Loading...
HomeMy WebLinkAboutNF Water Supply Plan 1983North Fork Water Supply Plan Suffolk County, New York Peter F. Cohalan David Harris, M.D., M.P.H. Herbert W. Davids Suffolk County Executive Commissioner Director, Division of Environmental Health I I ~ ERM-Northeast CAMP DRESSER & McKEE I 88 Sunnyside Blvd./Plainview, New York 11803 (516) 349-0050 CDM CAMP DRESSER & McKEE One WoHd TCade Cenle[ Suile 2637 New York, New York 10048 212 432-6700 22 April 1983 David Harris, M.D., M.P.H., Commissioner Department of Heal th Services County of Suffolk Hauppauge, New York 11788 Dear Dr. Harris: We are indeed pleased to submit herein our report "North Fork Water Supply Plan, Suffolk County, New York". This report concludes a most challenging assignment - to develop a plan to provide safe, dependable water supplies to the residents of the Towns of Riverhead and Southold in the North Fork. The plan was developed in close cooperation with members of your staff; their assistance is greatly appreciated. The innovative concepts developed in this plan are a reflection of your staff's high level of cap- ability and dedication to the public good. We wish to thank you for this opportunity and look forward to working with you and your staff in the future. Very truly yours, for ERM-Northeast /Sen/for V/f}ce Presi dent ~afnp Drc~ser & McKee (SPH12/3) NORTH FORK WATER SUPPLY PLAN SUFFOLK COUNTY, NEW YORK March 1983 Prepared For: Suffolk County Department of Health Services Hauppauge, New York Peter F. Cohalan, Suffolk County Executive David Harris, M.D., M.P.H., Commissioner Herbert W. Davids, Director, Division of Environmental Health Prepared By: ERM-Northeast Engineers, P.C. 88 Sunnyside Blvd. Plainview, New York Camp Dresser & McKee One World Trade Center New York, New York CONTENTS 1.0 EXECUTIVE SU~IARY AND REC(]~94ENDATIONS ..................... 1-1 1.1 Executive Summary .................................... 1-1 Area Characteristics ................................. 1-1 Groundwater Quantity ................................. 1-2 Groundwater Quality .................................. 1-4 Water Supply Alternatives ............................ 1-5 Implementation ....................................... 1-6 1.2 Recommendations ...................................... 1-7 General Recommendations .............................. 1-7 Area-Specific Recommendations ........................ 1-9 Preventive Measures .................................. 1-11 2.0 INTRODUCTION .............................................. 2-i 2.1 Scope ................................................ 2-1 2.2 Study Area ........................................... 2-2 2.3 Planning Approach--Water Supply Zones ................ 2-2 3.0 POPULATION AND LAND USE ................................... 3-1 3.1 Population ........................................... 3-1 3.2 Land Use ............................................. 3-5 4.0 GEOLOGY ................................................... 4.1 Geologic Units ....................................... 4.1.1 4.1.2 4.1.3 4.1.4 Bedrock ....................................... Raritan Formation ............................. Matawan Group--Magothy Formation Undifferentiated .................... Pleistocene Deposits .......................... 4.2 Aquifer Parameters ................................... 4.3 Fresh Water-Saltwater Interface ...................... 4-1 4-1 4-1 4-1 4-1 4-2 4-4 4-5 5.0 GROUNDWATER QUALITY[......... ........................... .. 5-1 5.1 Inorganic Parameters ................................. 5-1 5.2 Organic Parameters ................................... 5-6 6.0 EXISTING WATER SUPPLY SYSTEMS ............................. 6-1 6.1 Greenport Municipal System ........................... 6-1 6.2 Riverhead Municipal System ........................... 6-4 6.3' Private Water Systems ................................ 6-6 7.0 WATER USE AND AVAILABLE SUPPLY ............................ 7-1 7.1 Existing and Future Water use ........................ 7-1 7.1.1 7.1.2 7.1.3 Residential Use ............................... 7-1 Agricultural Use .............................. 7-3 Total Water Use ............................... 7-3 7.2 Available Groundwater Supply ......................... 7-6 7.3 Water Balances ....................................... 7-10 7.4 Well Yields and Spacing Considerations ............... 7-10 7.4.1 Well Yields ................................... 7-10 7.4.2 Well Spacing - Budget Area .................... 7-13 8.0 WATER SUPPLY ALTERNATIVES ................................. 8-1 8.1 Water Demand Centers ................................. 8-1 8.2 Water Supply Concepts ................................ 8-4 8.2.1 Level I Alternatives .......................... 8-5 8.2.1.1 8.2.1.2 8.2.1.3 8.2.1.4 8.2.1.5 8.2.1.6 Home Treatment Units ................. 8-5 Bottl ed Water ........................ 8-6 Local Delivery of Bulk Water ......... 8-6 Local Supply of Bulk Water (With Treatment) ........................... 8-7 Individual Home Wells ................ 8-7 Level I Cost Estimates ............... 8-7 8.2.2 8.2.3 8.2.4 8.2.5 Level II--Neighborhood Systems ................ 8-13 Level III--Subdemand Center Systems ........... 8-14 Level IV--Subregional Systems ................. 8-16 Level V--Regional Systems ..................... 8-16 8.3 Water Treatment Processes ............................ 8-18 8.3.1 Ion Exchange/Carbon Adsorption ................ 8-20 8.3.2 Reverse Osmosis ............................... 8-21 8.4 Alternatives for Demand Centers ...................... 8-22 8.4.1 General Development Criteria .................. 8-22 8.4.1.1 8.4.1.2 8.4.1.3 8.4.1.4 Supply ............................... 8-22 Transmission/Di stribution ............ 8-24 Treatment ............................ 8-25 Annual Operation and Maintenance ..... 8-25 8.4.2 8.4.3 Development of Cost Estimates ................. 8-25 Wading River/Northvill e ....................... 8-27 8.4.3.1 8.4.3.2 8.4.3.3 Level I Alternatives ................. 8-27 Level II Alternatives ................ 8-31 Level III Alternatives ............... 8-31 8.4.4 Riverhead/Jamesport ........................... 8-36 8.4.4.1 8.4.4.2 8.4.4.3 Level I Alternatives ................. 8-36 Level II Alternatives ................ 8-36 Level III Alternatives ............... 8-40 8.4.5 Mattituck/Cutchogue ........................... 8-40 8.4.5.1 8.4.5.2 8.4,5,3 8.4.5.4 8.4.5.5 Level I Alternatives ................. 8-43 Level II Alternatives ................ 8-43 Level III Alternatives ............... 8-43 Level IV Alternatives ................ 8-47 Level V Alternatives ................. 8-47 8.4.6 Southol d/Greenport ............................ 8-52 8.4.6.1 8.4.6.2 8.4.6.3 8.4.6.4 Level I Alternatives ................. 8-52 Level III Alternatives ............... 8-52 Level IV Alternatives ................ 8-62 Level V Alternatives ................. 8-62 8.4.7 Orient ........................................ 8-66 8.4.7.1 8.4.7.2 8.4.7.3 Level I Alternatives ................. 8-66 Level III Alternatives ............... 8-66 Level IV and V Alternatives .......... 8-69 8.4.8 Isolated Neighborhood Systems ................. 8-70 8.5 Comparison of Alternatives ........................... 8-70 8.5,1 8.5.2 8.5.3 8.5.4 8.5.5 8.5.6 Cost .......................................... 8-70 Reliability ................................... 8-73 Im~ementabil ity .............................. 8-73 Environmental Considerations .................. 8-74 Adaptability to Future Changes ................ 8-74 Summary Comparative Matrices .................. 8-75 9.0 RECOMMENDED WATER SUPPLY PLAN ............................. 9-1 9.1 Wading River/Northville .............................. 9-1 9.2 Riverhead/Jamesport .................................. 9-3 9.3 Mattituck/Cutchogue .................................. 9-3 9.4 Southold/Greenport ................................... 9-4 9.5 Orient ............................................... 9-4 9.6 Isolated Areas ....................................... 9-5 9.7 Groundwater Management ............................... 9-5 10.0 IMPLEMENTATION PLAN ....................................... 10-1 10.1 Implementation Agencies .............................. 10-1 10.2 Potential Funding Sources ............................ 10-6 10.3 Legislative Requirements ............................. 10-6 Table 1-1 Table 3-1 Table 3-2 Table 3-3 Table 3-4 Table 4-1 Table 4-2 Table 5-1 Table 5-2 Table 6-1 table 6-2 Table 6-3 Table 7-1 Tab1 e 7-2 Tab1 e 7-3 Table 7-4 Table 8-1 Table 8-2 LIST OF TABLES Water Budgets and Consumptive Use Projections ........ 1-3 Population by School District ........................ 3-2 Population by Groundwater Supply Zones ............... 3-3 Peak S~nmer Weekend Population ....................... 3-4 Existing Land Use .................................... 3-6 Upper Glacial Aquifer Parameters ..................... 4-6 Magothy Aquifer Parameters ........................... 4-7 Percentage of Well Samples Exceeding Standards or Guidelines for Inorganic Parameters .................. 5-2 Percentage of Well Sam~es Exceeding Guidelines for Organic Parameters ................................... 5-7 Summary of Well Supplies for the Greenport Municipal System ..................................... 6-3 S~nmary of Well Supplies for the Riverhead Water District ....................................... 6-5 North Fork Private Water Supplies .................... 6-7 Existing Water Usage--Greenport and Riverhead Water Districts ...................................... 7-2 Average Annual Water Usage ........................... 7-4 Irrigation Requirements for Crops Groom on the North Fork ....................................... 7-5 Smnmary of Water Budget Analysis ............... ~ ..... 7-9 Estimated Year 2000 Population and Water Consumption .......................................... 8-3 Water Supply Concepts ................................ 8-5 Table 8-3 Table 8-4 Table 8-5 Table 8-6 Table 8-7 Table 8-8 Table 8-9 Table 8-10 Table 8-11 Table 8-12 Table 8-13 Table 8-14 Table 8-15 Table 8-16 Table 8-17 Table 8-18 Table 8-19 Cost Estimate--Individual Home Well Systems (No Treatment) ....................................... 8-9 Cost Estimate--Individual Home Well Systems (With Treatment) .................................... 8-11 Level V Alternatives ................................ 8-19 Components of Alternatives Levels II Through V ...... 8-23 Example of Cost Determination for Alternatives ...... 8-26 Estimated Annual Dwelling Unit Costs of Water Supply Alternatives for the Wading River/ Northville Demand Center ............................ 8-30 Existing Public Water Systems in the Wading River/Northville Demand Center ...................... 8-32 Capital Cost Estimates--Wading River/Northville Demand Center ....................................... 8-35 Estimated Annual Dwelling Unit Costs of Water Supply ~ternatives for the Riverhead/Jamesport Demand Center ....................................... 8-37 Capital Cost Estimates--Riverhead/Jamesport Demand Center ....................................... 8-41 Estimated Annual Dwelling Unit Costs of Water Supply Alternatives for the Mattituck/Cutchogue Demand Center ....................................... 8-44 Level III Alternative Components for the Mattituck/Cutchogue Demand Center ................... 8-46 S~nmary of Level V Alternative Components for the Mattituck/Cutchogue Demand Center ................... 8-50 Capital Cost Estimate--Mattituck/Cutchogue Demand Center ....................................... 8-53 Estimated Annual Dwelling Unit Costs of Water Supply Alternatives for the Southold/Greenport Demand Center ....................................... 8-54 Level III Alternative Components for the Great Hog Neck and East Marion Subdemand Centers .......... 8-65 Cost Estimates--Greenport/Southold Demand Center .............................................. 8-67 Table 8-20 Table 8-21 Table 8-22 Table 8-23 Table 8-24 Table 8-25 Table 8-26 Table 8-27 Table 8-28 Table 8-29 Table 9-1 Estimated Annual Dwelling Unit Cost of Water Supply Alternatives for the Orient Demand Center .... 8-66 Capital Cost Estimates--Orient Demand Center ........ 8-69 Estimated Costs of Water Supply ~ternatives for Isolated Neighborhood Systems ................... 8-71 Capital Cost Estimates--Neighborhood Systems ........ 8-72 Comparison of Alternatives--Wading River/Northville Demand Center ...................... 8-76 Comparison of Alternative$--Riverhead/jamesport Demand Center ....................................... 8-77 Comparison of Alternatives--Mattituck/Cutchogue Demand Center ....................................... 8-78 Comparison of Alternatives--Greenport/Southold Demand Center ....................................... 8-79 Comparison of Alternatives--Orient Demand Center .... 8-80 Comparison of ~ternatives--Isolated Neighborhood Areas .................................. 8-81 Summary of Capital Costs for the Recommended Plan... 9-2 I I: I I I I I 2-1 4-1 5-1 5-2 6-1 7-1 8-1 8-2 8-3 8-4 8-5 8-6 8-7 8-8 8-9 LIST OF FIGURES 8-10 8-11 8-12 Study Area--North Fork Water Supply Plan ................. 2-3 Geologic Cross Section 4-3 Nitrate Contamination Areas 5-4 Aldicarb Contamination Areas 5-8 Existing Water Supply Systems ............................ Water Budget Area ........................................ 6-2 7-8 Major Demand Centers ..................................... 8-2 Level III--Subdemand Center Systems ...................... 8-15 Level IV--Schematic of Subregional Systems ............... 8-17 Average Construction Cost Breakdown ...................... 8-28 Piping Cost Per Dwelling Unit at Various Housing Densities ................................................ 8-29 Level III Alternative Components for Wading River/Northville ......................................... 8-33 Levels III and IV Alternative Components for Calverton ................................................ 8-38 Level III Alternative Components for Jamesport ........... 8-39 Level III Alternative Components for Mattituck/Cutchogue ...................................... 8-45 Level V(B) Alternative Components for Mattituck/Cutchogue and Southold/Greenport ............... 8-49 Level III Alternative Components for Great Hog Neck ...... 8-55 Level III Alternative Components for East Marion ......... 8-56 8-13 Level III Alternative Components for the Greenport Municipal System ............................... 8-58 8-14 Schematic of Level III(B) for Greenport .................. 8-61 8-15 Level III Alternative Components for Orient .............. 8-68 ACKNOWLEDGMENTS We would like to express our appreciation to the numerous individuals who assisted us in the preparation of this report. Specifically, we thank the members of the Steering Committee who worked with and guided us throughout the project. North Fork Water Supply Plan Steering Committee Mr. Robert Schneck New York State Department of Environmental Conservation Mr. Richard Hanley Community Development Agency Town of Riverhead Mr. William J. Schickler and Mr. August A. Guerrera Suffolk County Water Authority The Honorable Gregory J. Blass Suffolk County Legislator, District No. i Mr. Henry Raynor, Chairman Town of Southold Planning Board Mr. Francis J. Murphy Deputy Supervisor Town of Southold Mr. Franklin Bear North Fork Environmental Council Mr. David Newton Cooperative Extention Association of Suffolk County Dr. Edith Tannenbaum Long Island Regional Planning Board Ms. Ruth Oliva North Fork Environmental Council Mr. Robert A. Villa, P.E. Suffolk County Department of Health Services Dr. Aldo Andreoli, P.E. Suffolk County Department of Health Services Mr. Joseph H. Baier, P.E. Suffolk County Department of Health Services Mr. Dennis Moran, P.E. Suffolk County Department of Health Services THIS PUBLICATION IS FULLY OR PARTIALLY FUNDED BY SUFFOLK COUNTY- PETER F. COHALAN, COUNTY EXECUTIVE. SECTION 1.0 EXECUTIVE SUMMARY AND RECOMMENDATIONS 1.1 EXECUTIVE SUMMARY Water supply problems on the North Fork of Long Island are severe. Shallow, thin groundwater aquifers are extensively impacted by con- tamination from agricultural chemicals, primarily nitrates and pesticides, and are threatened by saltwater intrusion from over- pumping. Increasing pressures for development which will result in accelerated population growth will further stress the water supply aquifers. In response to the threats to groundwater quality and the increased denands for potable drinking water, the Suffolk County Department of Health Services (SCDHS) sponsored a study of water supply options for the North Fork. The objectives of this North Fork Water Supply Plan are to develop and evaluate several alternative water supply plans which could safely meet present and future potable water requirements and recommend a future course of action. The planning area includes the towns of Riverhead and Southold. Since hydrogeologic conditions are significantly different throughout the area, five water supply zones were established for planning pur- poses. Zone 1 starts at the western Town boundary of Riverhead and extends easterly to the hamlet of Riverhead. In this zone, the Magothy aquifer is saturated with fresh water all the way to bedrock. The presence of saltwater increases in the lower portions of the Magothy as one progresses east from the hamlet, until the fresh water 1 ens terminates at Mattituck Inlet, which is the eastern boundary of Zone 2. Zones 3, 4 and 5 are defined by three isolated fresh ground- water lenses. Zone 3 extends from the Mattituck Inlet, to Hasha- momuck Pond; Zone 4 extends from Hashamomuck Pond to Dam Pond; Zone 5 starts at Dam Pond and continues through Orient Point. Area Characteristics The Long Island Regional Planning Board estimates that permanent population in the study area will increase from approximat~y 39,000 today to 49,800 in the year 2000. On a peak summer weekend, an addi- tional 32,000 persons are anticipated. 1-1 Agricultural land use is the most prevalent in the study area today, accounting for approximately 42 percent of the land area (29,560 acres); current residential land use accounts for 12 percent of the area. Future land use projections by the Regional Planning Board indicate that low-density residential development will significantly increase in both Towns, while agricultural usage will decrease. The land use trends projected by the Planning Board forecast a decline in agricultural land of approximately 18 percent (5,300 acres). Fur- thermore, the Cooperative Extension Service anticipates a gradual change in the types of crops grown on the North Fork. In general, potatoes, sod and rye acreage will decrease while grapes, mixed vegetables, horse farms and nursery stock will increase in acreage. Groundwater Quantity The major geologic units in the two towns are bedrock, the Raritan Formation, the Matawan Group (Magothy Formation aquifer) and Pleistocene deposits (glacial deposits including the upper glacial aquifer). The principal water supply source in the study area is the upper glacial aquifer. It is a highly productive water-bearing unit with consistent physical properties. In the eastern part of water supply Zone 2 and in zones 3, 4, and 5, a very delicate balance exists between fresh groundwater lenses, saltwater intrusion, chemical contamination and water supply require- ments. The fresh groundwater lenses are relatively thin; therefore, the total available supply is limited. Over-pumping or improper well location causes lateral and vertical saltwater intrusion. The aqui- fers are extremely susceptible to chemical contamination because they are shallow and have limited dilution or assimilative capacity. Groundwater flow velocities are very low (on the order of one foot per day) so it takes long periods of time for contaminants to flush out of the aquifer lup to 100 years). These conditions mandate that groundwater resources be properly managed and protected in order to be able to support necessary water supply requirements. Consumptive water use patterns will change in the study area over the next 20 years. Domestic requirements lincluding commercial and in- dustrial) will increase by 25 percent from an annual average of ap- proximately 4 million gallons per day Imgd) today to over 5 mgd in the year 2000. Agricultural usage, however, will decline by 18 per- cent from 11 mgd to 9 m§d. During summer months, domestic usage is expected to be over 6 mgd; agricultural consumption will be approxi- mately 22 mgd. The trend to residential development from agriculture will result in a slight overall decline in average daily water use over the planning period, but the de~and for uncontaminated, potable water will increase. 1-2 I I I I I I I I I I I I I m In order to estimate the total quantity of groundwater that may be withdrawn from larger capacity public supply wells from each water supply zone, water budget areas were delineated. Substantial amounts of groundwater are available outside of the budget areas but, to avoid saltwater intrusion, can only be withdrawn by small, domestic capacity well s. In zones 1 and 2, the budget areas were defined as those locations where the groundwater level is 5 feet or more above sea level. In zones 3, 4 and 5, the availability of groundwater is more limited, so the budget area boundary was defined as the 2-foot groundwater con- tour. A total of approximately 41.2 mgd of fresh groundwater is available from the budget areas. An additional 10 to 20 mgd is available for domestic wells outside the budget areas. The results of the water budget analysis, by zone, are shown below in Table 1-1, which also includes consumptive use projections for the year 2000. Conclusion. Sufficient fresh groundwater is available to satisfy the needs of the overall planning area. However, critical water supply conditions exist in Zone 5 (Orient) where projected requirements are approximately equal to available supply. Groundwater supply condi- tions in Zone 4 (Greenport/Southold) are also critical although there is some extra available supply (0.9 mgd available versus 0.63 mgd re- quiredl · TABLE 1-1 WATER BUDGETS AND CONSUMPTIVE USE PROJECTIONS Permissive Domestic Agricultural Water Sustained Yield, Consumptive Use, Consumptive Use, Supply Budget Area Year 2000 Year 2000 Zone (mgd) (mgd) (mgd) I 29.4 2.25 3.06 2 5.6 0.97 3.06 3 4.9 1.18 2.80 4 0.9 0.59 0.04 5 0.4 0.11 0.35 TOTALS 41.2 5.10 9.31 1-3 Groundwater Quality The upper glacial aquifer throughout the planning area is already contaminated with nitrates and organic pesticides and herbicides. Nitrate levels are significantly elevated above natural background conditions of 0.1 to 1.0 milligrams per liter (mg/1) and exceed the drinking water standard of 10 mg/1 in many areas. Sampling of 639 wells for nitrates performed by SCDHS in the Town of Riverhead showed the following: 58 percent of the samples ranged from 0 to 5 mg/1; 26 percent were above 7.5 mg/1; and 16 percent were above 10 mg/1. Similar results were obtained in a sampling of 1,121 wells in the Town of Southold; 51 percent of samples ranged from 0 to 5 mg/1; 30 percent were above 7.5 mg/1; and 17 percent were above 10 m9/1. Organic contamination from pesticides and herbicides is also wide- spread. A comprehensive survey to define the extent of contamination from one pesticide--Aldicarb--was conducted by the SCDHS in 1979 and 1980. In the Town of Riverhead, 2,161 wells were sampled; 32 percent were contaminated by Aldicarb; 16 percent had Aldicarb concentrations above the health guideline of 7 parts per billion (ppb). In the Town of Southold, 3,160 wells were sampled; 23 percent showed Aldicarb contamination; 11 percent exceeded the health guideline of 7 ppb. Numerous other agricultural chemicals have also been found including carbofuran, dacthal, dichloropropane, oxamyl and dinoseb. Nitrate contamination is vertically extensive throughout the upper glacial aquifer. Aldicarb and other organics are presently limited to the upper 40 feet of groundwater; however, they are expected to distribute throughout the aquifer over time. Since velocities of flow are on the order of I foot per day in the North Fork aquifers, contaminants will be present for many decades before they are flushed out. The amounts of uncontaminated groundwater available for consumption, although not quantifiable, are substantially less than those shown in Table 1. When groundwater flow is considered, almost all of zones 3, 4 and 5 become suspect as well as the upper 50 feet of groundwater in zones i and 2. Conclusion. Groundwater contamination is currently extensive and will remain so for many years. As additional groundwater quality data is collected, more contamination problems will be discovered. Water supply impl eme~tation for the North Fork must proceed im- mediately. Alternative solutions must consider the fact that the limited volume of fresh groundwater is further limited because large portions have been contaminated by agricultural chemicals. 1-4 Water Supply Alternatives There are several levels of development on the North Fork, all of which had to be considered in the planning process. Five areas of population concentrations were defined as major water supply demand centers: (1) Wading River/Northville, (2) Riverhead/Jamesport, (3) Mattituck/Cutchogue, (4) Southold/Greenport, and (5) Orient. Dis- tinct individual communities (e.g., Calverton, Little Hog Neck, East Marion) comprise the demand centers. The remaining areas were con- sidered low-density residential and rural and were considered separately. The development of alternative water supply plans to serve this broad spectrum of needs proceeded in a building block fashion, beginning with an evaluation of individual home wells through community systems to large, regional systems. Five levels of water supply alternatives were developed and evaluated: Level I: Individual Home Systems--In areas where groundwater is degraded, these would include treatment of the home supply. Level II: Neighborhood Systems--These are small municipal sys- tems serving two to 50 homes. Level III: Community Systems--These are systems serving indi- vidual communities with local groundwater. Level IV: Subregional Systems--These are larger systems serving an entire demand center with local groundwater. Level V: Regional Systems--Wherein the supply is uncontamin- ated groundwater from Riverhead piped via a major transmission main to the eastern portions of the study area. In addition to the above listing, dual water supply systems, bottled water, clean water vending machines, truck-delivered water and cen- tral, community water supply taps were also evaluated as part of this study. Preliminary engineering studies and designs were developed for each alternative. Capital cost estimates were prepared and annual amorti- zation and operation/maintenance/administration costs were also esti- mated. The cost estimates ranged from $155 per year per home for in- dividual home wells in areas with potable groundwater to over $2,000 per year per home where extensive treatment is necessary or where clean groundwater has to be imported. Because of low density devel- opment, the relative costs of distribution systems are extremely high and preclude the development of municipal systems in some areas. 1-5 Commerical and industrial water usage was not included in the development of per home costs. Although this represents a small fraction of the total, the inclusion of the commerical and industrial base would reduce the cost per home. In addition to comparing the costs of alternatives, several other factors were considered in the screening and evaluation process: (1) Reliability of operation (2) Environmental consequences (3) Implementation problems (4) Flexibility to respond to future problems. Conclusion. The study found that a combination of residential public water systems and individual home treatment units is essential. Some form of direct governmental involvement is needed for implementation. Implementation Under the provisions of New York State law, there are four institu- tional entities that could implement all or portions of the s~ected alternatives: villages, towns, counties and the Suffolk County Water Authority. Implementation at the town level will give local government the abi- lity to provide public water systems in response to the needs and desires of the people. Expenditures and cash flow would remain in the control of the towns and they would retain a great deal of flexi- bility in the types of public water supply facilities to be provided for individual areas. Major legislation at the County or State level is not required if the towns implement the study's recommendations. It will be necessary, however, to have the State Comptroller or legislature define home treatment units as water supply facilities. If the towns do not choose to implement, the County would have to consider creating water districts for implementation; the County could operate the districts or contract to the Suffolk County Water Authority. County implement- ation would require legislative action at the County level. This investigation also included an analysis of funding sources to assist in implementing the recommendations. No such sources were found which are currently funded. Thus, the improvements will have to be paid for at the local level, primarily through user charges. 1-6 Conclusion. The towns of Riverhead and Southold can best implement the study's recommendations: {1) by forming town Water Management Programs, as administrative functions, to coordinate all water supply activities in the respective towns, {2) by forming Water Supply Dist ricts or extending existing district boundaries to purchase, own and operate private water companies or to construct new public water sys- tens, and {3) by forming Home Treatment Unit Districts {covering all or portions of the town area not served by public water) and owning and maintaining the home units. (The towns may allow private industry to sell home treatment units after approval of the units by the town, but the town should retain maintenance responsibility.) 1.2 RECOMMENDATIONS General Recommendations (1) This study has denonstrated conclusively that numerous technical and financial difficulties are encountered when att6mpting to provide potable water to the residents of the North Fork. Pre- ventive measures to minimize or eliminate additional groundwater contamination must be implemented. These measures are not immediate solutions to current water supply problems but are desireable and necessary long-term actions. (2) Because of severe groundwater quality problems in the study area, a safe, dependable water supply must be provided to the residents of the North Fork. A combination of individual home treatment units plus public water systems should be impl enented. Individual home treatment units can contribute significantly to solving current water supply problens but there should not be a long-term reliance on home units - instead, measures should be implemented to start cleansing the aquifers and protecting them against future contamination. {3) The towns of Riverhead and Southold should individually assume responsibility for implenenting water supply programs within their boundaries. Administratively, the towns should establish Water Management Programs, Water Supply Districts and/or water improvements encompassing parts of, or the entire areas of, the towns, for implementation of the plans and recommendations of this study. As an incorporated village, Greenport cannot be included in the Southold District unless it petitions the Town for inclusion. (4) The personnel operating the water systems in the hamlet of Riverhead and Village of Greenport have technical and adminis- trative water supply expertise which can be utilized by the towns. Since the Riverhead systen is currently a part of Town government, institutional arrangements would not be required to 1-7 (5) (6) (7) (8) (9) utilize this expertise. In Southold, the Town can do the fol- lowing: establish its own water supply staff; contract with Greenport for personnel services beyond those currently provided by the Village; or employ a combination of Town staff and Village services. The town Water Management Programs should include individual home treatment as part of their overall responsiblities as Home Treatment Unit Districts. In order to ensure safety and reli- ability, the town Home Treatment Unit Districts {not the home- owner) should own, operate and maintain the home treatment units, not the homeowner; private enterprise can also be allowed to supply units, as approved by the towns, but the towns should retain maintenance responsibilities. In areas of existing development where groundwater is contamina- ted, the Water Management Programs should provide public water systems, if economically feasible, through Water Districts. If public systems cannot be implemented, home treatment units should be publicized and recommended by the water programs and, if existing homeowners request service, the program should provide, own and maintain the home units. When new subdivisions are proposed in areas of groundwater con- tamination or potential contamination, connection to existing public water supplies of adequate capacity are required. If such supplies are not available, then new public supplies are to be constructed by the developer and deeded to a town Water Dist- rict. If an existing home is sold in an area of suspected groundwater contamination, the SCDHS or other appropriate agency should sample and test the home well and the results should be attached to the deed. If the water quality does not meet standards, the current homeowner or new owner should be required to make provi- sion through the Water Management Program to provide a safe water supply prior to the sale of the home. This requirement will necessitate a change in County Health Department regula- tions. The SCDHS is planning to locate a vending machine in the study area which will purify local groundwater and sell the purified water at a nominal cost. If the concept is accepted by the citizens of the area, other vending machines should be located by the Water Management Programs throughout the area as interim measures until the other recommendations of this study are implemented. 1-8 Area-Specific Recommendations (10) Wading River/Northville--The existing individual water systems in Wading River, Baiting Hollow/Woodcliff Park and Reeves Park should be connected to new supply wells which will separately serve each of the three individual subsystems. This is neces-' sary to improve the reliability of the supply and to assure potable water. The new facilities also should be located and sized to serve areas which do not currently have public water systems but the new service areas should be developed in a phased approach. The Riverhead Town Water District should own and operate the new supply wells and transmission lines, and should sell water to the individual water systems through metering facilities. If any of the private systems want to deed their facilities to the Town, the Water District should accept them. It is not economically feasible to provide public water systems to the remainder of the area; these residents should continue to be served by individual home wells, with treatment as required, provided by a Home Treatment Unit District. (11 Riverhead/Jamesport--The Riverhead system should be extended to serve the Calverton area and should actively attempt to further expand its system to other areas adjacent to the present system, including presently unsewered portions of Aquebogue. Major improvements are not required in the Riverhead system except for additional distribution system storage. The remainder of the area, including Jamesport, should continue to be served by indi- vidual home wells with treatment as required, since it is not economically feasible to serve those areas with public water facilities. (12) Mattituck/Cutchogue--The Town of Southold should acquire and operate the existing water system in Mattituck Hills (Captain Kidd); measures should be taken immediately to upgrade the performance and reliability of the system and to augment its source of supply. In the remainder of the Mattituck/Cutchogue demand center (including Cutchogue, Mattituck, Little Hog Neck, East Cutchogue, Fleets Neck, New Suffolk and Indian Neck) it is not economically feasible to provide public water supply sys- tems. These areas should continue to utilize individual home wells; when treatment is required, it should be provided through the Home Treatment Unit District. (13 Southold/Greenport--The Greenport Municipal System, which pre- sently serves Greenport and parts of Southold, should continue to ray on local groundwater sources. However, major improve- ments to the system are required. An agricultural well on 1-9 (14) (15) (16) County Rt. 48 (on the Donohue Farm) should be upgraded (450-gpm total capacity) and used for public water supply. A 2.2-mgd reverse osmosis treatment plant should be constructed in stages to treat the water from the Donohue Well and existing plants no. 6 and 7 used for removal of nitrates, pesticides and herbicides. After implenenting the improvenents, the Greenport system should actively attempt to further expand into other areas adjacent to the existing system. Such expansion should not be permitted, however, until the recommended modifications have been com- pl eted. The activities of any town Water Districts should not impact the existing customers of the Greenport system who are located in the Town; their charges should only reflect the costs of operat- ing the present Greenport system after it is upgraded. In the remainder of the Greenport/Southold area, including Great Hog Neck and East Marion, it is not economically feasible to provide public water supply systems. These areas should con- tinue to be served by individual home wells, with treatment as required, provided through a Home Treatment Unit District. Orient--The Orient area, with its relatively low density of development, cannot economically support public water supply. Existing development should continue to be served by individual home wells with treatment as required. The available fresh water supply is limited and future development should be tightly controlled and result in water requirements consistent with the permissive sustained yield of the aquifer in Zone 5. It is further recommended that only variances resulting in less water usage be approved. If future development is more water-use intensive, the permissive sustained yield in the area will be exceeded and other more costly solutions will be needed. Neighborhood Systems--Public water systems for existing isolated neighborhoods in areas of groundwater contamination are gener- ally not economically feasible. Therefore, it is recommended to continue to serve these areas by individual home wells, with re- quired treatment provided through a Home Treatment Unit Dist- rict. Regional Pipeline System--The development of a major supply of uncontaminated groundwater in eastern Riverhead and piping it via a major transmission main to the eastern portions of the study area was a major alternative considered in the study. It is not recommended for the following reasons: 1-10 I I I I (a) The pipeline would encourage levels of development which appear to be inconsistent wi th the current life style of the people and the general character of the area. (b) The construction of such a pipeline would require the early commitment of major financial resources and resolution of numerous institutional issues. Both types of problems would require a great deal of time for resolution; the water supply problems of the North Fork cannot tolerate significant delays. (c) The pipeline alternative is economically competitive with treating local sources of groundwater in Mattituck, Cutch- ogue, Greenport and Southold only if an uncontaminated supply can be found in the Mago-~h-y aquifer, below the clay layer and above the saltwater interface in eastern River- head. Preliminary data indicates such a supply exists, but its extent and yield needs to be verified with pump test information. To move further west in Riverhead into the Peconic Valley, where clean groundwater is known to exist in sufficient quantity, is not economically competitive with treating local sources of supply. (17) The cost estimates presented in the report for home treatment units should only be incurred by those homes requiring home treatment units. If a home's well tapped a clean supply, that home would not pay for any treatment costs. However, in the case of community-type water systems, all homes would have to pay for the water service, regardless of the quality or safety of the local groundwater. Preventive Measures 118) This plan has identi~ed numerous technical and financial prob- lens associated with providing water supply to areas of the North Fork where groundwater is contaminated. Preventive measures must be undertaken in parall~ with recommendations 2 through 16 in order to ~iminate or minimize additional con- tamination. The following preventive measures are recommended: (a) (b) Expand the SCDHS observation well network and home well sampling program. Support the Cooperative Extension Service, Cornell Univer- sity and U.S. Department of Agriculture (USDA) research and education programs directed to the homeowner and farmer r~ative to usage, dosages and timing of application of herbicides, pesticides and fertilizers. 1-11 m m (c) (e) (f) (g) (h) Support the testing of agricultural chemicals by State or federal agencies in the local environment as a precondition to use by the farm community. Prohibit or control the sale or use of products and chemi- cals which threaten the groundwater resources. Control industrial, commercial and residential activities which impact negatively on groundwater quality. Incorporate detail ed water quantity and quality considera- tions into rezoning and variance decisions because of the critical water supply problems which exist in most of the study area. If rezoning and variance decisions result in more intensive water usage than is currently anticipated, costly water supply treatment systems (desalinization, for exampl e) may be required. Encourage water conservation through public information programs and require water-saving fixtures in new home con- struction. Continue public information and education programs to enphasize the fragile nature of the area's water supply and to foster cooperation in the solutions to those problems. 1-12 SECTION 2.0 INTRODUCTION 2.1 SCOPE The water supply aquifers of the North Fork are threatened by several sources of contamination. Agricultural chemicals used on the land eventually find their way into the aquifers as demonstrated by the elevated nitrate and pesticide (Aldicarb) levels found in the upper glacial materials. In addition, over-pumping and incorrect well location can induce saltwater intrusion into the aquifers. Changing land use patterns further exacerbate the water supply prob- lems. As agricultural lands are converted to residential and commer- cial use, the need for potable drinking water increases. Planning officials have predicted that the trend from agricultural to residen- tial development will continue over the next 20 years, so the demand for potable drinking water will continue to grow. In response to the threat to groundwater quality and the increased demands for potable drinking water, the Suffolk County Department of Health Services (SCDHS) initiated a study of water supply options for the North Fork. The objective of the North Fork Water Supply Plan is to develop and evaluate several alternative water supply plans which could safely meet present and future potable water demands. This was accomplished in three phases. Phase I consisted of an analysis of present and future population levels and land uses. Present water consumption patterns were also evaluated to develop a complete understanding of water supply systems needs and to accurately project future water supply demands. Simul- taneously, hydrogeologic studies were performed to develop estimates of the quantities of fresh groundwater available and to ascertain the extent and degree of contamination of the groundwater by organic and inorganic constituents. The background investigations formed the basis upon which alternative plans were developed and evaluated in Phase II. Alternative re- gional, subregional and community water supply concepts were identi- fied and developed. These were then evaluated on the basis of cost, environmental consequences, reliability of operation, and flexibility to determine a future course of action. In Phase III legislative, management and financing constraints were studied and suitable recommendations for implementation were developed. 2-1 I I I I I I I I I I I I I I I I I I 2.2 STUDY AREA The plan covers the towns of Riverhead and Southold on the North Fork of Long Island (see Figure 2-1). The North Fork is essentially a peninsula which extends east for over 35 miles from the eastern terminus of the main land body of Long Island. The study area ex- tends from the community of Wading River on the west to Orient Point on the east, and covers approximately 109 square miles. It is bound on the north by Long Island Sound and on the south by the Peconic River and Peconic Bay. Offshore islands were not included in this study. 2.3 PLANNING APPROACH--WATER SUPPLY ZONES As one progresses west to east through the study area, fresh ground- water availability changes significantly. From the standpoints of planning and implementation, it is logical to organize the study area into discrete groundwater supply zones for analysis. Five such zones were defined (see Figure 4-1). Zone i starts at the western Town boundary of Riverhead and extends east to the area of the commercial hamlet of Riverhead. In this zone, the Magothy aquifer is saturated with fresh water all the way to bedrock. Doctor's Path is used as the eastern terminus of Zone 1. In the hamlet area, saltwater begins to intrude into the lower depths of the Magothy. The presence of saltwater in the Magothy increases as one progresses east, until the fresh water lens terminates at Mattituck Inlet, which is the eastern boundary of Zone 2 (see Figure 4-1 for a definition of the fresh/salt water boundary). Zones 3, 4 and 5 are defined by the east-west boundaries of three isolated fresh groundwater lenses. Zone 3 extends from the Mattituck Inlet to Hashamomuck Pond. A fresh water lens begins at the Inlet, increases in depth in an easterly direction, and then ends where the ground surface intersects the groundwater table at the Pond and salt- water intrudes. Zone 4 extends from Hashamomuck Pond to Dam Pond; Zone 5 starts at Dam Pond and ends at Orient Point. Each zone has its own hydrogeologic characteristics and water budget. There is some fresh water movement between zones 1 and 2; the other three function independently. Contamination problems vary among the zones and there are substantially different volumes of fresh water available in each zone. In other words, each zone is unique, and its distinguishing features have to be considered. Therefore, all basic inventory data was organized by zone, and the development and evalua- tion of alternatives also considered the zone boundaries. 2-2 SECTION 3.0 POPULATION AND LAND USE 3.1 POPULATION The U.S. Census indicates that the year-round population of the North Fork study area was 39,097 in 1980; 20,243 people resided in the Town of Riverhead, and 18,854 in the Town of Southold (excluding Fishers Island, which is outside the study area). As shown in Table 3-1, the population reached these levels after growing 7 and 15 percent, re- spectively, in Riverhead and Southold during the 1970's. During this period, almost 4,000 people were added to the two towns. Most of the growth occurred in the Shoreham/Wading River, Southold and Mattituck/ Cutchogue areas. Oyster Pond, Greenport, and New Suffolk experienced minor decreases in population during the period. Population projections developed by the Long Island Regional Planning Board on the basis of school districts forecast continued growth. It is anticipated that the population in Riverhead and Southold will in- crease to 49,800 people (a 27 percent increase) between 1980 and 2000(1), with all school districts experiencing growth. The popu- lation in Riverhead is expected to increase to 26,750 persons (a 32 percent increase) while the population in Southold is expected to increase by 22 percent, reaching a level of 23,050. These pro~ec- tions represent a refinement of earlier planning board estimates (prepared during the 208 Program), and pro~ect a lower rate of growth in the area than that originally forecast. Population data is presented in Table 3-2 by groundwater supply zone. The reorganization of the data, which was based on school districts, was accomplished with the aid of detailed census tract data, Long Island Lighting Company (LILCO) population estimates and planning board p~pulation density and land use maps. During the summer, the population on the North Fork increases signi- ficantly. The summer population increase consists of guests, second- home owners, campers and motel users. The Suffolk County Planning Board estimates that the population in Riverhead and Southold in- creased during a peak weekend by 10,705 and 19,760 people, re- spectively, during the summer of 1980 (see Table 3-3). (1)Future population, land use and water demand information in this report is presented for the year 2000 although the planning board data is developed for lg95. However, the timing of the pr~ected growth l'evels is not accurate, and it is entirely possible that they will not be reached until 2000 or 2005. 3-1 TABLE 3-1 POPULATION BY SCHOOL DISTRICT(I) 1970 U.S. Census 1980 U.S. Revised 208 % Increase Census Projections 2000(3) 1970-1980 Increase 1980-2000 Riverhead Shoreham-Wading River 1,411 Riverhead 17,277 Laurel 191 TOTAL, RIVERHEAO 18,909 2,392 4,300 66.0 79.8 17,643 22,000 2.1 24.7 208 450 8.9 116.3 20,243 26,750 7.1 32.1 Southold Oyster Pond 1,530 Southold 4,127 Mattituck-Cutchogue 5,303 Greenport 4,018 Laurel 872 New Suffolk 492 Fishers Island 462 TOTAL, SOUTHOLD 16,804 TOTAL, SOUTHOLD IN STUDY AREA (2) 16,342 TOTAL IN STUDY AREA (2) 35,251 1,509 2,000 (- 1.4) 32.5 5,233 6,500 26.8 24.2 6,794 8,000 28.1 17.8 3,952 4,500 (- 1.6) 13.9 971 1,500 11.4 54.5 395 550 (-19.7) 39.2 318 500 (-31.2) 57.2 19,172 23,550 14.1 22.8 18,854 23,050 15.4 22.3 39,097 49,800 10.9 27.4 (1) Permanent year-round population. (2) Excludes Fishers Island. (3) Source: L.I. Regional Planning Board. TABLE 3-2 POPULATION BY GROUI~DWATER SUPPLY ZONES(l) Zone 1 2 3 4 5 School Districts Included In Groundwater Supply Shoreham-Wading River, Riverhead (Part) · Riverhead (Part), Laurel, Mattituck-Cutchogue (Part) Mattituck-Cutchogue (Part), Southold, New Suffolk Greenport, Oyster Ponds (Part) Oyster Ponds (Part) TOTALS: 1980 Projected Year % Increase Population 2000 Population (1980-2000) 14,798 8,456 10,382 19,755 33 10,895 29 12,650 22 5,367 17 1,133 33 49,800 27 4,606 855 39,097 (1) Permanent year-round population. m m m m m m m m m m m m m m m m m m m TABLE 3-3 PEAK SUMMER WEEKEND POPULATION(I) Summer Second Total Summertime Guest Home Camping Motels Increase Southold 5,628 11,846 744 1,542 19,760 Riverhead 2,927 5,101 1,568 1,109 10,705 TOTALS 8,555 16,947 2,312 2,651 30,465 (1) Source: Suffolk County Planning Board, Based on 1980 U.S. Census. 3.2 LAND USE Land use impacts water supply requirements since individual land uses have different water quality and quantity needs. It was, therefore, important to consider existing land use on the North Fork and to examine projected future land use. Existing land use in the two towns is summarized in Table 3-4. With the exception of Zone 4, agricultural land dominates each zone, with 42 percent of the land area (29,360 acres) currently devoted to agri- cultural usage. Crops grown on the North Fork include potatoes, mixed vegetables, cabbage, cauliflower, rye, nursery stock, fruit trees, grapes and sweet corn. Lower density residential development (one to four dwellings per acre) dominates the residential land use category. High-density residential development (greater than five dwellings per acre) occurs significantly only in zones I and 4 (downtown Riverhead and the Village of Greenport). The vacant land category includes open space, unmaintained parkland, and undeveloped portions of Grumman's complex in Zone 1. Industrial development on the North Fork is ex- tremely limited (except for Grumman). Commercial usage can be found to a limited extent in each zone, with significant commercial devel- opment in the hamlet of Riverhead, Greenport Village and to a lesser extent in Mattituck and Southold. Future land use was projected by the regional planning board during the 208 Program. The projections indicate that low-density residen- tial development will significantly increase in both towns, while agricultural and open/vacant space will decrease. The land use trends rojected by the planning board forecast an 18~rcent decline in agricultural land use (approximately 5,300 acres). These pro jections are generally consistent with past trends, which resulted in the loss of 30 percent of the farm land in Suffolk County between 1964 and 1978. This trend is abating somewhat as the County's Farm- land Preservation Plan and open space policies are implemented. The Cooperative Extension Service anticipates a gradual change in the types of crops grown on the North Fork. In general, potatoes, sod and rye acreage will decrease while the other crops grown will in- crease in acreage. (1)The projected declines in agricultural land use, by Zone, over the next 20 years are as follows: Zone 1, decline of 1,750 acres; Zone 2, decline of 1,880 acres; Zone 3, decline of 855 acres; Zone 4, decline of 535 acres; and Zone 5, decline of 280 acres. 3-5 TABLE 3-4 EXISTING LAND USE (ACRES)(1) Residential Industrial/Comm- 0-1 2-4, 5+ ercial/Utilities/ (2) ZONE DU/Acre DU/Acre DU/Acre Total Agricultural Transportion Vacant Other Totals 1 422 1,035 472 1,929 9,725 4,532 10,373 1,512 28,071 2 539 1,915 77 2,531 9,865 620 5,107 452 18,575 3 351 2,450 47 2,848 8,155 660 3,818 163 15,644 4 79 620 119 818 635 314 2,245 200 4,212 5 116 203 0 319 1,180 104 1,645 10 3,258 TOIALS 1,50! 6,223 715 ~,445 29,560 6,230 2J,188 2,3~T 69,760 (1) Source: Cornell Univers'ity (Center for Environmental ~esearch) - Water Resources Model Grid (1982) These data are unofficial and may change slightly. This will not impact the results of the study. (2) Includes recharge basins, water, golf courses and small parks. I I I I I SECTION 4.0 GEOLOGY The geology of the study area has been described in several earlier investigations. The most recent detailed geologic investigation of the North Fork was performed by the Suffolk County Department of Health Services (SCDHS). A brief summary of the stratigraphic re- lationships found in the study area is presented below. 4.1 GEOLOGIC UNITS 4.1.1 Bedrock Crystalline bedrock of Pre-Cambrian age underlies the entire study area at depths ranging from 500 feet below sea level in eastern Southold to approximately 1,300 feet below sea level in southwestern Riverhead. The bedrock consists of gneiss and schist of very low permeability. It has virtually no value as an aquifer. 4.1.2 Raritan Formation Resting unconformably on the bedrock surface is the Raritan Formation of Late Cretaceous age. This unit is generally divided into the lower Lloyd Sand member and the overlying Raritan Clay member. Both units become progessively thinner to the east. The Lloyd decreases from a thickness of about 300 feet to 150 feet. Similarly, the Rari- tan Clay varies from 200 to 100 feet. The Lloyd Sand has been described as a grayish white, coarse sand and gravel with interbedded units of gray clay. The Raritan Clay is described as 9ray clay and silty clay. Fresh water is only found in the Raritan Formation in parts of Zone 1. Generally, the water quality and depth of the Raritan Formation make it an unlikely source of water supply within the study area. 4.1.3 Matawan Group--Magothy Formation Undifferentiated The Magothy Formation, like the Raritan, is a later Cretaceous deposit although an erosional unconformity separates the two units. The upper surface of the Magothy was also shaped by erosion forces. The irregular upper Magothy surface forms a circular high about 150 feet below sea level in western Riverhead and then drops to a rela- 4-1 I I I I tively constant surface about 350 feet below sea level. The con- figuration of the Magothy's upper surface is shown in Figure 4-1. The total thickness of the Magothy in the study area varies from about 150 to 600 feet. The Magothy generally consists of gray fine to coarse sand with interstitial silt and clay. Lenses of clay, silt and clayey sand are also commonly found in the Magothy, particularly the upper sections. The basal portion of the Magothy is generally coarser with coarse sand and gravel being characteristic. In the five western towns of Suffolk County, the Magothy is the principal source of public water supply. Within the study area, however, the Magothy is only available for supply in zones i and 2. East of Mattituck Inlet, in zones 3, 4 and 5, the Magothy is generally salty. 4.1.4 Pleistocene Deposits The most important geologic unit within the study area with respect to supplying drinking water are the glacial deposits of Pleistocene age. This material rests unconformably on the Magothy's eroded upper surface and reaches a maximum thickness of about 550 feet in north- eastern Riverhead. The physical properties and appearance of the glacial deposits vary with manner of deposition. The predominant type of glacial deposit in the study area is outwash. Deposited by glacial meltwater, the outwash is stratified and generally composed of clean sand and gravel. Till, the other major glacial deposit, was laid down by moving ice, and consists of unsorted clay, silt, sand, gravel and boulders. The till is primarily found along the north shore of the study area. Several previous investigations discuss the existence of a substan- tial clay unit interbedded with the outwash deposits. As part of this study, an attempt was made to collect all the geologic logs available from wells drilled in Southold and Riverhead. Based on the correlation of these logs (wells no. A through E shown on Figure 4-1) including recent drilling projects done by SCDHS, a continuous unit of clay and sandy clay is shown to exist between 60 and 150 feet below sea level. The variable thickness and horizontal extent of the clay is shown in Figure 4-1. The clay is particularly well docu- mented in zones 2 and 3. In these areas, every well that penetrated deep enough encountered the clay at approximately the same elevation. The clay was also found in zones 1, 4 and 5; however, the paucity of data makes a correlation in these areas somewhat problematic. The compete areal extent of the clay, the depositional environment and its relationship to other clays on Long Island (i.e., the Gardiners Clay, the 20-foot clay, and the Smithtown Clay) remain to be dete- rmined. 4-2 ZONE I ZONE 2 ZONE 3 ZONE 4 ZONE 5 HEAD RIVERHEAD LINE BOUNDARY 0 AND SANDY CLAY SLAY FRESH WATER )NAL TEST WELL GLACIAL This clay unit has negative and positive water supply ramifications; it is not, in itself, a good fresh water aquifer, but it does have an ability to inhibit the upconing of saltwater beneath high-yield sup- ply wells. The clay unit has been shown to underlie Greenport Water District's Plant 6, and this facility has never been affected by elevated chloride levels even during periods of peak demand. Ground- water modeling of the impact of prolonged dewatering at the Long Island Lighting Company's (LILCO) proposed Jamesport plant also indicated that saltwater upconing through the clay would not take place, although lateral saltwater intrusion would be a problem. 4.2 AQUIFER PARAMETERS To characterize an aquifer's ability to store and transmit water, physical parameters including porosity, permeability (horizontal and vertical), specific capacity, and transmissivity must be determined. This can be done in several different ways (theoretical hydraulic formula, lab tests, tracer tests); however, the most accurate method for evaluating local aquifer performance is through a carefully monitored pump test. The data available from pump tests conducted in the study area are somewhat limited and variable, but enough informa- tion does exist to generally establish the average aquifer proper- ties. One of the first attempts to empirically determine aquifer parameters was a series of pump tests conducted throughout the County as part of the Suffolk County Test Well Program. This included three individual sites in Riverhead where six different screen settings in the Magothy and one setting in the upper glacial were tested. In Southold, two upper glacial settings were also tested. The results of this work are presented in the "Comprehensive Public Water Supply Study--Suf- folk County, Volumes II and III" (Holzmacher, McLendon and Murrell, 1970). A theoretical analysis of aquifer parameters that relied on existing data was presented in "Water Transmitting Properties of Aquifers on Long Island, New York" (McClymonds and Franke, 1972, USGS Profes- sional Paper 627-E). The regional variations in hydraulic conducti- vity and transmissivity were estimated using specific capacity and lithologic data. A detailed analysis of aquifer properties in the study area is conz tained in the 1975 Geraghty and Miller report "Study of Groundwater Condition on the Long Island Lighting Company Tract, Jamesport, New York." As part of this study, several pump tests in the upper glacial aquifer were performed to determine the impact of projected heavy groundwater withdrawals. 4-4 Most recently, a pump test was run in the upper glacial aquifer at a site near the Village of Southold as part of the Nassau-Suffolk Re- gional Planning Board's 208 Program. The pump test results are sum- marized in the 1977 Woodward Clyde report "Groundwater Studies for Section 208 Plan, Long Island, New York." Table 4-1 presents a summary of the upper glacial aquifer parameters determined during the studies cited above. The upper glacial aquifer is a highly productive unit, with physical properties that are relatively consistent throughout the study area. The uniform composition of the glacial outwash deposits in which all the aquifer test wells were screened makes average parameter values applicable over the entire extent of the aquifer. The variation in specific capacities reflect differences in well construction and pumping rates as well as aquifer lithology. The differences in transmissivities are the product of changes in aquifer thickness (transmissivity = hydraulic conductivity × saturated aquifer thick- ness~. The Magothy aquifer pump test data developed during the Suffolk County Test Well Program are summarized in Table 4-2. The sites tested are all in zones I and 2 where the Magothy is actually available for water supply. Because the Magothy deposits and their physical properties vary considerably both horizontally and verti- cially, it is not as easy to make regional extrapolations from these data. McCl~q~onds and Franke (1972) estimated that the Magothy in Riverhead has a hydraulic conductivity of about 400 gallons per day per square foot (gpd/sq ft) and a transmissivity of 260,000 gpd/sq ft based on an aquifer thickness of 650 feet. Although the Magothy is generally composed of finer-grained material and is less permeable than the upper glacial, the data in Table 4-2 indicates that it is capable of supplying large quantities of water. The Osborn Avenue well in Riverhead is screened in the Magothy and is currently used for public supply by the Riverhead Water District. Any interpretation of the data in Table 4-2 must consider that the well screens were intentionally located in the most permeable hori- zons and not in strata that are necessarily representative of the entire aquifer. In general, the transmission of water in the Magothy in zones 1 and 2 is not a limiting factor to the development of public water supplies. 4.3 FRESH WATER-SALTWATER INTERFACE The configurations of the fresh water lenses beneath groundwater supply zones 3, 4 and 5 are in dynamic equilibrium with the saltwater that surrounds them. Their thicknesses and lateral extents remain 4-5 TABLE 4-1 UPPER GLACIAL AQUIFER PARAMETERS Source of Data Aquifer Test Location and Groundwater Supply Zone Suffolk County Riverhead Village (Zone 1) 35 3,480(1) Test Well Cutchogue (Zone 3) 30 3,000 Program - 1969 Southold (Zone 3) 33 3,280 Specific Hydraulic Trans- Storage missivity Coefficient Capacity Conductivity (gpm/ft) (gpd/ft~) (gpd/ft) 543,000(1/ 267,000 223,040 0.01 to 0.2 McClymonds and Regional Estimate Not 1,500-2,000 240,000 Not Franke, 1972 Calculated Calculated Geraghty & Miller Jamesport (Zone 2) 68 4,400 Horizontal 380,000 0.10 LILCO Report, 1975 660 Vertical 208 Study, 1977 Village of Southold 58 Not 24,370 Not Calculated Calculated (1) Hydraulic conductivity and transmissivity calculated from specific capacities. Well Location and Groundwater Supply Zone Osborn Ave., Riverhead (Zone 1) TABLE 4-2 MAGOTHY AQUIFER PARAMETERS Depth of Screen Setting (ft) 345-365 668-718 Specific Capacity (gpm/ft) 24.7 17.3 Transmissivity (gpm/ft2) 54,000 37,000 Baiting Hollow 370-390 (Zone 1) 720-740 14.6 14.1 33,000 32,000 Aquebogue (Zone 2) 440-460 9.5 21,000 From: Holzmacher, McLendon and Murrell, CPWS-24, 1970. stable over time because of the balance between recharge and dis- charge. The location of the fresh water-saltwater interface, the boundary between the fresh water lenses and the underlying saltwater, also tends to remain unchanged as long as the aquifer is not over- pumped. The position of the interface has been determined during exploratory drilling on various projects by SCDHS, test well installations associated with the 208 Program and CPWS-24, miscellaneous test wells (i.e., S189, S490) and private supply wells that were inadvertently screened in salty water. (For the purpose of this study, saltwater is assumed to have a chloride concentration of more than 250 milli- grams per lite~ (mg/1.) This information, in conjunction with esti- mates based on the Ghyben-Herzberg principle, were used to locate the position of the interface on an east-west longitudinal cross-section of the study area. The location and elevation of the interface is shown on Figure 4-1. The Ghyben-Herzberg relationship states that the distribution of fresh and saltwater at hydrostatic equilibrium is caused by the difference in density between fresh and saltwater. Based on this difference, the depth of fresh water below sea level is predicted to be 40 times the height of fresh water above sea level. This re- lationship is theoretical and when applied to the North Fork, Cran- dell ("Geology and Groundwater Resources of the Town of Southold, Suffolk County, 1963, USGS Water Supply Paper 1619-66) found that it consistently underestimates the actual depth of available fresh water. However, in a hydrologically sensitive area such as this, a conservative approach is appropriate for water supply planning and, therefore, Ghyben-Herzberg was used. The position of the interface deviates from the predicted location in some areas because of lithology. The interface is found below the clay unit throughout Zone 3, and in many areas this is much deeper than would be expected. The thick, relatively impermeable clay unit apparently prevents the interface or the associated zone of diffusion from responding to short-term fluctuations of the water table. The interface in zones 4 and 5 has not been accurately determined during drilling projects so the influence of the clay unit cannot be assessed. The lateral position of the interface is not influenced by the clay. There is no comparable barrier to check the horizontal, landward encroachment of saltwater. Groundwater withdrawals near the shore- line, over a prolonged period of reduced recharge, could have an effect on the position of the interface with respect to the shore- line. 4-8 m SECTION 5.0 GROUNDWATER QUALITY The groundwater aquifers underlying the North Fork are significantly contaminated by inorganic and organic chemicals. The nature and extent of groundwater contamination was evaluated in this study relative to its impact on water supply. The data used in the analy- sis was collected by the Suffolk County Department of Heal th Services (SCDHS) in their home- and monitoring-well sampling programs and their North Fork transect project. 5.1 INORGANIC PARAMETERS Table 5-1 summarizes the SCDHS data for inorganic parameters on the North Fork. The table characterizes groundwater quality by present- ing the percentage of well samples vYnich exceed established guide- lines or standards for nine inorganic parameters. Data is presented for 14 communities on the North Fork (grouped by zone) and for Suf- folk County as a whole. Using the data on Table 5-1 and other supporting data from the SCDHS, it was determined that the upper glacial aquifer throughout the plan- ning area is already contaminated with nitrates. Nitrate levels are significantly elevated above natural background conditions of 0.1 to 1.0 milligrams per liter (n~j/1) and exceed the drinking water stan- dard of 10 mg/1 in many areas. A sampling of 639 wells for nitrates by the SCDHS in the Town of Riverhead showed the following: 58 per- cent of the samples ranged from 0 to 5 mg/1; 26 percent were above 7.5 mg/1; and 16 percent were above 10 mg/1. Similar results were obtained in a sampling of 1,121 wells in the Town of Southold: 51 percent of samples ranged from 0 to 5 mg/1; 30 percent were above 7.5. mg/1; and 17 percent were above 10 mg/1. Figure 5-1 graphically depicts the areal extent of nitrate contamina- tion in the upper glacial aquifer. As expected, the affected areas are closely related to agricultural land use in zones 1, 2, 3, and 5. The areas of nitrate contamination shown on Figure 5-1 are generali- zed from individual well data. Specific nitrate concentrations at individual locations within the shaded areas can vary significantly from below 8 nKJ/1 to well above the 10 ~/1 nitrate standard. 5-1 TABLE 5-1 RERCENTAGE OF WELL S/~qPLES EXCEEDING STANDARDS OR GUIDELINES FOR INORGANIC PARAMETERS (1) Number of Nitrates Zone communit.~____..-~JLles (lOm__~]]_ 1 Baiting Hollow 6 50.0 Riverhead 239 16.5 Wading River 170 6.5 2 Aquebogue 132 16.7 Sodium Chlorides 20 mg/1 (26om/_~!L!]- (lOOm_5~Z]] o o 0.4 IB.6 (1.31 0 11.8 (0) MBAS Sul fates (detergents) Iron Manganese Copper Zinc (0.6~ (0.3 m/__~]]._ (0.3 m__~].~ (1.0 m /1) (5.0 m /1) (250 m /1) 0 33.3 0 0 16.7 0 0.4 37.8 14.0 0.8 4.7 0 0.6 59.3 0.6 2.3 4.5 30.1 (6.5) 0 19.4 (1.4) 2.6 22.3 6.1 28 (12) 0.8 33.2 17.5 3.0 10.7 0.7 0 0 Calverton 98 10.0 Jamesport 77 10.4 Laurel 49 21.6 0 33.7 3.0 2.1 6.1 0 1.3 46.8 14.3 1.3 1.3 0 0 3.6 18.3 0 2.0 0 CutChogue 228 16.5 2.6 37.0 (3.2) Mattituck 286 17.4 0.7 24.8 (2.6) New Suffolk 25 0 0 54.5 (4.5) Peconic 46 21.7 0 28.6 (2.6) Southold 338 15.6 1.8 36.2 (3.7) 0.9 38.6 8.7 8.3 4.9 0 1.3 29.3 14.1 3.5 4.6 0 0 16.0 4.0 0 0 0 0 36.9 6.5 4.4 4.3 0 0.6 42.6 8.9 3.6 1.5 0.7 TABLE 5-1 (continued) PERCENTAGE OF WELL SAMPLES EXCEEDING STANDARDS OR GUIDELINES FOR INORGANIC PARJ~4ETERS Sodium MBAS Number of Nitrates Chlorides 20 mg/1 (detergents) Iron Manganese Copper Zinc Sulfates Zone Community Samples (10 m~/1) (250 mg/1) (100 m9/1) (0.5 mg/1) (0.3 m9/1) (0.3 mg/1) (1.0 mo/l) (5.0 mg/1) (250 mo/l,), 4 East Marion 69 8.6 2.9 39.3 44.9 2.9 8.7 0 0 (4.5) Greenport 70 9.5 1.4 40.0 0 46 2.7 4.1 0 0 (1.4) 5 Orient 98 18.2 2 48.0 0 32.6 4.1 8.2 2.0 1.2 (6.3) Suffolk County 14,000 6.7 1.2 23.0 1.1 45.7 17.5 2.7 5.2 0.2 (1)Source: SCOHS Home Wells Sampling Program. (125/3) '---]-- AREAS HAVING NITRATE CONCENTRATION OVER 8mg/I The vertical extent of the contamination is also important in pro- jecting future migration rates through the aquifer system and future water supply impacts. The most detailed study of vertical concentra- tion gradients was done by SCDHS as part of their New York State De- partment of Health (NYSDOH) pesticide grant. The results of the ver- tical groundwater quality profiles compiled along a transect in Cutchogue {Zone 3) showed extensive distribution of nitrates verti- cally, permeating almost all of the upper glacial aquifer above the clay. These results were confirmed by limited vertical profiling at two other study sites also in Zone 3. The vertical nitrate concen- tration data from Zone 3 is consistent with groundwater quality find- ings in other supply zones. In zones I and 2, the fresh water which is available beneath the clay unit is not contaminated with nitrate and will probably remain un- contaminated because of the reduced rate of migration of contaminants through the clay. While only a small number of wells currently tap the Magothy aquifer, this formation beneath zones i and 2 has been shown to have acceptable water quality and low nitrate concentra- tions. Samples from Riverhead and Baiting Hollow Magothy wells had nitrate concentrations of less than 0.02 mg/1. A nitrate concentra- tion of 1.09 mg/1 was found in a Magothy well in Aquebogue in 1969; the most recent data from that well (December 1981) showed a small increase to 2.9 mg/1. As indicated in Table 5-i, a significant number of samples exceeded the 20 mg/1 sodium guidelines contained in the New York State drink- ing water standards. The sodium standard is relevant to consumers on severly restricted salt diets. The State guideline also indicates that water containing more than 270 n~j/1 of sodium should not be used for drinking by those on moderately restricted sodium diets. As in- dicated in the table, few samples approach this level. Iron and manganese are frequently found in levels that exceed State standards. These constituents occur naturally in the groundwater and their distribution on the North Fork is not related to land use acti- vities. These standards do not represent toxic or hazardous levels, but have been established for aesthetic reasons (staining of bathroom fixtures and clothes during washing). The chloride levels found by the SCDHS in the North Fork transect were higher than normal background levels (20 mg/1) but well below the drinking water standard of 250 mg/1. The 'elevated chloride levels correlate with elevated nitrate levels, suggesting that agricultural fertilizers are the mdor source of the contamination. Table 5-1 indicates that the chloride standard is normally not ex- ceeded in home water supply wells on the North Fork. MBAS (detergents) and sulfates are rarely found in North Fork water supplies in concentrations exceeding standards. Copper and zinc were similarl~ not detected in concentrations exceeding established stan- dards. 5-5 5.2 ORGANIC PARAMETERS Table 5-2 presents the results of the SCDHS home well sampling pro- gram for the pesticide Aldicarb and three chlorinated hydrocarbons. This data summarizes the results of the Department's surveillance efforts which were begun in mid-1979. The extent of Aldicarb contamination is similar to the distribution of nitrates because both have been used extensively by potato farmers on the North Fork. Figure 5-2 maps the areal extent of Aldicarb con- tamination on the North Fork. SCDHS correlated the presence of Aldicarb contamination with potato farming; 98 percent of the wells with more than 7 parts per billion (ppb) of Aldicarb were within 1,000 feet of a potato farm. The areas of Aldicarb contamination shown on Figure 5-2 are generali- zed from individual well data. Specific Aldicarb concentrations at individual locations within the shaded areas can vary significantly from zero or close to zero to above the 7 ug/1 Aldicarb standard. SCDHS data and their statistical summaries of Aldicarb sampling re- sults further demonstrate that organic contamination from pesticides and herbicides is widespread. In the Town of Riverhead, 2,161 wells were sampled: 32 percent were contaminated by Aldicarb; and 16 per- cent had Aldicarb concentrations above the health guideline of 7 ppb. In the Town of Southold, 3,160 wells were sampled: 23 percent showed Aldicarb contamination; 11 percent exceeded the health guideline of 7 ppb. Unlike nitrates, Aldicarb has been shown to be generally limited to the upper 30 to 40 feet of the aquifer, except in the central re- charge areas where it has been detected near the bottom of the aqui- fer. The eventual migration of Aldicarb throughout the aquifer can reasonably be expected in the future. Efforts to deepen wells to avoid current Aldicarb contamination will, therefore, provide only temporary relief. Most notable among the wells contaminated with Aldicarb are the Greenport Water District wells 6-1 and 6-2. The water derived from well 6-1 was found to consistently exceed the 7-ppb guideline while the water from well 6-2 periodically exceeds the guideline. A carbon filter was installed in August, 1980 to treat the contaminated water from well 6-1. Water from well 6-2 is blended with treated water from well 6-1. Additionally, wells 4-6, 4-?, well 7 and well 8 have been impacted by pesticide contamination. SCDHS has sampled for and confirmed the presence of other pesticides and herbicides in the groundwater beneath the North Fork (dichloro- propane, carbofuran, dacthal, oxamyl and dinoseb.I The data avail- able on these contaminants is not as comprehensive as that compiled for Aldicarb but does verify that additional contamination has oc- curred. 5-6 · = m m m m m ,= m m m m m m m m m m m m TABLE 5-2 PERCENTAGE OF WELL SAMPLES EXCEEDING GUIDELINES FOR ORGANIC PARAMETERS (I) Aldicarb Chlorinated Hydrocarbons No. No. 1,1,1Trichloroethane Tetrachloroethylene Zone Community of Trace 7 ppb(2) of Samples Samples Trace 50 ppb(2) ~b(2) Trace 50 ppb (2) 1 Baiting Hollow 5 0 0 1 0 0 0 0 Riverhead 604 16.9 8.4 85 4.7 1.2 1.2 1.2 Wading River 230 7.8 1.8 62 8 0 0 0 1,1,2 Trichloroethylene Trace 50 0 0 0 1.2 0 0 Aquebogue 261 16.8 19.2 33 21.2 0 0 0 0 0 Calverton 464 18.3 12.3 43 16.3 0 2.3 0 4.6 0 Jamesport 227 , 10.1 26.4 35 20.0 2.9 5.7 0 0 0 Laurel 29g 21.0 42.8 21 0 0 0 0 D 0 Cutchogue 579 12.1 16.2 89 4.2 2.2 7.8 0 0 Mattituck 984 12.3 12.3 123 8,2 2.4 4.1 0 1.6 New Suffolk 125 10.4 12.0 7 20 0 30 0 0 Peconic 225 15.1 13.8 8 D 0 0 0 0 Southold 714 14.7 9.8 80 3.8 0 7.5 0 2.5 0 0 0 12.5 0 4 East Marion 153 2.0 3.3 14 0 0 0 0 0 0 Greenport 45 2.2 11.1 11 42.9 14.3 9.1 0 0 0 5 Orient 335 8.0 5.4 17 0 0 0 0 0 0 (1)Source: SCDHS. (2)Guideline values. il $OUNO&VE NORTHVIL~E ~--AREAS WITH ALDICARB CONCENTRATIONS OVER 7,ug / I Figure 5-2 Aldicarb Contamination Areas Table 5-2 also presents the results of the SCDHS home well sampling program for three chlorinated hydrocarbons (1,1,1 trichloroethane, tetrachloroethylene, and 1,1,2 trichloroethylene). The use of these chemicals is generally not associated with agricultural activities. The greatest occurrence of these compounds was in Greenport, an area where agricultural activity is limited. The sources of this type of contamination are generally commercial activities and homeowner use of the compounds. After reviewing the water quality conditions on the North Fork, it was determined that a conservative approach to water supply planning must be taken because of degraded water quality and the need to pro- tect consumers. It is clear that the aquifers will remain contamin- ated with nitrates, Al dicarb and with other organic parameters through the year 2000. Even though additional inputs have ceased, concentrations of Aldicarb will continue to exceed the ?-ppb drinking water guideline. As additional water quality data is collected, more contamination problems will be discovered. 5-9 ! ! SECTION 6.0 EXISTING WATER SUPPLY SYSTEMS Within the study area, there are two municipal water supply systems: one serving the hamlet of Riverhead, and the other, the Village of Greenport. In addition, there are numerous (21) small, privately owned water systems. The Riverhead and Greenport service areas and the locations of the 21 private systems are shown on Figure 6-1. 6.1 GREENPORT MUNICIPAL SYSTEM The Greenport system presently serves the Village of Greenport and portions of the Town of Southold, as shown on Figure 6-1. The franchised area encompasses about 15 square miles extending generally from East Marion, west to Horton Neck and Jockey Creek. The system includes about 50 miles of piping serving about 2,300 connections and has one distribution system storage tank with a capacity of 300,000 gallons. The Village presently owns and operates six well fields with two (Plants i and 2) abandoned several years ago due to water quality problems. Table 6-1 summarizes information on the ~¢ell supplies. Presently, Plant 3 is only used for emergency purposes due to re- latively poor water quality. Also, it has been reported that the yield of Plant 4 has been reduced to 300 gallons per minute (gpm) due to the recent activation of Plant 8, since these well supplies are relatively close to each other and hydrologically connected. There- fore, the net available pumping capacity is about 1,710 gpm. At Plant 6, a granular activated carbon filter was installed on the larger of the two wells during the summer of 1980 (well no. 6-1) in order to remove pesticides. Union Carbide, the manufacturer of Aldicarb, supplied the filter and the initial 20,000 pounds of carbon. Because of the hydraulic restrictions imposed by the filter, the well capacity was reduced from 550 gpm to 400 gpm. Water from this well is blended with water from well no. 6-2 (150-gpm capacity) resulting in a total combined plant capacity of 550 gpm. Plant 6 represents the ma~or source of water for the system, accounting for up to 61 percent of the total water pumped over the last few years. Aldicarb contamination has also been detected, in lesser degrees, in the water from wells 4-6, 4-7, Plant 7 and Plant 8. 6-1 GREENPORT WATER DISTRICTI WADING RIVER WATERWORKS 3WER -WILDWOOD SHORES ASSOCIATION - HEROD POINT ASSOCIATION OUNOSHORE CLUB - RAMBLEWOOD MHP - HULSE FARMS MHP LIFF PARK CONDOMINIUM rEEVES BEACH WATER CO. WATER CO. IDD WATER CO.- CALVEFITON f PECONIC MHP RIVERHEAD ~I.G.F. { PECONIC RIVER VIEW) MHP AQUFBOGUE MHP Figure 6-1 Existing Water Supply Systems P1 ant No. No. of We11 s TABLE 6-1 SUMMARY OF WELL SUPPLIES FOR THE GREENPORT MUNICIPAL SYSTEM Total Pump Well Historical Water Capacity (gpm) Depth (ft) Quality Problems 3 3 340 45-57 4 3 510 79-80 5 1 160 60 6 2 550 77, 94 7 i 350 89 8 1 350 95 TOTALS 11 2,260 Iron, manganese, chlorides, pesticides Occasionally high chlorides, pesticides Nitrates Nitrates, pesticides Chlorides, pesticides Pesticides If, due to mechanical failure or maintenance, the carbon filter is inoperable at well no. 6-1, this supply would be removed from service, reducing the total system capacity from 1,710 gpm to 1,310 gpm. This latter capacity should be adopted for water supply planning purposes in order to assess future requirements. Accordingly, based on an estimated maximum day requirement of 1,780 gpm (see Section 8.0) it is estimated that a deficit of about 470 gpm (0.68 mgd) in dependable water supply capacity will exist by the year 2000. The Greenport system is well maintained and operated. Its major problems are the quality of its raw water supplies. However, the personnel operating the system have been able to cope wi th the quality problem because of their experience and expertise. 6.2 RIVERHEAD MUNICIPAL SYSTEM As shown on Figure 6-1, the Riverhead Municipal System (Riverhead Water District) serves an area north of the Peconic River between Calverton and Aquebogue. The system presently has about 53 miles of piping and provides service to 2,124 residential, 346 commercial and 57 institutional customers. Present population served includes 9,900 within the District and 2,200 outside of the District (Riverside Water District) for a total served population of 12,100. There are two elevated water storage tanks within the system; one having 750,000 gallons of active storage and the other 150,000 gallons. The larger tank automatically controls the operation of five of the District's six well pumping stations by use of a telemetering system that serves to sequentially start or stop the District's pumping stations according to the water level in the tank. The District presently has six well supplies; their capacities and depths are summarized in Table 6-2. Well no. 2, which is not automatically controlled by tank water level, is used for emergency purposes, and is powered by a diesel engine. Excluding this well, the total net available system pumping capacity is 5,150 gpm or 7.4 million gallons per day (m§d). Based on the estimated year 2000 maximum daily demand of 2900 gpm for River- head, it is apparent that sufficient well capacity exists to satisfy future requirements. Reportedly, there have been no significant water quality problems or concerns at the well supplies. Chlorine for disinfection and lime for pH control are added at each well, except for well no. 2 where only chlorine is added. 6-4 TABLE 6-2 SUN~4ARY OF WELL SUPPLIES FOR THE RIVERHEAD WATER DISTRICT Well No. Pump Capacity (gpm) Well Depth (ft) 1 2 3 4-1 4-2 5 TOTAL 750 800 1,000 1,000 1,200 1,200 5,950 120 140 120 715 390 25O 6-5 The system appears to be well maintained and efficiently operated. It is essentially 100 percent metered, with a meter replacement pro- gram recently being completed. Leakage detection and repair is an ongoing practice, as is hydrant and valve maintenance. These prac- tices are the major reason why the unaccounted-for water portion of the total average daily production is below 10 percent, a relatively low amount and representative of a well-maintained system. 6.3 PRIVATE WATER SYSTEMS Excluding the Riverhead and Greenport water districts, 21 privately owned water supplies were identified as serving residential areas in the towns of Riverhead and Southold. In all, an estimated population of approximately 5,400 is served by these systems. Under New York State Law (State Sanitary Code Pt. 5-1.1) these sys- tems are classified as community or noncommunity systems based on the number of service connections or seasonal nature of the service. Those designated as community systems are required to meet more stringent monitoring and quality requirements. Several of these water systems have experienced problems related to lack of standby power, water quality deterioration, and inadequate pressure. Since these difficulties may directly impact upon the availability and/or suitability of water, the Drinking Water Supply Section of the Suffolk County Department of Heal th Services has de- signated some of these systems as Marginal Water Suppliers {July 1981). Community water systems serving mobile home parks (M.H.P.), adult homes and apartments were excluded from this classification under the assumption that these facilities could be evacuated in the event of an emergency. Table 6-3 summarizes the water supply equipment, populations ser- viced, and classifications of each of the private supplies on the North Fork. Comments have been included describing the reasons for "marginal supply" classification, as appropriate. 6-6 TABLE 6-3 NORTH FORK PRIVATE WATER SUPPLIES Service Population Connections Served No. Well Well Wells Depth Capacity (ft) (gpm) Storage Marginal Community Capacity Water Water (gal) Supplier System Comments Wading River Waterworks 53 Wildwood Shores Assoc. 18 Sound Shore Club 59 Hulse Farms 74 Woodcliff Park 200 Baiting Hollow Cond. 38 Cliff & Ed's M.H.P. 23 Rabbit Lane Assoc. 24 Browns' Hills Estates 19 Oakwood-on-the- Sound 99 Herod Pt. Assoc. 22 Ramblewood M.H.P. 128 Thurm's Estates 208 Roanoke Water Co. 54 Reeves Beach Water Co. i84 30O 33 30O 30O 85O 180 lO0 45 5OO 8O 210 450 200 650 1 65 60 i 170 30 2 155 225 2 270 45 2 110 220 1 115 25 1 2 35 25 2 50 25 4 I 150 50 2 95/125 200 2 130 300 2 160/180 40+ 2 165 165 3500 X 6000 X 3000 1000 X X 1500 X 1500 1000 X 4000 X 1000 X 4000 6000 1500 X 5000 X 1,2,3 2,3 1,3 1,3,4 1,2,3,5 3,5 2,3,5 X 1,2,3 X X X 1,3,5 X 3,5 TABLE 6-3 (continued) NORTH FORK PRIVATE WATER SUPPLIES SeKvice Population No. Connections Served Wells Well Well Storage Marginal Community Depth Capacity Capacity Water Water (ft) (gpm~ (gal) Supplier System Comments Rollin M.H.P. 111 220 2 Aquebogue M.H.P. 50 120 2 Peconic River M.H.P 44 90 1 J.G.F.M.H.P. 19 40 2 Capt. Kidd Water Co. 154 580 2 Little Flower Inst. 150 70 300 2000 X 75 2000 X 110 200 100 I00 7500 X X 1,3,6 1. One-Man Operation 2. Single Well System 3. No Standby Power 4. Significant Water Quality Deterioration 5. Nitrates > 10 ppm 6. Inadequate pressure SECTION 7.0 WATER USE AND AVAILABLE SUPPLY 7.1 EXISTING AND FUTURE WATER USE Water use patterns were analyzed for residential use (including commercial and industrial), and agricultural use. 7.1.1 Residential Use Water for residential use is supplied through both public water sup- ply systems and individual home wells. The major municipal public supplies are the Riverhead and Greenport water districts. In addi- tion to the two major municipal systems, there are a n~nber of smaller systems serving residential developments and mobile home parks in the study area. Operating records from Riverhead and Greenport were reviewed to determine water use in the two districts. The results of the analy- sis, shown in Table 7-1, indicate that the Greenport District uses an average of 0.81 million gallons per Gay Imgd) or 110 gallons per capita per day (gpcd). The Riverhead District supplies 1.18 mgd or 120 gpcd to consumers in the study area and an average of 0.32 mgd to the Riverside section of Southampton. Average daily per capita usage was also examined .during the periods from January to April and October to December. Usage during these periods more accurately rejects domestic use by permanent residents. Both the Riverhead and the Greenport systems rem ected per capita use of 90 gpcd during these months. The relationship between annual average daily use and peak seasonal, monthly and daily demands were evaluated in both systems and found to be almost identical. Average daily use during the peak season (May to September) is 1.25 times annual average daily use. The peak month for consumption is July and average daily use during that month is 1.6 times the annual average. The peak daily demand experienced was 2.7 times the annual average daily use. Residential use in areas not served by a public supply was estimated to be 80 gpcd. The lower per capita rate in these areas (80 versus 110 and 120 gpcd) results from two factors: 7-1 TABLE 7-1 EXISTING WATER USAGE GREENPORT AND RIVERHEAD WATER DISTRICTS(i) Water Usage Average Daily Usage (Annual Basis) Population Served in Study Area Average Per Capita Use Average Daily Usage During Irrigation Season (May-September) Average Daily Usage During Peak Month Peak Day Usage Greenport District Riverhead District 0.81MG 1.50 MG (2) 1.18 MG (3) 7,225 9,635 110 gpcd 120 gpcd 1.02 MG 1.28 MG (July) 2.22 MG (1980) 1.90 MG 2.40 MG (July) 4.01MG (1980) (1) All figures except Peak Day are average during period 1979-1981. {2) Includes water supplied to Riverside Water District. (3) Supplied to the study area. (a) The water supplied to commercial development was included in the per capita rate calculated for the districts; commercial development outside of the water districts' service areas is limited. (b) The per capita rates for the water districts include use by summer visitors. This use was estimated separately in areas not served by public supplies, using a rate of 50 gpcd extending over a 90-day period (from June through August). The population projections presented in Section 3.0 were used to estimate the future residential water use shown in Table 7-2. The per capita rates w~re held at current levels in projecting future use. 7.1.2 Agricultural Use Irrigation rates required for each type of crop grown on the North Fork as well as estimates of the total acreage devoted to each type of crop w~re obtained from the Cooperative Extension Service {see Table 7-3). The average irrigation rate was then calculated to be 140,000 gallons per acre per year. Total irrigation requirements were developed by applying these rates to the total amount of agricultural land. Crop irrigation is normally required from May through September. Peak months are July and August, when 70 percent {35 percent during each month) of the total annual irrigation demand is required. Peak daily agricultural usage is assumed to be equal to the average daily demand during these peak months. 7.1.3 Total Water Use Total annual average water use {current and future) was obtained by adding the residential and agricultural usage in each zone (see Table 7-2). Currently, 15.35 mgd is used on the North Fork. Agricultural use accounts for 11.33 mgd or 74 percent of the total annual use. Residential water use will increase in the future as the population increases. However, this increase will be offset by the decline in water used for irrigation as the amount of land in agricultural use decreases. These changes will result in a slight overall decline in average daily water use over the planning period, although the domand for potable water will increase. Consumptive water use takes place when water is used and not returned to the groundwater system. Consumptive losses include evapotrans- piration from crop and lawn irrigation and residential water not re- 7-3 TABLE 7-2 AVERAGE ANNUAL WATER USAGE Existing Use (1980) (mgd) Future Use (2000) (mgd) Zone Residential Agricultural Total Total Residential Agricultural Total Total Consumptive Consumptive Consumptive Usage Consumptive Consumptive Consumptive Usage I 1.70 3.76 4.70 5.46 2.25 3.06 4.29 5.31 2 0.76 3.75 3.90 4.51 0.97 3.06 3.25 4.03 3 0.97 3.13 3.30 4.10 1.18 2.80 3.03 3.98 4 0.51 0.24(1) 0.50 0.75 0.59 0.04 0.38 0.63 5 0.08 0.45 0.47 0.53 0,11 0.35 0.37 0.46 TOTAL 4.02 11.33 12,80 15.35 5.10 9.31 11.32 14.41 (1)The relatively large projected decrease in agricultural consumptive use in Zone 4 (0.24 mgd to 0.04 mgd) is due to the anticipated significant reduction in agricultural activity in this part of the study area over the next 20 years. Crop Type TABLE 7-3 IRRIGATION REQUIREMENTS FOR CROPS GROWN ON THE NORTH FORK (1) Irrigation Requirements (3) Acreage (gallons per acre per year) Potatoes 12,000 125,000 Mixed Vegetables (2) 4,000 205,000 Cabbage and Cauliflower 3,500 205,000 Rye 2,000 .... Nursery Stock 2,000 165,000 Pastures 1,000 20,000 Sod 1,000 245,000 Sweet Corn 800 125,000 Fruit Trees 600 125,000 Grapes 350 40,000 Greenhouses 25 815,000 TOTAL 27,275(4) 140,000 (Weighted Averse) (1) Communication with D. Fricke et al. from Suffolk County Cooperative Extension Service. (2) Peppers, spinach, beans. (3) All figures rounded. (4) Acreage was estimated and does not coincide exactly with values listed in Table 3-4. charged to the ground (i.e., discharged to a sewer system with sur- face water outfalls). Therefore, all agricultural use is considered as consumptive use (leaving the system as evapotranspiration). Re- sidential use in sewered areas was assumed to be 100 percent con- sumptive, while residential use in unsewered areas was estimated to be 20 percent consumptive. Table 7-2 lists existing and future con- sumptive use figures for each zone. As with total use, consumptive use will decrease in the future as irrigation requirements decline. Peak water use rates were developed by applying the previously deter- mined peak use factors to the future annual average daily use rates. During summer months, domestic use will be over 6 mgd; agricultural use will be about 22 mgd. 7.2 AVAILABLE GROUNDWATER SUPPLY Groundwater reaches the aquifers underlying the North Fork from two sources: (1) annual recharge from precipitation and (2) groundwater in,ow from outside the study area. Of these two sources, annual recharge provides the majority of water reaching zones I and 2, and essentially all of the water in zones 3, 4 and 5. Groundwater in, ow from Brookhaven Township to zones I and 2 is estimated at 7 mgd. (CPWS-24, 1970). The aquifers underlying zones 3, 4 and 5 are iso- lated and do not receive in,ow from other groundwater sources. Average annual recharge is estimated to range between 19 and 21 inches per year or 42 to 46 percent of the total average preci- pitation (45 inches per year). The average recharge of 19 to 21 inches is equivalent to approximately i mgd per square mile of land area. In reality, however, not all of the recharge should be withdrawn for use. Some recharge must be all owed to mow across the fresh water-saltwater interface to prevent the movement of the inter- face and saltwater intrusion. CPWS-24 contains the following permissive sustained yields for the groundwater supply zones on the North Fork: Groundwater Supply Zone Permissive Sustained Yield (mgd/sq mile) I 0.7 2 0.4 3 0.35 4 0.25 5 0.25 The permissive sustained yield is the maximum rate at which water may be withdrawn from the aquifer perennially without bringing about the undesirable effects of saltwater intrusion. The permissive 7-6 sustained yields were developed during the preparation of CPWS-24 through an optimization procedure that balanced maximum withdrawal with storage capacity in each zone. Since groundwater elevations and aquifer storage volumes in zones 3, 4 and 5 are lower than in zones 1 and 2, the permissive sustained yields for these eastern zones were developed utilizing recharge rates reflecting drought conditions rather than the average recharge rates used for zones i and 2. The permissive sustained yields for each zone were examined in this study. Using the most recent data and information available, the impacts they would have on the position of the saltwater interface were estimated. It was found that these yields would have only a minor impact on the configuration of the fresh water-saltwater inter- face. In order to estimate the total quantity of groundwater that may be withdrawn for public supply from each zone, water budget areas were delineated (see Figure 7-1). The budget areas represent those loca- tions where there is sufficient groundwater to develop large public supply wells. Outside the budget area, there is only sufficient supply to allow the development of smaller home well systems. In zones i and 2, the budget areas were defined as those locations where the groundwater level is 5 feet or more above sea level. In zones 3, 4 and 5 the availability of groundwater is more limited; therefore, the budget areas in these zones were defined as the areas where the groundwater level is 2 feet or more above sea level. Allowable withdrawal rates and well capacities will be more limited in the three eastern zones because of the lower groundwater elevat- ion. The definitions of the water budget areas are consistent with those presented in CPWS-24. Table 7-4 presents summary data for the water budget areas in each groundwater supply zone. As one progresses west to east through the study area, the size of the water budget area generally decreases. Zone i has a water budget area of 42 square miles (approximately 95 percent of the total land area). In Zone 5, the budget area covers 1.6 square miles, representing only one-third of the total land area. These areas were based upon interpretation of the March 1981 ground- water contour map prepared by the SCDHS. Total permissive sustained yields were derived by applying the unit- yield rates listed above (mgd per square mile) to the area covered by the water budget areas. The total permissive sustained yields range from a high of 29.4 mgd in Zone 1 to a low of 0.4 mgd in Zone 5. It should be emphasized that these total permissive sustained yields only account for the quantity of water that may be withdravm from the water budget areas. Additional groundwater supply may be obtained from small er capacity, private home wells located outside the budget areas. It is estimated that approximately 10 to 20 mgd could be available outside the water budget areas for all five zones. 7-7 ii r---]-- SHADED AREA OUTSIDE WATER BUDGET BOUNDARY Figure 7-1 Water Budget Area TABLE 7-4 SUMMARY OF WATER BUDGET ANALYSIS Zone Water Permissive Present Potentially Total Budget Sustained Consumptive Available for Area Area Yield Use Future Use (sq. miles) (sq. miles) (mgd) (mgd/sq. ~ile) (mgd) (mgd/ ... I 44 42 29.4 0.7 4.7 24.7 2 28 14 5.6 0.4 3.9 1.7 3 24 14 4.9(1) 0.35 3.3 1.6 4 7 3.4 0.9(1) 0.25 0.5 0.4 5 5 1.6 0.40(1) 0.25 0.47 0(2) (1) Since the underlying aquifers in these zones have insufficient storage, these values are conservatively based on drought conditions and would be larger for a year of average precipitation. (2) The zero entry indicates that the present consumptive use is approximately equal to the permissive sustained yield in Zone 5 during drought conditions. 7.3 WATER BALANCES Table 7-4 compares the quantity of available groundwater with current consumptive water use in each zone. Each of the zones {except Zone 5) potentially has sufficient quantities of water to meet anticipated demands without causing adverse impacts from saltwater intrusion. Zone i has the largest surplus of water available (24.7 mgd); zones 2, 3 and 4 have a more modest surplus. In Zone 5, the analysis indi- cates that the current consumptive use is approximately equal to the permissive sustained yield. This represents a critical water supply condition in Zone 5. As explained later in this report, particular attention must be paid to future development patterns in Zone 5 to ensure that future consumptive use does not exceed permissive yield. However, it should be emphasized again that there is additional water available for use in Zone 5 outside the water budget area that can be developed utilizing small capacity wells. In reviewing the findings of the water budget analysis, several important factors must be considered: (1) Well location and design require special attention in ground- water supply zones 2, 3, 4, and 5 because of the low per- missive sustained yields and shallow aquifers. (2) Because of pesticide and nitrate contamination, all fresh groundwater potentially available may not be suitable for pot- able water supply without treatment, but could be used for irrigation. In the tained future development of water supply alternatives, the permissive sus- yields and the quantity of water potentially available for use were utilized to identify possible sources of supply. 7.4 WELL YIELDS AND SPACING CONSIDERATIONS The permissive sustained yields presented earlier represent average withdrawal rates that are applicable across an entire groundwater supply zone. Individual well capacities and well spacing must also be considered when developing specific water supply alternatives. 7.4.1 Well Yields The yield that can be obtained from a well is dependent upon a n~nber of interrelated factors: 7-10 Specific capacity of the aquifer Height of the water table above sea level - Proximity to saltwater bodies - Well screen location and length Well finishing and development techniques. While specific information can best be obtained from site investi- gations and pump tests, sufficient information currently exists to make planning decisions on well spacing, location and yields. The Suffolk County Test Well Program and data developed by the USGS indicate that the specific capacity for wells in the upper glacial aquifer averages 30 gallons per minute per foot (gpm/ft) of drawdown. (See Section 4.2 for a complete discussion of aquifer parameters.) This means that properly constructed wells can generally yield 30 gpm for each foot the water level in the well drops after pumping com- mences. Water table wells located close to saltwater bodies should not be designed and operated in such a manner that the continuous, average drawdown level over the year is below sea level, as this will induce saltwater intrusion into the well. As a general rule, wells are designed to meet peak day requirements (generally two to three times annual average pumping rates). There- fore, the well only pumps about 50 percent of the time to meet the annual average demands. It is, therefore, permissible to design a water table well so that it will have an intermittent pumping level below sea level, realizing that during the inactive pumping periods the well will recover. It is typically assumed that if a well is located several thousand feet from a saltwater body, it can be designed to have a drawdown twice the height of the static water level above sea level, and still operate safely without drawing saltwater from lateral or upconing intrusion. Such a well should not be pumped continuously; it should be used only 50 percent of the time and allowed to recover the other 50 percent of the time. Allowable well capacity can then be calculated: (1) if the height of the water table is known, (2) assuming a 30 gpm/ft specific capacity, and (3) al~lowing a total drawdown of two times the static water level above sea level. In zones I and 2, the water table ranges from 5 to 40 feet above sea level. Wells in the water budget areas within these zones can, therefore, be designed to provide 300 gpm to over 1,000 gpm. Wells in the 1,000-gpm capacity range, however, are not common in the North Fork and should be discouraged. A maximum well capacity in zones I and 2 should be approximately 750 gpm. In the water budget areas in zones 3, 4 and 5, the height of the water table ranges from 2 to 4 feet above sea level. Wells in these zones can be designed for capacities of 120 to 240 gpm. 7-11 The larger seasonal and weekend increases in the water supply re- quirements of the North Fork further complicate the design of wells. If the well-sizing criteria discussed above were followed, a large number of small-capacity wells would be required to meet peak weekend summer demands. Many of those wells would be completely inactive during the week and the off-peak season. This type of well network, comprised of numerous small wells, would be prohibitively expensive to construct and operate. It was, therefore, decided to allow peak- ing wells to pump intermittently at drawdowns up to 3 to 4 times greater than the static water level above sea level. The larger wells would only be operated on peak weekends during the summer. Since their period of operation will be limited, there would be no danger of drawing saltwater into the well. The well yields and negative saltwater impacts predicted through this type of analysis were verified by examining the existing installed capacities and saltwater impacts of the Greenport Water District wells. (The Riverhead Water District wells tap the Magothy aquifers in Zone I and are not subject to saltwater intrusion and, therefore, could not be used to further verify the assumptions.) The Greenport wells range in capacity from 200 to 400 gpm. The small er capacity Greenport wells produce yields which are consistent with those pre- dicted by the assumptions made in this study with no saltwater intru- sion problems. The 400-gpm wells, however, are larger than would be allowed by the specific capacity/drawdown criteria of this study (drawdowns greater than twice the height of the static water level above sea level). A review of the Greenport records indicates that their larger wells must be periodically rested because of increasing high chloride concentrations. In essence, then, the larger wells are sized and function similarly to the peaking wells assumed in this study, and substantiate the assumption that larger wells can be used on an intermittent basis to satisfy peak summer weekend requirements, if carefully monitored and controlled. Although the well-yield analysis has been presented for the water budget areas in each zone, the same general analytical approach can be used for locations outside the budget area. In summary then, wells tapping the upper glacial aquifer can be pumped intermittently at the following capacities without drawing saltwater: Height of Water Table Above Sea Level (feet) Capacity (gpm) 5 300 4 240 3 180 2 60-120 1-1/2 45-90(1) 1 15-30(1) Less than I 0-10(1) (1) Requires site specific investigations and analyses. 7-12 7.4.2 Well Spacing - Budget Area(1) Wells should be located and spaced based upon considerations of the effective radius of influence of each well. A well should not be located closer to a saltwater body than a distance equal to two times its effective radius of influence. Furthermore, wells within a group should be separated by a minimum of two radii of influence (center- to-center) so that their areas of influence do not overlap. These spacing criteria were used in subsequent phases of this investiga- tion. For steady-state conditions, the radius of influence of a pumping well describes an area over which the recharge rate is equal to the well pumping rate. This relationship is described by the following equation: QW= where, QW = effective well r 0 W ro2 W (2) pumping rate in gallons/minute = radius of influence in feet = recharge rate in gallons per squ~re foot per minute (18 inches per year: 2.13 x 10-~ gallons per square foot per minute) To normalize the pumping rate over the year, Qw was taken to be equal to the total annual quantity of water withdrawii from the well distri- buted evenly over the entire year (effective well pumping rate). Using the effective pumping rate and an average recharge of 18 inches/year, ro was calculated and used to define minimum well spac- ing distances, as described above. Example. As an example of the well-yield analysis and the use of the a o-~-6~-equation for well spacing, assume that a total well capacity of 900 gpm was required to satisfy a peak day requirement for a particu- lar service area where the groundwater table was 5 feet above sea level. Three wells would be required, each with a capacity of 300 gpm, based on a specific capacity of 30 gpm/foot and all owing each well to draw down to a level twice the static level above sea level {2 x 5 feet: 10 feet). Using a recharge rate, W, of 18 inches/ year and Qw equal to 300 gpm, r was calculated to be 2,050 feet. These well§ were then located 4~100 feet apart. {1)This analysis pertains to well spacing within the water budget area only. The investigation did not evaluate well spacing or yields outside of the water budget area. These latter con- siderations are very site specific and beyond the scope of this study. (2) Groundwater Technology, David K. Todd, 1964. 7-13 SECTION 8.0 WATER SUPPLY ALTERNATIVES 8.1 WATER DEMAND CENTERS The initial step in developing and analyzing water supply alterna- tives was to define the location and magnitude of water requirements. Accordingly, water demand centers v/nich represent the greatest con- centrations of population and/or commercial development within the study area were delineated. As shown on Figure 8-1, five major demand centers were identified. Within each demand center, subdemand centers wore also delineated; these represent the major concentra- tions of population within the larger demand centers. The future land use plan, as projected by the Long Island Regional Planning Board (LIRPB) during the 208 Program, was utilized to estimate the extent of future development and, accordingly, subdemand center boundaries. Separating each demand center into subdemand centers was primarily done for water supply planning purposes because it facili- tated development and analysis of a more localized level of altern- atives for comparison with larger subregional or regional options. Nine isolated neighborhoods that represent concentrations of popu- lation removed from the demand centers were also identified (see Table 8-1). They generally consist of single-family homes built as part of a subdivision, with each home having an individual well sup- ply. These neighborhoods were identified in order to develop and analyze water supply options that would be available at a smaller, neighborhood scale. The homes within these neighborhoods have wells that withdraw water from the upper glacial aquifer, which is, or has the potential to be, laden with objectional concentrations of nitrates and/or pesticides. Table 8-1 summarizes the estimated year 2000 populations and water consumption requirements for each subdemand center and isolated neighborhood. Future population was estimated, as discussed in Sec- tion 3.0, with detailed census tract data and LIRPB estimates uti- lized to apportion population among demand centers. Future water consumption was estimated by the methods presented in Section 7.0. Generally, historical water use trends wore utilized to develop the following estimates of water consumption given in gallons per capita per day (gpcd): 8-1 WADING RIVER/NORTHVILLE SOUND AVE MATTITUCK/CUTCHOGUE PECONIC RIV[R NORTHVlLLE RIVERHEAD/JAMESPORT GREENPORT/ SOUTHOLD ORIENT Figure 8-1 Major Demand Centers TABLE 8-1 ESTIMATED YEAR 2000 POPULATION AND WATER CONSUMPTION ESTIMATED YR 2000 POPULATION ESTIMATED YR 2000 CONSUMPTION, GPM DEMAND CENTER PERMANENI ADD'L SUMMER(I) TOIAL SUMMER AVE DAY MAX DAY PEAK HOUR Wadtn9 Rtver/Northville Wading River 5,655 3,271 8,926 430 900 1400 Baiting Hollow/Woodcltff 2,025 2,144 4,169 190 355 525 Park Reeves Park 1,915 821 2,736 135 300 450 Total 9,959 ~ 15,831 ~ Rtverhead/Ja~sport Riverhead/Aquebogue 12,800 7,018 19,818 1070 2900 3650 Calverton 500 216 716 35 80 20 Jamesport 1,959 849 2,808 140 300 465 Total ~ ~ ~ 1,245 Mattltock/Cutchogue Mattitock West 3,148 3,065 6,2t3 280 540 80 Mattttuck East 1,748 1,730 3,478 160 300 50 MAttttock South 2,886 2,841 5,727 260 500 740 Little Hog Neck 858 835 1,693 75 150 22 E. Cutchogue/Little Creek 692 673 1,365 60 120 175 Fleets Neck/Cutchogue 1,218 1,185 2,403 110 210 10 New Suffolk 650 535 1,085 50 95 140 Indian Neck 317 310 627 30 55 80 Total 11,~1! ~ ~ 1,025 Southold/Greenport Great Rog Neck 540 2,540 3,080 120 165 210 Greenport 8,601 8,885 17,486 660 1780 200 East Marion 867 843 1,710 80 150 220 Total ll)~Ol)l~ ~ 2~¥LrTG' ~ Orient 1,133 1,152 2,285 100 200 295 TOTAL DEMAND CENTERS 47,412 38,913 86,325 3,985 (5.74 MGD) Neighborhoods 1. Lake Panomoka 294 127 421 20 45 70 2. Rt. 25 S. of Scutle Hole 110 48 158 8 17 26 3. Rt. 25 N. of Grumman Airport 385 167 552 27 60 90 4. Baiting Hollow 110 48 158 8 17 26 S. Off Tuthill Lane 181 78 259 13 28 43 6. Duck Pond Point 135 58 193 10 21 32 7. Rt. 27 near Depot Lane 83 35 118 6 13 20 8. Goldsmith Inlet 260 110 370 18 40 62 9. Great Pond 330 140 470 23 50 80 TOTAL NEIGHBORHOODS 1,888 811 2,699 133 TOTAL DEMAND CENTERS & NEIGHBORHOODS 49,300 39,724 89,024 4,118 (5.93 MGD) (1) Due to continual updating, these estimates may vary slightly from those presented in Table 3-3. These variations will not impact upon the conclusions developed in this analysis. (SPH5/43) m m mmm m 80 gpcd - Residential use in areas not presently served by public supply 110 gpcd - 120 gpcd - Greenport Municipal System permanent residents Riverhead Municipal System permanent residents 50 gpcd - Summer residents. The above rates were used to estimate year 2000 average daily water use. Maximum daily water use, which is used to determine the design capacity of most waterworks facilities, was estimated using a maximum day to average day ratio (i.e., maximum day ratio) of 2.5, based on historical water use trends for the Greenport and Riverhead municipal systems. The maximum day use represents the largest 24-hour water use that is expected to occur within a year, usually during the sum- mer season when water use is higher. The maximum day ratio was applied only to the average daily water use for permanent residents, since summer residents would, most likely, use water at a more con- stant rate throughout the season, with lawn watering and car washing being less prevalent within this class of users. The adopted rate of 50 gpcd for summer population is, most probably, slightly conserv- ative and should be more than adequate to account for any fluctua- tions in average daily water use. Peak hourly water use was estimated using the same method that was established for maximum day use, but using a peak hour ratio of 4.0. Peak hour represents the largest water use expected to occur over a 1-hour period during the maximum day (i.e., the highest hourly flow expected to occur within a year), and was used in this study to size pumping facilities for systems that do not have available system storage. 8.2 WATER SUPPLY CONCEPTS Based on the present groundwater quality problems associated with the upper glacial aquifer, as discussed earlier in this report, the need for water supply planning is apparent. According to historical area groundwater quality data and the resulting mapping that was developed to show the extent of nitrate and pesticide contamination, it is evi- dent that most individual home well systems throughout the study area presently have, or in the future, may have objectional levels of nitrates and/or pesticides. The Greenport Municipal System has ex- perienced high levels of nitrates and pesticides, in addition to iron, manganese and chlorides. Additional water supply is also re- quired in Greenport. The Riverhead system has an adequate, high- quality supply primarily due to the utilization of water from the Magothy aquifer. Accordingly, water supply planning, from a quality and quantity standpoint, is not required for this system. However, recommendations associated with distribution system storage and con- tinuation of the present operation and maintenance practices will be addressed later in this report. 8-4 In order to evaluate all avail able options that would provide ample and high-quality water to North Fork residents, several water supply concepts were developed. These essentially represent a building block approach ranging from individual home well treatment systems to large-scale regional systems. Five levels of water supply alterna- tives were developed and evaluated: TABLE 8-2 WATER SUPPLY CONCEPTS Level I: Level II: Level III: Level IV: Level V: Individual Home Water Supply Systems Neighborhood Systems Subdemand Center {"Community") Systems Subregional Systems Regional Systems 8.2.1 Level I Alternatives Several Level I alternatives were developed and evaluated: Home treatment units Bottled water Local delivery of bulk water Local supply of bulk water {with treatment). 8.2.1.1 Home Treatment Units. Reverse osmosis (R-O) was chosen as the preferred method of treatment in individual home systems over ion exchange with and without carbon adsorption. Reverse osmosis can efficiently remove the three types of contaminants potentially pesent in the area's groundwater - nitrates, pesticides/herbicides and salt. Ion exchange would only remove the inorganic contamination; carbon adsorption would have to be used for the pesticides/herbicides. Reverse osmosis is a proven technolo!tY and units are currently avail- able at various capacities. The equipment is cost competitive with other treatment processes. The process is not subject to break- throughs as is carbon adsorption and will perform reliably between maintenance periods. (See Section 8.3 for a more complete discussion of treatment processes.) Two basic concepts were developed for individual homes: Treatment of the entire home supply Treatment of drinking/cooking supply only (kitchen tap). 8-5 m ! A1 though they are not commonly used, individual R-O systems compact enough to install in a home are available to produce up to 400 gal- lons per day (gpd) of treated water. These units are typically in- stalled in commercial establishments. For a domestic installation, water supplied from a contaminated home well would be piped to the R-O unit. A booster pump would be used to raise the water pressure to approximately 200 pounds per square inch gage (psig), the minimal working pressure of the unit. Treated water, or permeate, would pass through a pressure-reducing valve to a 50- to lO0-gallon storage tank maintained at 60 psig. Concentrated reject water, or brine, would be disposed of in a small dry well for percolation into the ground. This will not exacerbate groundwater contamination because the quan- tities of brine from home units are very minor; the brine consists only of materials that were previously present in-the groundwater; when the brine reaches the groundwater table, it will begin to dis- perse; and some of brine contaminants may be removed as the brine percolates through the unsaturated zone. Treatment of only the drinking/cooking supply would require the in- stallation of a compact under-sink unit. These units are capable of treating 5 to 20 gallons a day. In an installation of this type, water would be supplied to the entire home through an individual home well. The R-O unit, installed under the kitchen sink, would be tap- ped into the cold water supply line. Normal house pressure (30 to 50 psig) can operate these units with no booster pump required. The treated water is stored in a small hydropneumatic tank, which is con- nected to a third faucet. Reject water is piped directly to the kit- chen sink drain line for disposal. The unit will be serviced every six months to replace filters, clean the membrane, and collect quality control samples. 8.2.1.2 Bottled Water. As an alternative to home treatment units, a home supply can be augmented by periodic delivery of bottled water. It was assumed that the daily consumption of bottled water (for drinking and cooking purposes) would be limited to i gallon per day {gpd) per person. Although average levels for individual consumption are typically much higher, the high cost and restricted availability of bottled, as opposed to tap, water was assumed to restrict consump- tion. Bottled water would be delivered once or twice a week in re- turnable, plastic containers of 5-gallon capacity. A deposit would be required on each bottle delivered to ensure return of the con- tainers. A bottle stand, rented monthly from the water supplier, would dispense the water. 8.2.1.3 Local Delivery of Bulk Water. For this concept, a 5,000- gallon tanker truck would be employed to deliver water purchased from the Riverhead water supply system. The tanker would be driven to a central community station, where it would remain until depleted. At that time, it would be replaced with a full tanker and towed to Riverhead for refilling. Residents would be required to transport 8-6 I I I I I I I I I I their own water. It was assumed that the 5,000-gallon tanker would be replaced once a week, each household would make 120 roundtrips (at 6 miles} per year, and the community station would be located within 35 miles of the Riverhead Water District. Each station would serve approximately 1,000 people. 8.2.1.4 Local Supply of Bulk Water (With Treatment). In areas of groundwater contamination, the alternative of providing a treated supply of bulk water at a central source was investigated. A central supply of groundwater would be treated to remove nitrates and pesti- cides, and stored in a supply tank for distribution. As above, the residents would be required to pick up their water for drinking and cooking purposes. Each station would also serve about 1,000 people. 8.2.1.5 Individual Home Wells. It is important to note that in all Level I water supply concepts, even those which require the importa- tion of potable water, an individual home well system is required for each residence. For existing homes in areas of contamination, the well system would continue to be required either as a source of non- potable water and/or to supply an individual treatment system. Each well system would consist of the following major components: Well casing and screen Pump and motor Storage/pressure tank Piping and valving Electrical equipment and controls. Well systems in different areas of the North Fork have different well requirements. In Wading River, for example, a 200-foot well depth is typical, requiring a 4-inch casing and a submersible pump. In River- head, depth to water averages 100 feet and a 2-inch casing with a jet pump is adequate. Southold Township wells average 75 feet in depth; a 2-inch casing with a jet pump is also adequate for these installa- tions. Typically, the service life of a 4-inch well is about twenty years: a 2-inch well system would require replacement every 10 years. For comparative purposes, it was assumed that a 4-inch well system would require a replacement pump every 10 years, while a 2-inch system would require a new casing and screen to be driven and the pump to be replaced every 10 years. 8.2.1.6 Level I Cost Estimates. Costs for individual home well sys- tems were developed by averaging costs for new installations as ob- tained from local North Fork drillers. For 200-foot-deep, 4-inch wells typically installed in an area similar to Wading River, a com- plete installation averages $4,500, including pump, pressure/storage tank and all necessary piping and valves. Shallower wells, requiring a 2-inch casing, averaged $1,600 IlO0-foot depth) and $1,200 (75-foot 8-7 depth). Some areas may utilize slightly less expensive suction pumps in the well system. This should not significantly impact the accura- cy of the cost analysis. For these new installations, an amortized cost was developed under the assumption that the capital cost would be included in the cost of the home and resulting mortgage loan. Loan terms of 14 percent for 30 years were used for these calcula- tions. Additional annual costs include power, routine maintenance (estimated at $70 a year), insurance, and well replacement. An addi- tional insurance premium of $25 a year is typically charged for lack of fire protection in areas which are not served by water systems, for fire protection. A sinking fund factor was applied (10 percent, 10 years) to annualize the $750 pump replacement cost for 4-inch wells expected every 10 years. An identical factor was applied to the cost of replacing the 2-inch well casing and jet pump (estimated at 60 percent of first cost) every 10 years. Costs for existing homes presently served by individiual well systems were developed by subtracting the amortization of the capital cost from the total annual cost of a new well system. A summary of annual costs for new and existing homes in areas where well supplies do not require treatment is presented in Table 8-3. A new home in an area with a depth to water of 200 feet would incur an annual cost of $810. For an existing home with a well system, annual operating costs were estimated to be $155 to $172. Home owners in areas requiring treatment of groundwater supplies (using R-O systems) must incur additional costs to pay for the capi- talization, operation and maintenance of these systems. Under-the- sink R-O systems treating 8 to 20 gallons a day have an average capi- tal cost of $650. Annual operation and maintenance (O&M) costs for these systems include two service calls for replacement of filters and membrane cleaning. R-O units capable of serving entire homes were assumed to have a capital cost of $1,800. This figure may be significantly higher for homes which require additional storage and/or flow capacity to meet peak demands. Several capital cost estimates were obtained from different manufacturers - these ranged from a low of $1,800 to a high of $3,700, depending on the capacity of the unit and quality of con- struction. The lower value was used in this analysis because, in actual practice, most homes would use an under-the-sink unit (costing $600) instead of a central unit and, therefore, the $1,800 figure re- presented an approximately average value for comparative purposes. If the $3,700 figure had been used, it would not have affected the cost rankings and selection of alternatives. These units also re- quire two service visits a year. 8-8 TABLE 8-3 COST ESTIMATE INDIVIDUAL HOME WELL SYSTEMS (NO TREATMENT) Well Depth {ft)/ Diameter (in.) Installed Cost $ Amortized Cost(l) S/Yr. O&M(2) S/Yr. Total S/Yr. A. New Homes B. Existing Homes 200'/4" (e.g., Wading River) 4,500 100'/2" (e.g., Riverhead) 1,600 75'/2" (e.g., Southold Township) 1,200 200'/4" (e.g., Wading River) NA 100'/2" (e.g., Riverhead) NA 75'/2" (e.g., Southold Township) NA 645 230 170 NA NA NA 165 172 155 165 172 155 810 402 325 165 172 155 (1) Assumed to be included in mortgage; 14%, 30 years. Includes power; sinking fund (10%, 10 years) for pump replacement every 10 years (4") or well replacement every 10 years (2"), insurance. (122/8) In order to develop an annualized cost of individual systems which include treatment, it was assumed that these units would be purchased by a central authority that would own and maintain them. Costs would be passed on to the home owner. Amortized costs for these units were developed using terms typical to a municipal bond (12 percent, 40 years}. Operating costs for these systems were developed from manu- facturers' quotes. The annual O&M cost for an under-the-sink R-O unit was found to average about $100; the annual cost to maintain an entire home unit was about $170. For units serving an entire home, an additional $30 a year is required for power costs associated with the R-0 booster pump. As can be seen in Table 8-4, a new, complete individual home well system with treatment, includes the amortized costs of the well sys- tem and treatment equipment, plus the 0&M costs assigned to the R-0 unit and well system. For new homes equipped with under-sink units, total annual costs ranged from $505 to $990; using R-0 units to treat the total home supply increased these costs to $745 to $1,230 a year. For existing homes, the capitalization of the well system can be sub- tracted. This lowered the range of costs to $335 to $352 a year for under-sink units and $575 to $592 for entire home units. These costs also include the 0&M associated with the existing well system. The cost of using bottled water for drinking and cooking supply was developed using vendor quotes. Presently, water can be purchased for $1.19 a gallon, with a minimum delivery of five 5-gallon bottles. Assuming a per capita consumption of I gallon a day, annual water cost per household (average of 3.2 people) would be $1,510, including $10 a month for rental of a bottle stand. A fully refundable deposit of $100 would also be required. These costs must be added to those incurred for individual well systems, which must still be operated to supply the remaining water demand. Local delivery of bulk water to a central location incurs transporta- tion and operational costs in addition to the cost of the water. Assuming a 1-gpcd consumption, a 5,000-gallon tanker can serve 1,000 people for about 5 days. Using a bulk purchase price of $0.60 per 1,000 gallons for Riverhead water, a tanker cost of $40,000, and an annual labor cost of $30,000 per man-year, a total estimated annual cost of $200 per household was calculated. This includes a 15-year capitalization of the tanker at 12 percent, 0&M at $0.40 a mile, and a 35-mile round trip for refills. It was also assumed that each household would require 120 trips to the tanker per year at an aver- age round trip of 6 miles. Auto reimbursement was taken to be $0.25 a mile. As above, each household also would be required to maintain and operate an individual well system. Treatment of local water, with community pickup as above, would re- quire a central treatment plant and a storage tank. The cost of a facility to remove nitrates and pesticides (using ion exchange and 8-10 m m~ m m .m m m m m m m,m m .m mm m mm mm m mm TABLE 8-4 COST ESTIMATE INDIVIDUAL HOME WELL SYSTEMS (WITH TREATMENT) Well pepth/Diameter Treatment(I) Individual Well Installed ~morti'zed In"stal'l~d' Amo'rtfzed' Cost Cost(2) Cost Cost(3) O&M(4) Total System {$) (S/yr.) ($) (S/yr.) (S/yr.) (S/yr. A, New Homes 200'/4" 100'/2" 75'/2" Kitchen Tap 650 80 4,500 645 265 990 Entire Home 1,800 220 365 1,230 Kitchen Tap 650 80 1,600 230 272 582 Entire Home 1,800 220 372 822 Kitchen Tap 650 80 1,200 170 255 505 Entire Home 1,800 220 355 745 Be Existin9 Homes 200'/4" 100'/2" 75'/2" Kitchen Tap 650 80 NA NA 265 345, Entire Home 1,800 220 365 585 Kitchen Tap 650 80 NA NA 272 352 Entire Home 1,800 220 372 592 Kitchen Tap 650 80 NA NA 255 335 Entire Home 1,800 220 355 575' m m m. m m m. m m m m m m mm m. m m m m m (1) Treatment system for nitrates, pesticides and brackish water consisting of prefilter, reverse osmosis unit and carbon filter. (2) Assumed to be owned by central Authority; 12%, 40 years. (3) Assumed to be included in mortgage; 14%, 40 years. (4) Includes two service visits per year on R-O unit; power; sinking fund (10%, 10 yrs.) for pump replacement every 10 year~ (4") or well replacement every 10 years (2"); semi-annual replacement of filters, routine maintenance of well system, ~ insurance. Each service visit would be by one man for approximately one hour. One sample per year would be taken and be tested for nitrates plus one or two organics. (122/ I i i i I I carbon) at a flow rate of 25 gallons per minute (gpm) was estimated at $97,000. Capitalization of this cost, with associated O&M costs and consumer mileage costs, resulted in a per household cost of $522 year. Again, each household would require an individual well system. Based upon cost and convenience, individual home treatment units were selected as the optimum Level I alternative to compare with Level II, III, IV, and V alternatives presented in the following sections. 8.2.2 Level II--Neighborhood Systems This level of alternatives includes the following: (1) Providing a new public water supply system for the nine iso- lated neighborhoods listed on Table 8-1 (2) Upgrading the existing neighborhood water systems located in the Wading River/Northville demand center and the Mattituck West subdemand center. The required components associated with .the development of a new public water system for isolated neighborhoods include two production wells, each capable of providing the maximum daily demand. Each in- stallation would include a submersible vertical turbine well pump in- stalled in a vault. The well pumps would deliver water to a booster pumping station, which would serve to increase system pressures to desirable limits for distribution to consumers (50 psi to 60 psi) through a piping network. It was assumed that 6-inch-diameter cement-lined piping would be provided for water distribution. The booster station would house all controls, chemical addition facili- ties and standby power. A hydropneumatic tank would also be provided to assure proper system operation and control pumping cycles. In general, the neighborhood systems 1 through 5 in Table 8-1 could tap the Magothy aquifer, avoiding the need for treatment. The Magothy aquifer underlying systems 6 through 9 is brackish; thus, it was assumed that the upper glacial aquifer would be utilized, with treatment for nitrates and pesticides required. Upgrading the existing neighborhood water systems would include pro- viding additional supply, if required, and providing treatment faci- lities, since all wells presently tap the upper glacial aquifer. Supply augmentation was assumed for those systems expected to exper- ience supply deficits, based on estimated year 2000 maximum day de- mand. In addition, a minimum of two well supplies was considered a requisite; thus, a second well was provided for systems having only one at the present time. It was also assumed that the existing piping systems are adequate and their continuous use would be main- tained. 8-13 I I I i I 8.2.3 Level III--Subdemand Center Systems Level III alternatives generally consist of the development of new public water systems for each subdemand center as illustrated on Figure 8-2. Under this concept, each system would be a separate entity and essentially include components similar to those described for the isolated neighborhood systems, but at a larger scale. Simi- larly to Level II, treatment was assumed not to be required for those subdemand centers that could tap the fresh water portion of the Magothy aquifer. These include the Wading River/Northville and Riverhead/Jamesport demand centers. Also, treatment was assumed not to be required in the Little Hog Neck and New Suffolk subdemand centers. Water system components for the subdemand centers within the Wading River/Northville and Riverhead/Jamesport demand centers (excluding the Riverhead subdemand center) include production wells, well pump- ing stations, and transmission and distribution piping. Since there are several existing neighborhood systems located throughout the Wading River/Northville demand center and an existing system located in the Mattituck West subdemand center, two Level III options were developed for these areas, as follows: Level III (A) - A new public system would serve the presently unserved areas only. Existing systems would be upgraded as required (under Level II alternatives) for continued operation. Level III (B) - A new public system would serve the presently unserved areas and provide supply to existing neighborhood sys-i-6-ms. Existing neighborhood system supplies would be discontinued (or used as a back-up source), but existing distribution systems would remain in service. Two Level III options were also developed for augmentation of the Greenport Municipal System, which has been designated as a subdemand center within the Southold/Greenport demand center. They are summar- ized as follows: Level III (A) - Renovation and use of an existing farm well for municipal supply (Donohue's Farm Well, an agri- cultural well on County Rt. 48). 8-14 BAITING I WADING HOLLOW REEVES RIVER ~'IOOOCL [ FF PARK PARK :lAD I NG RIVER/NORTHVILLE CALVERTON RIVERHEAO AQUEBOGUE JA}IESPORT RIVERHEAD/JN1ESPORT MATTITUCK MATTITUCK WEST EAST FLEETS/ r~TTITUCK NECK SOJTH CUTCHOGUE LITTLE NEW HOG SUFFOLK NECK EAST INDIAN NECK MATTITUCK/CUTCHOGUE GREAT HOG , GREENPORT NECK EAST UARION SOUTHHOLD/GREENPORT ORIENT VILLAGE AND NORTH ORIENT ~'l DEUAND CENTER ~' I SUBDEMAND CENTER SYSTEM * EXCLUDED FROM LEVEL III ANALYSIS ** AUGNENTATION OF EXISTING SYSTEM Figure 8-2 Level III Subdemand Center Systems Construction of a water treatment facility for reduction of nitrates, pesticides and dissolved solids from plants 6 and 7 and the Donohue's Farm Well. Additional distribution system storage and piping. Level III (B) - Development of a water supply system at Great Pond. Continued use of existing supplies as required. Additional distribution system storage and piping. The specific alternatives as they apply to each subdemand center under Level III are described in detail in Section 8.4. Water treat- ment processes evaluated for this analysis and adopted for cost esti- mating purposes are discussed in Section 8.3. 8.2.4 Level IV--Subregional Systems Under this concept, a single public water system would be developed for each of the five demand centers. System components would include a central well supply and production facilities (multiple wells), treatment facilities as required, transmission piping and distribu- tion piping and storage facilities. For the Riverhead/Jamesport and Southold/Greenport demand centers, expansion of the existing munici- pal systems to serve adjacent subdemand centers would be analyzed under this level. The major intent of the investigation of this con- cept was to compare the differences in supply and treatment costs between Level IV and Level III, since piping costs would essentially be the same. As will be discussed in Section 8.4, it was found that detailed consideration of Level IV alternatives was not warranted, for the most part, for economic reasons, generally related to distri- bution system piping requirements and the resulting costs. However, consideration was given to development of a water system with central supply sources for the Wading River/Northville demand center for com- parison with Level III. Supplying Calverton from the Riverhead Water District was also evaluated under Level III. Figure 8-3 schemati- cally depicts the elements associated with these analyses. A detailed discussion of these analyses is presented in Section 8.4. 8.2.5 Level V--Regional Systems Regional systems evaluated for this report generally consist of the following major components: 8-16 CALVERTON /DISTRIBU'FION SYSTEH RIVERHEAD/ AQUEBOGUE (MUNICIPAL SYSTEM) METERED CONNECTION JAHESPORT RIVERHEAD/JAIfESPORT DEMAND CENTER Figure 8-3 Level IV Schematic of Subregional Systems m m - Supply wells plus production facilities from potable ground- water sources in Riverhead (Magothy aquifer) - Transmission system from supply to demand centers - Distribution system piping and storage for demand centers. The Level V analysis focused only on the Mattituck/Cutchogue and Southold/Greenport demand centers. Providing a supply from Riverhead to Wading River/Northville was not considered, since potable ground- water is available within the demand center itself; thus, a more re- mote supply could only result in a higher cost. Similarly, the Riverhead/Jamesport demand center was not considered under Level V since potable water is available from the Riverhead district which could serve Calverton (Level IV) and Jamesport. Orient was also ex- cluded from Level V primarily because of the extensive transmission system piping and pumping requirements and resulting high costs associated with providing water to this relatively remote area from a source in Riverhead. Several alternatives were developed and analyzed under Level V, rang- ing from augmentation only of the existing Greenport system, to sup- plying the entire Mattituck/Cutchogue and Southold/Greenport demand centers. In all cases, the supply sources would consist of new pro- duction wells in Riverhead. For alternatives involving Greenport, two scenarios were considered: Providing a supply to satisfy the estimated year 2000 maxi- mum daily demand, and discontinuing the use of Greenpo~ existing well supplies (except for emergencies) (2) Providing a supply to satisfy the estimated year 2000 aver- age daily demand, and continuing the use of Greenport'~ highest quality supply wells to serve as supplemental sources when required. The latter scenario is based on the assumption that the quality of water from the Greenport wells would improve with reduction in groundwater withdrawal requirements. A summary of the alternatives evaluated under Level V is presented in Table 8-5. 8.3 WATER TREATMENT PROCESSES Water treatment for the removal of nitrates and pesticides is a major component of Level I, II and III alternatives. Alternative III(A) for Greenport also includes reduction of total dissolved solids (TDS) from existing well supplies owing to occasionally high chloride levels. The following processes were considered for removal of these constituents: 8-18 I I I I i I TABLE 8-5 LEVEL V ALTERNATIVES Alternative Service Area (D.C. = Demand Center) (S.D.C. = Sub Demand Center) Year 2000 Supply (Max. Day or Ave. Day) V(A) V(B-~) V(B-2) v(c-1) v(c-2) V(D-1) V(D-2) Mattituck/Cutcho§ue D.C. Mattituck/Cutchogue D.C. Southold/Greenport D.C. Mattituck/Cutchogue D.C. Southold/Greenport D.C.(1) Greenport Municipal System S.D.C. Greenport Municipal System S.D.C.(1) Mattituck West & East S.D.C.'s(2) Greenport Municipal System S.D.C. Mattituck West & East S.D.C.'s(2) Greenport Municipal System S.D.C.(1) Max. Day Max. Day Max. Day Max. Day Ave. Day Max. Day Ave. Day Max. Day Max, Day Max. Day Ave. Day (1) Alternative includes continuing the use of Greenport's highest quality well supplies. (2) Evaluated these S.D.C.'s separately due to economic reasons which are discussed in Section 7.4. (122/10) I I I I I I I I I I I I I I I I I I I Constituent Treatment Process Considered Nitrates Reverse Osmosis, Ion Exchange Pesticides Reverse Osmosis, Carbon Adsorption TDS Reverse Osmosis As shown, the process of reverse osmosis is capable of reducing all of the constituents to acceptable levels. However, based on an analysis of treatment facility requirements and associated costs, it was found that removal of nitrates and pesticides could be accomp- lished at a lower cost utilizing the processes of ion exchange and carbon adsorption. Accordingly, these unit processes were adopted for development of water treatment cost estimates for nitrates and pesticide removal for Level II and III alternatives. For alternative III(A) {Greenport), TDS reduction is also required. For this alter- native, the process of reverse osmosis was adopted for cost esti- mating purposes. The proposed reverse osmosis system is identical to those used for desalinization of brackish water. It should be emphasized that the treatment processes considered and adopted for this study are based on average study area water quality parameters and reported process removal efficiencies obtained from historical operating data from existing facilities, manufacturers' information and literature. They do not reflect site-specific treat- ment requirements; these would be determined by water quality analy- ses and pilot plant testing that is beyond the scope of this study. However, the treatment costs presented in this report and utilized for development of alternative costs, sufficiently represent treat- ment costs in a level of detail comparable to other alternative components. A summary of water treatment processes adopted for this study follows. 8.3.1 Ion Exchange/Carbon Adsorption The processes of ion exchange and carbon adsorption were adopted for removal of nitrates and pesticides, respectively. At a treatment facility, the process unit reactors would be arranged in series and consist of an ion exchanger followed by a granular activated carbon (GAC) filter. Both units would operate as pressure vessels. The ion exchanger would contain a strong base anion exchange resin that would serve to displace the negative nitrate ion (NO~) with a negative chloride ion (C1-). The resin would be periodically regenerated with sodium chloride {NaC1) when the unit's exchange capacity has been ex- hausted. Recent testing by the U.S. Environmental Protection Agency (EPA) showed that nitrate removal is effective at an application rate of 20 gallons per minute per square foot of bed area. At raw water nitrate levels that averaged 23 milligrams per liter (mg/1), removals to less than 10 mg/1 were consistently realized. About 20 pounds of NaC1 per cubic foot of resin was required for regeneration of the bed and to remove the nitrate from the spent resin. 8-20 A GAC pressure filter would follow the ion exchanger for removal of pesticides by adsorption. This process has proven to be effective in removing the pesticide Aldicarb as shown by the successful operation of the existing filter at Greenport's well no. 6-1. It has been reported that about 107.447 million gallons of water were treated by the 20,000 pounds of GAC contained in the filter before replacement was required. Aldicarb levels were reduced from slightly over 7 micrograms per liter (ug/1) to below detectable limits. 8.3.2 Reverse Osmosis Reverse osmosis was adopted as the treatment process for removal of nitrates, pesticides and TDS because, as a membrane process in which the water passes through a barrier membrane, it has proven effective in rejecting ionic and a majority of non-ionic species such as pesti- cides, bacteria and viruses. For comparison, the processes of elect- rodialysis removes only ionic species, and thus was not considered. The reverse osmosis process was selected for treating water produced from existing wells in Greenport under alternative III(Al which have, or have the potential, to contain nitrates, pesticides and high TDS. It is also the process utilized in the individual home well treatment systems evaluated under Level I. Reverse osmosis (R-O} is a membrane process in which the semiperme- able characteristics of a membrane are utilized to separate contamin- ants from the feed water under high pressure. To overcome the natural osmotic pressure of a solution and provide sufficient driving force to obtain economical water production rates, R-O systems are typically operated at a pressures of 400 to 500 psi for treating brackish water. R-O membrane materials fall into two broad categories: cellulosic and polyamide. There are also two primary membrane configurations: spiral wound and hollow fiber. For all systems, the feed water is pressurized using high-pressure pumps and conveyed to the membrane units, where the contaminants are rejected at the membrane wall and directed out of the unit as a brine reject stream requiring disposal. The contaminant-free stream, after passing through the membrane, must be decarbonated to remove carbon dioxide that has been entrained by the pretreatment requirement of pH reduction by addition of acid. Other pretreatment requirements include sequestering for scale inhi- bition and cartridge filtration for particulate removal to protect the membranes. At the completion of the decarbonation process, the treated water is chlorinated and then pumped into the distribution system. 8-21 8.4 ALTERNATIVES FOR DEMAND CENTERS 8.4.1 General Development Criteria In accordance with the Level I through ¥ water supply alternatives presented in Section 8.2, several alternatives were developed and costs estimated for each demand center. The criteria associated with the development of Level I alternatives, which consist of individual home water supply systems, were presented in detail in Section 8.2.1 and will not be repeated in this section. The Level II through V alternatives generally consist of the development of new public water supply systems or augmentation of existing systems, with similar com- ponents for each alternative. These components are summarized on Table 8-6. The conceptual design criteria for each component that was adopted for this study in order to enable the development of re- presentative estimated costs is presented in the following sections. 8.4.1.1 Supply. The supply components include new production wells and well pumping stations. For the Wading River/Northville and Riverhead/Jamesport demand centers, the wells would be 350 feet to 450 feet deep into the Magothy aquifer and sized in accordance with the estimated year 2000 maximum day demands. For the other demand centers, new production wells would tap the upper glacial aquifer. Gravel-packed wells were assumed and 20 feet of stainless steel well screen was included for cost estimating purposes. The new wells were sized and spaced in accordance with the criteria presented in Section 7.4. Two types of well pumping facility arrangements were considered, depending upon the size and spacing of the wells. For relatively small-capacity wells spaced in accordance with their zones of in- fluence, it was assumed that each well would be equipped with a sub- mersible vertical turbine well pump with the motor installed in a below-ground concrete vault. The well pumps would discharge to a common header in which water would be conveyed to a booster pumping station, where system pressure would be increased to desirable levels for distribution to customers. For relatively large-capacity wells that are spaced further apart, double pumping in this manner would not be cost-effective; thus, separate well pumping stations were assumed, each delivering water into the system. The selection of the type of system most applicable to an alternative was on a case-by- case basis and will become apparent during the discussions of the specific alternatives. The booster pumping stations and the indivi- dual well pumping stations would include a structure, standby power, pump and motor and associated piping and valves, HVAC, metering, electric power and controls. Facilities for the addition of caustic soda for pH control and sodium hypochlorite for disinfection were also included, and would generally consist of storage tank, chemical feed pumps and controls. A hydropneumatic tank to control pump oper- ation and maintain proper system pressures was also included for those systems where distribution system storage is not warranted. 8-22 TABLE 8-6 COMPONENTS OF ALTERNATIVES LEVELS II THROUGH V SUPPLY 0 0 PRODUCTION WELLS WELL PUMPING STATIONS TRANSMISSION/DISTRIBUTION o TRANSMISSION PIPING o BOOSTER PUMPING STATIONS (If Required) o DISTRIBUTION SYSTEM PIPING 0 DISTRIBUTION SYSTEM STORAGE TREATMENT (If Required) o CARBON/ION EXCHANGE (PESTICIDES, NITRATES) o REVERSE OSMOSIS (PESTICIDES, NITRATES, BRACKISH WATER) ANNUAL OPERATION & MAINTENANCE o POWER o LABOR o CHEMICALS o EQUIPMENT REPAIR & REPLACEMENT (122/12) overflow elevation of the tank is based on providing a minimum system pressure of 35 psi at the highest ground elevation to be served by the system. Either a standpipe or elevated steel tank was selected based on area topography. 8.4.1.3 Treatment. Treatment requirements and associated facilities were discussed in Section 8.3. In general, an activated carbon/ion exchange system was adopted, for cost estimating purposes, for treat- ment of groundwater in the Level II and III alternatives, except for Greenport's alternative III(A). For that alternative, a reverse osmosis treatment facility was selected since groundwater produced from existing Greenport wells is occasionally brackish, in addition to containing nitrates and pesticides. Individual home well treat- ment systems, evaluated under Level I, also utilize a reverse osmosis treatment process. 8.4.1.4 Annual Operation and Maintenance. Annual operation and maintenance (O&M) requirements for new water systems would include power, labor, chemicals and equipment repair and replacement. Power primarily includes the energy required to operate pumping and water treatment equipment. Labor requirements for new water systems in- clude a water department staff and crew. The present size and type of staff personnel for the existing Riverhead and Greenport systems was used to estimate labor requirements for new water systems. Water treatment facility labor requirement estimates took into considera- tion the size and complexity of the system and were based on past ex- perience with facilities similar in nature. In general, it was assumed that water department personnel could operate and maintain the relatively small activated carbon/ion exchange systems with proper training; however, the large reverse osmosis treatment plant under Greenport's alternative III(A) would require a full-time, qualified operator and staff, since this is a large facility and a complex process. It was assumed that caustic soda for pH control and sodium hypo- chlorite for disinfection would be added at all new well supplies. 8.4.2 Development of Cost Estimates For each alternative, capital and annual O&M costs were estimated for the demand centers and/or subdemand centers relative to the alterna- tive components discussed in the preceding section. In order to facilitate the comparison of the cost-effectiveness of each alterna- tive, and to assess relative affordability, costs were expressed as the estimated annual cost per dwelling unit for each alternative and for each subdemand center. Table 8-7 presents a typical example of the methodology associated with this approach. 8-25 TABLE 8-7 EXAMPLE OF COST DETERMINATION FOR ALTERNATIVES Wading River Sub-Demand Center Estimated Year 2000 Water Demands -- Alternative III(A) -- Average Day 360 gpm -- Maximum Day 690 gpm CONSTRUCTION COSTS 0 0 Supply -- 2 Wells, each 350 gpm, 350 Ft. Deep -- 2 Well Pumping Stations, each 350 GPM Sub-total Transmission/Distribution -- Piping; 5.7 mi. 12" dia.; 32.4 mi. 8" dia. -- Storage; Active Storage = 300,000 gal. Sub-total Total Construction Cost Total Project Cost (including 30% for engineering and contingencies) Amortized Capital Cost (40 yrs., 12%) $. 165,000 400,000 $ 565,000 (6%)(1} $ 8,735,000 (89%)(1) 540,000 (5%)(1) $ 9,275,000 $ 9,840,000 $12,790,000 $ 1,550,000 ANNUAL O&M COSTS o Power, Chemicals, Labor, Maintenance Total Annual Capital + O&M Estimated No. of Dwelling Units Estimated Annual Cost per Dwelling Unit 75,000 $ 1,625,000 $ 1,812 $ 900 (1) Percent of total estimated construction cost. (122/14) As shown, estimated construction costs were increased by 30 percent to account for engineering fees during design and construction and other contingencies such as administrative and legal fees. The total project cost was then amortized over 40 years at 12 percent interest, which is representative of an annual debt service for repayment of a long-term revenue bond. Added to this are estimated annual O&M costs, resulting in a total estimated annual cost for the subdemand center. This cost was then divided by the estimated number of year 2000 dwelling units, resulting in an estimated annual cost per dwell- ing unit. Table 8-7 also shows the relative percentage of the major capital cost components, with piping accounting for 89 percent of the total cost. This trend is similar for all presently unserved subdemand centers because of the relatively low housing density in most areas of the North Fork. Figure 8-4 illustrates this trend and shows that, on the average, about 87 percent of the costs for alternatives in- volving new public water supply to serve presently unserved areas is piping. Figure 8-5 further illustrates the impact of housing density on pi ping costs. For example, a typical home owner in Reeves Park (which has an average density of 1/2 to 3/4 acres per home), would pay about $400 a year, representing his share of distribution system piping cost. This estimate is based on 8-inch ductile iron piping in- stalled in an existing paved roadway, including service connections. For comparison, a typical home owner in Jamesport would pay over $1,100 a year due to the much lower housing density. In Jamesport, there is an average of only 25 homes per mile of roadway, which must support the costs for installing the required pipe, as compared to 90 homes per mile, on the average, in Reeves Park. The subdemand center with the lowest density is Great Hog Neck, where there is an average of only 18 homes per mile of roadway, resulting in an estimated annual piping cost of over $1,400 per home. 8.4.3 Wading River/Northville The total estimated costs of water supply alternatives for the Wading River/Northville demand center are shown on Table 8-8. They are pre- sented as estimated annual costs per dwelling unit to allow compari- sons among alternatives. 8.4.3.1 Level I Alternatives. Level I represents individual home water supply systems for homes located in presently unserved areas. The Level I costs shown on Table 8-8 represent a weighted average of the cost of systems for new homes and the cost of augmenting existing home systems. It was assumed that the entire 1980 population of each area (less those served by public water systems) was supplied by ex- isting well systems. This portion of the population would only incur those costs associated with the addition of individual treatment units. The costs associated with a complete well system (including treatment) would be incurred by new development. An areawide average cost was developed using the existing population and the projected year 2000 population. 8-27 m m m m mm m m mm --m ,~ m m mm m mm m m mm m AREAS CURRENTLY NOT SERVED BY PUBLIC SUPPLY AREAS CURRENTLY SERVED BY PUBLIC SUPPLY TRANSMISSION AND DISTRIBUTION SUPPLY TREATMENT 77% TREATMENT 23% SUPPLY AUGMENTATION m m m m mm m m .m ,-- .m m mm --- -- m m m m m 1200 ~ 1000 800 600 400 200 I I 100' (1/4 ACRE) JAMESPORT ~, REEVES PARK 150' 200' 250' 300' 350' 400' (1/2 ACRE) (1 ACRE) (I 1/2 ACRE) (2 ACRE) (3 ACRE) (3 1/2 ACRE) AVERAGE LOT FRONTAGE (SIZE) m m ~m m m m m m mm mm mm mm mm mm mm mm mm mm mm TABLE 8-8 ESTIMATED ANNUAL DWELLING UNIT COSTS OF WATER SUPPLY ALTERNATIVES FOR THE WADING RIVER/NORTHVILLE DEMAND CENTER SUB-DEMAND CENTER ESTIMATED NO. OF YEAR 2000 LEVEL DWELLING UNITS I ESTIMATED ANNUAL COST PER DWELLING UNIT LEVEL II LEVEL III IliA IIIB* $ $ $ $ o Wading River -- Unserved Areas 1812 790 --- 900* 850 -- Existing Systems 394 --- 340 340 315 o Baiting Hollow/Woodcliff Park -- Unserved Areas 549 815 --- 740* 690 -- Existing Systems 238 --- 330 330 335 o Reeves Park -- Unserved Areas 510 820 --- 455* 430 -- Existing Systems 238 --- 260 260 190 *Treatment not required. (SPH5/45) 8.4.3.2 Level II Alternatives. Level II costs for the Wading River/Northville demand center represent augmentation of the existing small public water systems located in each of the three subdemand centers. Augmentation would include: (1) additional well supply, if necessary, to provide the expected year 2000 water requirements and/or to provide at least two well supplies (for reliability), and (2) an activated carbon/ion exchange water treatment system for re- duction of existing or potential future high levels of nitrates and pesticides. The existing distribution systems would remain in ser- vice. Table 8-9 summarizes information on the existing water sys- tems and augmentation requirements adopted for this study. It should be noted that the costs shown on Table 8-8 are average costs per existing system dwelling unit, and are based on total capital and annual costs for all existing systems within a subdemand center. Also, these estimated costs do not include the present annual cost each homeowner is paying. 8.4.3.3 Level III Alternatives. Two alternatives were considered under Level III: III(A) New public water systems to serve the presently un- served areas only Existing small public systems augmented for continued operation, as described under Level II III(B) New public water systems to serve the presently un- served areas and to supply the existing public systems Existing public system supplies discontinued; distri- bution systems to remain in service. Figure 8-6 presents the Level III alternatives for the three sub- demand centers and shows the locations of new well supplies, trans- mission mains and system storage. The alternative components would be the same for both III(A) and III(B); however, the supply capacity would be greater for III(BI in order to provide service to the exist- ing systems. As shown, the new system for Wading River would include two well supplies, a transmission network and a storage facility. The total supply capacity would be 700 gpm for III(A) (two wells, each rated at 350 gpm) and 900 gpm for III(B) (two wells, each rated at 450 gpm) based on expected year 2000 maximum day water demands. A pumping station would be provided at each well. The average well depth would be about 350 feet into the Magothy aquifer, thus treatment was as- sumed not to be required. A transmission main network of 12-inch diameter pipe would convey maximum day flows to the storage facility and to the higher consumption areas. A storage facility was provided for this subdemand center since it is a relatively large area with 8-31 m m m mm m m m m m mm m m TABLE 8-9 EXISTING PUBLIC WATER SYSTEMS in the WADING RIVER/NORTHVILLE DEMAND CENTER m m mm m System Wading River Estimated No. of Dwel 1 i ngs Estimated(i) Existing Supplies Yr. 2000 No. Total(2) Peak Hr. of Wells Capacity Level II Augmentation Requirements Supply(3) Treatment(4) (gpm) Wading River Water Works Wildwood Shores Assoc. Herod Point Assoc. Oakwood-on-the-Sound Hulse Farms Ramblewood Motor Homes 53 85 1 60 18 10 1 30 22 20 i 50 99 105 4 (unknown) 74 85 2 45 128 50 2 200 2nd supply 50 2nd supply 6 2nd supply 12 (Assume OK) 65 3rd supply 50 Adequate 30 Subtotal 394 Baiting Hollow/Woodcliff Pk Woodcliff Park Baiting Hollow Condo's Subtotal 200 225 2 22O 38 50 i 25 Adequate 135 2nd Supply 30 Reeves Park Reeves Beach Water Co. Roanoke Water Co. Subtotal 184 170 2 165 54 45 2 40 Adequate 110 Adequate 30 TOTAL 870 Notes: ----~-)Gallons per minute (gpm) (2)Total reported well pumping capacity, in gpm {3)Includes new well at depth similar to existing wells, and a well pumping station. Well pumping capacity adequacy based on peak hour since there is no distribution system storage. (4)All systems would require treatment consisting of a carbon/ion exchange system. Treatment capacity shown is based on max day water demands. (SPH5/46) ~EGEND: ____DEMAND CENTER BOUNDARY --.__SUBDEUANO CENTER BOUNDARY EXISTING SYSTEM PIPING NETWORK Figure 8-6 Level III Alternative Components for Wading River/Northville expected maximum day demands in excess of 1 mgd. Also, since the supplies are not centrally located, a tank would serve to equalize system pressures and increase system reliability. Fire protection would also be provided. A standpipe was selected and located off of Wading River Road, as shown on Figure 8-6. This site has a rela- tively high ground elevation (El. 235 +, USGS base) which results in a required tank height of about 80 feet based on a 315-foot overflow elevation to provide a minimum of 35 psi at all points within the system. The required active storage volume would be 300,000 gallons for III(A) and 370,000 gallons for III(B). Distribution system piping requirements include about 32 miles of 8-inch diameter pi~ing in addition to about 6 miles of 12-inch diameter transmission main. The systems for Baiting Hollow/Woodcliff Park and Reeves Park include several small-capacity wells, each equipped with a submersible pump in a vault and connected to a manifold-type piping system. The mani- fold system would convey flow to a booster pumping station, which would deliver water to the distribution piping network. The booster pumping stations would operate off a hydropneumatic tank in lieu of providing distribution system storage since these subdemand centers are small and have relatively low water demands. The supply capacity for Baiting Hollow/Woodcliff Park was estimated at 190 gpm for III{A) (4 wells at 50 gpm each) and 335 gpm for III(B) {4 wells at 55 gpm each; 2 wells at 75 gpm each), the higher capacity being required for III(B) to serve the estimated 238 dwelling units in the existing systems. The well depth would be about 270 to 300 feet into the Magothy aquifer; thus, it was assumed that treatment would not be required. About 9 miles of distribution system pi ping would be required to serve the presently unserved areas. The supply capacity for Reeves Park was estimated at 160 gpm for III(A) (3 wells at 55 gpm each) and 300 gpm for III{B) (5 wells at 60 gpm each). The well depth would be about 275 feet into the Magothy aquifer. About 5 miles of distribution system piping would be re- quired to serve the presently unserved areas. The Level III estimated annual cost per dwelling unit for the three subdemand centers are shown on Table 8-8. The Level III{B) costs are lower since the supply costs were proportioned between the unserved areas and the existing systems on the basis of maximum day water re- quirements; thus, they would share.the costs. Table 8-10 presents capital cost estimates for the Wading River/Northville Level III alternatives. A Level IV alternative was considered for the Wading River/Northville demand center which would involve a central groundwater supply source and a transmission system to provide supply to the three subdemand centers. It was found that the cost of supply and transmission under this concept was about 25 percent greater than the total of the sup- 8-34 m TABLE 8-10 CAPITAL COST ESTIMATES WADING RIVER/NORTHVILLE DEMAND CENTER (ENR-3800) Level III(A) Level III(B) Transmi ssion Total Total lransmi ssi on Total Total and Construction Project and Construction Project Subdemand Centers Supply Distribution Treatment Cost Cost Supply Distribution Treatment Cost Cost Wading River $ 860,000 $ g,275,000 $ 475,000 $10,610,000 $13,793,000 $ 630,000 $ 9,400,000 HOT REQUIRED $10,030,000 $13,040,U00 Baiting Hollow/ Woodcliff Park 420,000 2,100,000 380,000 2,900,000 3,770,000 685,000 2,100,000 NOT REQUIRED 2,785,D00 3,620,UUU Reeves Park 255,000 1,120,000 340,000 1,715,000 2,229,500 470,000 1,120,000 NOT REQUIRED 1,590,000 2,070,000 TOTALS $1,535,000 $12,495,000 $1,195,000 $15,225,000 $19,792,500 $1,785,000 $12,620,000 -- $14,405,000 $18,730,000 (122/3) I I I I I I I I I I I I I I I I I i ply costs for each subdemand center under Level III. This was due to the length of required transmission piping which was substantial since the centers are relatively far apart. Accordingly, this con- cept was eliminated from further consideration. 8.4.4 Riverhead/Jamesport The estimated total annual costs per dwelling unit of water supply alternatives for the Riverhead/Jamesport demand center are shown on Table 8-11. The Riverhead M¥~icipal System {i.e., Riverhead and Aquebogue subdemand centers)~ ~ was not considered for analysis since an adequate, high-quality supply presently exists. 8.4.4.1 Level I Alternatives. Level I represents individual home water supply systems consisting of a well and treatment unit for new homes, and only a treatment unit for existing homes. The Level I costs are a weighted average of the costs for existing homes and the costs for new homes. 8.4.4.2 Level III Alternatives. Level III consists of new public water systems for Calverton an'~ Jamesport, as shown on Figures 8-7 and 8-8. The Calverton system would include two 40-gpm-capacity wells and pumping facilities operating off a hydropneumatic tank, and about 3.5 miles of distribution piping. The Jamesport system would include two 150-gpm capacity wells and pumping stations, a distribu- tion system storage tank with 150,000 gallons of active storage, and about 30 miles of transmission and distribution system piping. A storage tank was provided for this subdemand center because of its size, and to assure equalized pressures throughout the system since the supplies are not centrally located. It would also provide fire protection throughout the area. As shown on Table 8-11, the estimated annual cost of the Level III alternative for Jamesport is $1,530 per dwelling unit. This high cost is primarily due to low housing density and the resulting length of distribution piping required to provide service throughout the area. For this area, about 90 percent of the total water system con- struction costs would be piping. Accordingly, an attempt was made to reduce public water supply system costs in Jamesport by limiting the service area to the relatively high-density developments located close to the supply source, as shown on Figure 8-8. This reduction in service area would reduce the estimated number of dwelling units from 764 to 440 and'the distribution system pi ping from about 30 miles to about 11 miles. The expected year 2000 maximum day demand would also reduce from 300 gpm to about 155 gpm, which weuld require two 80-gpm-capacity wells with submersible pumps, and one booster (1)Portions of Aquebogue are served by the Riverhead system; Aque- bogue is, therefore, not included in the Jamesport analysis and recommendations. 8-36 TABLE 8-11 ANNUAL DWELLING UNIT ESTIMATED COSTS OF WATER SUPPLY ALTERNATIVES for the RIVERHEAD/JAMESPORT DEMAND CENTER ESTIMATED ANNUAL COST PER DWELLING UNIT ESTIMATED NO. OF YR. 2000 LEVEL LEVEL LEVEL LEVEL SUB-DEMAND CENTER DWELLING UNITS I II III IV $ $ $ $ Calverton 195 650 --- 795* 720* Jamesport 764 775 --- 1530' --- *Treatment not required I <, __., ,)i:%'¢' ... 'x L~×,s~,.~ ,,.,VER,d~L:'~\ %¢" ~'*¢.: .%.. ,',.,., - . ", '%, 0 ., ,:~ # ! .-- "g-"';c::~Z~'-- , ~ ,,~(" - Il' ~:~ELL I~,ITH PUMP I j~: ~C~-~r_ - ..-:~' l~ --~-~ ;~ ~ -/ /::: - ---~ ~ .. ~.: ::~ ,' ~ BOOSTER PUMP NGI, ' / .--. ..-~"~ ..' ,, . ~ LEVEL~ I/ .-' : , '~- I .:' J I Figure 8-7 I Levels III and IV Alternative Components for Calverton Nor~hville SUBDEMAND CENTER BOUNDARY WITH LIMITED SERVICE AREA (SEE TEXT) % % · FLANDERS BAY WELL AND PUMPII~ STATION (TYP OF2) ~amesport N MAJOR TRANSMISSI¢ PIPING SYSTEM Figure 8-8 Level III Alternative Components for Jamesport pumping station with a peak hour pumping capacity of 230 gpm operat- ing off a hydropneumatic tank (a storage tank would not be warranted with the reduction in service area). From the results of this analy- sis, it was found that the annual cost per dwelling unit would be re- duced by about 33 percent to $1,020. The capital costs were substan- tially reduced by over 60 percent; however, a comparable reduction in cost per dwelling unit was negated by the reduction in the total number of dwelling units to be served. 8.4.4.3 Level IV Alternatives. Level IV was considered for Calver- ton only for comparison with Level III costs. Ninety percent of the Level III costs for Jamesport is for piping; a significant cost savings would not be realized in a Level IV option by utilizing Riverhead as a supply source (in lieu of new wells and pumping sta- tions) since piping requirements would essentially be the same. Accordingly, Level IV for Jamesport was not considered for further analysis. As indicated on Figure 8-7, Level IV for Calverton would simply con- si st of a metered connection to the existing Riverhead system and construction of about 3.5 miles of distribution system piping. A connection could be made to the existing 12-inch main located along Route 25 near its intersection with River Road. Ameter installed in a vault would also be required to monitor water use for billing pur- poses. The Level IV cost shown on Table 8-11 includes an estimated annual water cost that Riverhead would assess Calverton based on their present bulk rate of $0.60 per 1,000 gallons. This option has the advantage of providing Calverton wi th a reliable, high-quality supply and fire protection available from Riverhead's storage facilities. Table 8-12 includes capital cost estimates for Level III and IV alternatives for the Riverhead/Jamesport demand center. 8.4.5 Mattituck/Cutchogue The estimated total annual costs per dwelling unit of water suppl~ alternatives for the Mattituck/Cutchogue demand center are shown in Table 8-13. The concepts associated with each level of alternatives were discussed in Section 8.2, and can be summarized for this demand center as follows: Level I: - Individual home water supply systems Level II: Augmentation of the existing neighborhood sys- tem (Captain Kidd development) in Mattituck West Level III(A): New systems for presently unserved areas Captain Kidd system augmented as per Level II 8-40 TABLE 8-12 CAPITAL COST ESTIMATES RIVERHEAD/JAMESPORT DEMAND CENTER (ENR-3800) Sub-Demand Centers Supply Level III Level IV Transmission Total Total Transmission Total T6tal and Construction Project and Construction Project Distribution Treatment Cost Cost Supply Distribution Treatment Cost Cost Riverhead/Aquebogue -- $1,500,000 NOT REQUIRED $1,500,000 $ 1,950,000 .......... Calverton $160,000 775,000 NOT REQUIRED 935,000 1,215,000 $5,000 $775,000 $780,000 $1,015,UUD Jamesport 375,000 6,890,000 NOT REQUIRED 7,265,000 9,445,000 .......... TOTALS $535,000 $9,165,000 -- $9,700,000 $12,610,000 $5,000 $775,000 -- $780,D0~ $1,0kS,UUU Level III(B): Level V: - Level ¥(A): - Level V(B-1): Level V(B-2): Level V(C-1,2): - Level V(D-1): Level V(D-2): New systems for presently unserved areas and to supply Captain Kidd system Captain Kidd system supply to be discontinued (except for emergency use); distribution sys- tem to remain in service Supply from potable groundwater sources in Riverhead Transmission system from supply to demand centers Distribution~system piping and storage for demand centers Mattituck/Cutchogue demand center (year 2000 maximum day) Mattituck/Cutchogue and Southold/Greenport demand centers (year 2000 maximum day) Mattituck/Cutchogue demand center (year 2000 maximum day) Southold/Greenport demand center (year 2000 average day) Continued use of existing supplies, as re- quired (Regional alternatives for the Greenport Municipal System presented in Section 8.4.6) Mattituck West and East Subdemand centers (year 2000 maximum day) Greenport Municipal System (year 2000 maximum day demand) Mattituck West and East subdemand centers (year 2000 maximum day) Greenport Municipal System (year 2000 average day) Continued use of existing supplies, as re- quired. 8-42 8.4.5.1 Level I Alternatives. The costs shown on Table 8-13 are a weighted average of existing home and new home costs, based on the expected number of new homes by the year 2000. Treatment would not be required in the Little Hog Neck and New Suffolk subdemand centers since high levels of nitrates and/or pesticides are not expected in these areas. 8.4.5.2 Level II Alternatives. Level II for this demand center applies only to the Mattituck West subdemand center, which contains an existing neighborhood system serving the Captain Kidd development. It is estimated that by the year 2000 the population would be about 580, housed in 154 dwelling units, with estimated water demands of 32 gpm on an average day basis, 96 gpm on a maximum day and 160 gpm during a peak hour. Presently, the system has two wells with a com- bined pumping capacity of 300 gpm; therefore, the supply is consider- ed adequate. The estimated annual cost per dwelling unit of $300 re- flects the cost of a lO0-gpm capacity activated carbon/ion exchange treatment facility. 8.4.5.3 Level III Alternatives. Level III alternatives for this demand center are similar in nature to those developed for the Wading River/Northville demand center. This level involves development of individual public water systems for each subdemand center including supply, treatment (if necessary) and transmission and distribution pi ping. Layouts of the major system facilities for each subdemand center are shown on Figure 8-9. Table 8-14 summarizes information on the components of the alternatives for each demand center and in- cludes estimated supply and treatment system capacities and piping requirements. For each subdemand center, a system consisting of multiple wells (with submersible pumps) discharging to a header system for convey- ance to a booster pumping station was selected. Multiple wells would provide increased reliability and assure that the aquifer is not overstressed at one location, since groundwater withdrawals would be distributed over a wide area. This should prevent upconing and, accordingly, saltwater intrusion. For the subdemand centers requir- ing treatment, the booster pumping station would be contained within the treatment facility structure and would be operated off a hydro- pneumatic tank, except in Mattituck South, where a distribution sys- tem storage tank is suggested. As is evident on Figure 8-9, the pro- posed well location for Mattituck South is actually outside the demand center boundary at the western extremity. Accordingly, in order to equalize system pressures and to reduce water transmission requirements, a storage tank, centrally located at a high ground ele- vation as shown, was provided. The required active storage volume for fire protection and hourly fluctuations is estimated at 225,000 gallons. The remaining subdemand centers would not warrant system storage, either because of their relatively small size and resulting 8-43 TABLE 8-13 ESTIMATED ANNUAL DWELLING UNIT COSTS OF WATER SUPPLY ALTERNATIVES MATTITUCK/CUTCHOGUE DEMAND CENTER Sub-Demand Center Estimated No. Of Yr. 2000 Level Level Dwelling Units I II Estimated Annual Cost Per Dwelling Unit Level III Level LEVEL V(1) ~II(A) III(B'I IV V(A) V'('~'-l) V(B-2') V(C-1) V(C-'2)"V'('D-1) V{D-2) Mattituck West -Unserved Areas 1,447 -Existing System 154 Mattituck East 916 Mattituck South 1,512 Little Hog Neck 449 E. Cutchogue/Little Creek 362 Fleets Neck/Cutchogue 638 New Suffolk 288 Indian Neck 167 610 --- 770 --- 300 300 605 740 605 1,235 185(1) 960(1) 605 1,160 605 860 185{1) 1,145(1) 6O5 1,435 760 375(2) 925 920 935 (2) 685 685 (1) Treatment not requi"red (2) Does not apply to this Demand Center (See text) WEST · . ' k2 v~ .~.~, :~;,. ~ x MATTITUCK SOUTH MATTITUCK EAST FLEETS CUTCHOGUE p I", C IIA y LEGEND: · WELL WITH PUMP IN VAULT [] WATER TREATMENT PLANT AND PUMP STATION 0 STORAGE TANK RAW WATER TRANSMISSION PIPING .... FINISHED WATER TRANSMISSION PIPING -- DEMAND CENTER BOUNDARY ~.~ SUBDEMAND CENTER BOUNDARY STATION ONLY ION NOT REQUERI DEO NEW SUFFOLK O O INDIAN EAST CUTCHOGUE / LITTLE CREEK LITTLE HO6 NECK Figure 8-9 Level III Alternative Components for Mattituck/Cutchogue / / TABLE 8-14 LEVEL III ALTERNATIVE COMPONENTS for the MATTITUCK/CUTDHOGUE DEMAND CENTER Supply Treatment(1) Transmission/Distribution Sub-Demand Center Estim. Yr. 2000 Demands No. of Ave. Total Well 8ooster P.S. System Length of(2) Active Distribution Ave Max Peak Wells DepthlCapacity Capacity Capacity Piping, Miles System Storage Vol., Gal Mattituck West --Unserved Areas-IIIA 250 445 660 4 80' 460 660 460 22.7 (3) -IIIB 280 540 800 5 80' 575 800 575 22.7 (3) Mattituck East 160 300 450 3 90' 330 450 330 14.6 (3) Mattituck South 260 500 740 2 75' 500 500 500 43.5 225,000 Little Hog Neck 75 150 220 3 55' 150 220 (none) 10.1 (3) E. Cutchogue/ 60 120 175 2 65' 120 175 120 9.6 (3) Little Creek Fleets Neck/Cutchogue 110 210 310 2 70' 210 310 210 11.8 (3) New Suffolk 50 95 140 2 70' 100 140 (none) 8.3 (3) Indian Neck 30 55 80 2 65' 60 80 60 5.7 (3) Notes: General--All flows are in gallons-per-minute (gpm) {1)Treatment consists of a carbon/ion exchange system having a treatment design capacity equivalent to max day demands and a hydraulic design (2~capacity equivalent to peak hour demands. 'Includes the estimated length of required transmission and distribution piping to serve the estimated year 2000 population assuming con- {3)tinuation of historical grov~ch patterns. ~Distribution system storage not warranted--a hydropneumatic tank would be provided at the booster pumping station. low water demands, or a centrally located supply. For these areas, peak hour pumping capacity would be required, with the pumps operated off the hydropneumatic tank (which serves to control pump cycle time and system pressures}. As with the Wading River/Northville demand center, the costs of Level III(B} for Mattituck West include supplying the Captain Kidd system from a new, central source that would also supply the unserved areas. Accordingly, the supply and treatment costs were proportioned between Captain Kidd and the unserved areas based on maximum day water re- quirements. As shown on Table 8-13, Level III(B) would be more ex- pensive for Captain Kidd since it includes a portion of the total supply and treatment costs, whereas Level III(A) includes the addi- tion of only a treatment system to their present supplies. For this demand center, piping represents an average of 83 percent of the Level III costs shown on Table 8-13. Indian Neck has the lowest housing density within the demand center (average of 30 homes per mile of roadway} and the highest Level III costs due to piping re- quirements. Mattituck West and East have the highest density within the demand center (average of 60 homes per mile of roadway) which re- flects the lower Level III costs. However, in all subdemand centers, Level I proved to be less expensive than Level III. 8.4.5.4 Level IV Alternatives. Level IV for Mattituck/Cutchogue was not evaluated in detail since a significant cost savings would not be realized compared with Level III. Level IV would include a central supply serving the entire demand center; however, piping requirements and associated costs would be essentially the same as the sum of the piping costs for the subdemand centers under Level III. Since an average of 83 percent of the Level III costs is piping, and consider- ing that a central supply would require additional transmission piping and, most likely, distribution system storage, it is estimated that only a 5 to 10 percent cost savings would be realized under a Level IV option. Refering to Table 8-13, Level IV costs would still be significantly higher than Level I costs, except in Mattituck West and East where the difference would not be as significant. For these subdemand centers, additional analysis is warranted and was performed under Level V. This will be discussed in the following sections. 8.4.5.5 Level V Alternatives. Level V for Mattituck/Cutchogue in- cludes construction of groundwater supply and production facilities in Riverhead, a water transmission system from the supply sources to demand centers, and distribution system piping and storage within the demand centers. As discussed earlier, Levels V(B) and V(D) also in- clude service to Greenport and essentially involve a continuation of the water transmission system to that area. For these options, the supply and transmission costs would be shared by the demand centers, proportioned on the basis of design flows. As shown in Table 8-13, the cost of scenarios i and 2 under both Levels V(B) and V(D) are 8-47 essentially equal for Mattituck/Cutchogue since the scenarios apply only to Greenport. Scenario I would provide maximum day supply to Greenport, while scenario 2 would only provide up to average day supply, with Greenport's existing supplies remaining in service. Under all Level V alternatives, maximum day supply would be provided to Mattituck/Cutchogue since there are no existing supplies within this area. Level V for Greenport will be discussed in Section 8.4.6. Level V{D) was developed and evaluated primarily to determine if the Level III costs for Mattituck West and East could be reduced by uti- lization of a regional supply source. Since a regional supply would, most likely, not be developed only for these subdemand centers, the Greenport Municipal System was also incorporated into this Level for comparison with the costs for Greenport's Level V(C) (discussed in the next section). A layout of the components of Level V, specifically Level V(B), is shown on Figure 8-10. Table 8-15 summarizes pertinent information on each component. Supply would consist of deep production wells in Riverhead, preliminarily located as shown on Figure 8-10. Each well would tap the Magothy aquifer and would average 350- to 450-feet deep. Treatment for nitrates and pesticides removal was assumed not to be required since, reportedly, the Magothy aquifer in this loca- tion does not contain these constituents. Each well would include a pumping station equipped with a vertical turbine well pump, standby power, chemical addition facilities (for disinfection and pH adjust- ment) and other associated piping and control appurtenances. Total supply (and transmission system) capacity was based on the total estimated year 2000 maximum day demands for the areas to be served for Levels V(A), V(B-i~, and V(D-1). For Levels V(B-2) and V(D-2), supply and transmission system capacity was based on year 2000 maxi- mum day demands for Mattituck/Cutchogue plus year 2000 average day demands for Greenport. Transmission system piping would consist of cement-lined ductile iron pipe with a maximum pipe size for each alternative as indicated in Table 8-15. This pipe size would be required for the section between the most easterly production well and the connection to the first storage tank in the Mattituck/Cutchogue area, and would convey up to the total design flow for the entire service area. Beyond the first tank, the pipe size would reduce in size as it heads toward Green- port, in accordance with the reduction in required carrying capacity as demands are satisfied. Under Level V(D}, the transmission system would follow the same route as indicated in Figure 8-10; however, the pipe sizes would be less and only one new storage facility to serve Mattituck West and East would be required. Booster pumping stations would be provided along the pipeline in order to maintain system pressures, which would gradually decline due to pipe friction as the water flows east. llqe stations would include a centrifugal pump, standby power and other related appurtenances, housed in a super- 8-48 TABLE 8-15 SUMMARY OF LEVEL V ALTERNATIVE COMPONENTS for MATTITUCK/CUTCHOGUE DEMAND CENTER LEVEL V ALTERNATIVES ALTERNATIVE COMPONENTS VA VB-1 VB-2 VD-1(1) VD-2(1) Estimated Year 2000 Demands Ave Day, GPM 1,025 1,025 1,025 440 440 Max Day, GPM 1,970 1,970 1,970 840 840 Supply e No. of Production Wells(2) 3 5 4 4 2 · Total Capacity. GPM(2]' ' 2,100 4,100 2,850 2,620 1,500 · Capacity Allocated to Matt./Cutch., GPM 2,100 1,970 1,970 840 840 Transmission · Total Length of Piping, Miles(2) 12.8 22.7 22.7 19.5 15.2 · Largest Pipe Size, Dia. in inches~2]' ' 12" 20" 16" 16" 16" · Total No. of Bosster Pump Stations~2~' ' 1 3 2 2 2 e Percent of Total Cost Allocated to Matt./Cutch!3~' ' 100% 41% 56% 22% 31% Distribution/Storage · Total Length of Piping, Miles 130 130 130 36 36 · Estimated No. of Storage Facilities Required 2 2 2 1 1 · Estimated Total Volume of Active Storage Required, GAL. 8DO,O00 800,000 800,000 350,000 350,000 · Type of Storage Tank Elevated Elevated Elevated Elevated Elevated · Ave. Overflow Elev., Ft. {USGS Base) 202' 202' 202' 202' 202' Notes: )~(Mattituck West & East sub-demand centers only (in addition to Greenport) )3~Includes service to Greenport for Levels VB and VD ' 'Function of design flow an~d length of required piping I I :1 I I I I I I I I I I I structure. For this conceptual analysis, minimum transmission system pressures were assumed to be 35 psi and maximum pressures 100 psi. As presently conceived, the system would have the capability of fill- ing proposed storage facilities in the Mattituck/Cutchogue area. For Levels V(BI and V(D) the pressures at the eastern end of the system would be sufficient to supply the Greenport system while maintaining their present system pressures. System pressures within the Matti- tuck/Cutchogue demand center v~uld be controlled by the water level in the storage tanks, which would provide a minimum of 35 psi at the highest ground elevations. Water would be delivered to consumers through a distribution piping system containing about 130 miles of piping, required to serve the entire demand center. As shown on Figure 8-10, the transmission main route selected for this analysis generally follows Route 25, which traverses the areas to be served. Transmission piping costs include allowances for uti- lity crossings, pavement restoration and policing. Two alternative routes were also considered, as follows: Option i - Generally follows North Road. Option 2 - ~ong the existing right-of-way (ROW) of the Long Island Lighting Company (LILCO). This option assumes access could be obtained from the utility. These route options are also shown on Figure 8-10. An analysis of the estimated transmission system piping requirements and related costs associated with each option was performed only for Levels V(C) and V(D) since only these alternatives have the potential to be as cost-effective as Level I, as is evident in Table 8-13. The results of the analysis for Level V(C) {Greenport Municipal System) will be presented in the next section. For Level V(D), it was found that the amount of piping would be less for options i and 2, and the pipe in- stallation costs for option 2 would be about 10 percent less expen- sive since paving, utility crossing and policing would not be re- quired along the LILCO right-of-way. The impact on Level V(D) costs for the route options is summarized as follows (for Mattituck West and East only): Route 25 Option 1 Option 2 V(D-1) V(D-2) V(D-1) V(D-2) V(D-1) V(D-2) Length of Piping Required, miles* 19.5 15.2 18.8 14.2 15.7 12.5 Annual Cost per Dwelling Unit $685 $680 $665 *Includes service to Greenport also and includes transmission piping only. 8-51 A significant cost savings would not be realized since, as discussed earlier, over 83 percent of the total capital cost of Level V(D) for Mattituck West and East is distribution system piping, which tends to reduce the overall impact of cost savings associated with the trans- mission system route options. However, Level V(D) costs are lower than Level III costs, regardless of transmission system routing, for two reasons: (1) treatment would not be required, and (2) Greenport would pay for a portion of the supply and transmission system. Level I, however, still appears to be the lowest-cost option available to all subdemand centers within Mattituck/Cutchogue. Table 8-16 presents the capital cost estimates for Level III and V alternatives for the Mattituck/Cutchogue demand center. 8.4.6 Southold/Greenport The estimated total annual costs per dwelling unit of water supply alternatives for the Southold/Greenport demand center are shown on Table 8-17. The concepts associated with each level of alternatives were summarized in the preceding section and so will not be repeated here. 8.4.6.1 Level I Alternatives. Level I applies to Great Hog Neck and East Marion only and includes individual home water supply systems. In both subdemand centers, treatment would probably be required; therefore, it was included in the Level I costs. 8.4.6.2 Level Iit Alternatives. Level III consists of new public water systems for Great Hog Neck and East Marion, and augmentation of the existing Greenport Municipal System. Similar to the other demand centers, Level III for Great Hog Neck and East Marion includes a multiple well-type supply system feeding a booster pumping station and water treatment facility, as shown on Figures 8-11 and 8-12. For Great Hog Neck, a distribution system storage tank was provided due to the size of the area and the availability of a relatively high ground elevation (El. 70 feet, USGS base) near the supply source. Table 8-18 summarizes information on the components of Level III alternatives for the two subdemand centers. As is evident in Table 8-17, the Level III costs for Great Hog Neck and East Marion are relatively high, primarily because of the length of required piping and associated costs. Great Hog Neck has the lowest density of any subdemand center on the North Fork, with an average of only 18 homes per mile of roadway. To provide water ser- vice throughout the area, over 28 miles of piping would be required to serve only about 500 homes. Similarly, East Marion also has a very low density with an average of about 40 homes per mile of road- way. It is apparent that these areas are not conducive to a public water system because of the lack of an adequate population base. 8-52 TABLE 8-16 CAPITAL COST ESTIMATE MATT[TUCK/CUTCHOGUE DEMAND CENTER (ENR ~ 3800) LEVEL lEI LEVEL V(A) Trans. & Treat- Total Total Trans. & Total Total Dtstribu- merit Const. Project Distribu- Const. Project Subdemand Center Supply tion Cost Cost Cost Supply tion Cost Cost Mattituck West $410,000 $5,360.000 $1,150,000 $6,920,000 $8,995,000 Mattituck East 250,000 3,210,000 600,000 4,060,000 5,280,000 Mattttuck South 265,000 10,000,000 1,100,000 11,365,000 14,775.000 Little Nog Neck 390,000 2,230,000 Not Req'd 2,620,000 3,41D,000 E. Cutchogue/ Little Creek 150,000 2,120,000 250,000 2,520,000 3,275,000 Fleets Neck/ Cutchogue 180,000 2,720,000 400,000 3,300,000 4,290,000 New Suffolk 180,000 1,830,000 Not Req'd 2.010,000 2,615,000 Indian Neck 100,000 1,260,000 110,000 1,470,000 1.910,000 TOTALS: $1,925,000 $28,730,000 $3,610,000 $34.265,000 $44,550,000 $1,105,000 $32,315,000 $33,420,000 $43,446,~00(1) Subdemand Center Supply Mattituck West Mattttcuk East Mattituck South Little Hog Neck E. Cutchogue/ Little Creek Fleets Neck/ Cutchogue New Suffolk Indian Neck TOTALS (2) TOTALS (3) LEVEL V(B-1) LEVEL V(D) (1) Trans. & Total Total Trans. & Total ~otal Distribu- Const. Project Distribu- Const. Project tion Cost Cost Supply tion Cost Cost $ 956,000 $32,066,000 $33,022,000 $42,929,006 $455,000 $39,795,000 $10,250,000 $13,325,000 $1,040,000 $32,630,000 $33,670,000 $43,771,000 $425,000 $ 9,825,000 $10,250,000 $13,335,000 (1) Provides service only to Mattituck East and West in this demand center. (2) These totals assume the total supply is from Riverhead. (3) These totals assume that only the average daily demands are provided from Riverhead and peak needs are supplied from existing Greenport wells. TABLE 8-17 ESTIMATED ANNUAL D~ELLING UNIT COSTS OF WATER SUPPLY ALTERNATIVES for the SOUTHOLD/GREENPORT DEMAND CENTER ESTIMATED ANNUAL COST PER DWELLING UNIT ESTIMATED NO. ' ......................................................................... LEVEL V* OF YR. 2000 LEVEL LEVEL III ' SUB-DEMAND CENTER DWELLING UNITS I IIIA IIIB VAI VB-1I VB-21 VC-11 VC-2I VD-11 VD-2 $ $ $ $ $ $ $ $ $ $ Greenport Municipal System 2700 --- 545 490 500 395 530 430 515 410 Great Hog Neck 482 605 2365 --- 2135 2170 East Marion 454 605 1055 --- 1055 1045 *Treatment not required . , Figure 8-11 Level III Alternative Components for Great Hog Neck SUBOEM~ND CENTER BOUNDARY / WATER TRE PLANT AND PUMP STATION WELL WITH PUMP IN VAULT (TYP) 0 \ \ \ \ \ \ \ Figure 8-12 Level III Alternative Components for East Marion Two Level III alternatives were evaluated for the Greenport Municipal System, as follows: III(A) - Reverse osmosis water treatment plant Additional distribution piping and storage Great Pond Water Supply System Reverse Osmosis water treatment plant - Additional distribution piping and storage. The components associated with these alternatives are shown concep- tually on Figure 8-13. As indicated, additional distribution system piping (8 miles) and storage would be required under each alterna- tive. Year 2000 distribution system storage requirements wore estimated assuming that storage of 25 percent of the estimated year 2000 maxi- mum day demand would be provided to account for hourly demand fluc- tuations. In addition, a fire-fl ow volume was also included based on a fire flow of 1000 gpm for a 1-hour duration. The resulting total required storage volume is 700,000 gallons. The existing system storage facility, which is located on Moores Lane, has a capacity of 300,000 gallons of active storage, resulting in an estimated deficit of 400,000 gallons. Accordingly, alternatives III(A) and III(B) in- clude costs for a 400,O00-gallon elevated storage tank with the same overflow elevation (El. 178 feet, USGS base) as the existing tank. This would allow continuation of the present system operating condi- tions and would maintain current system pressures. The tank location would be somewhat dependent upon the water supply plan ultimately selected for Greenport and a hydrualic analysis of the piping system, which is beyond the scope of the present study. A possible tank location would be adjacent to the existing tank, which would serve to centralize system storage and equalize system pressures if a new sup- ply source wore located in the western portion of Greenport. The components associated with Level III(A) include a 1500-gpm capa- city reverse osmosis water treatment plant (WTP) which would serve to remove nitrates, pesticides and total dissolved solids from the raw groundwater. Under this option, the existing carbon filter would be removed at well 6-1, increasing the capacity from 400 gpm to its original 550 gpm. With woll 6-2, the total Plant 6 capacity would be 700 gpm. Plant 7 would also be utilized as a raw water source and has a pumping capacity of 350 gpm. llqe third proposed raw water source is an existing farm well located off of Middle Road near Tuckers Lane, known as Donohue's Farm well. It has a reported yield of about 450 gpm. It was assumed that renovations would include a new pump and motor and associated piping and appurtenances, electri- cal controls and a new structure. Cleaning of the well screen may also be required. 8-57 LEGEND: EXISTtNG WATER SUPPLY PLANT ---- EXISTING MAJOR SYSTEM PtPING LEVEL 22[ A ALTERNATIVE FigureS-13 Level III Alternative Components for Greenport Municipal System At plants 6 and 7, the piping connections to the existing 12-inch main in Middle Road would be blocked (or valved~ and a new raw water transmission system constructed from the wells, including Donohue's Farm well, to the WTP. It was assumed that the plant would be located along Middle Road opposite the Donohue's Farm well, as shown on Figure 8-13. The raw water would be metered and then chemically treated with sulfuric acid and sodium hexametaphosphate (SHMP} prior to filtration. The acid is required to lower the pH to prevent cal- cium carbonate deposition and fouling of the membranes. SHMP is added to prevent precipitation of iron, manganese and/or other con- stituents which would also foul the membranes. The chemically treated water would then pass through cartridge filters which would remove any suspended solids greater than 10 microns in size. This protects the membranes and increases the length of time between mem- brane cleaning and, eventually, replacement. High-pressure pumps would boost the pressure to about 300 to 400 psi and the chemically treated water would enter the R-O membrane units. The brine reject, which is that portion of the water that does not pass through the membranes, would be collected and transported out of the plant for disposal. Typically, the volume of this waste stream averages about 20 to 25 percent of the total plant flow, resulting in an average flow rate of about 300 to 375 gpm. It would contain nitrates, pesticides, total dissolved solids and other constituents found in the groundwater, in concentrations approximately four to five times greater than the raw groundwater. It was assumed for this study that the waste stream could be discharged into Jockey Creek since the dilution offered by the creek would, most probably, negate any poten- tial adverse effects. However, this would require detailed investi- gations including receiving water analyses and impact assessment, in addition to approvals from local, State and federal regulatory agen- cies. of the water that does pass through the mem- decarbonation to remove carbon dioxide and other may be entrained within the flow stream due to The 75 to 80 percent branes would require dissolved gases that the addition of acid and the lowering of the pH. Removal of carbon dioxide would serve to increase the finished water pH and prevent air buildup within the distribution system. Accordingly, a forced-draft degassifier, which operates similarly to an air-stripping tower com- monly used for ammonia removal, was included in the cost estimate of the plant. Sodium hypochlorite, or another form of chlorine, would then be added to the flow stream for disinfection prior to pumping the finished water into the distribution system. The treatment facilities would be contained within a one-story super- structure, assumed to be constructed of concrete block with a brick facade. The approximate building size would be on the order of 3,000 to 3,500 square feet. Staff requirements would include one full-time plant operator and one full-time assistant. It was assumed that Greenport Water Department personnel would also be avail able to as- sist in WTP maintenance and repair when required. 8-59 As discussed above, 20 to 25 percent of the raw water entering the plant would be rejected; thus, the plant's actual production capacity would be in the range of 1,100 to 1,200 gpm. Since the estimated year 2000 maximum day demand for Greenport is 1,780 gpm, some of the existing supplies must remain in service. It was assumed that plants 4, 5 and 8 would remain in service and be used, when required, for peaking purposes. It was also assumed that with reduced usage of these wells and blending with the WTP finished water, overall system water quality would comply with all County and State water quality standards. The Level III(B) alternative for Greenport is based on a concept that was discussed previously and was developed in more detail for this study in order to assess potential costs and how they compare with other Greenport alternatives. It involves the use of Great Pond, which is located in the northwest portion of Greenport's system, as a water supply source. It has been reported that the nitrate levels in Great Pond water are considerably lower than levels in the ground- water of the surrounding area. It has been further assumed that this was most likely due to a natural denitrification process occuring in the bottom sediments of the pond, where bacteria would convert nitrates into nitrogen gas. It has been estimated that the potential yield of the pond would be in the order of 500 gpm; however, it could be as high as 1,000 gpm. Level III(B) is based on the assumption that Great Pond water contains acceptable levels of nitrates and that no treatment, other than disinfection, would be required. A system capacity of 500 gpm was also assumed. A concept of utilizing Great Pond water without a high degree of treatment was developed for this study and is shown schematically on Figure 8-14. It involves the construction of a groundwater injection system including a pond water pumping station. Pond water would be withdrawn from the pond and injected into the upper glacial aquifer through a number of injection wells arranged around a central produc- tion well. The production well would be a typical well and pumping station containing chemical addition facilities and the necessary system controls. This arrangement serves two purposes: (1) the injection wells would increase the water table ~evation around the production well, creating a hydrologic barrier between the well and the aquifer, which would prevent "short-circuiting" by assuring that pond water and not contaminated aquifer water was delivered to the system, and (2) pond water would travel through several feet of soil before it reaches the production well, taking advantage of the natural purifying action of the soil to reduce turbidity and increase overall water quality. It must be emphasized that this is a concept only, and its applicability is dependent upon actual pond water qual- ity, groundwater conditions and movement in the area and subsurface soil conditions. A full engineering investigation would be required prior to serious consideration of this alternative. The estimated cost of such an investigation was included in the Level III(B) cost estimate (approximately $90,000/, based on performing the following work tasks: 8-60 I I I PLAN WATER TABLE WITH SYSTEM IN OPERATION ~ --/ NORMAL WATER TABLE GROUND /--PRODUCTION WELL __:[.~ ~~ /-- ~JECTION WELL (TYPICAL)~7 SECTION Fig u re 8-14 Schematic of Level I1! (B) for Greenpoint - Site investigation - Exploratory program (including test wells) - Installation of monitoring wells Extended duration pump test (4 months) Water quality sampling and analyses Analysis of results and development of recommendations Preparation of engineering report. Since this system would deliver only 500 gpm of finished water, Level III(B) also includes construction of a 1,050-gpm (1.5 mgd) reverse osmosis water treatment plant, similar to Level III(A), which would treat water from plants 6 and 7 only. Assuming a 25 percent reject by the R-O process, the WTP production capacity would be about 800 gpm, resulting in a total of 1,300 gpm of potable water production including the Great Pond system. Utilization of Greenport's other existing wells, to a limited extent, would be required under this alternative, similar to III(A), to provide up to maximum day produc- tion capacity (1,780 gpm). As shown on Table 8-17, the estimated cost of Level III(B) is slightly less expensive than III(A) since the Great Pond system does not include treatment. If the Great Pond system could be developed as presently conceived, which would be determined from the above- described engineering study, Level III(B) would be more cost-effective than III(A). 8.4.6.3 Level IV Alternatives. Level IV for the Southold/Greenport demand center was not considered for reasons similar to those dis- cussed in connection with the Matticuck/Cutchogue demand center. Since about 90 percent of the Level III costs for Great Hog Neck and East Marion are for distribution system piping, a significant cost savings would not be realized by supplying these areas from the Greenport system. Estimated costs per dwelling unit would still be over $2,000 for Great Hog Neck and close to $1,000 for East Marion. 8.4.6.4 Level V Alternatives. llne components and preliminary design criteria associated with Level V alternatives were described in Sec- tion 8.4.5.5, since most of the altrnatives also include the Matti- tuck/Cutchogue demand center. In summary, V(A) involves Mattituck/ Cutchogue only and does not include Greenport, V(B) provides service to Mattituck/Cutchogue and the Southold/Greenport demand centers, V(C) provides service only to the Greenport Municipal System, and V(D) provides service to the Mattituck West and East subdemand centers in addition to the Greenport Municipal System. For each 8-62 I I I I I I I I I I I i I I I I I I alternative, the water supply from deep wells in eastern Riverhead would be conveyed to the demand centers by a transmission piping and booster pumping system. A conceptual layout of the components of Level V alternatives is shown on Figure 8-10, which also shows the transmission pipeline route options. As indicated in Table 8-17, two scenarios were considered for alter- natives V(B), V(C), and V(D). Alternatives V(B-1), V(C-1) and V(D-1) would provide supply from Riverhead equal to Greenport's estimated year 2000 maximum daily demand. Under these options the use of Greenport'~ing well supplies would be discontinued. Alterna- tives V(B-2), V(C-2} and V(D-2) would provide supply equal to Green- port's year 2000 average, daily demand, with selected existing well supplies remaining in service, in order to provide maximum day pro- duction capability. Under all alternatives, transmission system connections would be made to Greenport's existing system at its western extremity. The hydrau- lic grade line (HGL) of the transmission system would exceed Green- port's HGL at that location in the system in order to convey water into Greenport's system. Distribution system piping and storage would be required for Greenport as discussed under Level III in Sec- tion 8.4.6.2, and also for Great Hog Neck and East Marion under alternative V(B). Supply costs were distributed to the demand centers (or subdemand centers/ on the basis of design flow which, except for Greenport under V(B-2), V(C-2) and V(D-2), was the esti- mated year 2000 maximum day demand. Transmission system costs were also distributed based on flow; however, they were also dependent on proximity of the demand center in relation to the transmission sys- tem. For example, referring to Figure 8-10, which depicts alterna- tive V(B~, estimated costs for the transmission system from the supp- lies to the eastern storage tank serving Mattituck/Cutchogue were proportioned between Mattituck/Cutchogue and Southold/Greenport based on year 2000 maximum day demands. However, costs for the remaining portion of the system, from the tank to the connection(s) with Green- port's system, were allocated to Southold/Greenport only. As mentioned above, scenario 2 (i.e., V(B-2~, V(C-2) and V(D-2)) in- cludes continuation of the use of certain existing supplies in Green- port, since only year 2000 average day demand would be supplied from the Riverhead wells. This scenario was considered in order to deter- mine the impact on costs-to Greenport by reducing the Riverhead sup- ply requirement and associated facilities. For alternatives V(C-2) and V(D-2), the estimated year 2000 average day that would be supp- lied to the Greenport Municipal System is 660 gpm. Since the esti- mated maximum day demand is 1,780 gpm, a system supply capability in Greenport of about 1,120 gpm must be maintained. Reportedly, the highest quality well supplies in Greenport are Plant 6, which is equipped with a carbon filter (550 gpm), Plant 7 (350 gpm) and Plant 8 (350 gpm). The total combined capacity of these supplies is 1,250 8-63 gpm. Reportedly, plants 7 and 8 occasionally experience high chlor- ide concentrations; however, with reduced usage, the quality of the Greenport wells remaining in service would improve and the overall system water quality would comply with all federal, State and County standards. Prior to serious consideration of this scenario, further engineering study and water quality analyses on Greenport's wells would be required. Referring to Table 8-18, the cost impact of distribution system piping is apparent for Great Hog Neck and East Marion, since there is essentially no difference in cost between Level III and Level V(B). Level III involves a local, treated groundwater supply while under Level V(B) the supply would emanate from Riverhead. Source of supply has little, if any, significant impact on costs in these subdemand centers since, under either level, piping represents about 90 percent of total capital costs. As mentioned earlier, these subdemand cen- ters are not conducive to support of a public water system. The Level V costs for the Greenport Municipal System are similar for each alternative. There is a slight cost advantage to Greenport if the supply and transmission costs were shared with Mattituck/Cutch- ogue under V(B) or with only Mattituck West and East under V(D). However, as discussed in Section 8.4.5, Level V(B) does not appear to be cost-effective for Mattituck/Cutchogue if compared with Level I. Levels V(C) and V(D) do appear to be cost competitive with other options, especially with Level III for Greenport. If Greenport's existing supplies were to remain in service under scenario 2, costs of Levels V(C) and V(D) would be reduced an estimated 20 to 25 per- cent. As discussed in Section 8.4.5, alternative transmission main routes were considered for Levels V(C) and V(DI, as shown on Figure 8-10. The route selected for detailed transmission system analysis gener- ally follows Route 25 and the costs shown in Table 8-17 reflect that route. The two options considered were: (1) generally along North Road (R~ute 27) and (2) generally along the LILCO right-of-way, as- suming access would be available. The cost comparisons for the Greenport Municipal System were summarized as follows: Greenport Municipal Estimated Annual Cost Per Dwelling Unit System Route 25 Option i Option 2 V(C-1) $530 $520 $475 V(C-2) 430 420 465 V{D-1) 515 510 395 V(D-2) 410 400 380 8-64 TABLE 8-18 LEVEL III ALTERNATIVE COMPONENTS for the GREAT HOG NECK AND EAST MARION SUB-DEMAND CENTERS SUPPLY TREATMENT(I) TRANSMISSION/DISTRIBUTION ESTIM. YR. 2000 DEMANDS NO. OF AVE. TOTAL WELL BOOSTER P.S. SYSTEM LENGTH OF(2) ACTIVE DISTRIBUTION SUB-DEMAND CENTER AVE. MAX PEAK WELLS DEPTH CAPACITY CAPACITY CAPACITY PIPING, MILES SYSTEM STORAGE VOL, GAL Great Hog Neck 120 165 210 2 60' 170 170 170 28.4 105,000 East Marion 80 150 220 2 70' 150 220 150 11.1 (3) Notes: General--All flows are in gallons-per-minute (gpm) (1)Treatment consists of a carbon/ion exchange system having a treatment design capacity equivalent to max day demands and a hydraulic design capacity equivalent to peak hour demands (2)Includes the estimated length of required transmission and distribution piping to serve the estimated year 2000 population assuming continuation of historical growth patterns (3)Distribution system storage not warranted--a hydropneumatic tank would be provided at the booster pumping station As indicated, a significant cost savings would not be realized if Option 1 were pursued; however, Option 2 costs do average about 10 percent less than the Route 25 costs since paving restoration would not be required along the LILCO right-of-way. Costs for utilizing the easement, however, were not included in the above estimates. If alternative V(C) or V(DI were considered for eventual implementation, Option 2 should be further investigated and LILCO contacted to deter- mine if use of their easement would be allowed. Table 8-19 presents the capital cost estimates for Level III and V Alternatives for the Southold/Greenport demand center. 8.4.7 Orient The estimated total annual costs per dwelling unit for water supply alternatives in the Orient Demand Center are summarized in Table 8-20. TABLE 8-20 ESTIMATED ANNUAL DWELLING UNIT COSTS OF WATER SUPPLY ALTERNATIVES FOR THE ORIENT DEMAND CENTER Estimated Annual Cost per Dwelling Unit Estimated Number of Level Level Year 2000 Dwelling Units I III 593 $620 $1,055 8.4.7.1 Level I Alternatives. As discussed in previous sections, Level I consists of individual home water supply systems and includes the installation of a treatment unit in existing homes, and the in- stallation of a well plus a treatment unit for new homes. Treatment would consist of a reverse osmosis system for removal of nitrates, pesticides and/or total dissolved solids. The costs shown in Table 8-20 reflect a weighted average of the costs of facilities for exist- ing homes and the costs of facilities for new homes, based on the estimated number of new homes by the year 2000. 8.4.7.2 Level III Alternatives. As with the other demand centers, the Level III alternative consists of a new groundwater supply, a water treatment system and a transmission/distribution system. Figure 8-15 shows a conceptual layout of the major system components. Two wells, approximately 90-feet deep, would be constructed, as 8-66 Subdemand Center Supply LEVEL III TABLE 8-19 COST ESTIMATES GREENPORT/SOUTHOLD DEMAND CENTERS LEVEL V(B) Trans. & Treat- Total Total Trans. & Treat- Total Distribu- ment Const. Project Distribu- ment Const. tion Cost Cost Cost Supply tion Cost Cost Greenport Great Hogs Neck East Marion TOTALS $130,000 $2,725,000 135,000 6,480,000 $160,000 2,450,000 $425,000 $11,655,000 $3,690,000 $6,545,000 $8,510,000 300,000 6,915,000 8,990,000 280,000 .2,890,000 3,760,000 $4,270,000 $16,350,000 $21,260,000 $1,015,000 $15,880 000 $16,895,000 $21,963,000(1) $ 450,000 $13,985,000 $14,435,000 $18,765,600(2) Subdemand Center Supply Greenport Great Hogs Neck East Marion Totals LEVEL V(C) LEVEL V(D) Trans. & Distribu- Total tion Const. Trans. & Total Distribu- Total Total Project Supply tion Const. Project $1,050,000 $6,985,000 $8,035,000 $10,445,000 $970,000 $6,835,000 $7,805,000 $10,145,000 $1,050,000 $6,985,000 $8,035,000 $10,445,000 $970,000 $6,835,000 $7,805,000 $10,145,000(1) $ 370,000 $5,425,000 $5,795,000 $ 7,535,000 $335,000 $5,205,000 $5,540,000 $ 7,200,000 IDEMAND CENTER BOUNDARY ~LL WITH PUMP IN VAULT (TYP) . WATER TI~EATMEN1: PLANT AND N Figure 8-15 Level III Alternative Components for Orient shown, and tap the upper glacial aquifer. Each well would be equip- ped with a submersible vertical turbine pump with the motor located in an underground vault. The wells would be sized for a total pro- duction capacity of 200 gpm {100 gpm each) which is equivalent to the estimated year 2000 maximum day demand for Orient. Water would be conveyed to a treatment facility which would employ the processes of ion exchange and carbon adsorption for removal of nitrates and pesti- cides. Chemicals for disinfection and pH control would also be applied. Treated water would then be pumped into the distribution system for delivery to consumers. The booster pumping station would operate off a hydropneumatic tank and would include standby power. It is estimated that about 14 miles of distribution system piping would be required to serve this area. As with Mattituck/Cutchogue and several other areas on the North Fork, the relatively low density of Orient results in significantly high piping costs. The density of Orient is essentially the same as East Marion (an average of about 40 homes per mile of roadway) resulting in comparable Level III costs which, in both cases, are almost double the Level I costs. 8.4.7.3 Level IV and ¥ Alternatives. Levels IV and V were not con- sidered for Orient since a preliminary analysis revealed that costs would be greater than Level III. This is primarily due to the transmission piping required to serve this relatively isolated area from the Greenport system. This piping requirement essentially negates any potential cost savings resulting from the elimination of the water treatment system required under Level III. Table 8-21 presents the capital cost estimate for the Level III alternative in Orient. TABLE 8-21 CAPITAL COST ESTIMATES ORIENT DEMAND CENTER (ENR-3800) LEVEL III Supply $ 185,000 Transmission and 3,180,000 Distribution Treatment 400,000 Total Construction Cost Total Project Cost $3,765,000 $4,895,000 8-69 8.4.8 Isolated Neighborhood Systems The estimated costs for Level I and Level II water supply alterna- tives for the isolated neighborhood systems are summarized on Table 8-22. As indicated, treatment would not be required for the neigh- borhoods located in the western portion of the North Fork because uncontaminated groundwater sources are available. As with several of the subdemand centers located throughout the North Fork, distribution piping represents a substantial portion of the Level II costs. Although these neighborhoods are not as sparsely populated as several of the eastern subdemand centers, the housing density is relatively low, averaging about 60 homes per mile of road- way. The required piping and associated costs to provide public water ser- vice to neighborhood homes is the m~or reason for the higher Level II costs, as compared to Level I. Table 8-23 presents the capital cost estimates for the Level II Alternatives for the neighborhood systems. 8.5 COMPARISON OF ALTERNATIVES Several criteria must be considered when comparing and evaluating water supply alternatives for the North Fork. The criteria used in the evaluation include the following: Cost Reliability Implementability Environmental considerations - Adaptability to future changes. 8.5.1 Cost The cost of the various alternatives is perhaps the most straightfor- ward criteria to evaluate, and certainly one of the most important. The cost is comprised of two components: The capital required to construct the components of the sys- tem (i.e., source of supply, transmission/distribution, and treatment) 8-70 TABLE 8-22 ESTIMATED COSTS OF WATER SUPPLY ALTERNATIVES for ISOLATED NEIGHBORHOOD SYSTEMS ESTIMATED NO. ESTIMATED ANNUAL COST PER DWELLING UNIT OF YR.2000 ........................................... NEIGHBORHOOD SYSTEMS DWELLING UNITS LEVEL I LEVEL II $ $ 1. Lake Panamoka 115 230* 485* 2. Rt. 25 S. of Scuttle Hole 44 230* 865* 3. Rt. 25 N. of Grumman Airport 150 230* 615' 4. Baiting Hollow 43 325* 1000' 5. Off Tuthill Lane 71 225* 675* 6. Duck Pond Point 70 605 785 7. Rt. 27 near Depot Lane 43 605 930 8. Goldsmith Inlet 132 605 665 9. Great Pond 165 605 805 I I I *Treatment not required m m~ m m m m, m m m m m mm m m m m m m m TABLE 8-23 CAPITAL COST ESTIMATES NEIGHBORHOOD SYSTEMS (ENR-3800) Neighborhoods 1. Lake Panamoka 2. Rt. 25 S. of Scuttle Hole 3. Rt. 25 N. of Grun~nan Airport 4. Baiting Hollow 5. Off Tuthill Lane 6. Duck Pond Point 7. Rt. 27 near Aluans Lane 8. Goldsmith Inlet 9. Great Pond TOTAL Supply $ 90,000 65,000 140,000 100,000 130,000 65,000 60,000 80,000 95,000 $825,000 Transmission And Distribution $ 220,000 120,000 390,000 120,000 120,000 160,000 80,000 305,000 550,000 $2,065,000 Treatment Total $ 310 000 185 000 530 000 280 000 350.000 285 000 185 000 475 000 745.000 Not Required Not Required Not Required $ 60,000 75,000 60,000 45,000 90,000 100,000 $430,000. $3,320,000 Total Project Cost (Inc. 30% E&C} $4,320,000 $ 400 000 240 000 690 000 365.000 425 000 370.000 240 000 620 000 970000 (2) The cost of administering, operating and maintaining the system: labor, power, chemicals, supplies and routine main- tenance; this cost is generally referred to as the O&M cost. In order to compare the total cost of each of the alternatives, it was necessary to place the capital and O&M costs on a common basis. The approach utilized in this study was to amortize the total capital cost (using an appropriate interest rate and payback period) to yield an annual cost required to finance the construction of the project. This annual amortized cost was then added to the annual O&M cost to determine the total annual expenditures required. The annual costs were then divided by the estimated number of dwellings served by the alternative (in the year 2000) to develop average annual costs per dwelling unit. These costs were then used as the basis of comparing the alternatives. 8.5.2 Reliability Reliability addresses a series of considerations which bear upon the ability of the water supply system to consistently operate in a satisfactory manner. The following aspects were considered: (1) The reliability of the source of supply to provide water which meets required quality standards (2) The reliability of the treatment processes employed to re- move objectional constituents found in the raw water (3) The reliability of equipment and hardware to consistently remove contaminants from a raw water supply (4) The reliability of the water supply system to consistently deliver water of adequate pressures and volumes. The number of supply sources (i.e., wells), the provision of auxiliary power, and the provision of competent system operators were important considerations in evaluating system reliability. 8.5.3 Implementability The potential for implementing the project is an important criteria used to evaluate the alternatives. Two issues related to project implementation were considered: Project scope--In general, capital-intensive projects cover- ing large geographic areas are more difficult to implement than projects with a more limited geographic scope. 8-73 I I I I I I I I I (2) Public and political acceptability--Alternatives which in- volve the transfer of water from one municipality to another are more difficult to implement from a political standpoint. This is due to potential conflicts in the development goals of the two municipalities and to the complications that arise from the need to involve additional government bodies in the decision-making process. Alternatives which provide needed service and meet local growth objectives would re- ceive greatest public acceptability. 8.5.4 Environmental Considerations The potential environmental consequences of the alternatives con- sidered are: (1) Consumptive water use--This criteria considered the anti- cipated consumptive use of water in each zone with respect to the permissive sustained yield. Transfer of water from one supply zone to another was a particularly important con- sideration, because such withdrawals are totally consumptive from the perspective of the zone supplying the water. (2) Adherence to land use objectives--All alternatives were designed to meet the land use objectives expressed in the Comprehensive Plan. However, because of differences in scope and the physical configuration of the supply and transmission systems, the alternatives differ in their potential for stimulating undesired growth. Therefore, alternatives which minimized the potential for these nega- tive "secondary impacts" were deemed more desirable. (3) Construction impacts--Water supply system components were located, as far as practical, to minimize encroachment on sensitive environmental areas. However, alternatives differ in their impacts on these types of areas, particularly with respect to stream crossings required for the construction of major transmission lines. 8.5.5 Adaptability to Future Changes Depending upon the type and character of each alternative, some are more flexible in responding to unforeseen future changes than others. Those alternatives that require major initial construction and imme- diate financing tend to be less flexible than those alternatives which are gradually implemented over time. Because future conditions in the study area cannot be predicted wi th a high degree of cer- tainty, adaptability of an alternative to future change is an admir- able, desirable characteristic. 8-74 m m ! m m m 8.5.6 Summary Comparative Matrices Tables 8-24 through 8-29 compare all levels of alternatives which were developed for each demand center on the basis of the five screening criteria. (121/5) 8-75 m ~Evaluatt on  riteria A1 ternative ~ Level I- Individual Systems TABLE 8-24 C04PARISON OF ALTERNATIVES WADING BIVER/NORTHVILLE DEMAND CENTER Cost Reliability Implementability Because of the depth to water, the cost of Level I systems are as high as the cost of providing public water supply. Individual systems would be reliable if they were included in a Home Treatment Unit District. The individual systems provided wi th treatment could be included in the townwide Water Manage- ment Program to facili- tate implementation. Environmental Considerations No major impacts are anticipated. Adaptability The commitment of funds to purchase individual treatment units might preclude the provision of public supplies in the future. Level II - Augment Existing Public Supply Systems The costs of individually augmenting extsttng public systems are generally higher than abandoning the supply and connecting the systems to new subdemand center systems. The operation of these small systems has not been good. The size of systems makes it dif- ficult to provide profes- sional operation and maintenance. Although the required improvements are rela- tively small, they will be difficult to implement because of the limited financial resources of the systems. NO major environmental impacts are anticipated. Little opportunity for phasing of facilities. Level III- Subdemand Center Systems The provision of subdemand center systems, which pro- vide services to both cur- rently unserved areas and existing systems, generally resulted in the lowest cost per dwelling unit. The provision of larger systems serving each of the three subdemand cen- ters will increase the reliability of service to the areas. The systems serve small, well-defined areas. Because of the obvious need for these systems, strong local support is anticipated. Adequate groundwater resources are available for public systems. No major stream crossings would be required to transmit water from sup- ply to point of use. System would be con- structed in phases to increase its ability to adapt to future changes. TABLE 8-25 COMPARISON OF ALTERNATIVES RIVERHEAD/JAMESPORT DEMAND CENTER  Evaluatton eria Cost Reliability Implementabtlity Level I - Individual Systems In Jamesport, the Level I costs were significantly lower than the other alter- natives. In other areas, it was as high as the cost of public supply. Individual systems would be reliable if they were included in a Home Treat- ment Unit District. The individual systems provided wi th treatment could be included in the townwide Water Management Program to facilitate implementation. Environmental Considerations No major impacts are anticipated. Adaptability. The commitment of funds to purchase individual treat- ment units might preclude the provision of public supplies in the future. Level III - Subdemand Center Systems Level III alternatives were only economically Justified for the Riverhead subdemand center. The existing Riverhead system is highly reliable. Only minor system improve- ments are needed to meet future requirements. The Riverhead system has No major impacts would be been operating for many associated wi th the Level years. In Calverton and III options in any of the Jamesport, the cost of new subdemand centers. systems would make imple- mentation difficult. The supply and major transmission systems in Jomesport & Calverton would have to be con- structed initially: other facilities could be phased. Level IV - Subregional System The Level IV costs were competitive with Level I costs only in Calverton. Supply for Calverton is from the Riverhead system and therefore considered more reliable than Level I or III. No implementation problems Transmission lines are are anticipated in short and, therefore, will Calverton. have minor impacts; sec- ondary growth impacts have not been identified. Distribution system can be phased over time as the financial resources become available. TABLE 8-26 ~ Evaluation  riterta Alternative~ COMPARISON OF ALTERNATIVES MATTITUCK/CUTCHQGUE DEMAND CENTER Environmental Cost Reliability Implementability Considerations Adaptability Level I - Lowest cost for all sub- Individual systems would The individual systems Individual wells are Individual demand centers, be reliable if they were provided with treatment well-suited for these Systems included in a Home Treat- could be included in the areas. merit Unit District. townwide Water Manage- ment Program. The commitment of funds to purchase individual treat- ment units might preclude the provision of public supplies in the future. Level III - Subdemand Center Systems Unaffordable for all areas except Mattituck East and West. Treatment would be reli- able if provided at a central facility. Major costs of construc- tion for distribution systems would hamper the implementability of this alternative in all subde- mand centers. Adequate water exists to supply future needs: transmission/distribution systems would cross many environmentally sensi- tive areas. Little opportunity for phasing of facilities. Level V - Regional Systems Unaffordable for all areas except Mattituck East and West. The quality of the groundwater beneath the clay in eastern Riverhead cannot be guaranteed in the future. Riverhead would have to consent to the withdrawal of water from zone 2. Construction of the trans- mission line would be dif- ficult to implement because of its expense and because it crosses political boundaries. Withdrawals from zone 2 could exceed permissive sustained yield. Many major stream crossings would be required. Depending on the route selected, major secon- dary impacts can be anticipated along route of transmission lines. Most required facilities cannot be phased. Major transmission line, stor- age and source of supply (in Riverhead) must be construc ted ilnmediately. TABLE 8-27  Evaluatton teria Cost COMPARISON OF ALTERNATIVES GREENPORT/SOUTHOLD DEMAND CENTER Environmental Beliability Implementability Considerations Adaptability Level I - Level I costs for Great Individual systems would The individual systems Individual wells are Individual Hogs Neck and East Marion be reliable if they were provided with treatment well-suited for these Systems were significantly lower included in a Home Treat- could be included in the areas. than Level III and V. merit Unit District. townwide Water Management Program. The commitment of funds to purchase individual treat- ment units might preclude the provision of public supplies in these areas. Level III - Subdemand Center Systems Cost of Level III for Greenport was comparable to Level V, Level 1II was unaffordable for other subdemand centers. Treatment will be reli- able because of the high level of competence dem- onstrated by the Green- port system personnel. Becaused required facil- ties are within Green- port's franchise area, the implementability is enhanced. Adequate supply exists in Zones 3 and 4 to meet Greenport's needs. Treatment facility would have to be constructed initially; transmission/ distribution can be phased. Level V - Regional. Systems Regional systems were not affordable for East Marion and great hog neck. level v was only cost competitive for the Greenport system if a clean source of water could be found (beneath the clay) in eastern Riverhead. The quality of the ground- water beneath the clay in eastern rtverhead cannot be guaranteed in the future. Rtverhead would have to consent to the withdrawal of water from zone 2. con- struction of the transmis- sion lines would be diffi- cult to implement because of its expense and because it crosses several politi- cal boundaries. Withdrawals from zone 2 could exceed permissive sustained yield, many major stream crossings would be required. Depending on the route selected, major secon- dary impacts can be anticipated along route of transmission lines. Most required facilities cannot be phased. Major transmission line, stor- age and source of supply (in Riverhead) must be constructed immediately. TABLE 8-28 COMPARISON OF ALTERNATIVES ORIENT DEMAND CENTER ~Evaluation  riteria Alternative~x Cost Reliability Implementability Environmental Considerations Level I - Individual Systems The cost of Level I was significantly less than the other alternatives considered. Individual systems would be reliable if they were included in a Home Treat- ment Unit District. The individual systems provided with treatment could be included in the townwide Water Manage- ment Program. Individual wells will spread out the pumpage patterns and better utilize the water re- sources of the demand center. Adaptability The commitment of funds to purchase individual treatment units might preclude the provision of public supplies in the future Level II - Neighborhood Systems The cost of the Level 1I alternatives was beyond the point of economic feasibility. The treatment required could be reliable if pro- vided at a central facility. A large public system would not receive wide public acceptance because it is not compatible with the rural character of the area. There are extremely limited water resources available for large public supply systems. Most major system com- ponenents would have to be provided initially. TABLE 8-29 COMPARISON OF ALTERNATIVES ISOLATED NEIGMBORHOOD AREAS  Evaluatton teria Cost Reliability Implementability Level I - Individual Systems The cost of Level I alter- natives is lower regard- less of whether or not treatment ts required. Individual systems could be reliable if they were included in a Home Treat- ment Unit District. The individual systems provided wi th treatment could be included in the townwide Water Management Program. Environmental Considerations Individual wells would be beneficial in those areas with limited water resources. Adaptability The commitment of funds to purchase individual treat- ment units might preclude the provision of public supplies in the future. Level II - Neighborhood Systems Level II costs were higher and, in some cases, beyond the point of economic fea- sibility. This was the major factor thfluencing the selection of the recommended plan for isolated areas. Treatment, if required, could be reliably pro- vided at a central facility. Level II could not be implemented in many of these areas because of the extremely high cost. No major environmental impacts would be antici- pated. SECTION 9.0 RECOMMENDED WATER SUPPLY PLAN Section 8.0 of this report has presented in detail all of the water supply alternatives studied and evaluated as part of this plan. Sec- tion 8.0 also described the evaluation criteria and procedures used to compare the alternatives. This section enumerates the water supply alternatives recommended for implementation in each demand center. Only brief descriptions of the recommended plans are pre- sented--the reader is referred to Section 8.0 for the detailed descriptions. Table 9-1 summarizes the capital costs of the major structural com- ponents of each recommended alternative. The table does not include the capital costs for individual home treatment units since the number of required units cannot be predicted. Section 8.0 of the report, however, includes detailed capital and operation and mainten- ance cost data for home treatment unit systems. 9.1 WADING RIVER/NORTHVILLE The Level III (B) alternative for Wading River/Northville, as shown on Figure 8-6, is recommended for implementation. The existing in- dividual water systems in Wading River, Baiting Hollow/Woodcliff Park and Reeves Park should be connected to new supply wells, which will separately serve each of the three individual subsystems. This is necessary to improve the reliability of the supply and to assure potable water. The new facilities should be located and sized to also serve areas which do not currently have public water systems; the new service areas should be developed in a phased approach. Table 9-1 includes the cost estimates to implement the Level III (B) alternative. The future distribution system costs shown in the table can be incurred over time as a town Water District becomes finan- cially able to provide public water service to those areas that are presently unserved. It is not economically feasible to provide public water systems to the remainder of the demand center outside the service areas shown on Figure 8-6. The residents in those areas should continue to be served by individual home wells, wi th treatment as required. 9-1 TABLE 9-1 SUMMARY OF CAPITAL COSTS FOR THE RECObiMENDED PLAN (ENR - 3800) System Components Wadin~ River Demand Center Initia) Future Total Project Construction Construction Cost Rlverhead Demand Center Initial Future Total Project Construction Construction Cost Mattituck/Cutchogue Demand Center Initial Future Total Project Construction Construction Cost Supply $1,785,000 Transmission $2, I00,000 Distribution $ 665,000 Treatment Total Con- struction Cost $4,550,000 ~_ ..... $1,785,000 ...... $2,100,000 $9,855,000(1) $10,520,000 Not ReoNi red $9,855,000 $14,405,000 $12,815,000(1) $18,730,000 Total Pro- ject Cost $5,915,000 Greenport Demand Center Supply $ 130,000 .... $130.000 Transmission $ 300,000 .... $300,000 Distribution .... $2,425,000 $2,425,000 Treai~ent $3,690,000 .... $3,690,000 Total Con- struction Cost $4,120,000 $2,425,000 $6,545,000 Total Pro- ject Cost $5,355,0(30 $3,155,000 $8,510,000 $5,000 ..... $5,000 $10,000 Not Required .... .__ $2,275,000 $2,275,000 .... $5,000 $2,275,000 $2,280,000 $250,000 $5,000 $2,275,000 $2,280,000 $260,000 $7,000 $2,957,000 $2,964,000 $340,000 $10,000 $250,000 $260,000 $340,000 Future potential distribution system costs to be incurred only if new service areas are financially capable o' supporting public water systems. Schedule. The connections of the existing systems to new supply we---~l~-~ould be accomplished within a 3-year period. Service to new systems along the route of the transmission mains should be accomp- lished within 7 years thereafter, providing the areas are financially capable of supporting the new facilities. 9.2 RIVERHEAD/jAMESPORT The Level III alternative is recommended for the hamlet of Riverhead; the only required major improvement is to increase distribution sys- tem storage for the cost shown in Table 9-1. The Riverhead system should al so actively attempt to further expand its system to other areas adjacent to the present system. The Level IV alternative is recommended for the Calverton Area and consists of extending the Riverhead system to serve the area shown on Figure 8-7. Jamesport and the remainder of the area in the demand center should adopt the Level I alternative (i.e., continue to be served by indivi- dual home wells with treatment as required, since it is not economi- cally feasible to serve those areas with public water facilities). The capital cost estimate for the Riverhead/Jamesport recommendations are included in Table 9-1. Schedule. Additional distribution system storage and the extension ~iverhead system to Calverton should be accomplishd within a 3-year period. 9.3 MATTITUCK/CUCTHOGUE The Town of Southold should acquire and operate the existing water system in Mattituck Hills (Captain Kidd); measures should be taken immediately to upgrade the performance and reliability of the system and to augment its source of supply. In the remainder of the Matti- tuck/Cutchogue demand center (including Cutchogue, Mattituck, Little Hog Neck, East Cutchogue, Fleets Neck, New Suffolk and Indian Neck) it is not economically feasible to provide public water supply sys- tems. The Level I alternative is recommended and these areas should continue to utilize individual home wells with treatment as required. Table 9-1 includes capital cost estimates for implementing the Mattituck/Cuthogue recommendations. Schedule. The Captain Kidd acquisition should be completed within ~ and system improvements provided within i year thereafter. 9-3 / i I I I I I i 9.4 SOUTHOLD/GREENPORT The Level III alternative as shown on Figure 8-13, is recommended for Greenport. The Greenport Municipal System, which presently serves Greenport and parts of Southold, should continue to rely on local groundwater sources. However, major improvements to the system are required. The existing Donohue Farm well should be upgraded to 450-gallon-per-minute (gpm) total capacity and used for public water supply. A 2.2-million-gallon-per-daY (mgd~ reverse osmosis treatment plant should be constructed in stages to treat the water from the Donohue well and existing plants no. 6 and 7 for removal of nitrates, pesticides, herbicides and salt. After implementing the improvements, the Greenport system should actively attempt to further expand its system to other areas adjacent to the existing system. In the remainder of the Greenport/Southold area, including Great Hog Neck and East Marion, it is not economically feasible to provide pub- lic water supply systems. These areas should follow a Level I ap- proach and continue to be served by individual home wells, with treatment as required. Schedule. The upgrading of the Greenport system should be staged as ~as possible to minimize the impacts of the required capital cost expenditures as shown in Table 9-1. The Great Pond feasibility study should be undertaken and completed within 18 months. Concur- rently, preliminary design efforts on the reverse osmosis plant should be done. The Donohue Farm well and the final construction of the reverse osmosis plant should be completed within 3 years. The distribution system and storage additions should be implemented within the next 5 to 10 years as the needs develop. 9.5 ORIENT The Orient area, with its relatively low density of development, cannot economically support public water, so the Level I alternative is recommended. Existing development should continue to be served by individual home wells with treatment as required. Since the avail- able fresh water supply is extremely limited, it is further recom- mended that development in the Orient area be tightly controlled to conform with projected growth patterns and result in water require- ments which are consistent with the permissive sustained yield of the aquifer in Zone 5. It is further recommended that only variances re- sulting in less water usage be approved. 9-4 9.6 ISOLATED AREAS Public water systems for existing isolated neighborhoods in areas of groundwater contamination are generally not economically feasible. Therefore, it is recommended to implement Level I alternatives and to continue to serve these areas by individual home wells wi th treatment as required. 9.7 GROUNDWATER MANAGEMENT Throughout this study and planning process, it was apparent that groundwater contamination is extensive and, even if all contaminant imputs to the groundwater were to immediately cease, the contamin- ation will remain for many years. Because of the contamination, numerous technical and financial problems associated with providing water supply to areas of the North Fork were identified. In order to protect the future groundwater quality and to reduce the water supply problems associated with contamination, it is imperative that preventive measures be taken in parallel with implementing the structural components of this plan. At a minimum, the following preventive measures should be considered: (1) Expand the SCDHS observation well network and home well sampling program. (2) Support the Cooperative Extension Service, Cornell University and U.S. Department of Agriculture (USDA) research and educa- tion programs directed to the home owner and farmer relative to usage, dosages and timing of application of herbicides, pesticides and fertilizers. (3) Support the testing of agricultural chemicals by State or federal agencies in the local environment as a precondition to use by the farm community. (4) Prohibit or control the sale or use of products and chemicals which threaten the groundwater resources. (5) Control industrial, commercial and residential activities which impac~ negatively on groundwater quality. (6) Incorporate detailed water quantity and quality considera- tions into rezoning and variance decisions. (71 Encourage water conservation through public information pro- grams and require water-saving fixtures in new home construc- tion. 9-5 Continue public information and education programs to empha- size the fragile nature of the area's water supply and to foster cooperation in the solutions to those problems. 9-6 SECTION 10.0 IMPLEMENTATION PLAN 10.1 IMPLEMENTATION AGENCIES A review was performed of institutional structures in order to select an organizational framework under which the s~ected technical alter- natives could be implenented. The objective of the institutional analysis was to describe the entity that should develop, finance, construct, own and operate the required facilities. Under the provisions of New York State Law, there are four institu- tional entities that could be utilized to implement all or portions of the s~ected alternative: the Village of Greenport, the towns, the County or the Suffolk County Water Authority. Villages (Village Law, Section 11-1102) are able to develop, own an-~ate a municipal water system, or to acquire an existing system by purchase or eminent domain. Towns (Town Law, Article 12-C) are able to dev~op, own and ~perate a "water improvement," by resolution of the Town Board, as a Town function servicing all or a particular area in the Town; a permissive referendum is not required under 12-C. Towns (Town Law, Articl es 12 and 12-A) and Counties (County Law, ~-F~-~-~l e 5-A) can establish special improvenent-~n-~--([T~ricts, ini- tiated by resolution or by petition of the residents of the pro- posed district, with the purpose of developing, owning and oper- ating a water system. The procedures for establishing such are explicitly detailed in the respective sections of the State statutes. Such procedures include allowances for a possible per- missive referendum, requirements for stipulating the maximum capital cost for district improvements, identification of the means of allocating costs within the district, and requirement of State Comptroller approval of the proposed district. Towns can also assume any or all responsibilities for water management as normal administrative functions. Thus, towns can establish departments, groups, programs, etc. to coordinate water dev~opment activities, monitor water quality, provide water testing services, provide info.rmational services and generally serve as the focus of water management activities. 10-1 Counties and Villages (General Municipal Law, Article 14-C) may develop, own and operate a water system as a revenue-producing undertaking. In addition to powers provided to such municipali- ties in other statutes, this statute explicitly authorizes the municipalities to develop water systems on a multi-municipal level by resolution of the governing body of the particular muni- cipality. Under such, no formal special district must be estab- lished. The Suffolk County Water Authority was established under New York State Law for the purposes o~ acquiring, constructing, maintain- ing and operating water supply and distribution systems within the boundaries of Suffolk County. Private water companies can also provide water supply services in New York State. The choice of an institutional structure for implementation should always be dependent upon the alternative course of action selected for implementation. The uniqueness of the recommended plan--a com~ bination of home treatment units and new and expanded municipal sys- tems using local groundwater supplies--further limits the available choices. There is only one incorporated village in the study area, Greenport. Since water quality problems exist throughout the two towns, and townwide water supply actions are required, it is not possible for Greenport to assume implementation responsibility except within its own borders, unless it is contracted to do so by either of the towns. The recommended technical alternative does not involve large, re- gional systems or the transfer of water across municipal boundaries. It focuses upon local sources of supply and control. Thus, there is no incentive for the County to become involved. Additionally, Suf- folk County government has historically left water supply responsibi- lities to local units of government. Similarly, because of the local nature of the recommended plan, it is inconsistent with the past practices of the Suffolk County Water Authority. The Authority, his- torically, has not implemented the type of plan recommended in this report. The Authority might have difficulty financing the plan since the Authority can only utilize revenue-bond financing mechanisms. It is not clear that home treatment units, small water systems for iso- lated neighborhoods and purchasing and interconnecting small, private systems are amenable to complete revenue-bond financing procedures. A1 though the Suffolk County Water Authority does function under fis- cal restraints, it has, however, developed means whereby the Author- ity in conjunction with the towns can provide water service for designated areas not presently being served. Consequently, the Suf- folk County Water Authority has expressed a willingness to partici- 10-2 I I I i i I i 1 i i i I i I i i I i / pate in any discussions regarding the possibility of its providing water service to portions of the North Fork, in accordance with its rules and regulations. Clearly, governmental entities that should be responsible for imple- mentation are the Towns of Riverhead and Southold. There are three alternative methods, as defined in town law, whereby the Towns can provide water supply services. Water supply districts may be estab- lished under Article 12 of Town Law (by petition of the peo@lel or under Article 12-A of Town Law (by motion of the Town Board). Water improvements can be established under Article 12-C. In addition to formally structured programs under Article 12, 12A or 12C, Water Management Programs can be established as administrative actions in the Towns. Under Article 12, a Water Supply District can be created upon receipt of a petition signed by owners of taxable real property owning at least one-half of the assessed valuation of real property. The dist- rict cannot be extended into an incorporated village unless requested by the village. This approach has the benefit that it is initiated by the people; however, it is highly unlikely that this would occur in the study area. Under Article 12-A of Town Law, a town board may, by resolution, establish a water supply district covering portions of the entire town. The action is subject to permissive referendum if 5 percent of the owners of taxable real property sign a petition asking for the referendum. Again, the district may not be established or extended into an incorporated village unless the village consents. This approach is desirable because the towns can initiate action; a dis- benefit is that referendums would probably be required. Article 12-C of Town Law authorizes a town board to provide for water improvements as a town function in the entire area of the town (out- side any villages) without the establishment of a district. The town board is responsible for the management, maintenance, operation and repair of the water supply facilities. The cost of providing water improvments may be on a benefit or ad valorem basis. This differs from Article 12 and 12-A, which provide that the expenses must be charged on a benefit basis. The cost of the water improvement can be borne: - Entirely by the area of the town outside of any villages - Partly by the area of the town outside of villages - Partly by the lands benefited - Entirely by the lands benefited. 10-3 The resolution adopted by the town board authorizing a water improve- ment is subject to a permissive referendum onl~ if all or any part of the costs are to be borne by the entire area of the town outside of any villages. Within the framework of the applicable laws as discussed above, the following approach appears to be the most desirable for the towns to follow. First, the town governments should create Water Management Programs as administrative functions and appoint individuals to direct the programs. These programs would assume responsibility for managing and coordinating all of the water supply programs in the towns. The Water Management Program in each Town would be funded out of general tax revenues and it should provide water services which are common to and which benefit all town residents. For instance, its staff would plan and coordinate all water supply activities; admin- ister all water supply activities; could provide or contract for or arrange with the County for laboratory services for testing of home well samples; serve as a vehicle for education and dissemination of public information; take the lead in forming Water Supply Districts or Home Treatment Unit Districts; interact with the County Health Department. Since these activities benefit all town residents, they should be paid for out of general tax revenues. The staff of the Water Management Program would be limited to a part-time Director (an existing Town employee devoting one-quarter of his time to the Water Management Program), half-time assistant and part-time secretarial personnel. This would result in an annual administrative budget approximately as follows: Salaries Director, one-quarter time Assistant, half-time Secretarial, part time Subtotal, salaries Overhead at 60% Miscellaneous expenses Total Annual Cost Cost $10,000 22,000 5,000 $27,000 16,000 5,000 This estimate represents anticipated costs during the early years of implementation. As Water Supply Districts or Home Treatment Unit Districts were formed, these costs would decrease and these same personnel would devote more of their time (and costs) to the Dist- rict's functions. As specific water supply needs develop, the Program would react to solve them. If private water systems were to be taken over by the town, for instance, the Program would be responsible for organizing 10-4 the establishment of a new Water Supply District or expanding the area of an existing district to purchase, own and operate the system. A number of such districts could be created as private water com- panies were taken over by the Town, or one Water Supply District could be created to own and operate all the private systems. Simi- larly, if new public systems were required, new districts could be formed for implementation. Home treatment units could be handled as a separate town Water Management Program function. A Home Treatment Unit District could be formed covering all areas of the town not served by public systems. Or, several Home Treatment Unit Distri~ could be established in specific areas where they are required, as they become necessary. The Home Treatment Unit Districts could own and maintain the indi- vidual home treatment systems or could create a set of regulations to allow private enterprise to do the same. Under the latter approach, the town should control which home treatment units are to be allowed in the town by retaining approval power. In addition, the town should control maintenance by performing it themselves or contracting it to a private company under control of the town. When Water Supply or Home Treatment Unit Districts are created, the people receiving service in the Districts would pay for the spec~-ffc costs of the services they receive. For instance, if a home needed a home treatment unit, it would pay for the amortized or capital cost of the unit as well as its maintenance, including paying for an annual sample of the well water to be analyzed by the County Health Department. If a home required no treatment, it would not pay any costs associated with a home treatment unit. In this example, both homes would pay for the common services provided to all residents by the Water Management Progra~ described in the previous paragraphs. Because of the use of home treatment units, the Water Management Pro- gram would have to include a continuing, widespread monitoring pro- gram of individual home wells to ensure the safety of water suppl les. (This monitoring program would be in addition to the annual sample taken from homes with treatment units.) This monitoring program would benefit all town residents and should be conducted as a joint effort with Water Management Program personnel collecting the sampl es and the Suffolk County Health Department providing laboratory analy- tical support. Schedule. Town Water Management Programs should be established imme- diately in both towns. Water Supply Districts for the ownership and operation of new or existing public systems should be established over a 1- to 5-year period, as the town Water Management Programs are able to incorporate them. Home Treatment Unit Districts should be established within 1 year and begin providing services at that time. The number of home treatment units will increase over time; initial emphasis should be focused on those areas where nitrate and pesticide contamination is known to exist. 10-5 I i i / i i i / i i / / / i i i i 10.2 POTENTIAL FUNDING SOURCES There are no State or federal grant programs from FmHA or HUD cur- rently available or funded to assist with the implementation of this study's recommendations. All funding, therefore, would have to be at the local level. There is a proposal for a New York State bond issue to assist local municipalities in funding water systems but the timing, fate and nature of the bond issue are all unknowns at this time. If Riverhead and Southold implement this plan's recommendations, financial support from a combination of ad valorem taxes and benefit charges and revenue-bond financing is appropriate. Those costs in- curred by the town Water Management Program which bene~t all town residents (general management, sampling, water quality management) could be paid for from the general tax base. Costs specific to an area Ihome treatment units, acquisition and operation of a private water company) would be charged directly to the users; these types of activities could be financed through revenue-bond issues. 10.3 LEGISLATIVE REQUIREMENTS New 1 egislation is not required to implement the recommendations of this plan. There is some question as to whether or not home treat- ment units can be considered as water supply facilities. An inter- pretation by the legislature or State Comptroller would be helpful in clarifying this issue. 10-6