Dutchess County Aquifer Recharge Rates & Sustainable Septic ...

Groundwater Resource Report

Dutchess County Aquifer Recharge Rates & Sustainable Septic System Density Recommendations Dutchess County Water & Wastewater Authority

April 2006

Prepared for: Dutchess County Water & Wastewater Authority 27 High Street Poughkeepsie, NY 12601

©2006 The Chazen Companies

Groundwater Resource Report

Dutchess County Aquifer Recharge Rates & Sustainable Septic System Density Recommendations Dutchess County Water & Wastewater Authority

April 2006

Prepared by: The Dutchess County Office The Chazen Companies 21 Fox Street Poughkeepsie, New York 12601 (845) 454-3980 Lead Author: Russell B. Urban-Mead, CPG-10328 Senior Hydrogeologist Dutchess County (845) 454-3980

Orange County (845) 567-1133

Capital District (518) 273-0055

North Country (518) 812-0513

Dutchess County Aquifer Recharge Rates and Sustainable Septic System Density Recommendations

Page i

TABLE OF CONTENTS FORWARD EXECUTIVE SUMMARY ............................................................................................. 1 1.0

INTRODUCTION ........................................................................................... 6 1.1

Summary of Aquifer Recharge Rates .................................................. 6

1.2

Summary of Sustainable Septic System Densities ............................. 8

2.0

AQUIFER RECHARGE ANALYSIS ............................................................ 10 2.1

Aquifer Recharge Evaluation Components ....................................... 10 2.1.1

Types of Recharge ........................................................................... 10

2.1.2

Summary of Prior Recharge Rate Estimates ................................. 11

2.1.3

Regional Scale of Recharge Estimates ........................................... 13

2.2

3.0

Dutchess County Aquifer Recharge Assessment .............................. 14 2.2.1

Dutchess County Geology and Aquifers ......................................... 14

2.2.2

Recharge Assessment Method Selection ........................................ 15

2.2.3

Aquifer Recharge Analysis for Dutchess County........................... 18

SUSTAINABLE SEPTIC SYSTEM DENSITY ANALYSIS........................ 21 3.1

Review of Models and Assumptions .................................................. 22

3.2

Septic System Density Analysis for Dutchess County ..................... 25

3.3

Conditions and Applications .............................................................. 26

4.0

RELATED TOPICS FOR FUTURE WORK................................................. 31

5.0

REFERENCES.............................................................................................. 33

The Chazen Companies April 2006


Page ii

LIST OF TABLES Document Tables (bound at rear of report) Table 1: Aquifer Discharge; Baseflow Separation in Dutchess County Watersheds Table 2: Aquifer Recharge Rates; by Watershed/Region & Hydrologic Soil Group Table 3: Recommended Minimum Average Septic System Densities LIST OF FIGURES Executive Figures (bound with Executive Summary) Executive Figure 1: Dutchess County Annual Aquifer Recharge Rates and Minimum Sustainable Septic System Densities Executive Figure 2: Dutchess County Hydrologic Soil Groups Document Figures (bound at rear of report) Figure 1: Dutchess County Watersheds, Precipitation and Geology Figure 2: Detail of Hydrologic Soil Groups versus Surficial Geology Figure 3: Dutchess County Annual Precipitation and Watershed Discharge Figure 4: New Jersey Septic System Density Model Figure 5: Dutchess County 1982 Gerber Septic System Density Model Figure 6: Dutchess County Hydrologic Soil Groups



Page iii

Foreword by Scott Chase, Executive Director Dutchess County Water & Wastewater Authority This report was commissioned by the Dutchess County Water & Wastewater Authority to better understand County-wide aquifer recharge rates and to give municipalities guidance in setting sustainable development densities specifically related to the use of individual wells and conventional individual septic systems based upon our better understanding of aquifer recharge. While I don’t like introducing a report with language that points out limitations, there is always a concern that some will misuse guidelines when carrying out decisions in favor or against development proposals. Therefore, it is critical for readers of this document to understand that the numerical minimum septic system densities recommended herein cannot be applied more broadly to limit any other manner of development in the County, and there are also variables in these calculations that can in some cases change the outcome of these septic system density calculations. Specifically, these development density recommendations for septic systems are not applicable in areas where development will use combinations of central water supplies, central sewer systems or septic systems with enhanced wastewater treatment processes that change nitrogen/nitrate discharges to groundwater. In addition, each development decision may involve other information and/or unique circumstances that will require different inputs to the calculations and result in different outcomes for the recommended density guidelines. With this clarification stated, the information contained within the report and the logic for establishing the mathematical formulas and carrying out the calculations should be extremely valuable to community decisionmakers, and particularly to municipal governing boards and their advisors charged with setting long-range sustainable development goals. Dutchess County and most of its communities have adopted Greenway Compact principles to guide future development. The septic system density guidelines in this report should be viewed as consistent with these principles since the density guidelines are average, rather than fixed-parcel size, densities. As such, they do not preclude use of smaller lots to conserve natural greenspace and farmland, as found in “conservation or cluster development”. Conservation development can and should still be recommended, as long as the total parcel and area-wide average build-out densities of septic systems are maintained and as long as wells are placed with consideration of any areas of concentrated septic system discharge.



Page iv

Concentrating most new development in and around traditional mixed-use, walkable cities, villages and hamlet centers is still one of the best strategies to protect natural resources and the rural countryside which provides significant filtering and recharge of our groundwater resources. This work presents important new information and refinements on our overall understanding of recharge rates for our groundwater supplies. Since aquifers and surface waters are so closely linked, this recharge information will have important implications for a wide range of groundwater and surface water resource management and protection strategies. I anticipate that the information in this report describing aquifer recharge rates in general, will be used in project reviews, in broader resource analysis and in establishing additional water resource management policy for a wider range of analyses than the septic system density calculations included in this report. Ultimately the source of the water to recharge our aquifers is the precipitation we receive. Communities must understand the limitations of their water resources and plan accordingly. Those communities in Dutchess County that are wholly dependent upon aquifer recharge may not have as many options as those with the ability to tap the Hudson River. Upstream communities must be required to be careful stewards of water resources on behalf of downstream communities. It is to our collective advantage to manage development so that we maintain the natural ability of soils to absorb and allow infiltration of precipitation, and to minimize runoff, as this will protect our surface water environments and our groundwater quality and quantity. Proper management of stormwater becomes increasingly critical as we develop more of our communities. Stormwater not only has important implications for aquifer recharge but it has significant implications for surface water quality, wildlife and our use of streams and lakes. Our goal in managing development should be to ensure that we do not degrade the natural ability of a site to infiltrate precipitation and store and filter runoff. In many cases we may have the ability to return or actually enhance these natural functions while accommodating new development. Development decisions involve making choices and balancing interests. Each of us has a responsibility to do our part to insure that we have access to adequate supplies of clean water both now and for the future. In areas dependent on individual wells and conventional septic systems we need to understand and respect the limitations of the natural environment to both provide water supplies and treat our wastewater. This report on aquifer recharge should help us to better understand the natural limitations and to make better decisions in managing development at sustainable levels so that future generations will continue to have access to clean water. Scott Chase, Executive Director, Dutchess County Water & Wastewater Authority


Dutchess County Aquifer Recharge and Sustainable Rural Density Analysis

Page 1

EXECUTIVE SUMMARY This report by The Chazen Companies describes updated understandings of aquifer recharge rates throughout Dutchess County and provides new guidance for sustainable densities of septic systems to preserve the quality of well water. Aquifer recharge supports groundwater supplies critical to various uses, including reliability of public water system and individual domestic wells, the adequate dilution of wastewater discharges from septic systems, and aquifer base flow contributions to streams. As such, the aquifer recharge rates provided here will have many applications. These aquifer recharge rate estimates have been calibrated against stream flow data from Dutchess County’s three major watershed systems and regional precipitation data. Total annual rainfall increases from west to east across the county, supporting greater stream flows and higher aquifer recharge rates in areas with higher total annual rainfall. For ease of use, the aquifer recharge rates presented here are correlated with readily-available Hydrologic Soil Groups which are mapped and periodically updated by the Natural Resources Conservation Service. Hydrologic Soil Groups C and C/D together cover fifty-three percent of Dutchess County, and Hydrologic Soil Groups B, C and C/D jointly cover approximately eighty percent of the County, as shown on Executive Figure 2. Executive Figure 1 posts regional aquifer recharge rates throughout Dutchess County. In summary, •

In areas with Hydrologic Soil Groups A and A/D, aquifer recharge rates vary from west to east across the County between 17.3 and 20.2 inches per year;

•

In areas with Hydrologic Soil Group B, aquifer recharge rates vary from west to east across the County between 12.6 and 14.7 annual inches per year;

•

In areas with Hydrologic Soil Groups C and C/D, aquifer recharge rates vary from west to east across the County between 6.5 and 7.6 inches per year;

•

In areas with Hydrologic Soil Group D, aquifer recharge rates vary from west to east across the County between 3.6 and 4.2 inches per year.

For each range of aquifer recharge rates above, the higher recharge rates occur in the Tenmile River watershed which typically receives approximately 44 inches of annual precipitation and the lower recharge rates occur in the northwest towns of Hyde Park, Rhinebeck and Red Hook which typically receive approximately 38



Page 2

inches of annual precipitation. Aquifer recharge rates in the Wappinger Creek and Fishkill/Sprout Creek watersheds lie between the values listed above. These aquifer recharge rates are expected to inform understanding of a wide range of future aquifer and water resource protection inquiries. Examples include water budgets and SEQRA reviews, build-out water supply analyses at local or regional scales, water budgets often provided with pumping test yield reports, and stream baseflow and riparian wetland management assessments. This report includes updated septic system density guidelines as the first of many likely applications of the newly available aquifer recharge rates. Extensive areas of Dutchess County are and will continue to be dependent on the use of individual wells and conventional septic systems for water supply and wastewater disposal. In such areas, guidelines for maximum sustainable densities of wells and septic systems will help insure availability of potable well water supplies. The Executive Figure 1 posts minimum average recommended sustainable parcel sizes for areas reliant on use of wells and septic systems. The calculations assume residential occupancy rates of 2.6 persons per system and include a 0.1 acre allowance for on-site impervious surfaces on each parcel. In summary, •

In areas with Hydrologic Soil Groups A and A/D, minimum sustainable average septic system densities vary from east to west across the County between 1.2 and 1.4 acres per system;

•

In areas with Hydrologic Soil Group B, minimum sustainable average septic system densities vary from east to west across the County between 1.6 and 1.9 acres per system;

•

In areas with Hydrologic Soil Groups C and C/D, minimum sustainable average septic system densities vary from east to west across the County between 3.0 and 3.5 acres per system;

•

In areas with Hydrologic Soil Group D, minimum sustainable average septic system densities vary from east to west across the County between 5.4 and 6.2 acres per system.

The ranges of recommended average minimum parcel sizes relate to the differing aquifer recharge rates identified across the County. The smallest average parcel sizes can be sustained in the Tenmile River watershed where precipitation and aquifer recharge rates are highest. Larger average parcel sizes are necessary in the northwest towns of Hyde Park, Rhinebeck and Red Hook where precipitation and aquifer recharge rates are lowest. Sustainable septic system densities on lands in



Page 3

the Wappinger Creek and Fishkill/Sprout Creek watersheds lie between the values posted above. These septic system density recommendations apply equally to minor and major subdivisions, as well as to more regional municipal or watershed-scale planning approaches. The present analysis of aquifer recharge rates and sustainable septic system densities builds on previous work by independent consultant Robert Gerber, retained by the Dutchess County Department of Planning in 1982. The intent of Gerber’s septic system density calculations was to ensure that residents using individual wells and septic systems could be assured potable water even as surrounding permitted lots reach full build-out. The present analysis builds on Gerber’s work and intentions by re-evaluating his estimated rates of aquifer recharge, and by updating his method of calculating sustainable septic system density. The guiding principle for sustainable septic system uses involves balancing the dual aquifer demands to supply potable well water and to adequately dilute wastewater coming from septic systems. Too many septic systems renders groundwater non-potable if not enough recharge is available to provide dilution. Current Health Department separation distances required between wells and septic systems are designed to ensure bacteria and viral die-off between wells and septic systems. But these measures were never intended to address or provide the dilution necessary to ensure that more persistent wastewater constituents such a nitrate remain below drinking water standards. The present septic system density recommendations complement Health Department separation distance regulations by planning for the dilution of such dissolved contaminants in a full build-out scenario. Used in conjunction with existing Health Department separation distance requirements, these density recommendations provide a greater assurance of available, sustainable, potable groundwater to Dutchess County citizens reliant on individual wells and conventional septic systems. These septic system density recommendations therefore apply specifically to areas which are dependent on, and are expected to remain dependent on, individual wells and conventional septic systems. It should be noted that the density guidelines here are average densities; thus, they do not preclude the use of smaller lots such as found in conservation or cluster developments which can still be recommended as long as the total parcel or area-wide build-out densities are met and as long as specific attention is given to the layout and design of more closely-installed wells or septic systems. This report work has been completed by The Chazen Companies under funding initiatives by the Dutchess County Legislature, endorsed by the County Executive’s office, enabling this work to be completed for the Dutchess County Water & Wastewater Authority.



Page 4

EXECUTIVE FIGURE 1 - DUTCHESS COUNTY ANNUAL AQUIFER RECHARGE RATES AND MINIMUM SUBSTAINABLE SEPTIC SYSTEM DENSITIES Northwest Region includes Hyde Park, Rhinebeck (V/T), Red Hook (T/V), Tivoli

Tivoli

Wappinger Watershed area includes Milan, Pine Plains, Clinton, Stanford, Pleasant Valley, Washington, Millbrook, part-LaGrange, Poughkeepsie (T/C), Wappinger, Wappingers Falls

Red Hook Red Hook Milan

Pine Plains

Rhinebeck Northwest Region

Millerton

Soil Inches Parcel Type Recharge Avg. Size

A B C D

17.3 12.6 6.5 3.6

Northeast

1.4 1.9 3.5 6.2

Tenmile Watershed area includes Northeast, Millerton, Amenia, Dover, Pawling (T/V)

Stanford

Fishkill Watershed area includes part-LaGrange, Union Vale, Beekman, Fishkill (T/V), East Fishkill, Beacon

Wappinger Creek Watershed Clinton

Hyde Park

Inches Parcel Soil Type Recharge Avg. Size

A B C D

18.2 13.3 6.8 3.8

1.3 1.8 3.3 5.9 Washington Millbrook

Pleasant Valley

Amenia Tenmile River Watershed Inches Parcel Soil Type Recharge Avg. Size

A B C D

20.2 14.7 7.6 4.2

Legend

1.2 1.6 3.0 5.4

Major Streams and Rivers Watersheds

Poughkeepsie Poughkeepsie

Union Vale La Grange

Dover

Hydrologic Soil Group, See NRCS Soil Survey or Executive Figure 2

Fishkill Creek Watershed Inches Soil Parcel Type Recharge Avg. Size

Wappingers Falls Wappinger

A B C D

19.2 14.0 7.2 4.0

Beekman East Fishkill

Pawling Pawling

A B C D

Fishkill Beacon

Recommended Minimum Average Parcel Size for Areas with Wells and Septic Systems, in acres per system

1.3 1.7 3.2 5.2

Fishkill

NOTES:

20.2 14.7 7.6 4.2

1.2 1.6 3.0 5.4 Inches of Annual Aquifer Recharge

For purposes of this study, HSG A soils include A/D soils and HSG C soils include C/D soils. Recommended average minimum parcel sizes for septic systems apply to areas without central water or sewer. They are calculated at 2.6 persons per system and include 0.1 acres per system for roof, driveway and other onsite impervious surfaces.



Page 5

EXECUTIVE FIGURE 2 - DUTCHESS COUNTY HYDROLOGIC SOIL GROUPS Tivoli

Red Hook

Milan

Pine Plains Millerton Northeast

Rhinebeck

Rhinebeck

Stanford Clinton

Amenia

Washington Hyde Park Millbrook

Legend

Pleasant Valley

Hydrologic Soil Groups and Dutchess County Percent Coverage

Poughkeepsie Poughkeepsie


Dover Waterbodies /Unclassified (4.0%) A and A/D Soils (10.5%) B (26.9%)

Beekman Wappingers Falls Wappinger

Fishkill Beacon Fishkill

C and C/D (53.2%)

Pawling East Fishkill

D (5.4%)

Pawling Source: U.S. Department of Agriculture, Natural Resources Conservation Service Soil Survey for Dutchess County, New York (2003)



1.0

Page 6

INTRODUCTION

A key objective of the present effort is to present data and findings that are easy for the general public to read. Section 1.0 of this report is therefore written to bypass most technical language and to most clearly provide the results of this study. Aquifer recharge rates and septic system density recommendations have been assigned to the four Hydrologic Soil Groups (HSG) described by the Natural Resource Conservation Service (NRCS). These soil groups can be identified from soil survey materials and normally require no further hydrogeologic interpretation. The distribution of hydrologic soil groups is shown on Figure 6. The most common soil groups are group C and C/D soils which together constitute fifty-three percent of Dutchess County. Hydrologic soil group B soils cover twenty-seven percent of the county. Hydrologic soil groups A and A/D cover ten percent, and hydrologic soil group D covers five percent of Dutchess County. Results of this investigation have been helpfully reviewed prior to public release by agency personnel in the Dutchess County Water & Wastewater Authority, Dutchess County Department of Planning, Dutchess County Soil & Water Conservation District, Dutchess County Department of Health, and the Cornell Cooperative Extension/Dutchess County, and the Chair of the Environment Committee of the Dutchess County Legislature. Levels of review among the above ranged from verbal briefings, to formal presentations, to close edits of draft text. An early draft of this report was also generously reviewed by United States Geological Survey geologist Allan Randall. Notwithstanding the above beneficial reviews, the work product here remains the responsibility of The Chazen Companies. 1.1

Summary of Aquifer Recharge Rates

Recharge rates for aquifers in Dutchess County have been estimated in this report on the basis of soil classes. Soils in Dutchess County have been assigned to Hydrogeologic Soil Groups (HSG) by the Natural Resources Conservation Service (NRCS). These classifications range from HSG group A soils with the highest infiltration rates to HSG group D soils with the most restrictive infiltration rates. The aquifer recharge analysis conducted in this study focuses on recharge most likely to reach aquifer horizons used for water well supply uses. The recharge rate estimates therefore exclude more ephemeral groundwater replenishment sometimes referred to as interflow, quick flow, or rejected recharge, all of which normally fails to reach deeper aquifer horizons due to diversion along buried clay seams, lodgment till, or shallow buried bedrock surfaces. The updated aquifer recharge rate estimates identified in this report therefore represent a subset of overall



Page 7

groundwater recharge which includes the more intermittent or ephemeral interflow groundwater components. A recent study of the Wappinger Creek watershed employs methods to identify both the interflow and aquifer recharge components of total groundwater recharge (Chazen, March 2006). This study does not distinguish between aquifer recharge to sand and gravel aquifers or to fractured bedrock aquifers. This is because although pumping rates from bedrock wells and pumping rates from sand and gravel wells can vary considerably, the long-term sustainable yield of all wells is ultimately predicated on available aquifer recharge. It is this aquifer recharge which is the focus of this investigation. Significantly different aquifer recharge rates occur across Dutchess County because of different precipitation patterns. Figure 1 documents the variability, from a low of approximately 38 inches per year in Hyde Park, Rhinebeck and Red Hook, to over 46 inches per year in parts of the Harlem Valley and the Fishkill watershed. These differences directly influence groundwater recharge and stream flows across Dutchess County. Table 2 lists aquifer recharge rates identified by this study. Aquifer recharge rates are highest in the Fishkill and Tenmile River watersheds where precipitation rates are highest. Lower aquifer recharge rates are identified in the Wappinger Creek watershed. Yet lower aquifer recharge is projected for the towns of Hyde Park, Rhinebeck and Red Hook due to proportionally lower available precipitation rates. The low aquifer recharge rates identified by this study in northwest Dutchess County are consistent with findings in a 2002 Dutchess County study (Chazen, 2003), documenting lowest County drought-stage stream flows in the Crum Elbow Creek in Hyde Park and the Little Wappingers Creek in adjacent parts of the Town of Clinton. As shown on Table 2, Dutchess County aquifer recharge rates for soils in HSG groups A and A/D range between 17.3 and 20.2 inches per year, recharge rates on HSG group B soils range between 12.6 and 14.7 annual inches, recharge rates on HSG groups C and C/D soils range between 6.5 and 7.6 inches per year and recharge on HSG group D soils range between 3.6 and 4.2 inches per year. The NRCS has identified some dual-HSG A/D and C/D soils. These assignments reflect sub-soil conditions which can slow the initial rates of infiltration, usually because of buried clay or shallow bedrock horizons or because the deeper sediments are periodically saturated at near-watertable conditions. In this analysis, the estimated aquifer recharge rates for such dual-category soils have been consolidated with their higher-infiltration category (e.g. A/D with A soils and C/D with C soils) since most aquifer recharge occurs during rain events of less than 1.25 inches



Page 8

(Charles, et al, 1993) when the initial rate of infiltration will not yet be slowed by the slowing effect which only begins to occur as the upper-most soil becomes fully saturated and the lower-permeability secondary characteristic begins to affect net infiltration rates. Precipitation recorded at the Institute of Ecosystem Studies Environmental Monitoring Program in Millbrook New York over a recent ten year period, from January 1, 1996 to January 1, 2006, indicates that precipitation events exceeding 1.25 inches occur on average just 6.7 days per year. This study estimates that only during such heavier events will the effect of the deeper horizon inhibit the otherwise higher immediate tendency of the soil to accept aquifer recharge. 1.2

Summary of Sustainable Septic System Densities

The sustainable residential density recommendations presented here apply only to sites or regions reliant on wells and septic systems. The formula used to develop these recommendations is shown on Table 3 and is printed below. It considers aquifer recharge rates, household occupancy rates, annual nitrate production per capita, impervious surface coverage, and a selected allowable residual nitrate concentration target. Recommended minimum average parcel sizes are summarized in shaded columns on Table 3. Where traditional site design is used, (e.g. without low-impact design features such as rainfall gardens or other focused efforts to enhance on-site recharge) between 1.2 and 1.4 acres per system are needed where HSG A and A/D soils are present, between 1.6 and 1.9 acres per system are needed where HSG B soils are present, between 3.0 and 3.5 acres per system are needed where HSG C and C/D soils are present, and as much as 5.4 to 6.2 acres per system are needed where HSG D soils are found. The regional variation reflects precipitation and hence aquifer recharge rate changes across Dutchess County. To provide perspective on regional implications of these recommendations and based on specific analysis of the three dominant County watersheds (Table 2), HSG B and HSG C and C/D soils together cover approximately eighty percent of Dutchess County and Hydrologic Soil Groups C and C/D alone cover fifty-three percent of Dutchess County, so the sustainable minimum average parcel size in areas with wells and septic systems should for the over fifty percent of Dutchess County exhibiting HSG C or C/D soils be no less than 3.0 to 3.5 acres per parcel. The distribution of Hydrologic Soil Groups is shown on Figure 6. These minimum average parcel size recommendations assume a target groundwater nitrate concentration of 5.2 milligrams per liter and a typical residential occupancy rate of 2.6 persons per household based on typical Dutchess County occupancy



Page 9

levels in 2000 census data. We note that average family sizes in Dutchess County are reportedly over 3.0 persons per household, but the overall average occupancy rate of 2.6 persons per household was judged to most closely describe typical residential occupancy rates and was used in the default density recommendations provided here. The calculated average parcel sizes also include an allowance of up to 0.1 acre of impervious surface per system and an average annual discharge of 10 pounds of nitrate-nitrogen per person. Aquifer recharge rates used in the calculations were selected for each watershed region and Hydrologic Soil Group from Table 2. Minimum regional density recommendations are also provided on Table 3. These density values differ from minimum average parcel size recommendations by including an additional allowance for proportionally-associated roads normally accompanying parcels. For these regional density recommendations, the impervious surface allowance is increased to 0.3 acres per septic system to account for both onsite roof and driveway impermeable surfaces, and a share of roads to support each, parcel, all designed assuming traditional stormwater management techniques which most commonly focus on removing stormwater from immediately adjacent areas and which usually include detention techniques rather than recharge mitigation techniques. All other inputs are left as for the minimum average parcel size calculations. If calculated impervious surfaces for proposed particular projects differ from 0.30 acres per septic system the regional density recommendations can be adjusted accordingly. For reference purposes, the guiding formula for the recommendations provided in this report is provided below, available for use a project scale, on for municipal or wider-regional analysis: A = 4.4186HM / CqR + Isc Where A = acres per average system, in acres H = persons per system M = pounds of nitrate-nitrogen per person per year, in pounds Cq = Nitrate-nitrogen target average groundwater concentration, in mg/L R = Annual Recharge Rate, in inches Isc = Impervious surface cover, in acres. More complete details describing the selected density model are discussed in Section 3.0 of this report, including a review of assumptions and conditions applicable to the model.



2.0

Page 10

AQUIFER RECHARGE ANALYSIS 2.1

Aquifer Recharge Evaluation Components 2.1.1 Types of Recharge

Aquifer recharge in Dutchess County comes from local rather than distant areas. A portion of precipitation infiltrates through soil layers to underlying saturated porespace or fractures. These saturated geologic formations constitute the aquifers which supply water to wells and which receive septic system liquid discharges. Clay horizons in some settings and the generally clay-rich soils found in some upland areas can each reduce infiltration rates; in general, however, all land in Dutchess County allows at least some measure of aquifer recharge. Investigators distinguish between various types of recharge (de Vries & Simmers, 2002; Hiscock, 2004; Scanlon et al, 2002). Under various names, these include direct recharge falling on a site, additional localized recharge flowing onto a site from adjacent areas to supplement direct recharge, and indirect or induced recharge which occurs where pumping wells draw surface water down into aquifers. There are also important distinctions between types of recharge and infiltration: infiltration refers to water movement into the subsurface, while recharge refers to the share of infiltration which reaches either the uppermost surface of the aquifer known as the watertable, recognized here as aquifer recharge, or at least reaches shallow groundwater flow pathways below root zones and so escapes prompt transpiration back to the atmosphere. This last fraction of infiltration follows clay layers or bedrock layers to streams or ponds in a matter of days or weeks without ever meaningfully replenishing the aquifer and is sometimes referred to as “interflow”, “quick flow” or “rejected recharge,” depending on mechanisms involved and investigators consulted. In this report this component of groundwater recharge is referred to as interflow. Thus, only a fraction of infiltration ends up as aquifer recharge successfully recharging underlying fractured bedrock or granular sediment aquifers. These aquifers serve as substantial groundwater storage reservoirs and are the geologic horizons from which wells intersect and extract water supplies. The present study focuses on the aquifer recharge rates, excluding evapotranspiration and interflow losses. The present study also estimates that indirect or induced recharge from pumping wells seldom occurs from use of lowcapacity domestic wells which are the primary subject of this evaluation. Finally, the present study does not distinguish between direct and localized recharge since most areas of at least a few acres in size likely benefit from each and seeking to distinguish between these is presently beyond the scope of this study and is unlikely to yield substantially different overall results.



Page 11

2.1.2 Summary of Prior Recharge Rate Estimates Various prior investigators have estimated groundwater recharge rates for Dutchess County or adjacent regions. These estimates have been developed using a range of methods and recharge rate estimates have generally been similar. All investigators have recognized that soil types crucially influence effective recharge rates to the underlying aquifers so most describe recharge rates in terms of recharge into surficial geologic formations without consideration of the various underlying bedrock formations. No known prior investigators focusing on Dutchess County have distinguished between aquifer recharge and interflow components of overall groundwater recharge. Another general convention among prior investigators has been to evaluate annual total recharge rather than identifying month-to-month or seasonal recharge rates. This study also focuses on annual aquifer recharge since most wells used for residential purposes are, or should be, installed deeply enough to access sufficient stored groundwater in fractures or pore spaces to support several months water demand without meaningful recharge. Such periods without recharge currently occur during most summers when evapotranspiration loss rates routinely exceed potential recharge rates. Robert Gerber (1982) provided some of the earliest recharge rate estimates commonly referenced in Dutchess County groundwater investigations. Gerber used published literature from other regions to estimate annual recharge for five prevalent Dutchess County geologic landscapes. In areas with sand and gravelbased soils, Gerber estimated annual recharge rates of 14.3 inches where deposits were relatively thin and 18.0 inches for thicker deposits. Gerber’s recognition of the importance of sediment thickness may allude to what today we recognize as interflow or rejected recharge along buried bedrock surfaces occurring where thinner sandy soils allow limited retention time during which fuller recharge of the bedrock formation could occur. In till-covered upland areas, Gerber estimated annual recharge rates of 6.75 inches and 3.2 inches for areas with thin and thick clay-rich till coverage, respectively. The higher recharge estimate for thin-soiled areas addresses the opportunity for some recharge to pass directly into fractured bedrock through bare-rock areas, whereas areas with thicker, clay-rich till has fewer such recharge windows. Finally, in Dutchess County locations where postglacial temporary lake beds collected layered silty-clay deposits, Gerber’s research suggested an estimated annual recharge rate of 2.25 inches. A study by Snavely (1980) offers less geologic detail than Gerber by focusing only on valley versus upland landscapes. This simplified dual-terrain study approach estimated an annual recharge rate of 21 inches for valley areas with sand and gravel deposits in the Fishkill-Beacon area, and 8 inches of annual recharge in



Page 12

adjacent hillside areas covered with glacial till. Written communication with United States Geological Survey geologist Allan Randall is cited as the source of the recharge rate estimates. The basis for Randall’s estimates is not provided so it is unknown how the estimates were determined. A 1995 evaluation of aquifer recharge rates in Putnam and Westchester Counties used a computer model to estimate recharge rates for sediment-filled valley deposits and upland till-mantled areas (Wolcott & Snow, 1995). The selection of a dualterrain model was similar to the approach used by Snavely. The model was calibrated to water level data in observation wells and to the groundwater component of streamflow, or baseflow. The analysis estimated annual aquifer recharge rates of 19.17 inches in valley areas and 8.45 inches in upland areas with clay-rich glacial till. The method of baseflow separation used in the analysis is likely to have identified the combined groundwater components of interflow and aquifer recharge. Methods of baseflow separation which can be used to separate these two components are addressed in Section 2.2.3 and elsewhere (Chazen, March 2006). New Jersey regulators have researched recharge rates for all soil types in the state (Charles et al, 1993). New Jersey’s Geological Survey has worked cooperatively with NJ Department of Environmental Protection to develop recharge rate estimates for soils in New Jersey, taking into consideration state-wide precipitation variations, existing land uses, runoff coefficients representative of typical rainfall events, and evapotranspiration losses. The resulting recharge estimates are calibrated using baseflow separation data from New Jersey’s rivers and streams. New Jersey’s Geological Survey recommends using a baseflow separation method devised by Posten (1984) for calibrating recharge rates (Hoffman, 1999). Recharge rates identified by New Jerseys program vary from less than 3 inches annually to over 20 inches annually for different soil types. These various prior investigations have generally converged on similar recharge rate estimates for many common Dutchess County terrains. Recharge rates through open sand and gravel formations are agreed to lie somewhere between 14 and 20 inches per year. Silt-rich glacial till landscapes are agreed to allow between 3 and 9 inches of annual recharge. Massive clays are agreed to be least conducive to aquifer recharge. All methods consider soil types or surficial geology to be the most significant factors in defining recharge rates. None of the prior methods have distinguished between the interflow and aquifer recharge components of overall groundwater recharge.



Page 13

2.1.3 Regional Scale of Recharge Estimates The investigations above evaluated groundwater recharge at several scales (e.g. size of the study area). Methods by Snavely (1980) and Snow & Wolcott (1995) identify recharge rates only between predominantly water-laid valley deposits and typically clay-rich upland deposits. The general recharge rates assigned by these studies are helpful at the landscape scale but become problematic where clay or silty deposits are found in valley settings, or where more granular tills are found in upland areas. Work completed in Connecticut assigns similar regional recharge rates to such monolithic “upland” versus “valley” deposits and considers also the ratio between such upland and valley-train deposits when estimating low-flow duration reliability in river systems (Cervione, et al, 1972). For review purposes only, these broad valley versus upland geographic type-areas are shown on Figure 1 of this report, demonstrating visually that the distribution of such glacio-fluvial deposits are somewhat uniformly distributed throughout Dutchess County, and demonstrating also the lack of refinement offered by such approaches. Gerber (1982) provided more distinct recharge estimates for smaller geographic units, assigning separate recharge rate estimates to five landscape terrains. The research underlying the recharge estimates was not extensive but the terrain distinctions provide more refinement than the broader categories assigned by Snavely, and Wolcott & Snow. Recharge rate analysis completed for the State of New Jersey is extensive, providing discrete recharge estimates for every soil type in the State. The research behind these assigned rates was not exhaustively reviewed for the present study but the peer review process followed by the United States Geological Survey and its State offices is extensive, so project output is likely to be among the highest quality currently available. Recharge rate values provided by the New Jersey model fall within the ranges of values identified by all prior investigations. This review indicates that recharge rates can be estimated at the broad landscape level as shown by the work of Snavely and Wolcott & Snow, or at a more focused but still broad basis as shown by Gerber’s five recharge classes, or can be estimated down to the individual soil group level as has been attempted in New Jersey.



2.2

Page 14

Dutchess County Aquifer Recharge Assessment 2.2.1 Dutchess County Geology and Aquifers

Bedrock aquifer formations underlying Dutchess County’s watersheds are varied in terms of types and distribution but generally consist of shale/greywacke and their metamorphic equivalents, limited crystalline, and dolomite limestone’s and their metamorphic equivalents. Evaluations by Gerber (1982) and Simmons (1961) among others indicate that median and average domestic well yields from bedrock aquifers in Dutchess County vary by less than one order of magnitude, consistent with their uniform reliance on secondary porosity for groundwater storage and transmission. Overlying the bedrock formations are a variety of unconsolidated materials. Siltdominated glacial till mantles all upland and some low-land areas within the watersheds. During glacial recession, melt water flushed sediment down watershed valleys, depositing high-permeability sand and gravel in many localities and depositing silt and clay in larger temporary lakes where flow slowed. These waterlaid deposits of glacial origin are shown in a simplified form on Figure 1 based on interpretation from New York State’s surficial geologic map. Each watershed contains approximately equivalent ratios of upland geology versus the valley glaciofluvial deposits which include outwash sand and gravel, lacustrine deposits and kame deposits. Watersheds and streamflow records were closely evaluated in this study to help estimate rates of aquifer recharge. The selection of watersheds and data years was based on the availability of historic USGS streamflow data. The three largest watersheds in Dutchess County are the Wappinger Creek, Tenmile River, and the Fishkill/Sprout Creek. The Tenmile river flows eastward to the Housatonic River in Connecticut while the Wappinger and Fishkill/Sprout Creek watersheds flow southward to the Hudson river. Available historic precipitation data from weather stations selected near the watersheds show that precipitation is normally evenly distributed throughout the year, although increasing slightly during the summer months. The mean annual precipitation at Dutchess County airport for the 109-year period (1896-2004) of record is approximately 40.4 inches. Figure 1 demonstrates a marked increase in average precipitation from northwest to southeast across Dutchess County, with annual averages ranging from 38 inches annually in Hyde Park, Rhinebeck and Red Hook to over 44 inches in portions of southeast Dutchess County and Massachusetts.



Page 15

The Wappinger Creek is the second largest stream in Dutchess County, flowing 32 miles from Pine Plains in the northeast to New Hamburg in the southwest (Figure 1). The drainage area of the watershed above the Red Oaks Mill USGS gauging station located upstream of the Village of Wappingers Falls is approximately 181 square miles. It has a total fall of 460 feet and an average fall of about 14 feet per mile. The watershed lies entirely in Dutchess County, draining much of the Towns Clinton, Stanford, Washington, Pleasant Valley, and portions of the Towns of Milan, Pine Plains, Hyde Park, Poughkeepsie, and LaGrange. Main Wappinger Creek tributaries generally follow a dendritic drainage pattern within a landscape of rolling hills and valleys. The drainage density (miles of blue USGS map streams per watershed square mile) is approximately 1.73 miles per square mile of watershed. The Fishkill/Sprout Creek watershed drains most of the central and southern parts of the Dutchess County (Figure 1). It flows southwesterly from Union Vale to the Hudson River at Beacon. It has a total fall of about 600 feet in 26 miles, or an average fall of about 23 feet per mile. The drainage area above the USGS gauging station at Beacon, NY is about 190 square miles. The Fishkill watershed is drained by two major tributaries. The Sprout Creek drains the western portion of the watershed while the upper reaches of the Fishkill Creek drain the eastern portions of the watershed. The resulting drainage density of approximately 1.95 miles per square mile of watershed would normally be predictive of more rapid baseflow depletion than that of lower-density streams but baseflow depletion curves for the Wappinger Creek, Fishkill Creek and Tenmile Rivers are quite similar (Ayer & Pauszek, 1968), so the higher drainage density in the Fishkill Watershed appears to be more a function of having two major tributaries than of a single more efficient overall drainage network. The Tenmile River has a drainage area of 203 square miles above the USGS gauging station near Taylorsville, Connecticut. The watershed lies in Dutchess County and part of the state of Connecticut (Figure 1). A unique characteristic of the Tenmile river is its occupation of a single and deep north-south valley. Tributaries generally follow a dendritic drainage pattern within this elongate valley with a drainage density of approximately 1.75 miles per square mile of watershed. 2.2.2 Recharge Assessment Method Selection Selection of a recharge assessment method for this study involved selecting between methods and data sets based on necessary levels of precision and budget constraints. Key selection decisions follow: Scale: Chazen and DCWWA selected a study scale more geographically focused than the simple dual terrain model used by Wolcott & Snow, and by Snavely but



Page 16

less geographically precise than the New Jerseys emphasis on individual soil types. Visual inspection of Figure 2 confirms that use of the Hydrologic Soil Groups provides a far greater level of detail than use of the simple Valley and Upland distinctions used by Wolcott & Snow and by Snavely. Estimation of recharge rates for individual soils was both beyond the scope of work envisioned for this study and was judged unwarranted given the applications envisioned for the recharge estimates. A choice remained between continued using five glacial geologic terrains as identified by Gerber (1982) or to convert to use of four Hydrologic Soil Groups. The advantages of the NRCS Hydrologic Soil Groups (HSG) are evident because current maps showing Hydrologic Soil Groups are accessible to planners and community members and so require no continued mapping by others. These units were therefore judged more accessible and less prone to challenge than continued use of Gerber’s geologic mapping boundaries. No specific geologic or hydrogeologic investigations would be needed to apply recharge rates if the report findings were keyed to Hydrologic Soil Groups. NRCS soil groups, and maps thereof, also reflect a more current interpretive effort than the Gerber maps, and NRCS may be expected periodically update these soils maps as new soil survey date become available. The soil units and boundaries delineated by Gerber already no longer conform to most recent soils and surficial geology maps in some localities and so are not readily updated. Finally, the four hydrologic soil groups are assigned on the basis of infiltration ranges, considered relevant to proportional net aquifer recharge rates. Supportive work published recently by Brands et al (2005) also confirms that hydrologic soil groups and watershed drainage density most reliably predict or model baseflow recession performance in watersheds. Therefore, for purposes of this study, aquifer recharge has been assigned to NRCS Hydrologic Soil Groups. Where soils are classified by NRCS as Hydrologic Soil Group A/D or C/D, these have been consolidated in this analysis with their respective primary A and C classes since infiltration rates associated with the primacy capacity is generally dominant during rainfalls of less than 1.25 inch even in areas with buried bedrock interfaces or shallow watertable conditions. Stated another way, it is only during the heaviest rains that the “/D” adjudicating classification diverts infiltration to interflow along buried clay or buried bedrock horizons, or leads to rejection of recharge if soils are already fully saturated in riparian wetlands or otherwise already flooded sediments. Since research indicates that most groundwater recharge occurs during lighter rather than heavier rainfalls (Hoffman & Canace, 2002), uppermost soil horizons with an A or C HSG ranking are likely to allow initial rapid infiltration before being



Page 17

slowed by deeper horizons warranting the “/D” modifying characteristic. There are some areas where infiltration reaching a buried “/D” horizon will not reach a significant underlying aquifer, but for purposes of septic system density recommendations, many such areas are close to aquifer discharge locations where the general importance of calculating sustainable septic density recommendations may be less crucial because of a lack of downgradient wells. Areas for which no Hydrologic Soil Group assignment is available are limited, consisting primarily of unmapped soils and/or open water bodies representing only small percentages of the total watershed areas. Recharge Rates: Chazen identified aquifer recharge rates by using baseflow separation data from the three watersheds in Dutchess County, calculating the fractional share of each Hydrologic Soil Group in each watershed, and performing an approximate “best fit” analysis to select recharge rates for each soil group, constrained by the existing literature citation recharge values including those cited in Section 2.1.2 of this report. Posten’s method (Posten 1984) was selected for baseflow separations because this method best identifies baseflow associated aquifer recharge/discharge. Recharge rate adjustments were applied individually to each watershed based on historic average precipitation rates in the individual watersheds, and projected recharge rates were assigned to the towns of Hyde Park, Rhinebeck and Red Hook based on precipitation--calibrated adjustment of the baseflow-based aquifer recharge rates. Advantages of the selected approach include:

Literature sources cited in Section 2.1.2 provide credible recharge estimate ranges which can be reasonably matched to the Hydrologic Soil Groups used by this study.

The use of baseflow data from streams in and near Dutchess County allowed calibration to select “best fit” proportional recharge rates for each hydrologic soil group from among the many values found in the literature. Posten’s baseflow separation method is a conservative baseflow method (Hoffman, 1999) and so is least likely to over-emphasize interflow which does not meaningfully reflect groundwater used by groundwater wells. As such, calibration using the Posten method provides a balanced and somewhat conservative planning bias without over-focusing on “worst-case” recharge conditions. The balance between average and worst-case conditions was judged appropriate because aquifer storage capacity provides impact buffering during extreme droughts so residential density based only on drought data is overly conservative, and because any short-term defect occurring droughts will be manifested by temporary nitrate elevations but no more catastrophic well impacts. Rutledge (1998) recommends not using



Page 18

baseflow separation on watersheds over 500 square miles since it becomes increasingly difficult to distinguish surface water from true baseflow; however, neither the Wappinger Creek, Tenmile River nor the Fishkill Creek are near this limit.

Use of USGS precipitation data records (Randall, 1996) allowed proportional correction of recharge rates for each hydrologic soil group based on approximate precipitation rates identified for each watershed and for towns in northwest Dutchess County. 2.2.3 Aquifer Recharge Analysis for Dutchess County

Precipitation records from the Dutchess County airport were examined to evaluate relationships between annual precipitation and total flow in the Wappinger Creek, Tenmile River, and Fishkill creek. Daily rainfall data from the Dutchess County Airport from 1948 from 2001 were accessed from the National Climate Data Center (NCDC). Annual rainfall totals were compared to annual hydrograph totals for the available associated water years of the three watersheds. Figure 3 demonstrates that annual stream flow correlates generally with precipitation. The lack of perfect correlation reflects changes in groundwater storage based on drought or flush aquifer storage condition in preceding years. The Wappinger Creek was selected as a control watershed for this evaluation because it flows through the center of Dutchess County and has a long hydrograph data record. Fifteen years were selected in which total annual stream flow volume fell within 10 percent of the mean streamflow value over the period of record. From these, ten also agreed with years during which annual total flow of the Tenmile River was also within 10 percent of mean stream flow values for that stream. The Tenmile River has an approximately equal record of available hydrograph data. The ten years of average total stream flow discharge common to both the Wappinger Creek watershed and the TenMile River watershed were then selected for baseflow separation. The ten years for which records were available in both the Wappinger Creek and Tenmile River were: 1933, 1943, 1948, 1956, 1961, 1978, 1982, 1986, 1993 and 1998. The record of available hydrograph data from the Fishkill Creek is much shorter than that of the other two watersheds. The period of available record was inspected and, of the ten median streamflow years evaluated for the Tenmile River and Wappinger Creek, data for Fishkill Creek flows were only available for 1948, 1956, and 1961. Additional years were not selected for the Fishkill Creek since any other years would likely fall outside of those years during which streamflow in the other watersheds fell within 10 percent of average flows over their longer periods of record. Only these three years of record in the Fishkill Creek watershed were



Page 19

therefore selected for baseflow separation. Data years and resulting baseflow separation findings for each of these years from each watershed are summarized on Table 1. Baseflow separations were completed for these years following methods outlined by Posten (1984). Resulting tabulated findings from Chazen’s baseflow separation work are provided on Table 1. The results provide the basis for calculating average recharge rates for each watershed, constituting between approximately 50 to 54 percent of total runoff for each watershed. Variations are attributed to aquifer storage differences between the three watersheds among other considerations. As listed on Table 2, these recharge estimates equate to approximately 9.53 net inches of aquifer recharge across the full expanse of the Wappinger Creek watershed, 10.50 inches across the Fishkill Watershed and 10.59 inches across the Tenmile River watershed. The differences in recharge in these watersheds closely follows the pattern of increased annual precipitation across the County, with approximately 40 inches of rainfall in the Wappinger watershed, 42 inches of rainfall in the Fishkill watershed and as much as 44 inches of rainfall in the Tenmile River watershed (Figure 1). Figure 1 also indicates that precipitation in the towns of Hyde Park, Rhinebeck and Red Hook near the Hudson River is only approximately 38 inches, predicting proportionally lower net aquifer recharge rates in these towns. To estimate recharge rates attributable to different soils, hydrologic soils groups were totaled in each watershed using a GIS. Total acreages in watershed and area are shown on Table 2. Trial-and-error, weighted average recharge assignments constrained by literature-guided recharge rate estimates were adjusted until predicted net watershed recharge rates reached a best possible fit, calibrating with the watershed-wide baseflow findings to within approximately 0.2 inches of actual annual recharge. Agreement to within 0.2 inches was judged to be a reasonably successful level of precision given the scale of the study and the range of potential additional variables contributing to watershed systems. Proportional changes in available rainfall in each basin were used to scale recharge estimates for each soil group. For example since rainfall in the Wappinger watershed is approximately 95% of that in the Fishkill watershed, the recharge values assigned to each hydrologic soil group in the Wappinger watershed is 95% of the recharge values in the Fishkill watershed. Calibration stream flow data were not available for all portions of Dutchess County. However, inspection of Figure 1 indicates where precipitation rates remain consistent with or suggest the need for proportionally adjusted aquifer recharge rate assignments beyond watershed boundaries. Precipitation in Hyde Park, Rhinebeck and Red Hook is approximately 10 percent less than rainfall in the Fishkill watershed. Rainfall in northern portions of Milan and Pine Plains are approximately similar to rainfall in the Wappinger watershed, so reasonable



Page 20

estimates of proportionally adjusted aquifer recharge for such areas could be projected. Similarly, rainfall and thus likely aquifer recharge rates in the Town of Poughkeepsie is likely to be comparable to that found throughout the Wappinger Creek watershed. Rainfall in the Town of Wappinger includes rates found in both the Wappinger Creek and the Fishkill/Sprout Creek watersheds, so recharge rates and associated aquifer recharge rates from either watershed might apply but a conservative assignment selects the Wappinger Creek values. By such means, reasonable estimates of aquifer recharge rates can be provided for all areas in Dutchess County. Aquifer recharge rates for each hydrologic soil group within the individual watersheds and for the northwest municipalities of Hyde Park, Rhinebeck and Red Hook and enclosed villages are shown on Table 2. Hydrologic soil groups are shown on Figure 6.



3.0

Page 21

SUSTAINABLE SEPTIC SYSTEM DENSITY ANALYSIS

The preservation of well yield capacity and groundwater quality are important factors for the sustainable use of domestic wells and conventional septic systems. Individual systems are prevalently used in suburban to rural areas, while community cores and some suburban areas rely on centralized water and/or sewer services. The following analysis of sustainable septic system densities applies only to areas using domestic wells and conventional individual or community septic systems. State regulations already require specific separation distances between wells and septic system leaching fields. The distances are primarily focused on ensuring reliable die-off of pathogenic coliform and viruses (10 NYCRR, Appendix 5-B). Less attention has so far been given to groundwater quality management addressing conservative inorganic wastewater constituents like sodium chloride and forms of nitrogen for which drinking water standards exist, or for recently-identified pharmaceutical product residues, endocrine disrupting steroids and other personal care chemicals for which no standards presently exist. The presence of high concentrations of nitrates may be an indication that there are other such persistent contaminants in groundwater originating from septic tanks which, like nitrate, do not readily break down and will normally be most cost-effectively managed by dilution. The relatively high concentration of nitrogen released from septic systems and the relatively low drinking water and groundwater standard for nitrate has made nitrate frequently-modeled compound for research and sustainable density analysis. Groundwater nitrate concentrations originating from natural sources are generally inconsequential while a wide range of human activities can increase nitrate levels substantially. The New York State Department of Health standard for nitrate in drinking water is 10 milligrams per liter (mg/l). The average human being releases approximately 10 pounds of nitrogen per year to septic systems (Hoffman & Canace, 2002). In a properly functioning septic system, nitrogen is converted to nitrate by biological activity occurring in the soil horizon under system leaching fields. Once nitrification processes are complete, nitrate is generally stable in groundwater, and since it does not bond to soil, it migrates freely with groundwater (Hoffman & Canace, 2002). Some natural processes reduce nitrate levels in groundwater, including denitrification to nitrogen gas under specific conditions or nitrogen uptake by vegetation; however, neither is considered a significant mitigating factor in this investigation because denitrification to nitrogen gas occurs most readily in



Page 22

anaerobic settings uncharacteristic of Dutchess County aquifers, and most vegetative nitrate mitigation occurs when the water table is within 5 to 10 feet of ground level (from Schoonover & Williard, 2002), a condition generally existing in Dutchess County only near aquifer discharge locations including wetland and stream riparian areas. The concentration of nitrate in septic system effluent both at its point of release from the septic system and at its point of mixing with an underlying aquifer has been estimated by various investigators, and likely ranges between 30 and 50 mg/l (e.g. Horseley Witten Hegemann, 1992). Other investigators concentrate on identifying the annual nitrate load to the aquifer rather than day-to-day effluent concentrations because the later tends to increase as water conserving plumbing fixtures become increasingly prevalent and fluctuate more on a day-to-day basis because of varying domestic activities. A general estimate of annual nitrogen waste discharge is 10 pounds of nitrogen waste per year per capita (Hoffman & Canace, 2002) As the nitrate impact of this concentrated release migrates away from septic field areas it persists as a nitrate plume exceeding groundwater and drinking water standards unless and until diluted by aquifer recharge. The essential components of a sustainable well and septic system density model based on nitrogen dilution, provides a general analog for approaches which could also be used to manage other conservative wastewater contaminants; however, no comparable ratios to ensure potability of other constituents are yet known to be available. Therefore, nitrate modeling was used in this analysis. 3.1

Review of Models and Assumptions

Various nitrate dilution models have been described in published literature or have been applied by consultants working in Dutchess County. •

The earliest known residential density model applied to Dutchess County is that of Gerber (1982) who used a mass balancing equation to ensure that sufficient recharge was available near each septic system to ensure dilution of the wastewater concentration estimated to flow from each septic system. The formula and its conclusions are reproduced at Figure 5. Gerber’s estimated septic effluent concentration were 300 gallons per day conveying 30 mg/l nitrate and Gerber’s target concentration after dilution by local aquifer recharge was 10 mg/l. The necessary number of acres of land calculated by Gerber to provide the necessary nitrate dilution ranged from 0.5 acres for areas with readily-recharged thick sand and gravel deposits to 4.0 acres for areas with glacial-era lake deposits of silt and clay that allow minimal aquifer recharge. The strength in Gerber’s method is its simplicity. Weaknesses of the Gerber method include the use of a precise estimates of



Page 23

wastewater gallons and effluent concentrations both of which are subject to change as conservation water fixtures become more common, use of a target nitrate concentration of 10 mg/l rather than using a more conservative target which ensures that wells will remain below rather than at the standard of 10 mg/l, an accessible method to adjust the calculation based on household occupancy levels, and the lack of any consideration of impervious surfaces on or associated with the recommended acreages. A broader weakness with the Gerber model is that, as published, it requires continued reference to his geologic terrain maps which no longer coincide with published Natural Resources Conservation Service (NRCS) soils maps. •

In 1992, Horsley Witten & Hegemann completed nitrogen loading analyses assessing water quality threats to specific wells supporting centralized community water systems in Dutchess County. The study was not conducted to recommend individual septic system densities, but its consideration of potential sources of nitrogen contamination concluded that individual septic systems were estimated to release 124 gallons of wastewater daily at an effective concentration of 33 mg/l nitrate (Horsley Witten Hegemann, 1992). The nitrate concentration closely matched that of Gerber (1982) but the effluent volume reaching the watertable sharply differed from that used by Gerber. Where the Horsley, Witten & Hegemann report referenced aquifer recharge rates, they appear to be based on the work by Gerber (1982).

•

In 1998 and 1999, The Chazen Companies recommended minor modifications to Gerber’s calculations to ensure greater assurances that groundwater nitrate concentrations would reliably remain below 10 mg/l. The specific recommendation was to replace Gerber’s groundwater quality target of 10 mg/l with a planning objective of 5 mg/l. The Gerber target essentially conceded that the resulting distribution, or bell-curve, of actual groundwater quality findings would fall both below and above the groundwater and drinking water standards of 10 mg/l once full build-out was achieved. To avoid this conclusion, Chazen recommended recalibrating the Gerber density calculation to a target output of 5 mg/l (Chazen, 1998, 1999). The change effectively doubled the recommended acreages provided by Gerber. No other modifications to the Gerber model were proposed by the Chazen Companies at that time. Chazen did suggest that localized recharge by water run-on from valley walls (see vocabulary discussion in Section 2.1.1) was supplementing direct recharge along the steep valley walls in the Harlem Valley. However, since most groundwater recharge occurs during lowintensity storms which generate little runoff, the volume of such supplemental recharge may not be great as previously thought, and the significance of such run-on during heavy rains would likely result only in



Page 24

near-stream interflow or quickflow recharge rather than significant additional aquifer recharge. •

Various investigators working primarily in New Jersey have developed a detailed groundwater recharge and nitrate dilution residential density model for use in the State of New Jersey (Charles, et al, 1993). The methods used to identify recharge rates are addressed briefly in Section 2.1.2 of this report. The New Jersey density calculations are stipulated to apply to aquifer recharge rates and aquifers in New Jersey and include additional variables described here to recommend sustainable well and septic densities. Specifically, the calculation recommends using a target nitrate concentration of 5.2 mg/l to statistically ensure that the distribution or bell-curve of probable groundwater quality outcomes remains below 10 mg/l. The calculation avoids the vagaries of effluent concentration and volume estimates by instead using a reference per capita waste generation factor of 10 pounds of nitrate-nitrogen per person per year. Residential occupancy rates can thus also be addressed in the formula. Finally, the formula includes a natural-log variable interpreted from air photo evaluation of developed parcels of differing sizes to account for the inverse percentage relationship between impervious surfaces and lot size. The New Jersey density formula is shown on Figure 4. This model is detailed and has been peer reviewed and so is judged to be one of the most defensibly developed of the formulae surveyed here. It is somewhat difficult to use because simultaneous solution of two variables in the calculation is required to identify recommended acreages relative to a given recharge rate.

•

A nitrate-based residential density model applied by the New York Rural Water Association (NYRWA, 2005) uses general approaches from all prior investigations but separately calculates and then combines effluent dilution and well recharge areas for each parcel. The effluent dilution area is calculated using a target nitrate concentration of 5 mg/l and an average household wastewater discharge volume of 200 gallons daily at 40 mg/l nitrate, based on a typical household occupancy of 2.5 persons discharging 80 gallons daily per capita. The additional well recharge area is specified based on withdrawals of 250 gallons daily. It is not clear why the dilution acreage cannot meet both the dilution objective and the well recharge requirement. The NYRWA method adds an impervious surface factor from the USDA Soil Conservation Service 1986 TR-55 small watershed Urban Hydrology program to reach final density recommendations. There is little discussion of the source of recharge rates used in the NYRWA study but recharge rate values appear comparable to estimates developed by Gerber (1982).



3.2

Page 25

Septic System Density Analysis for Dutchess County

This study has adopted use of a variation of the NJ septic system density calculation with a modification to allow simpler, user-selected input of an impervious surface factor. The resulting method is shown and discussed below. A = 4.4186HM / CqR + Isc Where A = recommended minimum acres per system, in acres H = persons per system M = pounds of nitrate-nitrogen per person per year, in pounds Cq = Nitrate-nitrogen target average groundwater concentration, in mg/L R = Annual Recharge Rate, in inches Isc = Impervious surface cover, in acres. Use of an Isc factor selected on a case-by-case basis rather than New Jersey’s automatically-calculated standard is recommended for Dutchess County, not only to simplify use of New Jersey’s mathematically complex model, but also to allow local analysis and adjustment of the impervious surface factors for particular project sites, impervious coverage allowances in local zoning, or to suit regional design allowances for road drainage design or other stormwater management approaches. Table 3 and the Executive Figure in this report use a default Isc factor of 0.1 acres per septic system for minimum average recommended parcel sizes and a factor of 0.3 acres for minimum average regional densities. These factors are applied in Table 3 but the variables may be modified where warranted by site conditions. Table 3 summarizes the resulting recommended minimum septic system densities for each Hydrologic Soil Group in Dutchess County’s 3 major watersheds and the Towns of Hyde Park, Rhinebeck and Red Hook. Recommended minimum average parcel sizes and regional septic system densities assume roads with stormwater controls generally diverting runoff to surfacewater bodies and site design which does not emphasize on-site recharge of runoff from driveways and roofs. Use of design features which compensate for aquifer recharge lost due to impervious surfaces offers potential opportunities or incentives for denser but still sustainable septic system uses.



3.3

Page 26

Conditions and Applications

Various conditions apply to the present septic system density recommendations. These include the following:

Recharge rate estimates are based on best available stream hydrograph records over a period of record of approximately 70 years, manually completed baseflow separation analysis of median hydrograph flow years, consolidation of hydrologic soil groups, an approximation of typical long-term rainfall for Dutchess County’s three watersheds and northwest towns, and best-fit calibration of data to identify approximate recharge rates.

In selecting a septic system density model, a planning target for groundwater nitrate concentrations from regional use of conventional septic systems of 5.2 mg/l was selected to ensure that the vast majority of actual groundwater nitrate concentrations would vary around the target concentration such that an anticipated natural divergence of actual values (e.g. the statistical bellcurve of possible outcomes) will remain below the drinking water standard of 10 mg/l at the downgradient property line.

Additional sources of nitrate do exist but are not addressed by this septic system density model. Lawn chemical applications are frequently the nextmost common source of nitrate in aquifers in heavily settled areas. If properly applied, lawn fertilizers are fully utilized by site vegetation and need not contribute to elevated regional groundwater nitrate concentrations. Moreover, wastewater discharges are released at every home, not just those that elect to use law fertilizers and are released below the lawn level and so can only be partially and seasonally mitigated by vegetative uptake in limited areas over septic system leaching fields. In contrast, elevated nitrate resulting from lawn fertilization can be readily addressed or mitigated by modified practices and community best management practice education. Nitrate concentrations released from conventional septic systems can only be mitigated by dilution measures recommended here or by using more costly advanced onsite treatment facilities seldom considered for broad regional use. If advanced treatment unit septic systems are used in focused regions to effectively reduce nitrate releases to groundwater, adjustments to the density model could be considered since the effective per capita nitrate-nitrogen release per person to the groundwater would be reduced below 10 pounds per year.

Density recommendations based on nitrate dilution rely on various fundamental hydrogeologic and operational assumptions. These are listed in detail in Hoffman & Canace (2002). Most are judged to fully apply to



Page 27

Dutchess County such that the model approach is judged to be applicable in all but rare circumstances throughout Dutchess County. These include: o Wastewater releases and on-site groundwater recharge occur in the same aquifer and at least seasonally there is a high likelihood of complete and uniform mixing of the two aspects. During wet periods, some components of wastewater and groundwater may leave sites as interflow, but during dry seasons, both components will fully mix in the common aquifer on or near the site. The model selected here may generally be considered to predict the average nitrate concentration in the aquifer at the downgradient property line. o The only water available to dilute wastewater is on-site recharge. The assumption ignores mixing of the plume with upgradient groundwater since the calculation is intended for sub-regional applications (e.g. build-out, subdivision, or zoning district scale applications) where density cannot rely on other areas to provide necessary dilution water streams. o Once in the aquifer, nitrate is effectively inert and not prone to decomposition by any methods. Dilution, therefore, is presently the most cost-effective quality management technique. This assumption would also apply to other wastewater constituents not prone to biological breakdown (e.g. pharmaceutical residues, caffeine, etc.), but the dilution calculations here have been applied only to nitrate. Within residential areas, nitrate is generally not selectively removed by root systems since upland depths to watertable during critical dryseason periods normally exceed at least 30 feet, only becoming shallower in riparian or near-wetland settings. A literature search indicates that denitrification from vegetation normally only occurs where the watertable is within 10 feet or less of grade (Schoonover & Williard, 2002) o There is a one-to-one correspondence between homes and disposal systems. Where community septic systems are used, the numbers of users per system can easily be adjusted in the model calculation. o By adjusting the Impervious Surface Cover factor, the density calculation may be revised to address any beneficial on-site recharge design features (e.g. any Low Impact Development LID or other good design features which enhance on-site recharge) or revised to reflect precisely known or allowed impervious surface acreages from roofs, driveways and roadways. Where good design practices are used and



Page 28

storm drainage does not intentionally channel roof, driveway and other runoff away from the site, most precipitation may continue to recharge sites during lower-volume rain events by flowing to nearby lawn or natural areas to recharge the underlying aquifer. o The model is not intended to accurately identify precise nitrate concentrations along groundwater flow paths, but rather to address broader impact estimates of regional use of conventional septic systems on aquifers also used to support domestic wells. The general assumptions are judged to model conservative and realistic average nitrate concentrations, but in doing so will underestimate nitrate concentrations immediately downgradient of a system leaching field and overestimate nitrate concentrations in areas most distant from a leaching field plume.

The density recommendations found here do not preclude use of cluster subdivision models as recommended in the Dutchess County Greenway Connections guide and many municipal zoning ordinances and Comprehensive Plans. As long as overall site density objectives are met, and with proper site design and engineering practices, the model will continue to manage groundwater nitrate concentrations while allowing clustered construction techniques. For example, in some cases, one could increase the length of casing for added protection if clustered wells and septic systems must lie near one another, or quality risks may be reduced by locating wells upgradient from septic systems, or wells may be sited in open space areas to protect well water quality.

The present model derives ground water recharge rates from precipitation and steam-flow calibration data derived over an approximate period of up to 70 years. This period accounts both for wet and drought periods and so represents long-term averages. Some investigators use specific drought conditions, which could result in more conservative minimum parcel size recommendations, however, use of the extended period of record provides a strong record for use of typical effective aquifer recharge rates in Dutchess County, and the baseflow method selected by the study is considered sufficiently conservative to represent average recharge rates to the waterbearing aquifer formations supporting domestic wells without interference by more ephemeral interflow recharge volumes.

Although in summer very little aquifer recharge occurs which can dilute wastewater nitrate concentrations, groundwater moves very slowly so summer-time nitrate loads travel a long time before reaching a downgradient well. This travel time is usually longer than a full summer, so the autumn,



Page 29

winter, and spring wet periods provide the necessary recharge and nitrate dilution before a nitrate plume from one septic system will normally reach an adjoining downgradient well assuming typical geology and adherence to the average density recommendations found herein.

This model does not address yield reliability of wells, focusing instead on recommending regional water balances needed to ensure potable groundwater quality. The effect of the model, however, is to recommend large enough parcel densities that recharge volumes needed for dilution purposes also exceed normally rates of well-water consumption, therefore effectively providing a measure of protection preserving most well yields during extended drought periods. In some cases, however, specific study will still be required to ensure that well capacity needs can be met on sites.

Soil cover rather than bedrock formation is an effective predictor of net aquifer recharge. Surficial aquifers and other unconsolidated pore space provides temporary storage retaining groundwater over buried bedrock surfaces, facilitating recharge to the underlying deeper aquifer zones normally intersected by domestic wells. Neither recharge nor septic system wastewater releases are observed seeping directly into streams during dry periods, verifying that during all but the wettest periods, a majority of recharge passing through soils reaches the underlying aquifer.

This study does not factor in changes in aquifer recharge that might be attributable to slopes. Recharge values during the low-intensity storms responsible for most aquifer recharge does not generate much runoff so even precipitation on steep slopes is able to penetrate into the ground.

No correction factors have been applied to calibration data to account for existing land uses in Dutchess County. This is because the vast majority of land uses in the Wappinger Creek, Tenmile River, and Fishkill/Sprout Creek watersheds, with potential exception of lower-watershed portions of the Fishkill watershed including the Village of Fishkill, remain substantially in conditions of open land or residential development with impermeability coverage of less than 15 percent. Under such conditions, stormwater peaks (which are not the subject of this study) may be increasing due to more rapid times of runoff concentration; however, overall runoff volumes generally do not substantially increase relative to pre-development natural conditions during the more routine rainfalls events of under 1.25 inch which provide up to 80 percent of annual recharge (Charles, et al, 1993; Chazen March 2006). Until regional impervious surface coverage increases markedly, soil infiltration capacity rather than present land uses is judged to be the



Page 30

principal ranking parameter for recharge capacity (Hoffman & Canace, 2002).

This investigation does not include overt factors for evaluating the potential impacts of global warming. If, as has been outlined and/or predicted in other professional literature, future rainfall patterns become more torrential with longer periods without precipitation between heavy rains, recharge rates will decline since a majority of present recharge is normally attributed to less torrential rainfalls. The present model is nonetheless somewhat conservative in its use of a 5.2 mg/l nitrate planning target, allowing some latitude to compensate for precipitation pattern changes; however, the present recommendations may cease to be adequately protective of the quality of groundwater if extreme weather pattern shifts occur.

Less than 2 percent of Dutchess County underlies existing urban or residential centers which have received no NRCS hydrologic soil group assignment (Table 2 and Figure 6). To calibrate the present model and resulting estimates, this study assigned these soils to the HSG B ratings since much early development of urban centers in Dutchess County occurred over the County’s more permeable soils, judged arbitrarily to likely fall most commonly in the hydrologic soil group B category. If investigations of these limited developed areas are needed, hydrogeologists or soil scientists should be able to assign more precise hydrologic soil groups based on topography, surficial geologic maps or older soils maps. Some of these unassigned areas are also known to be open water bodies.



4.0

Page 31

RELATED TOPICS FOR FUTURE WORK

Analysis summarized in this investigation provides aquifer recharge data and sustainable septic system density recommendations with various immediate applications. Additional areas of investigation have been identified here which could build on or supplement the present work. Examples of such future work includes: •

Analysis of groundwater quality in existing developed neighborhoods throughout Dutchess County. Securing such data may help identify existing concentrations of nitrate and other conservative (persistent) wastewater constituents in groundwater. The septic system density calculations found in this report reflect calibration by mass balance approaches and are supported by reference data documented in aquifers in New Jersey; notwithstanding, groundwater quality data from Dutchess County could help further calibrate the septic density recommendations provided here.

•

Guidance for parcel-scale development of cluster subdivisions. Projects may meet the overall density recommendations provided by this study, but will require close design consideration when wells and septic systems are positioned closely in a cluster or conservation subdivision on a parent parcel.

•

Preparation of water budget estimates for existing or proposed projects. The aquifer recharge data developed by this report will contribute to recharge analysis often provided with aquifer pumping test reports or with water use studies for new projects. Recharge data provided by this report will support water budget analyses at such project sites, and so help predict potential offsite impacts to other wells, or streams, wetlands, and associated habitat areas.

•

Preparation of Municipal groundwater resource plans. The aquifer recharge rate estimates developed by this study and the interflow recharge rate estimates also recently developed by Chazen (March 2006) for the Wappinger Creek watershed will provide essential background for evaluations of groundwater recharge and availability in municipal groundwater planning studies.

•

Minimum stream baseflow studies. Habitat and minimum stream-flow evaluations will benefit from the baseflow data developed in this report, helping identify critical minimum baseflow needs for stream flow preservation. Where off-site habitat impacts are assessed, the interflow as



Page 32

well as aquifer recharge baseflow components will be needed (Chazen, March 2006). •

Stormwater management design. Baseflow discharge to streams is reduced by not only human water consumption but also as a result of stormwater management designs which route all water from impervious surfaces directly to streams. Data found in this report and a recent study of the Wappinger Creek watershed (Chazen, March 2006) may contribute to stormwater management planning as it relates to stream flow protection initiatives.

•

Precipitation Trend and Aquifer Recharge Rate Analysis. Impacts of global warming on precipitation patterns and hence aquifer recharge rates is poorly understood and a subject of future consideration to ensure viability of groundwater supplies under changing climatic conditions.

•

Provision of Recharge Enhancement Protocols. The present study points to the importance of aquifer recharge not only as a source of water supply to wells but as a source of dilution for below-grade wastewater discharges. Opportunities or measures which enhance recharge will contribute to sustainability of potable groundwater sources of water supply and may allow greater density of septic system installations. Density bonuses calculated interpreting the septic density formula introduced in this report may be provided where such measures are employed. Protocols to ensure introduction of clean water to the subsurface will also be necessary.

•

Exploration of Advanced Septic System Treatment Designs and Uses: Septic system modifications which reduce nitrate loading to aquifers would require less dilution capacity from aquifers, admitting the possibility of smaller average minimum parcel sizes. Septic system treatment districts using such modifications could potentially avoid use of central water or central sewage treatment while still allowing smaller average parcel sizes.



5.0

Page 33

REFERENCES

Brandes, D., Hoffmann, J.G., Mangarillo, J.T., 2005. Base Flow Recession Rates, Low Flows, and Hydrologic Features of Small Watersheds in Pennsylvania, USA. in Journal of the American water Resources Association, Vol 41, No 5, October 2005, pages 1177-1186. Cervione Jr., M., Mazaferro, D.L., Melvin, R.L., Water Resources Inventory of Connecticut. 1972. Part 6, Upper Housatonic River Basin, USGS Connecticut Water Resources Bulletin No. 21. Charles, E.G., Behroozi, C., Schooley, J., Hoffman, J.L., 1993. A Method for Evaluating Ground-Water-Recharge Areas in New Jersey. New Jersey Geological Survey Geological Survey Report GSR-32. 95 p. Chazen Companies, The. 1998. Water Resource Assessment for the Town of Dover. For the Town of Dover. Chazen Companies, The. 1999. Harlem Valley Watershed Investigation, Dutchess County, NY. For the Town of Dover and Dutchess County Water & Wastewater Authority. Chazen Companies, The, 2003, County-wide Groundwater Monitoring Program, 2002 Annual Report, Dutchess County, prepared for the Dutchess County Water & Wastewater Authority. Chazen Companies, The, March 2006, Wappinger Creek Watershed Groundwater Recharge and Stream Baseflow Evaluation Assessment, Wappinger Creek Intermunicipal Counsel. de Vries, J.J., Simmers, I., 2002, Groundwater recharge: an overview of processes and challenges, in Hydrogeology Journal, Vol 10, No. 1, February 2002, pages 5-17. Gerber, R. 1982, Final Report: Water Resources Study for Dutchess County. For Dutchess County Department of Planning. 40 p. and various maps on file the Dutchess County. Hiscock, K., 2004. Hydrogeology: Principles and Practice. Blackwell Publishing, 389 p., Malden MA.



Page 34

Hoffman, J.L., 1999, New Jersey Geological Survey, Technical Memorandum 99-1 (supplement to GSR 32: A method for evaluating ground-water-recharge areas in New Jersey). Hoffman, J.L., Canace, R.J., 2002, A Recharge-Based Nitrate-Dilution Model for New Jersey, New Jersey Geological Survey and New Jersey Department of Environmental Protection. 8/12/02 draft. Horsley Witten Hegemann, Inc., 1992. Water Supply Protection Program for Dutchess County, New York. For Dutchess County Water & Wastewater Authority. New York Rural Water Authority, 2005. Draft Ground Water Protection Plan for the Town of Saugerties. May 2005. Posten, S.E., 1984, Estimation of Mean Groundwater Runoff in Hard-rock Aquifers of New Jersey: Columbia University Seminar Series on Pollution and Water Resources, v. 16, Halaski-Kun, G.J., editor, Pergamon Press, NY, p. 109-154. Randall, A. 1996. Mean Annual Runoff, Precipitation, and Evapotranspiration in the Glaciated Northeastern United States, 1951-80, USGS Open-file report 96-395. Rutledge, A.T., 1998, Computer Programs for Describing the Recession of Groundwater Discharge and for Estimating Mean Ground-water Recharge and Discharge from Streamflow Data – update. USGS Water-Resource Investigation Report 98, 4148:43. Scanlon, B.R., Healy, R.W., Cook, P.G., 2002. Choosing appropriate techniques for quantifying groundwater recharge, in Hydrogeology Journal, Vol 10, No. 1, February 2002, pages 18-39. Schoonover, J.E., Williard, K.W.J, 2002. Ground Water Nitrate Reduction in Giant Cane and Forest Riparian Buffer Zones, in Journal of the American Water Resources Association. Vol 39, No. 2, pages 347-353. Snavely, D., 1980, Ground-water appraisal of the Fishkill-Beacon Area, Dutchess County, NY. US Geological Survey. Water Resources Investigation Open-File Report 80-437. 14 p. Wolcott, S.W., Snow, R.F. 1995. Computation of Bedrock-Aquifer Recharge in northern Westchester County, NY and Chemical Quality of Water from Selected Bedrock Wells. USGS. Water Resources Investigations Report 92-4157.


Tables

Table 1 - Aquifer Discharge Baseflow Separation in Dutchess County Watersheds

Total Watershed

Baseflow

Year Average cfs/day Average cfs/day

Wappinger Creek

Tenmile River

Fishkill & Sprout Creek

1933 1943 1948 1956 1961 1978 1982 1986 1993 1998 Average 1933 1943 1948 1956 1961 1978 1982 1986 1993 1998 Average 1948 1956 1961 Average

269.69 247.18 263.08 260.95 254.27 232.39 258.44 264.69 260.48 245.90 256 286.48 308.88 298.87 302.41 301.21 291.57 298.57 288.32 306.45 279.81 296 291.90 304.33 273.33 290

126.34 135.24 122.16 135.24 125.00 113.55 129.16 126.71 118.12 138.65 127 159.33 167.08 161.69 159.26 171.53 153.64 152.91 145.86 159.37 152.89 158 123.60 170.01 147.38 147

% Aquifer Baseflow 46.8% 54.7% 46.4% 51.8% 49.2% 48.9% 50.0% 47.9% 45.3% 56.4% 49.7% 55.6% 54.1% 54.1% 52.7% 56.9% 52.7% 51.2% 50.6% 52.0% 54.6% 53.5% 42.3% 55.9% 53.9% 50.7%

Baseflow separations were conducted using methods defined by Posten (1984) to identify annual total aquifer discharge contribution to annual flow volume of the Wappinger and TenMile watersheds during average flow years from the available periods of record. The Fishkill Creek period of record is shorter than that of the other two watersheds so only Fishkill Creek years matching flow years selected for the Tenmile River and Wappinger Creek longer period of record were analyzed.


Table 2 - Aquifer Recharge Rates by Watershed/Region and Hydrogeologic Soil Group

Aquifer Recharge Rates - Hydrologic Soil Group Analyses

Fishkill Watershed Wappinger Watershed Tenmile Watershed WEIGHTED AVERAGES2 Northwest Municipalities3

147 127 158 N/A

10.50 16,030 9.53 14,082 10.59 5,065 10.23

13% 12% 4% 11%

acres

%

19.2 34,034 28% 14.0 63,220 52% 18.2 28,472 25% 13.3 64,080 55% 20.2 43,678 34% 14.7 71,266 55% 19.0 29% 14.1 54% 17.3 12.6

7.2 6.8 7.6 7.2 6.5

5,554 8,169 8,459

%

5% 7% 7% 6%

4.0 3.8 4.2 4.0 3.6

1

1

D acres

other

%

2,357 178 1,931

2% 0% 1% 2%

recharge rate (inches/year)

%

1

1

acres

aquifer recharge (inches/year)

%

D and other

aquifer recharge (inches/year)

acres

C and C/D aquifer recharge (inches/year)

1

B aquifer recharge (inches/year)

Actual inches annual recharge

cfs/day (from seperation)

A and A/D

14.0 13.3 14.7 14.3 12.6

Total watershed (acres)

121,600 115,840 129,920

Predicted inches annual recharge from shaded recharge rates

Annual Baseflow

10.65 9.56 10.37 10.20

Notes: 1. Aquifer recharge rate variations between the Wappinger, Fishkill and Tenmile watersheds reflect percentage precipitation differences as follows: Wappinger Creek recharge (95% of Fishkill ); Tenmile River recharge:(105% of Fishkill) 2. Weighted aquifer recharge rate averages for the Fishkill, Wappinger and Tenmile watersheds 3. Aquifer recharge rates for Hyde Park, Rhinebeck and Red Hook and enclosed villages are projected from calibrated rates above, proportionally adjusted to 90% of Fishkill watershed aquifer recharge rates based on 90% precipitation rates. Portions of Milan and Pine Plains north of the Wappinger Creek watershed are expected to have aquifer recharge values comparable to those within the Wappinger watershed. Portions of the Town of Poughkeepsie west of the Wappinger Creek watershed are expected to have aquifer recharge values comparable to those within the Wappinger watershed. Portions of the Town of Wappinger west of the Fishkill Creek watershed are expected to have aquifer recharge values comparable to those within the Fishkill watershed.


Table 3 - Recommended Minimum Average Septic System Densities

Hydrologic Soil Groups: Region (Municipalities) Fishkill Watershed (part-LaGrange, Union Vale, Wappinger, Beekman, Fishkill, East Fishkill) Wappinger Watershed (Milan, Pine Plains, Clinton, Stanford, Pleasant Valley, Washington, Millbrook, LaGrange, Poughkeepsie, Wappinger, Wappingers Falls) Tenmile Watershed (Northeast, Millerton, Amenia, Dover, Pawling) Northwest Municipalities (Hyde Park, Rhinebeck, Red Hook)

Acres per System - Recommended Minimum Average Acreage per Septic System A and A/D B C and C/D D Average Min. Regional Average Min. Regional Average Min. Regional Average Min. Parcel Size* Density** Parcel Size* Density** Parcel Size* Density** Parcel Size* 1.3

1.5

1.7

1.9

3.2

3.4

5.6

5.8

1.3

1.5

1.8

2.0

3.3

3.5

5.9

6.1

1.2

1.4

1.6

1.8

3.0

3.2

5.4

5.6

1.4

1.6

1.9

2.1

3.5

3.7

6.2

6.4

Governing Formulas *Average Minimum Parcel Size includes 0.1 acre allowance for impervious surfaces: ** Regional Density includes 0.1 acre allowance for site, plus 0.2 acre allowance for road: Selected Inputs to Governing Formulas: Occupancy Rate (H): Nitrate Discharge (M) Target GW concentration (Cq): Natural Recharge rate (R): Impervious allowance per parcels (Isc): Impervious allowance for region (Isc):

Regional Density**

2.6 10.0 5.2 (see Table 2) 0.10 0.30

A = 4.4186HM / CqR + Isc (where Isc = 0.1 acre) A = 4.4186HM / CqR + Isc (where Isc = 0.3 acre)

Persons/household (2000 Dutchess County average) Annual pounds per year per person mg/L, Nitrate Inches per year Acres per developed parcel, including roofs, driveways and other onsite impervious areas. Acres associated with each developed parcel, including roofs, driveways, etc. PLUS a share of new roads.

In general, most applications of this analysis will focus on the average minimum parcel size recommendation (shaded values). Some investigations consider overall regional density (clear values).


Figures

Tivoli

Red Hook Red Hook Milan

Pine Plains Millerton

Rhinebeck Rhinebeck

46

38

Northeast

Stanford

Wappinger Creek Watershed

48

Clinton

40

Amenia

Hyde Park

Washington Millbrook

Tenmile River Watershed

42

Pleasant Valley

Poughkeepsie


Dover

44

Poughkeepsie

Beekman

42

Wappingers Falls Wappinger

Fishkill Creek Watershed East Fishkill

Pawling Pawling

Fishkill Beacon

Fishkill

46

Legend 40

Mean Annual Precipitation Isopleths, inches Major Streams and Rivers Glacial Fluvial Sand, Gravel Silt or Clay Deposits Watershed Boundaries

Dutchess County Office: 21 Fox St. Poughkeepsie, NY 12601 Phone: (845) 454-3980 Orange County Office: 356 Meadow Ave. Newburgh, NY 12550

ENGINEERS/SURVEYORS PLANNERS ENVIRONMENTAL SCIENTISTS

Capital District Office: 547 River Street Troy, NY 12180 Glens Falls Office: 110 Glen Street Glens Falls, NY 12801

40130.00-old\20060429_USGS_Prec_Watershed.mxd) 4/27/2006

FIGURE 1 - DUTCHESS COUNTY WATERSHEDS, PRECIPITATION AND GEOLOGY Dutchess County, New York Precipitation Source: USGS Mean annual runoff, precipitation, and evapotranspiration in the glacial northeastern United States, 1951-1980

Date: April 2006 Scale: Not To Scale Project #: 40130.00

Legend

Legend

Hydrologic Soil Groups

Upland Bedrock and Glacial Till Valley Glacial Fluvial Sand, Gravel or Clay Deposits

other A B C A/D C/D D

Decimal Degrees 00.0015 0.003 0.006 0.009 0.012

Created by:

CHAZEN ENGINEERING & LAND SURVEYING CO., P.C. Dutchess County Office: 21 Fox Street Poughkeepsie, New York 12601 Phone: (845) 454-3980

Engineers/Surveyors Planners Environmental Scientists GIS Consultants

Orange County Office: 356 Meadow Avenue Newburgh, New York 12550 Phone: (845) 567-1133

Capital District Office: 547 River Street Troy, New York 12180 Phone: (518) 273-0055

North Country Office: 110 Glen Street Glens Falls, New York 12801 Phone: (518) 812-0513

This map is a product of The Chazen Companies. It should be used for reference purposes only. Reasonable efforts have been made to ensure the accuracy of this map. The Chazen Companies expressly disclaims any responsibilities or liabilities from the use of this map for any purpose other than its intended use.

40130.00-old\20060426_40130_WappingersCreek_SoilComparison.mxd) 4/27/2006

Sources: Dutchess County Soil Survey. Surficial Geology Interperated from NYS Surficial Geology Map Note: Sample area taken from Town of Pine Plains

FIGURE 2 - DETAIL OF HYDROLOGIC SOIL GROUPS AND SURFICIAL GEOLOGY

STF Date:

April 2006 Scale:

1 in equals 5,000 ft Project #:

40130.00

Period of Record for Figure 1 Precipitation Isopleths Figure XX Precipitation and Streamflow

1951

1980

600

200

500

150

100

300

Average Annual Streamflow (cfs)

Annual Precipitation (inches)

400

200

50

100

2004

2002

2000

0 1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

Two Year Moving Average for Precipitation* 1978

1976

1974

1972

1970

1968

1966

1964

Wappinger Creek at Red Oaks Mill 1962

1960

1958

1956

Tenmile River 1954

1952

1950

1948

1946

1944

Fishkill Creek at Beacon 1942

1940

1938

1936

1934

1930

1928

1926

1924

1922

1932

Precip. at DC Airport

0

Time

* Moving Average Reflects Current and Prior Year. Dutchess County Office: 21 Fox St. Poughkeepsie, NY 12601 Phone: (845) 454-3980 Orange County Office: 356 Meadow Avenue, Newburgh, NY 12550 Capital District Office: 547 River Street, Troy, NY 12182 ENGINEERS/SURVEYORS PLANNERS ENVIRONMENTAL SCIENTISTS

Glens Falls Office: 110 Glen Street Glens Falls, NY 12801

Date:

FIGURE 3 - DUTCHESS COUNTY ANNUAL PRECIPITATION AND WATERSHED DISCHARGE Source data: Raw USGS stream flow data. Fishkill data collection discontinued in 1967. Notes: Graphic presentation identifies relationship between annual two year average precipitation to annual stream flow (consisting of runoff and baseflow).

April 2006

Scale:

Not to Scale

Project #:

40130.00

5/4/2006

Graphic Solution of New Jersey's Nitrate Dilution Model Using Selected Varables Listed Below and a Default Allowance of +/- 0.2 to 0.35 Acres Impervious Area per Septic System.

Governing Formula 0.5708

R=4.4186HM/(CA(1 - 0.179A )) R = Aquifer Recharge in inches per year A = Calculated minimum regional septic system density in acres per system C = Target groundwater nitrate concentration, 5.2 mg/L nitrate-nitrogen M = Per capita nitrate loading, 10 pounds nitrate-nitrogen per person/year H = 3 persons per system (home) Dutchess County Office: 21 Fox St. Poughkeepsie, NY 12601 Phone: (845) 454-3980 Orange County Office: 356 Meadow Avenue, Newburgh, NY 12550 Capital District Office: 547 River Street, Troy, NY 12182 ENGINEERS/SURVEYORS PLANNERS ENVIRONMENTAL SCIENTISTS


Date:

FIGURE 4 - NEW JERSEY SEPTIC SYSTEM DENSITY

April 2006

Scale:

Not to Scale

Source data: NJGS/NJDEP Technical guidance recharge-based Nitrate-Dilution Model for New Jersey, 8/02

Project #:

40130.00

Allowable Residential Densities for Homes on Septic Tanks as Limited by Water Quality Impacts

Dutchess County Office: 21 Fox St. Poughkeepsie, NY 12601 Phone: (845) 454-3980 Orange County Office: 356 Meadow Avenue, Newburgh, NY 12550

Date:

FIGURE 5-GERBER NITRATE DILUTION MODEL


Scale:

Not to Scale

Capital District Office: 547 River Street, Troy, NY 12182 ENGINEERS/SURVEYORS PLANNERS ENVIRONMENTAL SCIENTISTS

April 2006

Source data: Dutchess County Department of Planning Water Resources Study for Dutchess County, 1982

Project #:

40130.00

Beacon Washington

Poughkeepsie

Poughkeepsie Union Vale

La Grange

Wappingers Falls

East Fishkill

B

Dover

Beekman Pawling

Pawling

Fishkill

Fishkill

Legend

Waterbodies/Unclassified

C and C/D

A and A/D Soils

D

This map is a product of The Chazen Companies. It should be used for reference purposes only. Reasonable efforts have been made to ensure the accuracy of this map. The Chazen Companies expressly disclaims any responsibilities or liabilities from the use of this map for any purpose other than its intended use.

Source: United Stated Department of Agriculture Soil Survey of Dutchess County

Stanford

Engineers/Surveyors Planners Environmental Scientists GIS Consultants

Wappinger Northeast

North Country Office: 110 Glen Street Glens Falls, New York 12801 Phone: (518) 812-0513

Rhinebeck

Capital District Office: 547 River Street Troy, New York 12180 Phone: (518) 273-0055

Millerton

Orange County Office: 356 Meadow Avenue Newburgh, New York 12550 Phone: (845) 567-1133

Rhinebeck Pine Plains

FIGURE 6 - DUTCHESS COUNTY HYDROLOGIC SOIL GROUPS

40130.00

Project #:

Not To Scale

Scale:

April 2006

Date:

STF

Created by:

Red Hook

Dutchess County Office: 21 Fox Street Poughkeepsie, New York 12601 Phone: (845) 454-3980

Milan

CHAZEN ENGINEERING & LAND SURVEYING CO., P.C.

Tivoli

Clinton

Amenia

Hyde Park Millbrook

Pleasant Valley