,000 to 150,000 gal/d during the growing season (May to September). This is equivalent to about 25,000 to 60,000 gal/d averaged for 1 year, assuming a 150-day irrigation season from May to September. Data from the Stanwich Club indicated that some irrigation also takes place during April and October (Aleksandra Moch, Greenwich Conservation and Inland Wetlands Department, written commun., 2002).

SIMULATION OF GROUND-WATER FLOW IN THE GREENWICH AREA Ground-water flow in the Greenwich study area was simulated using MODFLOW-2000 (Harbaugh and others, 2000). MODFLOW-2000 is a three-dimen-sional, finite-difference ground-water-flow model that

28 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 includes a nonlinear regression algorithm to find the best fit among model parameters and hydrologic obser-vations. The purpose of the ground-water-flow model is to test hypotheses about hydrologic properties and provide a framework to estimate ground-water budgets and water availability in small basin-based zones in the Greenwich area. The calibrated model can be used to simulate hypothetical conditions involving variation in consumption of water, changes in recharge, or other changes to the hydrologic system. Two model simula-tions were run: (1) the first simulation was for calibra-tion to conditions during late April 2002 using winter water-use rates, and (2) the second simulation used the calibrated model with ground-water withdrawal rates adjusted to average water-use conditions.

Description of Flow Model and Model Assumptions The ground-water model uses a finite-difference grid with 57 rows and 46 columns (fig. 15). Individual cells in the model are 1,017 by 1,017 ft. The model has five layers of constant thickness (fig. 16). The upper-most two layers simulate the surficial aquifers that consist of thin till, thick till, glacial stratified deposits, and bedrock underlying areas with thin till (fig. 3). The top of layer 1 is the land surface. Layer 1 is 15 ft thick, and layer 2 is 30 ft thick. The lower three layers (each 200 ft thick) simulate the bedrock aquifer to a depth of 645 ft below land surface. The depth was selected because it is greater than the average depth of wells in Greenwich. The average depth of 400 wells compiled from well completion reports was about 450 ft. Model layers are assumed to be confined, including layer 1, due to numerical instability in the model with thin uppermost layers. The model is a steady-state ground-water-flow model that is calibrated to a stable period (April 2002) when recharge and discharge were in balance and no major changes were taking place in ground-water storage. Steady-state ground-water-flow models are used to simulate long-term average condi-tions, or stable periods of increased or reduced recharge. The ground-water-flow model of the Green-wich area was calibrated to 34 ground-water levels measured during April 16-20, 2002; water levels from 402 well-completion reports (fig. 17); and streamflow measurements at 8 sites (table 2). Water levels that were measured during April 2002 were assumed to represent average conditions because water levels in wells from the USGS statewide-monitoring network with monthly measurements (table 7; fig. 6) were at average (period of record) conditions during this time period. Normally during April, water levels in wells would be at above-average conditions; however, the spring of 2002 was very dry, and water levels in wells in the USGS statewide-monitoring network in Fairfield County were close to average conditions. Ground-water levels and streamflow in Greenwich during April 16-25 were generally stable, with little change (fig. 18).

The model also was calibrated to streamflow measurements from eight locations (table 1) in the Greenwich area on April 24-25, 2002. Streamflow during this time period was stable and represents base flow (fig. 18). These observations included seven measurements of discharge, and one determination of discharge from a rating curve at a continuous stream-flow-gaging station number 01211110 (fig. 2).

The bedrock aquifer is simulated in this model as homogeneous and isotropic. This model is designed primarily to understand the water budget and should not be used for studies that require detailed flow-path analysis. In reality, the frequency and spatial orienta-tion of fractures in the bedrock are likely to affect local ground-water-flow directions and possibly ground-water budgets. A similar approach modeling the water budget and hydraulic properties of fractured crystalline bedrock aquifers as an equivalent porous medium was used by Tiedeman and others (1997) and Wolcott and Snow (1995).

The steady-state ground-water-flow model was calibrated to observations of ground-water levels and streamflow. The calibration process is a nonlinear regression to provide optimal estimates of parameters based on the best fit to measured (or observed) values of hydraulic head and streamflow. Model parameters were estimated using a nonlinear-regression method developed by Cooley and Naff (1990) and applied to MODFLOW 2000 by Hill and others (2000).

Simulation of Ground-Water Flow 29 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' Putnam Lake Rockwood Lake Converse Lake Bargh Reservoir Kensico Reservoir EXPLANATION Constant-head cell 5

l 10 l

15 l

20 l

25 l

30 l

35 l

40 l

45 l

1 l

--1 5--

10--

15--

20--

25--

30--

35--

40--

45--

50--

55--

No-flow and active model area boundary Model cell with simulated stream 40--

Row number 15 l

Column number STAMFORD GREENWICH Westchester County, New York Westchester County, New York Long Island Sound 14--

Figure 15. Extent of finite-difference ground-water-flow model grid, Greenwich area, Connecticut.

30 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002

-500

-300

-100 100 300 500 4

5 6

7 8

9 10 11 12 13 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Column number 0

1 2 MILES 0

1 2 KILOMETERS Layer 5 (bedrock)

Layer 4 (bedrock)

Layer 3 (bedrock)

Layer 1 Layer 2 ALTITUDE, IN FEET Till and thick till Coarse-grained glacial stratified deposits Fine-grained glacial stratified deposits Model layer boundary No-flow boundary EXPLANATION Figure 16. Cross section through row 14 (see fig. 15), in the finite-difference ground-water-flow model, Greenwich area, Connecticut.

Simulation of Ground-Water Flow 31 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD GREENWICH Westchester County, New York Westchester County, New York EXPLANATION Water-level observation from well-completion report Water level measured April 16-20, 2002 Streamflow observation Figure 17. Location of observations for ground-water levels and streamflow, Greenwich area, Connecticut.

32 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 March 4

8 12 16 20 24 28 April 4

8 12 16 20 24 28 May 4

8 12 16 20 24 28 2002 0

5 10 15 20 25 30 35 40 45 0

10 20 30 40 50 60 GW21 hilltop GW22 valley bottom GW23 hillside 01211110 01211699 Ground-water levels (mean daily)

Streamflow (mean daily)

Ground-water-flow-model calibration period for ground-water levels and stream base flow GROUND-WATER LEVEL, IN FEET BELOW LAND SURFACE STREAMFLOW, IN CUBIC FEET PER SECOND GW21 hilltop GW22 valley bottom GW23 hillside EXPLANATION 01211699 01211110 Ground-water level Streamflow Figure 18. Ground-water levels and streamflow, March to May 2002, Greenwich area, Connecticut.

Simulation of Ground-Water Flow 33 Table 7. Water levels in April 2002 and average water levels at selected U.S.Geological Survey network wells in Fairfield County, Connecticut.

[Water levels in feet below land surface; USGS, U.S. Geological Survey]

USGS local well identifier and station number (see fig. 6)

Town April 2002 water level Average water level Period of record used Aquifer FF23 411256073153101 Fairfield 8.13 7.89 1966-2002 Glacial stratified deposits FF30 411124073172201 Fairfield 4.87 4.94 1993-2002 Crystalline bedrock FF31 411118073175801 Fairfield 7.04 8.47 1993-2002 Crystalline bedrock FF32 411030073181301 Fairfield 6.31 7.71 1993-2002 Crystalline bedrock FF33 411058073182001 Fairfield 5.18 5.30 1993-2002 Till BD8 413007073250501 Brookfield 32.37 30.56 1966-2002 Glacial stratified deposits Boundary Conditions The lateral model boundaries are based mostly on physical boundaries of the Greenwich area, including basin boundaries and streams (fig. 15). Basin boundaries on the perimeter of the model are treated as no-flow boundaries, and streams near the model boundaries are treated as head-dependent boundaries simulated with the MODFLOW stream package (Prudic, 1989). Other model boundaries include Long Island Sound on the southern side of the model, simu-lated as a constant-head boundary with altitude zero; the Byram River (fig. 2) in the southwestern part of the model, simulated as a head-dependent boundary; and the Mianus River on the eastern side of the model (head-dependent boundary) (fig. 15). To limit the size of the active model, three areas of the model were terminated across basin boundaries at a narrow point.

These areas include the northeastern corner of the model area, and the Blind Brook Basin (fig. 2) in Westchester County, where streamflow leaves the model. The model also is terminated across a narrow point in the Byram River Basin (fig. 2) where stream-flow enters the model. The large reservoirs, including the Kensico Reservoir, part of the Bargh Reservoir, Putnam Lake, Rockwood Lake, and Converse Lake (figs. 2 and 14), were simulated as constant head boundaries, because of their large size and typically large quantities of water in storage at all times other than extreme drought periods.

The lower model boundary is a no-flow boundary that is 645 ft below land surface. The upper boundary of the model is the land surface. Recharge is applied to layer 1 in the model and is considered a spec-ified-flux boundary.

Ground-Water Recharge Ground-water recharge to areas with glacial till was estimated using optimal values based on nonlinear regression modeling in MODFLOW. Ground-water recharge to glacial stratified deposits was held constant at previously estimated values (table 8). Recharge rates in the model are effective recharge rates and account for the effects of ground-water evapotranspiration. A recharge parameter was defined in the model using a zone array (fig. 19A) corresponding to surficial geology, and a multiplication array corresponding to the percent pervious area in the model cell. The recharge parameter multiplied by the percentage of pervious area in each model cell determined the recharge value for the model input. This was done because peak streamflows (normalized to drainage area) in urbanized basins are higher than in basins with less urban development (fig. 20). The increased over-land runoff limits the amount of water available for recharge, ultimately reducing the base flow of streams.

These recharge estimates are conservative, because not all impervious areas drain directly into watercourses.

Water from rooftops and driveways may infiltrate onto lawns or other unpaved areas, and stormwater drainage from some areas may infiltrate to the water table if the stormwater is discharged to retention basins or is allowed to flow onto pervious areas.

34 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD GREENWICH Westchester County, New York Westchester County, New York Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' STAMFORD GREENWICH Westchester County, New York Westchester County, New York EXPLANATION Glacial stratified deposits (coarse grained)

Active model area boundary Glacial stratified deposits (fine grained)

Model cells not highlighted inside active model are thin till EXPLANATION Glacial stratified deposits Active model area boundary Areas of thick till Model cells not highlighted inside active model area are thin till in layer 1 A.

B.

Recharge rate used in ground-water-flow simulations (inches per year) 20 7.9 6.9 Hydraulic conductivity used in ground-water-flow simulations (feet per day) 0.5 0.10 100 Figure 19. Zone array based on geology used for (A) application of recharge to the model and (B) application of hydraulic properties to layers 1and 2, steady-state ground-water simulation, Greenwich area, Connecticut.

Simulation of Ground-Water Flow 35 The percentage of impervious area in each model cell was estimated by two methods. For model cells completely within the town of Greenwich, impervious areas were estimated from 1997 Greenwich GIS layers as the sum of paved roadway areas, building areas, and driveway and sidewalk areas. In Greenwich GIS layers, driveways and sidewalk areas are linear features and have no area associated with them. To estimate the areas of these features, the length was divided by 2 and multiplied by the estimated width (driveways-13 ft; sidewalks-5 ft).

For model cells in areas outside Greenwich, impervious areas were estimated from a grid of land use/land cover based on 1995 Landsat thematic mapper images interpreted by Civco and others (1998). A regression model (fig. 21) was used to relate land-use/land-cover characteristics from Civco and others (1998) (table 9) to the calculated impervious areas in each of 120 randomly selected model cells in the Greenwich part of the model area. Predictions based on the regression model were applied to the model area outside Greenwich. Significant variables in the regres-sion at the 95-percent confidence level included the urban categories Commercial/Industrial/Paved, Residential and Commercial, and Turf and Tree Complex. Other urban variables, including Turf and Grass and Rural Residential were not found to be significant predictors of impervious area. Model resid-uals were determined to be reasonably normally distributed (fig. 21).

Table 8. Recharge estimates from historical values and nonlinear regression for April 2002 calibration.

[*, value presented in model calibration section of report]

Recharge parameter Method Recharge (inches per year)

Till Nonlinear regression Glacial stratified deposits, 

coarse grained Historical 

values 120 1Value from Mazzaferro, 1986.

Glacial stratified deposits, 

fine grained 2Estimated 2The area of fine-grained glacial stratified deposits represents a very small area in the ground-water flow model; a recharge rate similar to those for glacial till was used because of the low permeability of these deposits.

7.9 Table 9. Parameter estimates, standard errors, t-statistics and p-values for multiple linear regression model of percentage of impervious area for randomly selected cells of the finite-difference ground-water-flow model, Greenwich area, Connecticut.

[Coefficient of determination (R2)=0.87]

Variable Units Parameter estimate Standard error t-statistic p-value Intercept Dimensionless 2.0814 0.5464 3.8094 0.0002 Commercial/

industrial/paved Percentage of model cell area 0.6631 0.0475 13.9696

<.0001 Residential and 

commercial Percentage of model cell area 0.3331 0.0167 19.9993

<.0001 Turf and tree

complex Percentage of model cell area 0.1232 0.0166 7.4342

<.0001

36 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND PER SQUARE MILE March 3

6 9

12 15 18 21 24 27 30 April 3

6 9

12 15 18 21 24 27 30 2001 0

0 5

5 10 10 15 15 20 20 25 25 30 30 35 35 01211699 ~4% impervious area1 01211110 ~10% impervious area1 EXPLANATION

1. Estimated using equation based on table 12. % Impervious area = 2.0814 +.6631(C) +.3331(R) +.1232(T), where C = Commercial/industrial/pavement, R = Residential and commercial, and T = Turf and tree complex from Civco and others (1998).

Figure 20. Instantaneous streamflows per square mile at two continuous streamflow-gaging stations, Greenwich area, Connecticut, March to April 2001.

I I

Simulation of Ground-Water Flow 37 0

10 20 30 40 50 0

10 20 30 40 50 PREDICTED IMPERVIOUS AREA, IN PERCENT OF MODEL CELL AREA MEASURED IMPERVIOUS AREA, IN PERCENT OF MODEL CELL AREA 1:1 LINE 60 Aquifer Properties Hydraulic conductivity of aquifer materials either was assigned based on information from previous investigations or was estimated using nonlinear regression. Hydraulic conductivity of the crystalline bedrock was estimated by nonlinear regres-sion in MODFLOW 2000. Hydraulic conductivity of glacial stratified deposits, thin till, and thick till were based on values summarized in table 10. Values from table 10 were applied using zone arrays (fig. 19) for layers 1 and 2. Layers 3 to 5 were completely in bed-rock, and a zone array was not used.

Table 10. Simulated values of hydraulic conductivity in finite-difference ground-water-flow model, Greenwich area, Connecticut.

[*, value presented in model calibration section of report]

Aquifer material Method Hydraulic conductivity (feet per day)

Median values from historical summary by Melvin and others (1992)

(feet per day)

Thin till Historical values 0.5 0.6-2.7 Thick till (drumlin till)

Historical values

.10

.06 Glacial stratified deposits (combined coarse and fine grained)

Historical values 100 170 

(coarse grained)

Crystalline bedrock Nonlinear regression

.6 Figure 21. Measured and predicted impervious areas for a random sample of model cells, Greenwich area, Connecticut.

38 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Internal Sources and Sinks of Water Ground water is added (source) or removed (sink) from the modeled area through streams, reser-voirs, wells, and septic systems. Simulation of these sources and sinks is described below.

Streamflow Streams were simulated using the stream package created for MODFLOW by Prudic (1989). The stream package is an accounting program that tracks the flow in streams that interact with ground water. The package allows for diversions of water among basins and provides output for comparisons with streamflow measurements made for model calibration.

The active model area (fig. 15) contains 427 stream segments and 1,375 reaches. Streamflows are estimated for every cell in the ground-water-flow model that contains a stream reach and are accumulated in a downstream direction. Discharge between each stream reach and the adjacent aquifer is calculated using Darcys Law and the equation from Prudic (1989):

(3) where:

Q is leakage to or from the aquifer through the streambed Hs is head in the stream Ha is head in the aquifer CSTR is conductance of the streambed, defined as KLW/M, where K is the hydraulic conductivity of the streambed, L is the length of the reach, W is the width of the stream in the reach, and M is the streambed thickness.

Streambed hydraulic conductivity was assumed to be 1 ft/d, and streambed thickness was assumed to be 1 ft. Streambed altitude was determined for each reach from a grid of 3,652 points regularly spaced every 225 ft along all streams. The initial altitude of each point was determined using a digital elevation model. The values were then adjusted by visually inspecting each in comparison to USGS 10-ft contour quadrangle maps. Stream stage was assumed to be 2 ft higher than streambed altitude.

Streamflows predicted by the stream package downstream from the large reservoirs (Putnam and Rockwood Lakes) may be over-or underestimated because of the lack of information on daily diversions, withdrawals, and releases. Simulated streamflows for each zone represent the amount of water discharged from the ground-water system to the stream. With-drawals from and diversions to these reservoirs will, in reality, affect the actual streamflow. Streamflows esti-mated for streams that are boundaries (fig. 15) in the model (that is, the mainstems of Mianus and Byram Rivers, fig. 2) do not include ground-water discharge or tributaries that may flow into these rivers from outside the model area.

Reservoirs Reservoirs (Kensico, Converse Lake, Rockwood Lake, Putnam Lake, Bargh (figs. 2 and 14) were simu-lated as constant heads. This allows for an unlimited amount of water to enter the aquifer if heads in the aquifer are below those of the constant head. Another approach to simulating the reservoirs would be to use a general head boundary. Constant head boundaries were chosen for this simulation for the following reasons: (1) the discharge from reservoirs to the ground-water system was small, indicating that reservoirs were not a major source of recharge; (2) the hydraulic conduc-tivity of the till deposits is small, and of the same order of magnitude or lower than reservoir bottom sediments, therefore it is likely that the flow of water to or from reservoirs is primarily controlled by the properties of the underlying aquifers; and (3) large quantities of water are stored at all times of year in these reservoirs.

Ground-Water Withdrawals and Return Flow from Septic Systems In the calibration model simulation, ground-water withdrawals from residential land-use areas were summed for each model cell based on the water-use multiple-linear regression models described previ-ously. The ground-water model was calibrated using streamflow and water-level data collected during mid-to late April 2002; therefore, ground-water with-drawals were assumed to represent nongrowing season conditions. Ground-water was withdrawn from layer 3 of the model, which represents the top of bedrock.

Several recent studies have shown that a large percentage of the discharge from wells in fractured rock comes from shallow fractures near the surface of the bedrock (Mullaney and others, 1999; Wolcott and Snow, 1995).

Q CSTR HS Ha





=

Simulation of Ground-Water Flow 39 Return flow from septic systems was simulated as recharging wells summed for each model cell and was assumed to be equivalent to indoor water use; therefore, winter water-use rates were used. Return flow from septic systems includes water returned in areas with private wells and in areas with public water supply and septic systems. Water from septic systems was returned to layer 1 in the model. It should be noted that based on the assumptions described above, net ground-water consumption (in areas with wells and septic systems) is considered to be negligible during April 2002.

In the second simulation, in which average pumping conditions were simulated, the regression model estimates for average daily water use were applied to model cells; however, return flow was again assumed to be similar to winter water-use rates.

Model Calibration The calibration process is a nonlinear regression to provide optimal estimates of parameters based on the best fit to measured (or observed) values of hydraulic head and streamflow. Two parameters in the model were estimated in this wayhorizontal hydraulic conductivity of the bedrock aquifer and recharge to till deposits (table 11). These two parameters are important because the majority of ground-water flow takes place in the bedrock aquifer, and the majority of recharge takes place through till deposits. Other parameters that were not estimated using the nonlinear-regression method were based on values from previous investiga-tions in Connecticut.

The ground-water-flow model of the Greenwich area was calibrated to 34 ground-water levels measured during April 16-20, 2002, water levels from 402 well-completion reports for wells (fig. 17) and streamflow measurements at 8 sites. Water levels from well-completion reports represent a wide range of water-level conditions. It is assumed that this range of condi-tions represents long-term average annual conditions.

Data from water-level measurements and well-comple-tion reports are subject to errors, including inaccuracy in determining location and land-surface elevation.

Data from wells measured during April 2002 were given a larger weighting in the calibration process.

The weighting process for water level and streamflow measurements was accomplished as follows. Data that have a higher degree of accuracy are given a larger weight. The weights reflect the possible measurement error in the data and are equal to the inverse of the variance of the measurement. The vari-ance in the measurements for both streamflow and water-level measurements was determined subjec-tively, based on an assessment of, in the case of water level measurements, the accuracy of the location and the land-surface elevation. The measurement error was assumed to be half the contour interval of the topo-graphic map provided by the town of Greenwich (2 ft).

This translates to a variance in water level measure-ments for wells measured in April 2002 that was esti-mated to be 0.37 ft2 at the 90-percent confidence level.

Measurements obtained from well-completion reports were assumed to be less accurate because the wells were not located in the field. The weight of these measurements was assumed to be one, which translates to a measurement error of 1.65 ft at the 90-percent confidence interval. The methods of estimating model observation weights are described by Hill (1998).

The weighting of the streamflow measurements in the calibration process was dependent on the assign-ment of whether the measurement was determined to be good, fair, or poor based on width of the river and stream cross-section characteristics. This weighting process for observations of streamflow loss or gain is described by Hill (1998).

A plot of the weighted simulated equivalents and weighted observations is shown in figure 21. This plot contains information on the fit among the observations and values predicted by the calibrated ground-water model. The model residuals appear to be normally distributed (fig. 22), but there are some wells with water-level observations in the ground-water-flow model where predicted water levels are much higher or lower than observed. There does not appear to be a geographic bias in water-level residuals. The average weighted error in the water-level measurements was 1.9 ft (table 12). There was generally good agreement among streamflow observations (weighted and unweighted) and predictions (table 13, fig. 22).

Table 11. Parameter estimates of hydraulic conductivity and recharge from nonlinear regression, finite-difference ground-water flow model, Greenwich area Connecticut.

Estimated parameter Value Unit Horizontal hydraulic 

conductivity crystalline 

bedrock 0.05 Feet per day Recharge to till deposits 6.9 Inches per year

40 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Table12. Summary of error statistics and comparison of observed and simulated ground-water levels, Greenwich area, Connecticut, April 16-20, 2002.

Statistic Value Sum of squared weighted errors 4.98 X 105 Average weighted error 

(ground-water level measurements) 1.90 feet Average unweighted error 

(ground-water level measurements) 1.82 feet Table 13. Observed and simulated streamflows, Greenwich area, Connecticut, 

April 24-25, 2002.

[USGS, U.S. Geological Survey; ft3/s, cubic feet per second]

USGS station identification number (table 1, fig. 2)

Observed streamflow (ft3/s)

Simulated streamflow (ft3/s)

Difference (ft3/s) 01211010 1.01 0.99 0.02 01211040 0.54 0.71

-.17 01211110 1.60 1.63

-.03 01211140

.22

.25

-.03 01211699 1.09 1.28

-.19 01212550

.84

.75

.08 01212600 excluding the contribu-tions from 01211450, 01212550, and 01211699 7.12 5.79 1.33

Simulation of Ground-Water Flow 41 0

200 400 600 800 0

200 400 600 800

-100

-100 1,000 1,000 WEIGHTED SIMULATED EQUIVALENT WEIGHTED OBSERVATION Water-level measurements made April 16-20, 2002 Water-level measurements from selected well completion reports Streamflow (baseflow) measurements made April 24-25, 2002 1:1 line Hydrologic Budget The ground-water budget was summarized for two simulations. The first simulation was the model-calibration phase, based on measurements of water levels and streamflow collected during mid-to late April 2002. For the second simulation, the calibrated model was used with residential ground-water with-drawals adjusted to average daily residential with-drawal rates (tables 14 and 15). The ground-water budget for each basin-based zone (first shown in 

fig. 13) (tables 14 and 15) was summarized using ZONEBUDGET (Harbaugh, 1990), a computer program that summarizes ground-water budgets for specific zones from MODFLOW-2000 output. Ground-water budgets for this report only include residential withdrawals. Further information is necessary on nonresidential properties.

Components of the ground-water budget for each basin include inputs and outputs. Inputs of water to each basin include recharge from precipitation, discharge from streams and reservoirs to the aquifer, discharge from residential septic systems to the aquifer, and underflow from other basins. Outputs of water from each basin include underflow to other basins, discharge to reservoirs or to Long Island Sound, discharge to streams, and ground-water withdrawals.

Consumptive use of ground-water withdrawn from the bedrock aquifer was considered to be negligible during April 2002; therefore, estimates of the residential net consumption of ground water are included only in the model adjusted to average daily ground-water with-drawals (table 15).

Recharge rates from precipitation ranged from 3.9 to 7.5 in/yr for zones in the model (fig. 23) and are based on the surficial geology and percentage of pervious area of each zone. Leakage from reservoirs or streams to the aquifer was small, and added 0.1-

0.8 in/yr of recharge to the aquifer in some zones.

Figure 22. Weighted simulated equivalent plotted against weighted observations of hydraulic head and streamflow for model simulation of the Greenwich area, Connecticut, April 16-24, 2002.

42 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Table 14. Annual ground-water budget for modeled zones, based on calibration data from April 18-25, 2002, Greenwich area, Connecticut.

[Based on calibration data from April 16-25, 2002; all values in inches per year; mi2, square miles]

Zone (fig. 23)

Area (mi2)

Recharge from pre-cipitation Inflow from reservoirs Inflow from streams Inflow from residential septic sys-tems Inflow from other basins Outflow to other basins Outflow to reservoirs or Long Island Sound Outflow to streams Residential ground-water withdrawals 1

1.6 5.3 0.8 0.0 0.4 2.9 2.9 0.6 5.5 0.4 2

1.2 6.5 0

.5

.8 2.0 2.7 0

6.2

.8 3

.6 6.8 0

0 1.2 1.0 2.9 0

4.9 1.2 4

3.1 6.4 0

0

.9 2.1

.9 0

7.9

.6 5

.9 5.9 0

0

.1 1.2 4.9 0

2.2

.1 6

1.7 7.5 0

.5

.8 2.9

.3 0

10.6

.8 7

.8 6.4 0

0 1.5 1.2 4.3 0

3.3 1.5 8

1.6 6.4 0

0

.9 1.9 1.0 0

7.2

.9 9

.3 5.5 0

0 1.1 1.3 1.9 0

4.9 1.1 10 1.4 6.6 0

.2 1.4 1.5 2.2 0

6.2 1.4 11 1.1 6.9 0

.9 1.0

.7 3.8 0

5.1

.5 12 1.4 6.6 0

0 1.4 1.8 1.1 0

7.3 1.4 13 2.3 6.3 0

0

.7 1.4

.5

.4 6.7

.7 14 1.2 6.8 0

0 1.7 2.0

.5 0

8.9 1.1 15 3.5 6.7 0

.1

.9 1.2

.9 0

8.0 0

16 2

6.8 0

0

.9 2.0 1.6

.5 6.8

.9 17 1.7 6.7 0

0 1.9 1.4 1.3 0

8.5

.3 18 3

4.3 0

0

.5

.8

.9

.5 4.3 0

19 3

6.7

.1 0

1.2

.8

.8 0

7.0 1.1 20 1

6.1 0

0 2.1 1.5 1.3 0

7.0 1.5 21 2.3 6.7 0

0 1.8 1.1 1.5 0

7.9

.2 22 1.8 6.2 0

0 1.5 1.1 1.0 0

7.3

.5 23 1.2 4.1 0

0 1.0

.7

.4

.8 4.5 0

24

.4 7.1 0

0 2.6 2.0 2.2 0

6.9 2.6 25 1.3 6.7

.1 0

.6 3.0 2.6

.6 6.6

.6 26 1.5 7.0 0

.2

.9

.9 1.6 0

6.6

.9 27 1.1 5.7 0

.1 1.5 3.7 1.9 0

8.4

.7 28 1.2 7.2 0

0 1.9 1.4

.9 0

8.0 1.6 29 1.6 5.5 0

0 1.0 1.0 1.0 0

6.2

.2 30 2.7 4.9 0

0 1.4

.9 1.6

.4 4.9

.3 31 2.7 4.6 0

0

.5 1.3 1.1 1.2 3.9

.1 32 1.8 3.6 0

0

.5

.4 1.4 1.3 1.8 0

Simulation of Ground-Water Flow 43 Table 15. Annual ground-water budget for modeled zones, based on calibration data from April 18-25, 2002 and adjusted to average residential water withdrawals, Greenwich area, Connecticut.

[All values in inches per year; mi2, square miles]

Zone (fig. 23)

Area (mi2)

Recharge from pre-cipitation Inflow from reservoirs Inflow from streams Inflow from residential septic sys-tems Inflow from other basins Outflow to other basins Outflow to reser-voirs or Long Island Sound Outflow to streams Residential ground-water withdrawals Net residential water consumption 1

1.6 5.3 0.8 0

0.4 2.9 2.8 0.6 5.5 0.5 0.1 2

1.2 6.5 0

.5

.8 1.9 2.7 0

5.8 1.1

.3 3

.6 6.8 0

0 1.2

.8 2.8 0

4.3 1.7

.5 4

3.1 6.4 0

0 1.0 2.1

.9 0

7.7

.8

-.2 5

.9 5.9 0

0

.1 1.1 4.9 0

2.1

.2 0

6 1.7 7.5 0

.6

.8 2.8

.3 0

10.3 1.2

.3 7

.8 6.4 0

0 1.5 1.1 4.2 0

2.4 2.3

.8 8

1.6 6.4 0

0

.9 1.7

.9 0

6.8 1.3

.4 9

.3 5.5 0

0 1.1 1.3 1.8 0

4.6 1.4

.4 10 1.4 6.6 0

.1 1.4 1.4 2.0 0

5.5 2.1

.6 11 1.1 6.9 0

.9 1.0

.7 3.8 0

5.0

.7

-.3 12 1.4 6.6 0

0 1.4 1.8 1.1 0

6.6 2.2

.7 13 2.3 6.3 0

0

.7 1.4

.5

.4 6.5 1.0

.3 14 1.2 6.8 0

0 1.7 2.0

.5 0

8.4 1.6

-.1 15 3.5 6.7 0

.1 1.0 1.2

.9 0

8.0

.1

-.9 16 2.0 6.8 0

0

.9 1.9 1.5

.4 6.4 1.2

.3 17 1.7 6.7 0

0 1.9 1.3 1.2 0

8.3

.4

-1.5 18 3.0 4.3 0

0

.5

.8

.9

.5 4.3 0

-.5 19 3.0 6.7

.2 0

1.2

.8

.8 0

6.5 1.5

.3 20 1.0 6.1 0

0 2.1 1.4 1.2 0

6.5 1.9

-.2 21 2.3 6.7 0

0 1.8 1.1 1.5 0

7.8

.2

-1.6 22 1.8 6.2 0

0 1.5 1.1 1.0 0

7.1

.7

-.8 23 1.2 4.1 0

0 1.0

.7

.4

.8 4.5 0

-1.0 24

.4 7.1 0

0 2.6 2.1 2.1 0

6.1 3.6

.9 25 1.3 6.7

.1 0

.6 3.0 2.6

.6 6.2 1.0

.4 26 1.5 7.0 0

.2

.9 1.0 1.5 0

6.3 1.2

.3 27 1.1 5.7 0

.1 1.5 3.6 1.9 0

8.1

.9

-.6 28 1.2 7.2 0

0 1.9 1.3

.8 0

7.5 2.1

.2 29 1.6 5.5 0

0 1.0

.9 1.0 0

6.1

.3

-.7 30 2.7 4.9 0

0 1.4

.9 1.5

.4 4.8

.4

-1.0 31 2.7 4.6 0

0

.5 1.3 1.1 1.2 3.9

.2

-.3 32 1.8 3.6 0

0

.5

.4 1.4 1.3 1.8 0

-.4

44 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Estimated long-term average ground-water recharge rates, in inches per year 3.6 - 4.9

>4.9 - 5.9

>5.9 - 6.8

>6.8 - 7.3

>7.3 - 7.9 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' GREENWICH STAMFORD Westchester County, New York Westchester County, New York Long Island Sound Basin number EXPLANATION Figure 23. Simulated long-term, average annual ground-water recharge from precipitation aggregated by zones, Greenwich area, Connecticut.

Simulation of Ground-Water Flow 45 Inflow from residential septic systems is derived from ground-water withdrawals in the northern part of the Greenwich area and is not a net additional source of recharge over long-term average conditions. The redis-tribution of this water may be hydrologically impor-tant, because it is withdrawn from the deeper aquifer and discharged near the land surface, and therefore may change ground-water flow patterns. For instance, the water may discharge more quickly to a nearby surface-water body.

In parts of basins with public water supply and septic systems, the discharge from septic systems represents an additional source of recharge. The net amount of additional recharge can be obtained from table 14 or 15 by subtracting residential ground-water withdrawals from inflow from residential septic systems in the calibration model.

In many cases, the outflow to streams in the cali-bration model (table 14) does not balance with the input of recharge from precipitation. Differences in these two values can be attributed to (1) input of addi-tional sources of recharge described above, (2) under-flow among basins, and (3) discharge of water to reservoirs or to Long Island Sound. In the simulation that was adjusted to average residential ground-water withdrawals, differences also can be attributed to consumptive use of ground water. Water-use totals may differ slightly compared to those in the water-use section because of discrepancies in delineating basin boundaries in the model grid and assigning a zone/basin code to each model cell.

Net annual consumptive residential water use was determined by subtracting the simulated input from septic systems from the average annual ground-water withdrawals in each zone (table 15). Some basins that have withdrawals of ground water have greater input from septic systems supplied by public water supply; therefore, the net residential consumption of ground water is zero (fig. 24) or is shown as a negative number in table 15.

46 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Estimated long-term average net residential ground-water consumption 0

>0 - 0.2 >0 - 0.01

>0.2 - 0.4 >0.01 - 0.02

>0.4 - 0.6 >0.02 - 0.03

>0.6 - 0.9 >0.03 - 0.04 Base from Connecticut Department of Environmental Protection 1994 Digital Line Graph Projection State Plane Feet Zone 3526 41 o

41 o5' 73 o40' 73 o35' GREENWICH STAMFORD Westchester County, New York Westchester County, New York Long Island Sound Inches/

year Million gallons/day/

per square mile Basin number Basins with potential of 50 percent or more increase in average daily water use due to commerical, or public, or institutional properties

(-1.6) - (-0.1)*

• Basin has greater return flow from residential septic systems served by public water supply than consumptive ground-water withdrawals EXPLANATION Figure 24. Simulated long-term average net residential consumptive water use for zones, Greenwich area, Connecticut.

Ground-Water Availability 47 GROUND-WATER AVAILABILITY IN THE GREENWICH AREA Ground water is an important part of the water budget in Greenwich providing, at minimum, an esti-mated 35 percent of the annual streamflow (see eq. 1).

This base flow contributes to maintaining river habitat and aquatic life, providing inflow to public water-supply reservoirs, and providing water to ponds used for fire protection during periods of little overland runoff.

Determination of ground-water availability should consider what changes to the water budget of a particular basin might be acceptable. Changes to the water budget due to withdrawals may include decrease in base flow of streams, increase in the recharge from surface-water bodies, or the capture of ground water that would have originally been lost through ground-water evapotranspiration in wetlands and other areas with shallow water tables. One method to determine the status of each basin is to examine various streamflow criteria in comparison to the current water budget. To show this current status, several equations that predict statistical low streamflows were compared to the conditions described in the above section on ground-water budgets.

Weiss (1983) produced equations based on streamflow statistics in Connecticut to determine the 7-day 10-year low flow (Q7,10), the 7-day 2-year low flow (Q7,2), and the 30-day 2-year low flow (Q30,2). The (Q7,10) is the lowest flow for 7 consecutive days with a recurrence interval of 10 years, the (Q7,2) is the lowest flow for 7 consecutive days with a recurrence interval of 2 years, and the (Q30,2) is the lowest flow for 30 consecutive days with a recurrence interval of 2 years.

The (Q7,10) in Connecticut is considered to be the minimum amount of water required to assimilate wastewater (Connecticut Department of Environmental Protection, 1997). The (Q7,2) and (Q30,2) are statistical determinations of the typical summer flows of streams in Connecticut and probably represent periods of typical annual low base flow.

Equations for these streamflows from Weiss (1983) are:

(Q7,10)/A = 0.0065(%Asd+1) - 0.001 (4)

(Q7,2)/A = 0.0104(%Asd+1) - 0.001 (5)

(Q30,2)/A = 0.0124(%Asd+1) - 0.001 (6) where Q

is the specified discharge, in cubic feet per second, A

is basin drainage area, in square miles, and

%Asd is the percentage of the basin contain-ing coarse-grained glacial stratified deposits (glacial stratified deposits).

The specified flows are larger for basins containing larger percentages of glacial stratified deposits because of the greater recharge rates and the ability to store and transmit water.

Long-term average base flows predicted by the ground-water-flow simulation based on residential ground-water withdrawals are, in most cases, substan-tially greater than the low streamflow statistical criteria (fig. 25). The water budget for many of the basins prob-ably is different now compared to pre-development conditions because of ground-water consumption and increasing urbanization.

In general, basins with consumptive residential use of ground water have long-term estimated average base flows that are lower than basins with little or no consumptive use. Some of the basins with little or no consumptive use of ground water (for example, basins 31 and 32; fig. 21), however, are estimated to have reduced recharge rates due to urbanization, and there-fore the estimated base flows approach low-streamflow criteria.

48 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 21 17 30 23 15 22 29 27 18 32 11 20 4

14 5

1 28 19 16 2

6 13 26 9

8 25 3

10 12 7

24 15 0

2 4

6 8

10 12 14 PERCENT OF BASIN UNDERLAIN BY COARSE-GRAINED GLACIAL STRATIFIED DEPOSITS 0

12 0

2 4

6 8

10 STREAMFLOW IN INCHES PER YEAR EXPLANATION Q30, 2 Calculated baseflow (equation 1)

>0 - 0.4 in/yr

>0.4 - 1.0 in/yr Q7,10 Q7, 2 Estimated ground-water discharge to streams for basins with net residential1 consumptive use of ground water 17 Basin number 31 Calculated low flows (equations 4-6)

-1.6* - 0 in/yr

1. The effects of non-residential water use are not included

• Negative number indicates basin has greater return flow from residential septic systems served by public water supply than consumptive ground-water withdrawals Figure 25. Simulated ground-water outflow to streams, calculated base flow, and low flow for basins in the Greenwich area, Connecticut. [Q7,10 is the 7-day 10-year low flow, Q7,2 is the 7-day 2-year low flow, and Q30,2 is the 30-day 2-year low flow.]

Ground-Water Availability 49 When current estimated net residential ground-water consumptive use is compared to recharge rates for each zone in the study area (table 16), it can be seen that ground-water consumption is, in most cases, small.

One criteria that might be used to make comparisons among zones in the Greenwich area is to subtract the Q30,2 from the recharge (from precipitation) for each basin normalized to drainage area. This sample criteria is conservative in the respect that it does not include additional recharge from septic systems or underflow from other basins. The basins with the largest relative estimated net residential consumptive water use include zones 7, 10, 12, and 24. It is estimated that for these zones, about 10 percent or more of the difference between long-term average recharge and the Q30,2 is being used (table 16).

This sample criteria, although useful on a comparative basis, does not take into account other issues including (1) the additional nonresidential water use; (2) the physical plausibility of being able to capture the difference between the recharge and the Q30,2 by pumping; (3) the localized effects of lowering the water table on nearby wells, or on the habitat contained within wetlands and watercourses; and (4) the effects on summer streamflows (for example, if the average base flow is lowered to the Q30,2, summer base flow also will be substantially lower).

Output data from these ground-water models can be used to determine the current status of ground-water use relative to recharge or other water-budget criteria.

In order for this data to be used properly, several quali-fications and assumptions should be discussed.

In the ground-water flow simulations conducted for this study, it was assumed that impervious areas do not receive any recharge. Estimates of recharge from the nonlinear parameter estimation algorithm in MODFLOW were matched to streamflow observations and water levels to obtain the best fit for recharge to till deposits assuming no recharge beneath impervious areas. The recharge to urbanized basins is therefore reduced. It is believed that these estimates are conser-vative and that more research is needed in this area to determine the effects of impervious cover and alterna-tive stormwater management on recharge and base flow.

Table 16. Estimated net annual residential consumptive water use and the difference between estimated long-term average recharge and the 30-day 2-year low flow.

[Zones not shown in this table are estimated to have no net consumptive residential use of water; Q30,2, 30-day 2-year low flow]

Zone (fig. 22)

Recharge (inches)

Water consump-tion (inches)

Q30,2 (inches)

Difference between recharge and Q30,2 (inches)

Percentage of difference consumed 1

5.3 0.13 0.9 4.4 2.9 2

6.5 0.33 1.5 4.9 6.6 3

6.8 0.49 0.2 6.7 7.4 5

5.9 0.03 0.2 5.8 0.6 6

7.5 0.33 0.9 6.6 5.0 7

6.4 0.81 0.2 6.2 13.0 8

6.4 0.38 0.6 5.7 6.7 9

5.5 0.35 0.3 5.2 6.8 10 6.6 0.63 0.7 6.0 10.6 12 6.6 0.74 0.2 6.3 11.6 13 6.3 0.34 0.4 5.9 5.7 16 6.8 0.32 0.3 6.5 5.0 19 6.7 0.31 1.3 5.5 5.6 24 7.1 0.94 0.5 6.6 14.3 25 6.7 0.38 1.8 4.8 8.0 26 7.0 0.35 1.6 5.5 6.4 28 7.2 0.19 0.9 6.3 3.0

50 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Residential water use was documented for this study, but there may be additional substantial ground-water withdrawals for commercial, institutional, and public buildings and golf courses. As shown in 

figure 12, additional water is used in some basins that requires further research.

Water that has been returned to the aquifer by septic systems is assumed to remain in the basin as part of the water budget. In reality, some of this water may never recharge the bedrock aquifer, and may discharge to a nearby surface-water body and not be available for reuse. In the model simulations, return flow from septic systems was greater than ground-water withdrawals in nearly half of the basins studied (because of input from private septic systems at properties served by public water supplies), indicating that this is an important component of the water budget in the Greenwich area.

Wells drilled into the bedrock underlying glacial stratified deposits may have more water available under pumping stresses, if there is a good hydraulic connec-tion between these coarse-grained deposits and the underlying bedrock, due to the storage characteristics of this aquifer.

Data and interpretation of the water budget are provided for 32 basin-based zones (tables 14, 15).

Human activities, including new development and ground-water withdrawals, may have different effects depending on their location in these basins. For instance, concentrated development with ground-water withdrawals near the headwaters of a basin may have a more localized effect on the water budget (that is, streamflow reduction or reduction in ground-water levels) than the same development if farther down-stream in the basin, even though the overall change in pumping might not be large relative to the water budget for the entire basin. The potential for large fluctuation in water levels is greater in the hilltop setting and is less apparent in the valley bottom, due to ground-water contributions from upgradient parts of the basin 

(fig. 10).

Recharge rates and water budgets presented in this report are not fixed numbers, and the ground-water budget for any basin is dynamic. Changes in develop-ment in a particular basin may have consequences in addition to changes caused by increased withdrawal and use of ground-water. These changes include changes to recharge rates caused by impervious surfaces, changes to the water budgets of other basins (in the case where underflow from one basin to another is important), and the possibility of increased recharge if new development takes place in areas with public water supply and septic systems. The addition of sani-tary sewers to areas with private wells can export water from a basin, and therefore affect ground-water avail-ability. Other changes to the water budget include long-term changes in annual precipitation, changes in the seasonal variability of precipitation, or prolonged drought.

WATER QUALITY IN THE GREENWICH AREA Analyses of seven surface-water-quality samples collected during base flow showed that concentrations of some indicators of water-quality degradation were higher in more urbanized basins than in less urbanized basins (fig. 26; table 17). Concentrations of total nitrogen, total phosphorus, chloride, indicator bacteria, and the number of different pesticide detections were generally higher with increasing urbanization of the basin. Samples from one site (USGS station number 01211210), collected below a point of water diversion and storage, may not be entirely representative of the base-flow conditions of the lower part of Horseneck Brook (fig. 2).

The U.S. Environmental Protection Agency (USEPA) has set regional nutrient-criteria guidelines for State governments for determination of impaired waters (U.S. Environmental Protection Agency, 2000; 2001). The criteria for total nitrogen for Ecoregion XIV (which includes Connecticut) is 0.7 mg/L for rivers and streams and 0.32 mg/L for lakes and reservoirs. The criteria for total phosphorus is 0.031 mg/L for rivers and streams and 0.008 mg/L for lakes and reservoirs.

These criteria are considered a starting point in deter-mining problem areas, and State governments may set more stringent criteria.

Nitrogen and phosphorus concentrations above ambient concentrations may originate from atmo-spheric deposition, septic systems, lawn fertilizers, waterfowl, and from point-source discharges. Elevated nitrogen concentrations are of concern for eutrophica-tion and hypoxia in Long Island Sound (Long Island Sound Study, 1998) because nitrogen is the limiting nutrient in saltwater bodies. Data interpreted from Mullaney and others (2002) indicate that concentra-tions of total nitrogen in mostly forested basins range on average from 0.35 to 0.5 mg/L. As basins become

Water Quality 51 0

20 40 60 0

1 2

0 20 40 60 0.00 0.05 0.10 TOTAL PHOSPHORUS IN MILLIGRAMS PER LITER 0

20 40 60 0

20 40 60 80 TOTAL NITROGEN CHLORIDE 0

20 40 60 1

10 100 1,000 FECAL COLIFORM ENTEROCOCCI 0

20 40 60 URBAN LAND USE IN PERCENT OF BASIN AREA 0

1 2

3 4

5 NUMBER OF PESTICIDES DETECTED URBAN LAND USE IN PERCENT OF BASIN AREA COLONIES PER 100 ML CONCENTRATION, IN MILLIGRAMS PER LITER Figure 26. Percentage of urban land use and concentrations of selected water-quality constituents, Greenwich area, Connecticut.

Table 17. Percentage of urban land use and concentration of selected water-quality constituents in surface-water base-flow samples collected October 2000, Greenwich area, Connecticut.

[Urban land use determined from Civco and others (1998) by summing total area of each basins from the following land use/land cover categories: Com-mercial, industrial, pavement; residential and commercial; rural residential; turf and tree complex; and turf and grass; USGS, U.S. Geological Survey; 

mg/L, milligrams per liter; mL, milliliters]

USGS station identification number (fig. 2)

Urban land use (percentage of basin area)

Total nitrogen concentration (mg/L)

Total phosphorus concentration (mg/L)

Chloride concentration (mg/L)

Fecal coliform bac-teria (colonies per 100 mL)

Enterococci bacteria (colonies per 100 mL)

Number of different pesticides detected 01211600 32.0 2.00 0.064 73.2 100 920 2

01211140 21.1

.380

.022 20.4 40 42 1

01211210 40.3

.600

.105 27.4 110 112 3

01211010 27.7

.530

.026 25.7 16 11 1

01211699 17.9

.290

.012 17.0 3

26 1

01211110 63.1

.880

.059 29.0 110 132 4

01212100 30.0

.500

.020 26.4 110 68 3

52 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 more urbanized, concentrations of total nitrogen due to nonpoint sources of nitrogen may range from less than 0.5 mg/L to 2.0 mg/L. Basins containing municipal wastewater-treatment facilities may have total nitrogen concentrations as high as 5 mg/L or more if the waste-water is a large part of the total streamflow. Grady and Mullaney (1998) and Grady (1994) determined that median nitrate (the most common form of nitrogen in ground water) plus nitrite nitrogen concentrations in shallow ground-water beneath urbanized areas ranged from 1.1 to 2.4 mg/L, as compared with 0.11 to 0.14 mg/L in shallow ground water beneath forested areas.

Heisig (2000) determined that nitrate nitrogen concen-trations in stream base flow were positively correlated to unsewered housing density in the nearby Croton Basin in New York.

Elevated phosphorus concentrations are of concern for algal blooms and eutrophication in fresh-water bodies, where phosphorus is the limiting nutrient. Concentrations of total phosphorus in excess of 0.1 mg/L in moving water in streams may cause algal blooms or excessive nuisance plant growth (Litke, 1999). Concentrations lower than 0.1 mg/L may be able to cause algal blooms in lakes and ponds.

A water sample from an unnamed tributary to Greenwich Creek (USGS station number 01211110, fig. 2) was analyzed using an experimental procedure for wastewater compounds (Kolpin and others, 2002) to determine if this location (the most urbanized basin sampled) showed signs of input from septic systems.

The compounds that were detected were present at very low concentrationsat, near, or below the reporting limits. Concentrations of most of the compounds are estimated due to the experimental nature of the anal-ysis. The compounds detected (table 18) indicate that wastewater from septic systems or leaking sanitary sewer lines has entered this stream. The same compounds were detected in a national survey of phar-maceuticals, hormones, and other organic wastewater compounds (Kolpin and others, 2002).

Table 18. Wastewater compounds detected on October 25, 2000 at station number 01211110, Greenwich Connecticut.

[E, estimated concentration]

Wastewater compound Concentration (micrograms per liter)

Use or indication Para-nonylphenol-total E 0.223 Nonionic detergent metabolite Nonylphenol monoethoxylate- (NPEO1) (total)

E 0.531 Nonionic detergent metabolite Nonylphenol diethoxylate- (NPEO2) (total)

E 0.933 Nonionic detergent metabolite Triclosan 0.051 Anti-bacterial/disinfectant 3-beta-coprostanol E 0.428 Fecal indicator

Water Quality 53

SUMMARY

AND CONCLUSIONS Water use, ground-water availability, and quality of water were studied by the U.S. Geological Survey (USGS) in cooperation with the town of Greenwich, Connecticut, during 2000-02. The study area is 

52.8 square miles and includes Greenwich, part of Stamford, Connecticut, and adjacent parts of Westchester County, New York. Increasing develop-ment and the lack of large glacial stratified aquifers for public water supply have led to a need to study ground-water availability and water use in basins in the Green-wich area. Self-supplied ground water from a fractured crystalline-bedrock aquifer is the source of water supply to about 12 percent of the households in the Greenwich area.

Households with public water supply (from a surface-water source) in Greenwich used from 219 to 1,082 gal/d of water during 2000. Water use at these properties was less during the winter season (204 to 848 gal/d) and more during the summer season (230 to 1,389 gal/d). Water use was strongly dependent on the type and size of the residential property. Three multiple-linear regression models were used to esti-mate self-supplied water use (from a ground-water source) for average daily, winter, and summer condi-tions in 32 basin-based zones. The winter water-use model also was used to estimate return flow in areas with public water supply and septic systems. The regression models were developed with GIS character-istics of properties that were matched to a database of public water-supply information. Statistically signifi-cant predictors of water use included the amount of unforested area, the area of outside swimming pools, the sum of areas covered with buildings, and lot size.

Estimates of average residential daily ground-water withdrawals by basin ranged from 0 to 0.16 million gallons per day per square mile (Mgal/d/mi2), and totaled about 2.1 million gallons per day (Mgal/d). The estimated return flow from residential septic systems on properties served by public water supply ranged from 0 to 0.08 Mgal/d/mi2, and totaled 1.4 Mgal/d.

Residential consumptive water use in Greenwich was estimated to be about 20 percent of average daily water use, based on comparison of data aggregated by basin for seasonal regression models. Additional water use at nonresidential properties, including golf courses, churches, schools, and commercial properties, may amount to a 50-percent or more increase in average daily ground-water use in 5 of the 32 basin areas studied.

A steady-state finite-difference ground-water-flow model was used to study the water budgets of the 32 basins in the study area. The ground-water-flow model was calibrated to streamflow measurements, water levels measured in private wells during April 2002, and additional water levels reported on well-completion reports. Optimal values for hydraulic conductivity of the bedrock aquifer and recharge to till were estimated in the model using nonlinear regres-sion. Ground-water recharge was not applied to imper-vious areas in the model. Impervious areas were estimated for each model cell from GIS information provided by the town of Greenwich, or from extrapola-tion of this information using interpreted LANDSAT imagery. The ground-water recharge to till deposits that cover most of the study area was estimated to be 6.9 inches per year. Average (bulk) hydraulic conductivity of the bedrock was estimated to be 0.05 feet per day.

The calibrated ground-water-flow model was used to simulate average daily residential ground-water withdrawals, and water budgets were summarized by basin. Simulated long-term ground-water recharge by basin ranged from 3.9 to 7.9 in/yr. Basins with the smallest amount of recharge were primarily along the coastal sections of Greenwich where the degree of urbanization is greatest. Net consumptive residential ground-water use ranged from 0 to 0.9 in/yr. The largest net consumptive residential ground-water use was from small basins in the upper reaches of the East Branch of the Byram River Basin and in part of the Mianus River Basin in the Banksville section of Green-wich.

Water-budget components were compared to statistically based calculations of average base flow and several low flows, including the 30-day 2-year flow (Q30,2), the 7-day 2-year flow (Q7,2), and the 7-day 10-year (Q7,10) so that relative comparisons could be made among basins. A criteria of subtracting the Q30,2 from the estimated recharge for each basin was used as one possible estimate of ground-water availability. Based on this criteria, net residential ground-water consump-tion ranged from 0 to 14.3 percent of available water.

Simulated streamflows (long-term average base flows) for each basin also were compared to the statistical calculation of base flow and low flow to determine if any of the long-term average simulated base flows approached these criteria. Possible reductions in base

54 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 flow were noted in many basins that had either rela-tively large ground-water withdrawals or in basins with a high degree of urbanization. Two coastal basins and one basin in north-central Greenwich had simulated streamflows of less than 3 in/yr and approached statis-tical low flows.

Users of this information should note that a percentage of the recharge in each basin (including, for example, artificial recharge from septic systems) may never enter the bedrock aquifer and may be unavailable for use. Changes to the water budget caused by new development in any basin may be more pronounced in the headwaters of a basin than in the downstream valley bottom, in terms of the localized effect on streamflow and ground-water levels.

The water budgets estimated in this report are not fixed because the aquifer is a dynamic system. Changes in development in one basin may have effects on the water budgets of other basins, or may not be apparent for many years. New and existing development may have effects, in addition to ground-water withdrawals, on the water budget, such as loss of recharge, change in recharge patterns, or re-routing ground water through septic systems.

Results of the limited base-flow water-quality sampling indicate that the water quality is related to the degree of urbanization. Concentrations of nitrogen, phosphorus, chloride, and indicator bacteria appear to be related to the percentage of urban land in each of the basins sampled. More urbanized basins had detections of a larger number of pesticides, but concentrations were very low. In three basins, phosphorus concentra-tions were higher than regional nutrient criteria for rivers and streams and may cause excess algal growth.

Total nitrogen concentrations exceeded regional nutrient criteria for lakes and ponds in water samples from six of seven basins sampled, indicating concentra-tions above ambient levels.

REFERENCES CITED American Water Works Association Research Foundation, 1999, Residential end uses of water: Denver, Colorado, 310 p.

Bauer, H.H, and Mastin, M.C., 1997, Recharge from precip-itation in three small glacial-till mantled catchments in the Puget Sound Lowland, Washington: U.S. Geolog-ical Survey Water-Resources Investigations Report 96-4219, 119 p.

Civco, D.L., Arnold, C.L., Hurd, J.D., 1998, Land use and land cover mappingfor the Connecticut and New York portions of the Long Island Sound Watershed: Office of Long Island Sound Programs, Connecticut Department of Environmental Protection, Research Project CWF 330-R Technical Completion Report, unpaginated.

Cohn, T.A., DeLong, L.L., Gilroy, E.J., Hirsch, R.M., and Wells, D.K., 1989, Estimating constituent loads: Water Resources Research, v. 25, no. 5, p. 937-942.

Cooley, R.L., and Naff, R.L., 1990, Regression modeling of ground-water flow: U.S. Geological Survey Techniques of Water Resources Investigations, book 3, chap. B4, 232 p.

Grady, S.J., 1994, Effects of land use on quality of water in stratified-drift aquifers in Connecticut-Chapter B, Analysis of nonpoint-source ground-water contamina-tion in relation to land use: U.S. Geological Survey Water-Supply Paper 2381, 56 p.

Grady, S.J., and Mullaney, J.R., 1998, Natural and human factors affecting shallow water quality in surficial aqui-fers in the Connecticut, Housatonic, and Thames River Basins: U.S. Geological Survey Water-Resources Investigations Report 98-4042, 81 p.

Gregory, H.E., and Ellis, A.J., 1916, Ground water in the Hartford, Stamford, Salisbury, Willimantic, and Saybrook areas, Connecticut: U.S. Geological Survey Water-Supply Paper 374, 150 p.

Harbaugh, A.W., 1990, A computer program for calculating subregional water budgets using results from the U.S.

Geological Survey modular three-dimensional ground-water flow model: U.S. Geological Survey Open-File Report 90-392, 46 p.

Harbaugh, A.W., Banta, E.R., Hill, M.C., and McDonald, M.G., 2000, MODFLOW-2000, The U.S. Geological Survey modular ground-water modelUser guide to modularization concepts and the ground-water flow process: U.S. Geological Survey Open-File Report 00-92, 121 p.

Heisig, P.M., 2000, Effects of residential and agricultural land uses on the chemical quality of baseflow of small streams in the Croton watershed, southeastern New York: U.S. Geological Survey Water-Resources Investi-gations Report 99-4173, 15 p.

Helsel, D.R., and Hirsch, R.M., 1992, Statistical methods in water resources: Amsterdam, Elsevier, 529 p.

Hill, M.C., Banta, E.R., Harbaugh, A.W., and Anderman, E.R., 2000, MODFLOW-2000, The U.S. Geological Survey modular ground-water modelUser guide to the observation, sensitivity, and parameter-estimation processes and three post-processing programs: U.S.

Geological Survey Open-File Report 00-184, 209 p.

Hill, M.C., 1998, Methods and guidelines for effective model calibration: U.S. Geological Survey Water Resources Investigations Report 98-4005, 90 p.

References Cited 55 Hunter, B.D., and Meade, D.B., 1983, Precipitation in Connecticut, 1951-80: Connecticut Department of Environmental Protection Bulletin 6, 92 p.

Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M.,

Zaugg, S.D., Barber, L.B., and Buxton, H.T., 2002, Pharmaceuticals, hormones, and other organic waste-water contaminants in U.S. streams, 1999-2000A national reconnaissance: Environmental Science and Technology, v. 36, no. 6, p. 1202-1211.

Kontis, A.L., 2001, Computer program for simulation of variable recharge with the U.S. Geological Survey modular finite difference ground-water flow model (MODFLOW): U.S. Geological Survey Open-File Report 00-173, 75 p.

Litke, D.W., 1999, Review of phosphorus control measures in the United States and their effects on water quality:

U.S. Geological Survey Water-Resources Investiga-tions Report 99-4007, 38 p.

Long Island Sound Study, 1998, Phase III actions for hypoxia management: U.S. Environmental Protection Agency, 28 p.

Mazzaferro, D.L., 1986, Ground-water availability and water quality in Southbury and Woodbury, Connecticut:

U.S. Geological Survey Water-Resources Investiga-tions Report 84-4221, 105 p.

Mazzaferro, D.L., Handman, E.H., and Thomas, M.P., 1979, Water resources inventory of Connecticut, part 8, Quin-nipiac River Basin: Connecticut Water Resources Bulletin 27, 88 p.

Melvin, R.L., 1986, Connecticut observation wellsGuide-lines for network modification: U.S. Geological Survey Water-Resources Investigations Report 85-4079, 24 p.

Melvin, R.L., Stone, B.D., Stone, J.R., and Trask, N.J., 1992, Hydrogeology of thick till deposits in Connecticut: U.S.

Geological Survey Open-File Report 92-43, 43 p.

(Reprinted as part of Connecticut Department of Envi-ronmental Protection Bulletin 20.)

Morrison, Jonathan, Davies, B.S., 3rd, Martin, J.W., Norris, J.R., 2003, Water resources data, Connecticut, water year 2002: U.S. Geological Survey Water-Data Report CT-02-1, 377 p.

Morrison, Jonathan, Provencher, P.L., Martin, J.W., Norris, J.R., 2002, Water resources data, Connecticut, water year 2001: U.S. Geological Survey Water-Data Report CT-01-1, 353 p.

Mullaney, J.R., Mondazzi, R.A., and Stone, J.R., 1999, Hydrogeology and water quality of the Nutmeg Valley area, Wolcott and Waterbury, Connecticut: U.S.

Geological Survey Water-Resources Investigations Report 99-4081, 90 p.

Mullaney, J.R., Schwarz, G.E., and Trench, E.C.T., 2002, Estimation of nitrogen yields and loads from basins draining to Long Island Sound, 1988-98: U.S. Geolog-ical Survey Water-Resources Investigations Report 02-4044, 84 p.

Prudic, D.E., 1989, Documentation of a computer program to simulate stream-aquifer relations using a modular, finite-difference, ground-water flow model: U.S.

Geological Survey Open-File Report 88-729, 113 p.

Rantz, S.E., 1982a, Measurement and computation of streamflowVolume 1. Measurement of stage and discharge: U.S. Geological Survey Water-Supply Paper 2175, p. 1-284. (Online at http://water.usgs.gov/pubs/wsp/wsp2175/)

Rantz, S.E., 1982b, Measurement and computation of streamflowVolume 2. Computation of discharge:

U.S. Geological Survey Water-Supply Paper 2175, p.

285-631. (Online at http://water.usgs.gov/pubs/wsp/wsp2175/)

Rodgers, John (compiler), 1985, Bedrock geological map of Connecticut: Connecticut Geological and Natural History Survey, Natural Resources Atlas Series Map, 2 sheets, scale 1:125,000.

Rutledge, A.T., 1998, Computer programs for describing the recession of ground-water discharge and for estimating mean ground-water recharge and discharge from streamflow recordsUpdate: U.S. Geological Survey Water-Resources Investigations Report 98-4148, 43 p.

Ryder, R.B., Cervione, M.A., Jr., Thomas, C.E., Jr., and Thomas, M.P., 1970, Water resources inventory of Connecticut, part 4, southwestern coastal river basins:

Connecticut Water Resources Bulletin 17, 54 p.

Solley, W.B., Pierce, R.R., Perlman, H.A., 1998, Estimated water use in the United States, 1995: U.S. Geological Survey Circular 1200, 71 p.

Stone, J.R., Schafer, J.P., London, E.H., and Thompson, W.B., 1992, Surficial materials map of Connecticut:

U.S. Geological Special Map, scale 1:125,000, 2 sheets.

Thomas, M.P., 1966, Effect of glacial geology upon the time distribution of streamflow in eastern and southern Connecticut, in Geological Survey Research 1966: U.S.

Geological Survey Professional Paper 550-B, p. B209-B212.

Tiedeman, C.R., Goode, D.J., and Hsieh, P.A., 1997, Numer-ical simulation of ground-water flow through glacial deposits and crystalline bedrock in the Mirror Lake area, Grafton County, New Hampshire: U.S. Geological Survey Professional Paper 1572 50 p.

U.S. Department of Commerce, 1991, 1990 census of popu-lation and housing, summary population and housing characteristics, Connecticut: Washington D.C., Bureau of the Census, publication 1990/CPH-1-8. 64 p.

U.S. Environmental Protection Agency, 2000, Ambient water quality criteria recommendationsInformation supporting the development of State and Tribal nutrient criteria; rivers and streams in nutrient ecoregion XIV:

U.S. Environmental Protection Agency, Office of Water EPA 822-B00-022, 84 p.

56 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 2001, Ambient water quality criteria recommenda-tionsInformation supporting the development of State and Tribal nutrient criteria; lakes and reservoirs in nutrient ecoregion XIV: U.S. Environmental Protection Agency, Office of Water EPA 822-B01-011, 123 p.

U.S. Geological Survey, 1995, National handbook of recom-mended methods for water-data acquisition. Chap. 11--

Water use: U.S. Geological Survey Office of Water Data Coordination, Available online at http://water.usgs.gov/pubs/chapter11/ accessed 

February 20, 2003.

Weiss, L.A., 1983, Evaluation and design of a streamflow-data network for Connecticut: Connecticut Water Resources Bulletin no. 36, 31 p.

Wilde, F.D., and Radtke, 1998, National field manual for the collection of water quality data--Field measurements:

U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A6, variously paginated.

Wilde, F.D., Radtke, D.B., Gibs, Jacob, and Iwatsubo, R.T.,

1998a, National field manual for the collection of water quality data--Preparation for water sampling: U.S.

Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A1, 47 p.

Wilde, F.D., Radtke, D.B., Gibs, Jacob, and Iwatsubo, R.T.,

1998b, National field manual for the collection of water quality data--Selection of equipment for water sampling: U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A2, 94 p.

Wilde, F.D., Radtke, D.B., Gibs, Jacob, and Iwatsubo, R.T.,

1998c, National field manual for the collection of water quality data--Cleaning of equipment for water sampling: U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A3, 75 p.

Wilde, F.D., Radtke, D.B., Gibs, Jacob, and Iwatsubo, R.T.,

1999a, National field manual for the collection of water quality data--Collection of water samples: U.S. Geolog-ical Survey Techniques of Water-Resources Investiga-tions, book 9, chap. A4, 156 p.

Wilde, F.D., Radtke, D.B., Gibs, Jacob, and Iwatsubo, R.T.,

1999b, National field manual for the collection of water quality data--Processing of water samples: U.S.

Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A5, 149 p.

Wolcott, S.W., and Snow, R.V., 1995, Computation of bedrock-aquifer recharge in northern Westchester County, New York, and chemical quality of water from selected bedrock wells: U.S. Geological Survey Water-Resources Investigations Report 92-4157, 57 p.

References Cited 57 Appendixes 1-4.

58 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Appendix 1. Records from two U.S. Geological Survey streamflow-gaging stations, Greenwich, Connecticut, March 2001-September 2002.

Station 01211699 E. Br. Byram R. Below Lake Mead At Round Hill STATION NUMBER 01211699 E. BR. BYRAM R. BELOW LAKE MEAD AT ROUND HILL SOURCE AGENCY USGS STATE 09 COUNTY 001 LATITUDE 410558 LONGITUDE 0734101.5 NAD27 DRAINAGE AREA 1.65* CONTRIBUTING DRAINAGE AREA 1.65 DATUM 330 NGVD29 DISCHARGE, in CFS, WATER YEAR OCTOBER 2000 TO SEPTEMBER 2001. DATA COLLECTION BEGAN FEBRUARY 2001.

DAILY MEAN VALUES DAY OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 1 --- --- --- --- --- 4.3 8.5 1.9 1.1 1.1 0.03 0.57 2 --- --- --- --- --- 4.1 7.2 1.9 11 1.5 0.02 0.31 3 --- --- --- --- --- 4.0 6.4 1.7 6.0 0.88 0.01 0.17 4 --- --- --- --- --- 4.0 5.7 1.7 3.2 0.79 0.07 0.11 5 --- --- --- --- --- 4.4 5.2 1.4 2.1 1.5 0.14 0.08 6 --- --- --- --- --- 4.5 5.2 1.2 1.6 2.4 0.16 0.04 7 --- --- --- --- --- 3.9 4.9 1.1 1.3 1.2 0.11 0.03 8 --- --- --- --- --- 3.7 4.7 1.1 1.2 1.4 0.06 0.02 9 --- --- --- --- --- 4.1 5.8 1.2 0.95 1.4 0.03 0.02 10 --- --- --- --- --- 4.1 12 1.1 0.88 0.84 0.04 0.40 11 --- --- --- --- --- 4.2 6.1 0.99 0.83 0.79 0.06 0.61 12 --- --- --- --- --- 4.4 5.9 0.95 0.94 0.62 0.17 0.38 13 --- --- --- --- --- 12 5.3 0.78 0.90 0.49 0.67 0.22 14 --- --- --- --- --- 10 4.4 0.61 0.83 0.40 0.61 2.4 15 --- --- --- --- --- 7.8 4.0 0.58 0.74 0.31 0.32 1.7 16 --- --- --- --- --- 6.9 3.8 0.54 0.66 0.27 0.17 0.67 17 --- --- --- --- --- 8.1 4.0 0.59 18 0.24 0.12 0.34 18 --- --- --- --- --- 7.0 3.6 0.70 8.4 0.33 0.12 0.21 19 --- --- --- --- --- 5.5 3.3 0.75 3.3 0.31 0.08 0.15 20 --- --- --- --- --- 4.7 3.1 0.62 2.2 0.25 2.4 0.52 21 --- --- --- --- --- 5.9 3.2 0.59 1.8 0.20 3.0 5.2 22 --- --- --- --- --- 15 3.9 3.6 1.7 0.14 0.86 2.2 23 --- --- --- --- --- 15 3.3 4.9 2.0 0.11 0.71 0.84 24 --- --- --- --- 3.3 8.2 3.0 3.7 2.7 0.10 1.1 0.49 25 --- --- --- --- 4.9 6.7 2.6 2.1 1.9 0.10 0.54 1.9 26 --- --- --- --- 8.2 5.7 2.3 2.1 1.3 0.15 0.25 1.5 27 --- --- --- --- 6.1 5.2 2.4 2.2 1.1 0.21 0.15 0.78 28 --- --- --- --- 5.0 4.7 2.2 1.8 0.88 0.13 0.13 0.53 29 --- --- --- --- --- 4.3 1.9 2.0 0.67 0.11 0.10 0.41 30 --- --- --- --- --- 20 1.9 2.2 0.60 0.09 0.06 0.36 31 --- --- --- --- --- 14 --- 1.5 --- 0.06 0.32 ---

2001 M

A M

J J

A S

O N

D 2002 J

F M

A M

J J

A S

0 5

10 15 20 25 30 35 40 45 MEAN DAILY DISCHARGE, IN CUBIC FEET PER SECOND 01211699

Appendixes 59 Appendix 1. Records from two U.S. Geological Survey streamflow-gaging stations, Greenwich, Connecticut, March 2001-September 2002.Continued Station 01211699 E. Br. Byram R. Below Lake Mead At Round HillContinued STATION NUMBER 01211699 E. BR. BYRAM R. BELOW LAKE MEAD AT ROUND HILL SOURCE AGENCY USGS STATE 09 COUNTY 001 LATITUDE 410558 LONGITUDE 0734101.5 NAD27 DRAINAGE AREA 1.65* CONTRIBUTING DRAINAGE AREA 1.65 DATUM 330 NGVD29 DISCHARGE, in CFS, WATER YEAR OCTOBER 2001 TO SEPTEMBER 2002 DAILY MEAN VALUES DAY OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 1 0.78 0.33 0.55 0.36 1.2 0.39 2.2 3.3 3.3 1.2 0.02 0.20 2 0.72 0.29 0.47 0.34 1.3 0.33 1.9 3.6 2.7 1.4 0.50 3.7 3 0.52 0.64 0.43 0.32 1.1 11 1.9 5.3 2.2 1.3 0.71 2.4 4 0.36 0.58 0.37 0.31 0.90 4.7 2.2 3.0 2.0 2.2 0.37 3.6 5 0.29 0.63 0.38 0.33 0.72 2.1 1.8 2.1 2.4 1.4 0.22 1.2 6 0.55 0.35 0.36 0.44 0.64 1.6 1.6 1.9 7.1 1.1 0.14 0.37 7 0.44 0.18 0.40 1.1 0.62 1.5 1.5 1.9 22 0.52 0.05 0.18 8 0.27 0.26 0.44 0.93 0.64 1.3 1.5 1.9 7.2 0.48 0.02 0.12 9 0.22 0.27 0.98 0.83 0.62 1.2 1.5 1.8 3.8 0.64 0.01 0.10 10 0.26 0.29 1.6 0.84 0.65 1.8 1.9 1.8 2.8 0.73 0.01 0.08 11 0.28 0.31 1.6 1.2 1.5 1.3 1.8 1.6 2.1 0.48 0.01 0.12 12 0.29 0.27 0.79 1.4 1.3 0.97 1.6 3.4 1.8 0.41 0.01 0.05 13 0.30 0.26 0.59 1.3 0.94 0.91 1.8 14 1.9 0.32 0.01 0.02 14 0.36 0.33 0.59 1.0 0.67 1.3 2.1 23 3.8 0.25 0.0 0.02 15 1.0 0.43 0.74 0.97 0.60 1.5 2.2 7.3 6.3 0.22 0.0 0.46 16 0.83 0.45 0.63 0.87 0.61 1.5 1.9 5.0 4.2 0.11 0.02 2.2 17 0.71 0.44 0.66 0.84 0.67 1.4 1.6 4.3 3.2 0.05 0.01 1.2 18 0.75 0.44 1.6 0.79 0.61 1.8 1.6 15 2.2 0.03 0.01 0.48 19 0.55 0.51 1.6 0.77 0.54 2.1 1.6 9.5 1.7 0.04 0.01 0.25 20 0.62 0.60 1.1 0.85 0.54 5.2 1.7 5.5 1.5 0.08 0.01 0.17 21 0.85 0.54 0.60 0.81 0.83 7.4 1.5 4.8 1.3 0.05 0.0 0.15 22 0.91 0.53 0.36 0.78 0.89 3.3 1.9 4.4 1.2 0.03 0.01 0.13 23 0.84 0.58 0.31 0.84 0.70 2.2 1.9 4.1 1.1 0.52 0.01 0.12 24 0.71 0.64 1.8 1.4 0.56 2.0 1.1 3.8 0.96 0.86 0.01 0.08 25 0.73 0.96 2.3 1.6 0.50 1.8 0.75 3.5 0.85 0.43 0.01 0.06 26 0.59 1.2 1.3 1.3 0.48 1.8 1.2 3.3 1.3 0.18 0.01 0.14 27 0.49 0.86 0.81 1.0 0.52 4.4 0.70 3.4 2.2 0.14 0.01 2.1 28 0.41 0.80 0.64 0.87 0.50 3.2 4.1 3.9 2.5 0.13 0.01 2.1 29 0.34 0.74 0.56 0.80 --- 2.2 10 3.7 1.7 0.11 0.58 0.97 30 0.33 0.67 0.48 0.81 --- 1.9 3.7 3.2 1.3 0.06 0.39 0.55 31 0.30 --- 0.41 0.87 --- 1.9 --- 3.0 --- 0.04 0.31 ---

60 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Appendix 1. Records from two U.S. Geological Survey streamflow-gaging stations, Greenwich, Connecticut, March 2001-September 2002.Continued Station 01211110 Unnamed Tributary To Greenwich Creek Nr Cos Cob STATION NUMBER 01211110 UNNAMED TRIBUTARY TO GREENWICH CREEK NR COS COB STREAM SOURCE AGENCY USGS STATE 09 COUNTY 001 LATITUDE 410233.5 LONGITUDE 0733659.3 NAD27 DRAINAGE AREA 2.19* CONTRIBUTING DRAINAGE AREA 2.19 DATUM 65 NGVD29 DISCHARGE, in CFS, WATER YEAR OCTOBER 2000 TO SEPTEMBER 2001. DATA COLLECTION BEGAN MARCH 2001.

DAILY MEAN VALUES DAY OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 1 --- --- --- --- --- 4.7 13 2.2 3.2 e1.3 0.10 0.75 2 --- --- --- --- --- 4.3 9.6 2.2 20 e1.7 0.07 0.56 3 --- --- --- --- --- 3.9 7.8 1.9 12 e1.1 0.04 0.36 4 --- --- --- --- --- 3.7 6.6 2.2 4.8 e1.0 e0.25 0.30 5 --- --- --- --- --- 4.3 5.9 1.5 2.8 e1.8 e0.25 0.33 6 --- --- --- --- --- 4.5 5.6 1.3 2.4 e2.6 e0.25 0.29 7 --- --- --- --- --- 4.0 5.3 1.2 2.2 1.0 0.31 0.26 8 --- --- --- --- --- 4.0 4.8 1.1 2.5 e1.4 e0.23 0.21 9 --- --- --- --- --- 4.8 10 0.97 1.6 e1.5 0.18 0.18 10 --- --- --- --- --- 5.1 17 0.99 1.3 e1.1 e0.17 0.42 11 --- --- --- --- --- 5.2 8.1 0.82 1.0 e1.1 e0.19 0.68 12 --- --- --- --- --- 5.3 8.4 0.76 1.2 e0.94 1.6 0.36 13 --- --- --- --- --- 26 7.0 0.85 1.2 e0.73 5.9 0.17 14 --- --- --- --- --- 15 5.9 0.79 1.2 e0.60 2.5 6.2 15 --- --- --- --- --- 9.5 5.3 0.53 1.2 e0.52 1.5 2.0 16 --- --- --- --- --- 7.5 5.0 0.50 1.2 e0.49 1.3 0.98 17 --- --- --- --- --- 8.2 4.8 0.59 23 e0.44 1.3 0.66 18 --- --- --- --- --- 7.1 4.4 0.55 10 e0.44 1.3 0.49 19 --- --- --- --- --- 5.1 3.8 0.60 5.2 e0.47 1.0 0.39 20 --- --- --- --- --- 4.1 3.7 0.58 e3.4 e0.41 25 1.3 21 --- --- --- --- --- 6.6 3.8 1.1 e2.7 e0.39 12 19 22 --- --- --- --- --- 28 5.2 6.1 e2.3 e0.39 2.8 5.0 23 --- --- --- --- --- 26 3.8 8.6 e2.6 e0.33 4.0 3.0 24 --- --- --- --- --- 11 3.3 5.7 e3.1 e0.36 4.8 2.0 25 --- --- --- --- --- 7.3 2.9 4.0 e2.2 e0.31 1.8 9.9 26 --- --- --- --- --- 6.2 2.8 5.1 1.5 e0.39 1.2 4.0 27 --- --- --- --- --- 5.2 2.7 4.8 e1.2 e0.39 1.8 2.5 28 --- --- --- --- --- 4.3 2.5 4.0 e1.0 0.38 1.6 2.0 29 --- --- --- --- --- 3.9 2.2 3.4 e0.84 0.18 1.1 1.6 30 --- --- --- --- --- 41 2.3 3.1 e0.81 0.18 0.77 1.5 31 --- --- --- --- --- 24 --- 3.0 --- 0.10 0.68 ---

e Estimated 2001 M

A M

J J

A S

O N

D 2002 J

F M

A M

J J

A S

0 5

10 15 20 25 30 35 40 45 MEAN DAILY DISCHARGE, IN CUBIC FEET PER SECOND 01211110

Appendixes 61 Appendix 1. Records from two U.S. Geological Survey streamflow-gaging stations, Greenwich, Connecticut, March 2001-September 2002.Continued Station 01211110 Unnamed Tributary To Greenwich Creek Nr Cos CobContinued STATION NUMBER 01211110 UNNAMED TRIBUTARY TO GREENWICH CREEK NR COS COB STREAM SOURCE AGENCY USGS STATE 09 COUNTY 001 LATITUDE 410233.5 LONGITUDE 0733659.3 NAD27 DRAINAGE AREA 2.19* CONTRIBUTING DRAINAGE AREA 2.19 DATUM 65 NGVD29 DISCHARGE, in CFS, WATER YEAR OCTOBER 2001 TO SEPTEMBER 2002 DAILY MEAN VALUES DAY OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 1 3.7 0.88 0.92 0.88 1.8 0.70 4.3 6.8 e4.0 e2.6 e0.92 0.93 2 2.1 1.1 0.68 0.82 1.8 0.82 3.1 8.5 e3.4 e2.1 e1.1 25 3 1.6 1.0 0.62 0.80 1.4 13 3.2 7.6 e2.9 e2.3 e1.5 7.6 4 1.2 1.3 0.66 0.85 1.4 4.9 3.4 4.8 e2.9 e2.8 e1.2 17 5 0.94 1.2 0.67 0.72 1.2 3.2 2.7 4.1 e3.5 e2.2 e0.70 5.5 6 0.93 1.1 0.72 1.7 1.1 2.6 2.4 3.4 e9.4 e1.7 e0.40 3.6 7 0.81 1.0 0.89 4.5 1.0 2.2 2.1 3.2 e24 e1.4 0.02 3.0 8 0.64 1.8 1.2 1.9 1.0 2.0 2.1 2.8 8.9 e1.1 0.02 2.2 9 0.63 1.6 2.9 1.6 0.91 1.9 2.3 2.4 4.9 e1.3 0.02 2.4 10 0.56 1.5 1.6 1.7 1.5 1.9 3.2 2.3 3.6 e1.3 0.02 2.1 11 0.55 1.3 1.0 2.2 4.1 1.5 2.2 1.9 2.9 e1.1 0.02 1.9 12 0.69 1.2 --- 2.0 2.1 1.5 2.1 5.0 2.7 e0.92 0.02 1.4 13 0.77 1.3 --- 1.8 1.8 1.7 2.3 17 2.3 e0.85 0.02 1.5 14 0.67 1.4 --- 1.5 1.4 1.7 3.1 24 5.6 e0.77 0.01 1.9 15 2.1 1.5 1.2 1.4 1.3 1.5 2.8 7.9 5.9 e0.77 0.01 3.2 16 1.1 1.5 0.87 1.3 1.4 1.7 2.1 5.1 6.2 e0.70 0.05 9.5 17 0.95 1.5 0.95 1.3 1.4 1.2 1.7 4.3 5.1 e0.63 0.27 4.5 18 0.75 1.2 3.9 1.2 1.3 2.5 1.2 23 3.5 e0.63 0.06 2.9 19 0.57 1.3 2.1 1.1 1.0 2.7 1.5 11 2.8 e0.63 0.04 2.3 20 0.59 1.3 1.5 1.3 1.1 8.1 1.4 e7.0 2.4 e0.55 0.27 2.7 21 0.56 1.3 1.2 1.3 1.6 7.8 1.2 e5.6 2.2 e0.48 0.08 1.8 22 0.71 1.2 1.0 1.3 1.2 4.2 2.2 e5.0 1.8 e0.55 0.10 1.3 23 0.78 1.0 0.94 1.4 1.0 3.2 1.9 e4.6 1.8 e1.4 0.27 1.1 24 0.78 1.4 5.6 2.9 1.1 2.9 1.6 e4.3 1.7 e1.6 0.43 0.84 25 1.1 2.3 2.6 2.6 0.94 2.5 3.3 e4.0 1.7 e1.4 0.31 0.84 26 0.98 3.4 1.8 1.9 0.95 3.0 3.1 e3.8 2.7 e0.77 0.11 0.95 27 0.91 1.3 1.6 1.8 0.93 9.7 1.9 e4.0 5.0 e0.77 0.04 6.2 28 0.82 0.98 1.3 1.6 0.78 4.8 18 e4.4 4.6 e0.77 0.04 3.4 29 0.53 0.97 1.2 1.5 --- 3.8 20 e4.3 e3.8 e0.85 5.3 1.9 30 0.60 1.0 1.1 1.4 --- 3.4 8.1 e3.8 e3.0 e0.85 1.7 1.3 31 0.72 --- 0.96 1.2 --- 3.5 --- e3.8 --- e0.85 0.87 ---

e Estimated

62 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Appendix 2. Estimated average daily ground-water use for the Greenwich area, Connecticut.

[gal/d, gallon per day]

Zone Average daily water use Greenwich and Stamford Conn.

(gal/d)

Lower boundary of the 90-percent confidence interval (gal/d)

Upper boundary of the 90-percent confidence interval (gal/d)

Average daily water use in Westchester County, N.Y.

(gal/d)

Estimated total (gal/d) 1 41,000 33,000 50,000 6,500 47,000 2

67,000 50,000 88,000 0

67,000 3

51,000 37,000 67,000 0

51,000 4

110,000 95,000 130,000 0

110,000 5

5,000 2,800 8,000 0

5,000 6

40,000 30,000 52,000 49,000 88,000 7

78,000 58,000 100,000 0

78,000 8

110,000 93,000 140,000 0

110,000 9

19,000 12,000 26,000 9,500 28,000 10 140,000 110,000 170,000 0

140,000 11 35,000 26,000 45,000 0

35,000 12 120,000 97,000 150,000 14,000 140,000 13 99,000 78,000 120,000 17,000 120,000 14 76,000 63,000 91,000 0

76,000 15 9,900 6,000 15,000 0

9,900 16 100,000 86,000 130,000 0

100,000 17 37,000 27,000 49,000 0

37,000 18 1,700 600 3,600 0

1,700 19 240,000 210,000 280,000 0

240,000 20 96,000 84,000 110,000 0

96,000 21 16,000 11,000 23,000 0

16,000 22 64,000 53,000 77,000 0

64,000 23 0

0 0

0 0

24 41,000 32,000 51,000 19,000 60,000 25 65,000 28,000 120,000 0

65,000 26 67,000 56,000 79,000 0

67,000 27 59,000 49,000 69,000 0

59,000 28 120,000 100,000 130,000 0

120,000 29 25,000 19,000 32,000 0

25,000 30 48,000 41,000 56,000 0

48,000 31 23,000 18,000 30,000 0

23,000 32 2,900 1,000 6,100 0

2,900

Appendixes 63 Appendix 3. Estimated average daily summer ground-water use for the Greenwich area, Connecticut.

[gal/d, gallon per day]

Zone Average daily summer water use Greenwich and Stamford Conn., in gal/d Lower boundary of the 90-percent confidence interval (gal/d)

Upper boundary of the 90-percent confidence interval (gal/d)

Average daily summer water use in Westches-ter County, N.Y.

(gal/d)

Estimated total (gal/d) 1 53,000 41,000 67,000 7,700 61,000 2

95,000 65,000 130,000 0

95,000 3

71,000 48,000 100,000 0

71,000 4

150,000 120,000 180,000 0

150,000 5

6,400 3,300 11,000 0

6,400 6

55,000 39,000 75,000 59,000 110,000 7

110,000 78,000 160,000 0

110,000 8

160,000 120,000 200,000 0

160,000 9

25,000 16,000 38,000 11,000 36,000 10 190,000 150,000 250,000 0

190,000 11 46,000 33,000 63,000 0

46,000 12 170,000 130,000 230,000 18,000 190,000 13 140,000 110,000 190,000 22,000 170,000 14 100,000 81,000 130,000 0

100,000 15 13,000 7,400 21,000 0

13,000 16 150,000 120,000 180,000 0

150,000 17 51,000 35,000 70,000 0

51,000 18 2,200 680 4,900 0

2,200 19 330,000 270,000 410,000 0

330,000 20 120,000 110,000 150,000 0

120,000 21 21,000 13,000 32,000 0

21,000 22 85,000 67,000 100,000 0

85,000 23 0

0 0

0 0

24 52,000 37,000 70,000 23,000 75,000 25 110,000 35,000 240,000 0

110,000 26 87,000 72,000 100,000 0

87,000 27 75,000 61,000 91,000 0

75,000 28 150,000 130,000 170,000 0

150,000 29 32,000 24,000 41,000 0

32,000 30 60,000 50,000 72,000 0

60,000 31 30,000 22,000 39,000 0

30,000 32 4,000 1,200 8,900 0

4,000

64 Water Use, Ground-Water Recharge and Availability, and Quality of Water in the Greenwich Area, 2000-2002 Appendix 4. Estimated average daily winter ground-water use for the Greenwich area, Connecticut.

[gal/d, gallon per day]

Zone Average daily winter water use Greenwich and Stamford Conn.

(gal/d)

Lower boundary of the 90-percent confidence interval (gal/d)

Upper boundary of the 90-percent confidence interval (gal/d)

Average winter season water use in Westchester County, N.Y.

(gal/d)

Estimated total (gal/d) 1 31,000 25,000 37,000 4,900 36,000 2

48,000 38,000 59,000 0

48,000 3

35,000 27,000 44,000 0

35,000 4

82,000 71,000 94,000 0

82,000 5

4,200 2,500 6,400 0

4,200 6

27,000 21,000 35,000 37,000 64,000 7

51,000 41,000 63,000 0

51,000 8

80,000 67,000 94,000 0

80,000 9

13,000 9,400 18,000 7,300 20,000 10 89,000 76,000 100,000 0

89,000 11 24,000 19,000 30,000 0

24,000 12 80,000 67,000 94,000 10,000 90,000 13 67,000 54,000 81,000 13,000 80,000 14 54,000 46,000 63,000 0

54,000 15 6,900 4,500 10,000 0

6,900 16 73,000 62,000 85,000 0

73,000 17 26,000 20,000 33,000 0

26,000 18 1,200 470 2,300 0

1,200 19 170,000 150,000 190,000 0

170,000 20 72,000 64,000 81,000 0

72,000 21 11,000 7,800 15,000 0

11,000 22 45,000 38,000 53,000 0

45,000 23 0

0 0

0 0

24 29,000 24,000 36,000 15,000 44,000 25 40,000 22,000 65,000 0

40,000 26 54,000 46,000 63,000 0

54,000 27 44,000 38,000 51,000 0

44,000 28 90,000 82,000 100,000 0

90,000 29 19,000 15,000 23,000 0

19,000 30 38,000 32,000 43,000 0

38,000 31 17,000 14,000 21,000 0

17,000 32 2,000 800 4,100 0

2,000