Loading...
HomeMy WebLinkAboutSCWA Water Supply Devel in Southold Area 1985FACTORS AFFECTING WATER-SUPPLY DEVELOP~IENT IN THE SOUTHOLD AREA LONG ISLAND, NEW YORK PREPARED FOR SUFFOLK COUNTY WATER AUTHORITY Prepared For Suffolk County Water Authority ,. ' ~ . March 1985 LEGGETTE, BP3%SHEARS & GRAHAM, INC. Consulting Ground-Water Geologists 72 Danbury Road Wilton, CT 06897 TABLE OF CONTENTS INTRODUCTION. OCCURRENCE OF SALINE WATER. HYDROGEOLOGY OF THE STUDY AREA. ANALYTICAL METHODS OF DETERMINATION OF UPCONING Schmorak and Mercado (1969). Dagan and Bear (1968). Bennett, Mundurff, and ~ussain (1968) DETERMINATION OF ~JkXIMUM PUMPING RATE . DISCUSSION OF RESULTS . IMPACT ON THE STUDY AREA. WELL FIELD SITING AND DESIGN. BIBLIOGRAPHY. Page 1 1 4 5 9 9 10 12 13 14 15 17 LEI3{3ETTE, BRASHEA~S ~ {~RAHAM, IN{:. Figure 1 2 3 4 5 6 7 LIST OF FIGURES Water Supply Study Area Cross Section Of A Fresh/Saline Aquifer Upconing Of Saline Water Below A Well 1983 Water Table. 1959 and 1983 5-Foot Water Table. 1966 Drought Water Table. Bennett et al Graphical Technique . Page 2 3 3 6 7 8 1I ! ! I I I ! I I I I i I I I [ I i I Table 1 LIST OF TABLES Parameters Used in Calculations Calculated Maximum Sustained Pumping Rates Which Limit Upconing To 1/3 The Distance From The Static Interface To The Screen Bottom. Page 12 13 I I ! I I ! t ! I I 1 I I I i i ! FACTORS AFFECTING WATER-SUPPLY DEVELOPMENT IN THE SOUTHOLD AREA LONG ISLAND, NEW YORK PREPARED FOR SUFFOLK COUNTY WATER AUTHORITY INTRODUCTION Ground-water development in the North Fork of Long Island is constrained by recharge from precipitation being the only source of fresh water, and by the need to avoid salt-water intrusion to the wells. The salt-water intrusion can be due to lateral intrusion from adjacent salt-water bodies or from the upconing of salt water which underlies the fresh water aquifer. A proper management plan will balance pumpage from the aquifer with recharge, avoiding salt-water intrusion so that a supply of fresh ground water will be available indefinitely. Factors that need to be determined are the number of wells to be installed, their proximity to salt-water bodies, well spacing, screen depths, and pumpage rates. This report is written for water devel- opment for the Seacroft Subdivision, however the general conclusions and management concerns are valid for the entire North Fork area. Conclusions at the end of this report on the available water for pumping are based on a study area from Mattituck Creek to Richmond Creek, within the 2-foot water-table contour. The general study area is shown on figure 1. OCCURRENCE OF SALINE WATER Saline water, because it is more dense than fresh water, occurs below the fresh water (figure 2). Although saline and fresh water generally occur as two distinct phases, there is a zone of diffusion between the two phases consisting of a mix of the two types of water. However, for simplicity and because on a large scale the zone of LE[313ETTE, BRAShEARS & i~'~AHAM, INa. 8EACROFT ~ ' 'b 'u FIGURE 2 CROSS SECTION OF A FRESH/SALINE AQUIFER FIGURE 3 UPCONING OF SALINE WATER BELOW/ A WELL -3- FROM TODD (1980) I I I I I I I I I I ! i I I I I iL ! i -4- diffusion is small, salt water interfaces are usually treated as being abrupt, when performing calculations. The fresh water flows towards the ocean, as shown in figure 2. This outward flow helps to keep the interface towards the ocean. On the North Fork rainfall is the only natural s~urce of recharge, as shown in figure 2. For saline water with a density of 1.025 gram/cm3, which is typical for salt water, the Ghyben-Herzberg formula predicts that for every foot the static water level is above sea level, the saline water interface will be 40 feet below sea level; also for every I foot change in fresh water ele- vation, the interface will move 40 feet in the opposite direction. This relationship often underpredicts the depth to saline water, however it is useful for making a conserva- tive analysis. The relationship holds for horizontal flow (constant vertical head) only, and can't be used at the shoreline and under a pumping well. As shown in figure 3, pumping a well will cause the interface to rise under the well. This is known as upcon- ing. It has been found that there is a critical rise, after which the interface will rapidly rise to the well screen. For planning purposes, the upconing must be kept below this critical rise. A more detailed discussion of this follows in another section. HYDROGEOLOGY OF THE STUDY AREA The glacial aquifer is the primary fresh water aquifer underlying Southold. The Magothy Aquifer, which underlies the glacial, is salty in the Southold area. Some well logs indicate a substantial clay layer in the Glacial aquifer, but there is not enough data to conclude if the clay laye~ is areally extensive. A test well was installed at the Seacroft site in March and April 1984. Wash samples ob- tained during drilling showed no clay to a depth of 68 feet. A two-inch well was then driven from 68 to 98 feet and clay LeGGETTE, BI~ASHEARS ~{ {~RAHAM, INC. I I I I I I I I -5- was not indicated. At Alvah's Lane, approximately 4500 feet northwest of Seacroft, a test well encountered clay from 87 to 177 feet. Ground-water elevations over a large part of the study area are often over 5 feet above sea level. Figure 4 shows the 5 an~ 4-foot contours for 1983. Figure 5 shows the 5-foot contour for 1983 and 1959. This reflects a typical range of the 5-foot elevation. In 1984, the position was closer to 1959's. In drought situations, elevations have dropped significantly. Figure 6 shows the 3-foot contour from 1966, during the major drought. Its position roughly corresponds to the 5-foot contour of 1983. Agricultural land use and pumpage is decreasing with time, thus drought could have less of an impact than in the 1960's. Based on the Ghyben-Herzberg formula, a 5-foot water-table elevation corresponds to a fresh/saline water interface at 200 feet below sea level; 160 feet for an elevation of 4 feet. ANALYTICAL METHODS OF DETERMINATION OF UPCONING As previously stated, upconing of saline water will occur below a pumping well. The peak of the cone should be kept below a critical height to prevent a rapid rise of saline water to the well screen. This critical rise is reported to range from 0.3 to 0.5 of the distance between the original interface position and the bottom of the screen; or z/d = 0.3-0.5 in figure 2 (Dagan and Bear, 1968; Todd, 1980; Schmorak and Mercado, 1969). For the calcu- lations in this report, a value of z/d = 1/3 will be used. This compensates both for the assumption of a sharp inter- face and provides a safety margin. Use of the Ghyben-Herzberg equation alone is inappro- priate for the determination of upconing beneath a well. Because of vertical flow, upconing due to drawdown around a well will be less than predicted by Ghyben-Herzberg SCALE 1:24000 FIGURE~ 4. SEACROFT Cutchogue Harbor . 1983 WATER'TABLE' SCALE 124000 · FIGURE 5 SE^CROFT -._ . ~, : ~ ' ~ : 1959 and 1983 5-FOOT' WATER TABLE SCALE ]:24000 SEACROFT .4-' FIGURE 6 C Harbor %' 1966 DROUGHT WATER TABLE']' (McWhorter and Sunanda, 1977). Therefore, requiring the use of sea level as the minimum pumping level is not correct. Schmorak and Mercado (1969) Schmorak and Mercado present an equation to determine the rise ~f the interface under a well: - 2 ~ dK(A p/pf) where: Z = rise of interface Q = pumping rate d = distance from bottom of well screen to static interface K = permeability p = density of fresh and salt water. For z/d = 1/3 = ~d2K (P/Ps) Q max 3 ~ where: Qmax is the maximum pumping rate allowable without upconing occurring above the point of critical rise. Da~an and Bear (1968) Dagan and Bear present the following equation: ~For z/d = 1/3, = ~ TK (b-l)2 K Qmax 3 (~ P/P f) -10- where: b = saturated thickness of fresh aquifer 1 = distance between the top of the aquifer and the well screen. Bennett, Mundorff, and Hussain (1968) Bennett et al presented a graphical technique for determining the maximum allowable drawdown and pumping rate. The graphs are based on the results of several electric analog experiments. For their experiments, the top of screen was set 0.95 of the fresh water saturated thickness above the interface. The screen bottom ~'e~ting is a variable. The results are based on the following flow net constant: 2 b Kh Ky where: re = radius of influence of the pumping well Kh and Kv = horizontal and vertical permeability. For this study, Kh/Kv was set at 10/1, which is reasonable for the area and also results in a flow net constant similar to one presented by Bennett et al. The radius of influence varies for each situation analyzed, depending on projected pumping rate. The graphs are shown on figure 7. This method was used~as an order of magnitude verifica- tion of the previous two methods, due to specific well an~ aquifer configurations that the curves are based on. These are: the fixed position of the top of well screen; the ratio of saturated thickness to radius of influence equals 1:5.78; and the ratio of well radius to radius of influence equals 1:2896. 8.0 7.0 BENNETT et al GRAPHICAL TE.C_HNIQUE~.' 8O ' I r' Screen top ~ =0.95/ 095 0 0 0.1 0,2 0.3 0.4 05 0.6 0.7 0.8 0.9 DIMENSIONLESS ELEVATION OF SCREEN Bo~roM Graphs of maximum permissible drawdown /'unction, versus elevation of screen bottom, ~'''-~, , , , , , , , I maximum permissible dischnrse func[ion, ~, .versus C m -12- DETERMINATION OF MAXIMUM PUMPING RATE Before calculating the maximum pumping rate, the well penetration must be determined. Chandler and McWhorter (1975) determined that the .optimum degree of well pene- tration i~to the fresh water aquifer ranges from 30 to 55 percent. The lowest value is for an isotropic aquifer, where horizontal and vertical permeability are equal, and the highest value is for an aquifer where the ratio of horizontal to vertical permeability is 20:1. Because the two formulas used do not account for different horizontal and vertical permeabilities, a 30 percent well penetration was -used. When the vertical permeability is less than the horizontal, the allowable maximum pumping rate will actually be higher than that calculated with the formulas. For the graphical procedure, the same well penetration was used for consistency. Calculations have been carried out for four water-table elevations. The table below shows the values required for use in the formulas: TABLE 1 Parameters Used in Calculations Water Interface Saturated Depth to Distance table depth thickness top of from static elevation 20-foot interface to screen screen bottom (feet) (feet below (feet) (feet below (feet) sea level) static water table) 5 200 205 41.5 143.5 4 160 164 29.5 114.5 3 120 123 16.9 86.1 2 80 82 4.6 57.4 , , , ~. ~ ~ ~ ~ ~ ~ , , I , I l.. , ~ , '- I -13- For the two methods previously discussed, and for per- meabilities of 2000 and 3000 gpd/ft2 (typically quoted values for the area), maximum sustained pumping rates have been calculated. For the method of Bennett et aI, the calculated flow net constant is 0.011, which is close to their value of 0.0256. Table 2 shows the results of the analyses. TABLE 2 Calculated Maximum Sustained Pumping Rates Which Limit Upconing To 1/3 The Distance From The Static Interface To The Screen Bottom Schrnorak & Mercado Dagan and Bear Bennett et a1 Permeability Perrneabili ty Permeability (gpd/ft2) 2 (gpd/ft2) (gpd/ft ) 2000 3000 2000 3000 2000 3000 Water table elevation Pumping rate Pumping rate Pumping rate (feet) (gprn) (gpm) (gpm) 5 1500 2250 1940 2920 1170 1750 4 950 1430 1320 1970 750 1120 3 540 810 820 1230 420 630 2 240 360 440 650 190 280 DISCUSSION OF RESULTS Resul ts from the formula by Schmorak and Mercado are lower than those by Dagan and Bear, and should be used for this study to provide a more conservative estimate. Results of the graphical method of Bennett et al confirm the rela- tive magnitude of pumpage determined by the two analytical methods. The maximum drawdown as determined by the graph- ical method is equal to at least three times the distance that the water table is above sea level, for sustained pumpage. LEGGETTE, BRASHEARS & GRAHAM, INC. , ,. , , , , , , , , , , , , , , , 0- f I i -14- A permeability of 3000 gpd/ft2 is most often quoted for wells in the area. Therefore, even with a 3-foot drought water table, a well can safely provide at least 1 mgd (million gallons per day), or 700 gpm of sustained pumpage. This assumes no clay layer, thus the existence of a clay layer above the fresh/salt water interface will provide more protection. IMPACT ON THE STUDY AREA Maximum quoted recharge for the area is 0.9 mgd per square mile, which is approximately 625 gpm. However, because of low storage capability of the study area, the Comprehensive Public Water Supply Study for Suffolk County (CPWS-24) recommends a maximum sustained yield of 0.35 mgd per square mile to allow for safe pumpage during a drought. The North Fork Water Supply Plan (ERM-Northeast) states that based on the 0.35 mgd permissive sustained yield, 1.6 mgd of additional ground water is available for their zone 3 study area, which extends from Mattituck Creek to Hashamomuck Pond. The study area for this report, from Mattituck Creek to Richmond Creek, is estimated to have an additional 1.0 mgd available for development within the 2-foot water-table contour. Therefore, it is clear that the amount of recharge water available for pumping is the limiting factor and not upconing beneath a well. Since one well can safely supply 1 mgd, the use of several wells to develop the remaining 1 mgd over the study area will not result in upconing. It is important to emphasize that based on the recommended maximum sustained yield of 0.35 mgd per square mile and estimate of 1 mgd of additional ground-water supply, all new pumpage from throughout the study area should not exceed 1 mgd. )' , LEGGETTE, BRASHEARS & GRAHAM, INC. , , :1 I '. 4 , I ~i , ~ , , il , 1 I t , , , , "- , I L - - I -15- These estimates are based on consumptive use of the water pumped. If septic systems are to be installed and used permanently, the allowable pumpage over the study area would be higher, due to the recycling of wastewater. The 1 mgd of additional available ground water is based on a year-round average. Thus to meet this average, peak summer usage would have to be offset by pumpage lower than the average annual rate during the off-season. Because of the intermittent nature of the peak pumpage, upconing of saline water into the well would not be a problem at the higher pumpages. The increased upconing will subside when pumpage decreases again. WELL FIELD SITING AND DESIGN A 1 mgd well field, assuming a recharge rate of 13.5 inches per year, will have a radius of influence of approxi- mately 3500 feet. This recharge rate is from Crandell (1963) and is one of the lowest mentioned in the literature, thus it provides a safe estimate. A well field of this type, therefore, should not be located closer to a salt- water body than 3500 feet. If more than one well field is used to develop the 1 mgd of available water, the well fields could be closer to salt water, but not less than 2000 feet. If possible, the location of the well field(s) should be close to the peak of the water table (figures 4-6). This will provide maximum protection from lateral salt-water intrusion and upconing during drought years. Because one well could theoretically supply 1 mgd, the spacing of two or more wells in a 1 mgd well field is not cri tical. It would not be a problem for the drawdmm cones ~ LEGGETTE, BRASHEARS & GRAHAM, INC. I 1 ~ ~- ~ I j , I .i ~ ~ ~ , { { , { { I f:- , I -16- to intersect. A minimum spacing of 300 feet is recommended for a 1 mgd well field, but not less than 100 feet for a small capacity well field. A three well, 1 mgd well field would require 9 acres, two wells would require 6.5 acres, allowing for a 200 foot sanitary protection zone. LEGGETTE, BRASHEARS & GRAHAM, INC. ~w..~ ....,. < Cff'l:OFI!~~' ~ pO ...... ..q ~ ';;'~'~(CS.T!!~~ Robert Lamonicat~~' 5~149 .~;\~~ . Jr'.,: ~ .~ Assoc~ate j:EI AjP(;. ~ Y~V:J ,0<9/:' .(\~~ 28, 1985 "RrlA!~ rgs6c dmt March Disk: LEGGETTE. BRASHEARS & GRAHAM. INC. , I I ~- I J I i , ~ - :i I I I i ~ -17- BIBLIOGRAPHY 1. Bennett, G. D., M. J. Mundorff, and S. A. Hussain, "Electric Analog Studies of Brine Coning Beneath Fresh-Water Wells in the Punjab Region, West Pakistan", Geological Survey Water-Supply Paper 1608-J, 1968. 2. Bouwer, Herman, "Groundwater Hydrology", McGraw-Hill Book Company, 1978. 3. Chandler, R. L., and D. M. McWhorter, "Upconing of the Salt-Water-Fresh-Water Interface Beneath a Pumping Well", Ground Water, Vol. 13, No.4, 1975. 4. Crandell, H. C., "Geology and Ground-Water Resources of the Town of Southold, Suffolk County New York", Geological Survey Water-Supply Paper 1619-GG, 1963. 5. Dagan, G. and J. Bear, "Solving the Problem of Local Interface Upconing in a Coastal Aquifer by the Method of Small Perturbations", Journal of International Association of Hydraulic Research, Vol. 6, No.1, 1968. 6. Erm-Northeast, Camp Dresser & McKee, "North Fork Water Supply Plan, Suffolk County, New York". 7. Henderson and Bodwell, Consulting Engineers, "Water Supply Report, Seacroft at Cutchogue, Town of Southold, Suffolk County, New York", 1984. L , I , I t I ~ -- I "- I - I 8. Holzmacher, McLendon & Murrell, "Comprehensive Public Water Supply Study, Suffolk County, New York, CPWS-24", Vol. I-III, 1970. 9. McWhorter, D. M. and D. K. Sunada, "Ground-l'iater Hydrology and Hydraulics", Water Resources Publications, 1977. 10. Perlmutter, N. M. and F. A. DeLuca, "Availability of Fresh Ground Water, Montauk Point Area, Suffolk County, Long Island, New York", Geological Survey Water-Supply Paper 1613-B, 1963. 11. Schmorak, S. and A. Mercado, "Upconing of Fresh-Sea Water Interface Below Pumping Wells, Field Study", Water Resources Research, Vol. 5, No.6, 1969. 12. Suffolk County Department of Health Services, "Contour Map of the Water Table and Location of Observation Wells in Suffolk County, New York, March 1983". 13. Todd, D. K., "Groundwater Hydrology", John Wiley & Sons, Inc., 1980. LEGGETTE, BRASHEARS & GRAHAM, INC.