Analysis of NYC Reservoir Safe Yield and Operations;

Opportunities for Meeting Multiple Objectives

New Jersey Department of Environmental Protection,

Division of Water Supply

December 2010

1
TABLE OF CONTENTS PAGE

Executive Summary …………………………………………………………… 5

A. Introduction ………………………………………………………..…… 9

B. Background ……………………………………………………..……… 10

Good Faith Agreement Operating Assumptions ……………………. 11

Flexible Flow Management Program (FFMP) ……………………. 12

Significant Historic References, Sources and Studies ……………………. 12

C. Critical Issues in Effectively Assessing Safe Yield of NYC

Reservoir Systems ……………………………………………………………... 13

Conjunctive Use of Delaware and Hudson Systems ……………………... 13

Nominal Safe Yields for the Delaware and Hudson Systems,

Past and Present……………………………………………………………. 17

Table 1: Compilation of safe yields of NYC's reservoirs obtained under

various design presumptions and operating criteria ………………………... 18

Table 2: Estimation of NYC’s Catskill and Croton Systems’ safe yields

at various reserve storages ...……………………………………………… 20

Demand and Demand Patterns ...…………………………………………….…. 20

NYC’s Historic Reservoir Operating Rules Criteria ……………………………… 23

OASIS Model and Daily Flow Model (DFM) ………………………………….… 23

.
Reservoir Operating Rule Curves ……………………………………………….… 25

1983’s Lower Basin Parties Assessment of the

Good Faith Agreement Operating Plan (Rev 1)…………………………………… 26

Effects of Safe Yield Determination on the ERQ and IERQ ……………………... 27

Reserve Storage Quantity ………………………………………………………… 29

2
D. NJDEP Preliminary Findings on Safe Yield Based Operations of Delaware
Reservoirs Compared to Over-Draft Operations under Current Good Faith
Agreement Assumptions. ………………………………………………………… 30

Safe Yield, Delaware Reservoirs …………………………………………………... 30

Example: Minimum Montague Flow Maintainable at

520 mgd Safe Yield Based Plan ……………………………………………………. 32

NYC Per Capita Need, Now and for Ultimate Build-Out

Population (Design Year 2110) …………………………………………………… 33

Use Exceeding a Sustainable Safe Yield ………………………………………….. 33

Estimated Safe Yield under Rev 1 and FFMP Over-Draft Operating Plans ……… 33

Typical NYC Water Year Ops, June 1st to May 31st, under Existing Rules ……. 34

Current Off-Peak Season Operations …………………………………………. 34

Example: Worst-Case Delaware System Operations …………………………. 35

Trial Estimate of Safe Yield of FFMP & Previous Operating Plans…………… 36

E. Other Projects …………………………………………………………………… 37

New Storage

Cannonsville Reservoir Increase ……………………………………… 37

F. E. Walter & Prompton Reservoir Projects…………………………. 38

Mongaup System Purchase of Storage Capacity...……………………. 38

New Aqueducts and Connection of Delaware-Catskill Aqueducts …………... 38

Related Issue: Hydroelectric Power ………………………………..………….. 39

F. Preliminary Supporting Example:

Graphs for Delaware System Operations with Over-Draft Compared to
Safe Yield-Based Operations ………………………………………………. 39

Figure 1: Comparing a Safe Yield Based Diversion of 520 mgd, for a Total Conjunctive
Safe Yield of 620 mgd, to an Estimated 695 mgd Safe Yield
under the FFMP …………………………………………………….......... 40

3
 Figure 1a: A Sample of an Optimized Safe Yield-Based
Diversion Pattern …………………………………………………………. 42

Figure 2 Ten-Year Record of Rondout Daily Diversions, 1998-2007….. 43

Figure 3; Ten-Year Record of Rondout Daily Diversions, 1998-2007

w/Model Diversion Curve Superimposed.....…………………………… 44

G. Conclusion …………………………………………………………………… 45

Sources/References/Bibliography ……………………………………………… 46

Plates of NYC’s Delaware Basin and Hudson Basin Water Supply System…….. 54

4
Executive Summary

The Delaware River Basin (Basin) provides water supply for millions of people and is an integral
component of the region’s economy. The Basin’s water resources also support diverse natural
communities and provide recreational opportunities which enhance the quality of life. A critical
component of these water resources is the operation of reservoirs throughout the Basin.

The Parties to the 1954 Supreme Court Decree (Decree), New York (NYS), Pennsylvania (PA),
New Jersey (NJ), Delaware (DE), and New York City (NYC), as well as the Delaware River
Basin Commission (DRBC), consisting of NYS, PA, NJ, DE and the Federal Government, and
the United States Geological Survey (USGS), are responsible for managing and protecting the
water resources in the Delaware Basin so that a wide range of human and natural resource
objectives are met. NYC operates the three major Delaware Basin reservoirs in conjunction with
its Hudson Basin Reservoir.

The Decree imposed diversion and release requirements on the Decree Parties for the purpose of
equitably apportioning the waters of the Delaware Basin in relation to NYC’s total combined
Delaware and Hudson Reservoir Systems’ safe yields. The Decree enjoined NYC from diverting
800 mgd out of the Basin, the 1954 estimate of safe yield of the Delaware Basin reservoirs
based on the drought of the 1930’s, unless a minimum basic rate of flow of 1,750 cfs was
maintained on the Delaware River at Montague, New Jersey. The Decree also required
approximately 83 percent of the volumetric difference between NYC’s total combined systems’
unused safe yield and the expected coming year’s demand be made available to the down-basin
states. This Excess Release Quantity (ERQ) condition is a critical component of the principle of
equitable apportionment and was imposed to prevent NYC from under-utilizing its Hudson
Reservoir Systems and over-drafting or exceeding the safe yield of the Delaware reservoirs,
while its demand was less than its total combined safe yield.

Under the commonly accepted definition of safe yield, a reservoir system will be drawn down to
its minimum usable water level when the specified design drought occurs. However, operating
based on safe yield allows for reservoir rule curves to be established that would impose demand
or other operational controls on the system so that the reservoirs are not intentionally taken to
their minimum usable water levels. Although the nomenclature, such as “safe yield,” “reliable
yield,” “dependable yield” or “sustainable yield” may vary slightly from state to state, in general
the following definition is typical:
“The maximum amount of water that can be withdrawn from a source without depleting
that source during a prolonged drought. Usually the drought of record or some other drought
with a stipulated return period of 25, 50 or 100 years is used in determining the safe yield.”
(Taken from, “Water Resources Management Strategy, Delaware-Lower Hudson Region,”
prepared for NYS Water Resources Planning Council by NYSDEC, NYS Dept of Health &
Hazen & Sawyer P.C. Consultant, January, 1989.)

The New Jersey Water Supply Management Act (NJSA 58:1A) defines safe yield as:
"that maintainable yield of water from a surface or ground water source or sources
which is available continuously during projected future conditions, including a repetition of the

5
most severe drought of record, without creating undesirable effects, as determined by the
department."

Safe yield can usually be increased by optimizing conjunctive use of storage between reservoirs
and reservoir systems to minimize wasted spill, more efficiently make releases, credibly
represent current and projected seasonal demand and demand patterns, and allow for efficient
alternating use between reservoirs. The engineering and design of an effective safe yield based
operating plan of any reservoir entails applying appropriate physical laws and the operational
characteristics that aptly represent the water supply systems’ flow regimes, uses, demands, needs
and infrastructure limitations. The resultant quantification of safe yield establishes the volume of
water available under drought of record conditions and identifies how risks, benefits and
responsibilities are assigned if the safe yield is exceeded.

The Decree also limited New Jersey to no more than an average of 100 mgd diversion out of the
Delaware Basin via the Delaware and Raritan (D&R) Canal unless offset by storage. This
apportionment was run-of-river and, as such, dependent on NYC’s maintenance of an adequate
minimum Montague flow target. In 1954, the 1,750 cfs minimum Montague flow target was
designed to protect the lower Delaware Basin’s dependent in-stream uses, among which were
public water supplies. Another important facet of such flow maintenance is the salinity of the
estuary and its affect on run-of-river dependent estuary diversions, such as the City of
Philadelphia’s at Torresdale and New Jersey American Water’s at Delran.

After the 1960’s drought established a new “drought of record,” it was apparent that the
hydrological assumptions used in forming the Decree were no longer valid because less water
was available to achieve the conditions imposed by the Decree. As a result, the Decree Parties
entered a voluntary accord, the 1983 “Good Faith Agreement” (Agreement). However, the
Agreement operating plan, the over-draft program of Revision 1 (Rev 1) of Docket D-77-20-CP
adopted in 1983 and its subsequent revisions, Revs 2 through 9, significantly departed from the
original safe yield-based operating assumptions and the standards of hydrologic, hydraulic and
water supply engineering applied in the Decree, originally established in the Special Master’s
Report, adopted as Condition I of that Decree. It appears that very little of the crucial information
needed to make a fully informed decision was considered in 1983. Even if the information was
considered, the criteria used in 1983 are now obsolete and, accordingly, NYC's safe yields and
actual operating alternatives under both the Rev 1 and under the current Flexible Flow
Management Plan (FFMP) operating criteria are not assessed accurately or equitably. For
example, while NYC and NYS have indicated that under the FFMP the safe yield of the
Delaware reservoirs is only 480 mgd and that they have reduced their over-draft of that safe yield
from 800 mgd to 765 mgd to allow for increased fishery releases, in fact, the effective,
conjunctive use based safe yield of the Delaware reservoirs has increased due to the significant
reduction in the Montague flow target, to as low as 1,100 cfs, in Rev 1 and the FFMP.. This
increase has been preliminarily estimated by the NJDEP to be from an actual effective safe yield
of 570 mgd under Rev 1 to about 590 mgd under the FFMP. Further, the fact that the OASIS
Model is based on an unrealistic and unsustainable flat average draft of 800 mgd (or the 765 mgd
used temporarily in the FFMP) means that, not only are the theoretical "drought days" over-
predicted, but that the flood risks are under-estimated.

6
The Good Faith Agreement allows NYC to divert 800 mgd out of basin and requires NYC to
maintain the Montague, NJ, flow target of 1,750 cfs only when it diverts at 800 mgd instead of
requiring it as a design minimum at all times as per the Decree. Now, when NYC’s diversion is
required to be curtailed to as low as 520 mgd, the Montague flow target is reduced to as low as
1,100 cfs. The effects of these changes to safe yield have not been properly addressed. The
Agreement did not appropriately assess the sustainable, effective safe yield of NYC’s water
supply system as affected by conjunctive use of its entire reservoir system. While the diversion
curtailments resulted in little impact to NYC’s water supply (since NYC’s actual demands and
needs are significantly lower than the modeled amounts), the release curtailments seriously
impact the Delaware Basin. In essence, the Agreement placed the risks and burden of NYC’s
reduced yield on the lower Basin by reducing minimum flow (Montague flow target), stressing
lower basin reservoirs to make up that flow, effectively reducing New Jersey’s allocable
diversion from D&R Canal from 100 mgd to 65 mgd, while increasing the risk of elevated
salinity at the Philadelphia Water Department’s and New Jersey American’s Delaware River
intakes at Torresdale and Delran, respectively.

In September 2007, the Decree Parties agreed to the temporary FFMP that modified the
operating plan developed during the 1983 Good Faith Agreement negotiations. To a limited
degree, the FFMP enables the Decree Parties to adjust NYC releases from Basin reservoirs in
response to varying conditions such as low flows or when high storage enables spill relief via
additional discharges. However, the agreement continues to support decisions that are based on
operational scenarios that deviate from the safe yield principle and are derived from modeling of
conditions that are unrealistic or no longer exist. It also results in decisions that unnecessarily
default to an unsustainable 800 mgd water supply need for NYC over Delaware Basin needs
Because of these deficiencies, New Jersey predicated its approval of the FFMP on a
comprehensive reassessment of the safe yield and operations of the NYC reservoir system.

The New Jersey Department of Environmental Protection (NJDEP) performed a preliminary

analysis of NYC's Delaware and Hudson basin reservoir systems to evaluate how reservoir
operations since the 1983 Good Faith Agreement might optimize beneficial operating
alternatives when using the same accepted standards and principles appropriately used for
establishing the safe yield based operating conditions in the 1954 Decree. The NJDEP's
preliminary analysis was also designed to determine if a comprehensive reassessment of
reservoir operations would identify opportunities whereby multiple objectives in the Delaware
Basin can be achieved. The NJDEP's preliminary analysis included actual demand scenarios,
both current and projected, considered the hydrological and hydraulic limitations of the NYC
water supply system, reserve storage impacts, and conjunctive use of NYC’s Delaware and
Hudson River Basins’ reservoir systems.

The NJDEP utilized USGS inflow records, the Delaware River Master Reports, the OASIS flow
model data and records of safe yield reports and various water supply reports from 1944 to 2009
to apply the same universally accepted safe yield based engineering principles and design
methodologies that form the basis for the Decree’s equitable apportionment. The NJDEP
provides an example of a sustainable diversion for an annual average demand of 520 mgd. This
diversion, equivalent to the current minimum draft on the Delaware reservoirs during drought
emergency, enables NYC to meet peak summer diversion needs at or near 800 mgd during a

7
repeat of the drought of record and meet intermittent off-season peak demands as well as
intermittent peak drafts for sustaining water quality. Further, considering continued
improvements in water conservation, the assessment indicates that the 520 mgd is likely more
than NYC needs, now or at its ultimate future build-out population of 11 to 12 million people.

As an example of a safe yield based operating scenario, if NYC was restricted to an average
annual diversion of 520 mgd, it may maintain a minimum passing flow at Montague of
approximately 1,650 cubic feet per second (cfs), at all times, through a repeat of the drought
of record (550 cfs above the minimum of 1,100 cfs established under the current Flexible Flow
Management Plan). Further, this may allow for developing alternate Montague flow objectives
that are more beneficial such as lowering the yearly minimum to 1,500 cfs to create a permanent
storage bank of nearly 55 BG for fisheries enhancement. Also, because more releases, more
evenly distributed between the Delaware basin reservoirs would be promoted under a safe yield-
based operating plan, the probability of larger void spaces more frequently and for longer
duration should increase the probability of flood mitigation. “Yo-yoing” releases; that is, severe
release fluctuations to the West Branch Delaware River because of the nearly exclusive use of
Cannonsville Reservoir to maintain Montague flows should be reduced because a safe yield
based operating plan should promote even distribution of all three Delaware reservoirs’
diversions and releases so as to promote equal probability of refill.

The NJDEP analysis also finds that NYC under-utilizes its Hudson Systems’ reservoirs because
of NYC’s deference to over-drafting the safe yield of the Delaware Basin reservoirs beginning
with adoption of the 1983 Good Faith Agreement. Since that time, nothing prevents NYC from
over-drafting the Delaware Basin reservoirs even when its Hudson Basin reservoirs are full or
spilling. By departing from the fundamental principle of equitable use between the systems that
safe yield based programs entail, unexpected impacts have occurred consisting of unnecessary
drought emergencies, unreliable fisheries protection, environmental degradation, and increased
flood risk.

To withstand the rapid depletion of the reservoirs caused by seasonal over-drafting, NYC's
Delaware reservoirs are often kept at storage levels much higher and for longer periods than
would be needed to sustain a reliable and usable safe yield to NYC in a repeat of the drought of
record, especially since the City's change from a high flat rate to a lower seasonal demand
pattern. The over-drafts result in unnecessarily and pre-maturely triggering the severe reduction
or complete cessation of fishery releases and reduced Montague flows. The bulk of the 5,600-
5,900 “drought days” occurring under Rev 1 and the FFMP are not necessary to provide a usable
and sustainable supply to the City but are only an artifact of unrealistic modeling and operating
assumptions. The unrealistic overdraft model, which disregards the conjunctive effect of the
West Branch and Rondout reservoirs, masks periods when the Delaware reservoirs are fuller than
assumed. This reduces the effective modeling of alternative realistic, safe yield based operations
and increases flood risk over and above what is needed to provide usable, sustainable safe yield
to NYC.

Moreover, the NJDEP's preliminary study finds that future addition of new transmission tunnels
and incorporation of computer-guided operations (OST) will increase the current ability to over-

8
draft the reservoirs unless safe yield based conditions are applied, even though no additional safe
yield will actually be derived from such changes,

An accurate, flexible and optimized model that includes, at a minimum, the conjunctive effect of
the Hudson System on the Delaware Basin reservoirs’ effective safe yield should be constructed
so that future reservoir management decisions are made with a clear understanding of the
impacts on Delaware Basin water supply, the regional economy, natural resources, conjunctive
use effects, flooding or shared risk between the Parties. A complete combined total conjunctive
use safe yield assessment that optimizes the water resources should be a prerequisite to any
consideration by the Delaware Basin Parties for continuation of reduced Montague flows.

The findings in this paper are intended to serve as a rationale and basis for a comprehensive,
objective and transparent reassessment of the effective safe yield that includes the conjunctive
use of the entire NYC reservoir system. It is not intended to be prescriptive or establish absolute
values. The recommended comprehensive re-assessment would be the avenue for establishing
firm operating parameters. Such an analysis would result in a more realistic and optimizing
operations that realistically respond to and balance the multiple-objectives of water supply,
fisheries and flood mitigation.

A. Introduction

The Delaware River Basin (Basin) provides water supply for millions of people and is an integral
component of the region’s economy. The Basin’s water resources also support diverse natural
communities and provide recreational opportunities which enhance the quality of life. A critical
component of these water resources is the operation of reservoirs throughout the Basin. NYC
operates the three major Delaware Basin reservoirs of Neversink, Pepacton, and Cannonsville in
conjunction with the Hudson Basin’s Rondout Reservoir. Together these reservoirs are known
as the Delaware System. The Parties to the 1954 Decree, New York State, Pennsylvania, New
Jersey, Delaware, and New York City as well as the United States Geological Survey are
responsible for managing and protecting this resource so that it meets a wide range of human and
natural resource objectives. The DRBC, consisting of NYS, PA, NJ, DE and the Federal
Government, was formed in 1961, in part, to help preserve the interstate comity needed to fulfill
the Decree's operating criteria, but since the more severe drought of 1961-66, the DRBC has
facilitated negotiations between the Decree Parties to respond to the more limited water
resources than the 1954 Decree anticipated. However, these negotiations have excluded the
impacts of NYC’s Hudson basin reservoir operations on the water supply, economy and ecology
of the Delaware Basin. Today, many of the decisions made regarding reservoir operations and
drought management since the 1961-66 drought have been based on assumptions that, upon
further review, are invalid or no longer applicable and should be revisited using the appropriate
standards for widely accepted safe yield-based operating criteria.

The NJDEP performed a preliminary evaluation of the safe yields and operations of the NYC
reservoir system based on both current and projected demand conditions in an effort to help
identify water supply management strategies that enhance our ability to meet multiple objectives
in the Delaware Basin. These objectives include: improved water supply management that
meets the water supply needs of NYC and the Delaware Basin States, Decree Party equity,
natural resource and environmental protection, enhanced flood mitigation, and improved lower

9
basin water quality. This report describes the NJDEP evaluation and provides findings to assist
in making those water supply decisions that are of extreme importance to the Decree Parties and
the millions of people in the region who depend on the Basin to meet their water supply,
economic, ecological, fishery, flood mitigation, and recreational needs.

Preliminary results indicate that a full reassessment of NYC’s Delaware and Hudson Basin
reservoir systems operations and safe yield has the potential to substantially improve our ability
to meet multiple objectives in the basin; e.g. improve flows while meeting water supply needs.
An accurate and flexible model that includes, at a minimum, the effects to the Delaware
reservoirs’ conjunctive use of Rondout and West Branch reservoirs to derive an effective,
sustainable safe yield should be constructed so that future reservoir management decisions are
made with a clear understanding of the impacts on lower basin water supply, the regional
economy, natural resources, flooding and shared risk between the Parties. A complete combined
total conjunctive use safe yield assessment that optimizes the conjunctive water resources of both
basins, including alternate reserve storage scenarios should be a prerequisite to any consideration
by the lower basin Parties of continuing reduced Montague flow objectives.

B. Background

In 1929, NJ brought an action in the U.S. Supreme Court seeking to restrain NYC and NYS from
diverting waters from the Delaware River and its tributaries. The Commonwealth of
Pennsylvania and State of Delaware joined the suit as intervenors. On May 25, 1931, the Court
granted NYC the right to divert 440 mgd from the then proposed Neversink and Pepacton
reservoirs.

In 1952, NYC petitioned the Court to amend the 1931 Decree to allow an increase of diversion to
800 mgd after construction of a third Delaware Basin reservoir, Cannonsville Reservoir. The
matter was referred to a Special Master, who recommended to the Court via the Special Master's
Report, filed May 27, 1954, the operating conditions adopted in the Decree on June 7, 1954.
(Entered: 347 U.S. 995, 74 S. Ct. 842, 98 L. Ed. 1127)

Most surface water systems in the United States operate on the basis of safe yield as measured by
the most severe drought of record or other similar statistical criterion. A safe yield-based
operating plan establishes the volume of water available under drought of record conditions, and
clarifies how risks, benefits, and responsibilities are assigned when such a yield might be
exceeded during the onset of a drought period potentially worse than the design drought.

The Decree established the Decree Party rights and responsibilities as to the management of the
shared waters of the Basin and imposed diversion and release requirements for the purpose of
equitably apportioning the waters of the Delaware Basin in relation to NYC’s Delaware and
Hudson Reservoir Systems' total safe yields. The Decree enjoined NYC from diverting 800 mgd
out of the Delaware Basin, the safe yield of the Delaware Basin reservoirs based upon the
drought of the 1930’s, unless it designed its use to maintain a minimum downstream flow of
1,750 cfs on the Delaware River at Montague, New Jersey and made available each year the
Excess Release Quantity (ERQ). In general, the ERQ is 83% of the volumetric difference
between NYC’s total combined systems’ unused safe yield and the upcoming year’s expected

10
demand. The ERQ condition is a critical component of the principle of equitable apportionment
and was imposed to prevent NYC from under-utilizing its Hudson Reservoir Systems and over-
drafting the safe yield of the Delaware system while its demand was less than its total combined
safe yield. (See “Report of the Special Master,” 1954 Supreme Court Decree, pp.73-79; and
“Comprehensive Public Water Supply Study for the City of New York and County of
Westchester,” NYS Dept. of Health, pg. 163 "Supreme Court of the U.S.1952-1954 Case, Notes
of Chief Engineer Kennison on provisions of Decree June 7,1954 and reports and memoranda by
Witnesses not Called to Testify," pp. 3-18, ) Although NYC’s yearly use has been declining,
ERQ computations have often been erroneously based on the assumption that NYC’s demand
would reverse this decline and rise, in one year, to the highest demand ever recorded (1980). The
ERQ is supposed to return a major portion of NYC’s total combined systems’ unused safe yield
relative to a realistically projected demand. However, even if the projected demand is properly
estimated, if the total combined safe yield is under-estimated, the ERQ is under-estimated and
less water for releases is available for lower basin flow needs.

The Decree, (Condition V. B. 3.) also limited New Jersey to no more than an average of 100 mgd
diversion out of the Delaware Basin via the Delaware and Raritan Canal without compensating
releases, regardless of whether NJ constructs storage. The 100 mgd apportionment was run-of-
river dependent and, as such, is dependent on NYC’s maintenance of an adequate minimum
Montague flow target. In 1954, the 1,750 cfs minimum Montague flow target was designed to
protect the lower Delaware Basin’s dependent in-stream uses, among which were public water
supplies. An important facet of such flow maintenance is also the salinity of the estuary and its
affect on run-of-river dependent estuary diversions, such as the City of Philadelphia’s at
Torresdale and New Jersey American Water’s at Delran.

Good Faith Agreement Operating Assumptions

After the 1960’s drought established a new “drought of record”, it was apparent that the
hydrological assumptions used in forming the Decree were no longer valid: Less water is
available to achieve the conditions of the Decree. As a result, the Decree Parties approved the
1983 “Good Faith Agreement.”(Agreement) The Agreement operating plan, Revision 1 (Rev 1)
of Docket D-77-20-CP adopted in 1983 and its subsequent revisions, Revs 2 through 9,
significantly departed from the original safe yield-based operating assumptions and the standards
of hydrologic, hydraulic and water supply engineering applied in the Decree as demonstrated
within the Special Master’s Report, adopted as Condition I of that Decree. The Agreement
protocol modeled a continuous draft of 800 mgd for NYC, a rate that was beyond the
reservoirs’1960’s drought safe yield and is currently unnecessary for meeting existing and
projected demands. The Agreement did not quantify the sustainable, effective safe yield of
NYC’s water supply system. The rule curves developed as part of the Agreement are intended to
trigger diversion curtailments and Montague flow target reductions that are overly protective of
NYC’s current or projected ultimate build-out demands. While the diversion curtailments
resulted in little impact to NYC’s water supply (since NYC’s actual demands and needs are
significantly lower than modeled), the release curtailments seriously impact the lower Basin. In
essence, the Agreement placed the risks and burden of NYC’s reduced yield on the lower Basin
by reducing minimum flow (Montague flow target), stressing lower basin reservoirs to make up
that flow, reducing New Jersey’s ability to allocate 100 mgd to 65 mgd, while potentially

11
increasing the risk of elevated salinity at the Philadelphia Water Department’s and New Jersey
American’s Delaware River intakes at Torresdale and Delran, respectively.

The Agreement adopted a policy allowing NYC to attempt a continuous draft of the Delaware
reservoirs’ safe yield if Montague were to be maintained at 1,750 cfs; this over-draft rate was
800 mgd, the rate designed for the safe yield of the drought of the 1930’s that is now not
sustainable under the new design drought, the drought of the 1960’s. Operating rule curves were
developed to set storage zones to trigger curtailed diversions and reduced Montague flow targets
in response to the presumption of prior over-drafting. As a result, the modeled storage levels
appear lower than they would if NYC were to autonomously design its annual average diversion
to be less than 800 mgd, but significantly greater than 480 mgd. To obtain more than a 480 mgd
safe yield, the operating rule curves are set to reduce Montague flow targets and NYC diversions
by the end of the peak season, after which NYC’s demand is well below 800 mgd and recharge
of storage begins. During this recharge period, NYC may often be nominally “curtailed” to 520
mgd even though its demand is much less so that recharging the reservoirs is assured at the same
time as NYC can meet its water supply needs with virtually no conjunctive system stress.

Flexible Flow Management Plan (FFMP)

In September 2007, the Decree Parties agreed to a temporary FFMP that modified the existing
operating plan developed during the 1983 Good Faith Agreement negotiations. The FFMP
enables the Decree Parties to adjust NYC releases from Basin reservoirs in response to varying
conditions such as low flows, or conversely impending flood conditions. The NJDEP performed
a preliminary analysis of the operating criteria of the FFMP and the Operational Analysis and
Simulation of Integrated Systems (OASIS) model assumptions (the OASIS Model was updated
from the original 1981 ACE Daily Flow Model). In the analysis NJDEP found that the FFMP
and the 1983 Good Faith Agreement operating plans result in an inefficient, inequitable use of
the Basin waters that over-estimates drought days, reduces fishery maintenance flexibility, and
increase flood risk. Flood risk is increased not because the reservoirs cannot be operated to
significantly mitigate flooding but because they are not operated in accordance with similar
accepted water supply safe yield based standards that the Decree apportionments and conditions
were predicated upon.

The FFMP provides a program by which reservoir operations can be minimally adjusted in
reaction to specific conditions. However, such reactive measures are arbitrary because the
accepted fundamentals of safe yield for determining reliable and equitable operations of the
NYC reservoirs have never been applied. While the Decree Parties, federal government and
DRBC have attempted to respond to changing conditions and needs in the basin over the years,
many of these actions have been taken without the full knowledge, understanding or application
of the hydrological and hydraulic limitations and opportunities inherent to the system, changing
demand patterns, equitable apportionment, effects of conjunctive use, and other issues paramount
in the Decree. Because of these deficiencies, New Jersey predicated its approval of the FFMP on
a comprehensive reassessment. Although, a post FFMP approval Work Group put some effort
toward development of a scope of work for a comprehensive reassessment study, no consensus
was achieved; the temporary FFMP expires in 2011.

12
A safe yield-based operating plan that includes Rondout and West Branch reservoirs in the
Hudson Basin, as well as the analysis of conjunctive use with NYC’s Hudson reservoirs to
equitably optimize multiple uses and reapportion the waters between the basins consistent with
the Decree should surpass the relatively nominal and ineffective short-term fixes culled from the
inaccurate assumptions of the FFMP’s over-draft policy, such as the “spill mitigation” program,

Significant Historic References, Sources and Studies

The NJDEP reviewed the assumptions made under various safe yield-based operating analyses,
reports and studies prior to the Drought of the 1960’s and compared them to the assumptions
used after the Drought of the 1960’s. Other studies, documents, related data were also reviewed,
such as diversion and use reports, flood reports, wave action, water quality reports and models,
operations programs (such as Rev 1, the FFMP and the OST), multiple use alternatives, and other
related reports on the Delaware and Hudson Systems’ and similar systems.
Key documents used to support this analysis were: “Special Master’s Report,” adopted as
Condition I in the Decree; “Task Group Report for Docket D-77-20, Appraisal of Upper Basin
Reservoir Systems, Drought Emergency Criteria and Conservation Measures,” March, 1979;
“Notes of Chief Engineer Kennison on Provisions of the Decree, June 7, 1954, Reports and
Memoranda by Witnesses Not Called to Testify,” by Karl R. Kennison, 1955; “Probability
Analysis of Allowable Safe Yield of New York City Reservoirs in the Delaware Basin,” by
Clayton H. Harrison, USGS, 1968; “Reports of the Delaware River Master,” 1955 to 2008;
“Interim Report of the Technical Responses to “Good Faith” Recommendations #3 and #4,” by
the Flow Management Technical Advisory Committee, June 10, 1983; “Study of the Water
Demands on New York City System-Final Report,” by Hazen & Sawyer P.C. 1989;
“Northeastern United States Water Supply Study: Hydraulic Analysis of the New York City
Water Supply System,” prepared for the US Army Corps of Engineers Northeastern United
States Water Supply Study by Quirk, Lawler & Matusky, Engineers, 1974; and “Comprehensive
Public Water Supply Study for the City of New York and County of Westchester,” by Metcalf &
Eddy, Hazen & Sawyer, and Malcolm Pirney, Engineers, sponsored by the New York State
Department of Health, 1967.

A list of most of the essential sources used to prepare the NJDEP evaluation is at the end of this
paper.

C. Critical Issues in Effectively Assessing Safe Yield of NYC Reservoir Systems

Conjunctive Use of Delaware and Hudson Systems

Since the Drought of the 1960’s, all analyses of NYC reservoir operations have considered the
safe yields of the Delaware and Hudson basin reservoirs separately. This may have been because
after the drought, all through the Good Faith negotiations and even today, it had been considered
that, since the Hudson Basin reservoirs were not directly under the DRBC’s jurisdiction, the
Decree Parties could no longer apply the rights they obtained under the Decree.. The NJDEP
believes this is an invalid assumption that has effectively prevented an appropriate and equitable
safe yield assessment from ever being performed. The Decree, which pre-dated the creation of

13
the DRBC by 7 years, comprehensively addressed inter-state and inter-basin water issues,
including issues of combined total safe yield and equitable apportionment. The actual demands,
diversion options, reserve storage options, release options and equitable use of both the Hudson
and Delaware systems need to be transparently examined and open to periodic verification and
reassessment, especially if NYC storage is to be subsidized by Montague flow reductions that
severely impact the Delaware Basin.

Isolating an individual reservoir or a sub-group of reservoirs for safe yield assessment in a multi-
reservoir system will typically result in under-estimating the effective safe yield of that single
reservoir or sub-group. That is, the effect of a reservoir in the total safe yield calculation may be
greater than the safe yield of any reservoir alone. The combined consideration of the two
(Hudson and Delaware) reservoir systems was included in the Special Masters Report (May 27,
1954) and the Decree (see Decree III. B. 1. (c)). The Decree established that the combined total
safe yield for “all of its sources obtainable without pumping” as the criteria for setting the Excess
Release Quantity (ERQ). The ERQ was devised in the Decree to ensure equitable use between
Hudson and Delaware reservoirs and prevent over-drafting the Delaware reservoirs’ safe yields.
At that time, the safe yields of the Hudson reservoirs were conservatively under-estimated by
excluding 25 percent reserve storage from the Hudson Basin reservoirs’ yield analyses,
excepting the 17 BG Schoharie Reservoir; this amounted to about 66 billion gallons (BG). In
1954, the disparity between reserve storage in the Delaware, where no reserve storage is
assigned, and the Hudson reservoirs appears to have been accepted because the Hudson System
criteria pre-existed the Delaware reservoirs and had been designed in the early 20th Century
without an extensive flow history to assure reliability. Further 1,750 cfs, the minimum flow at
Montague agreed upon when negotiating the terms of the Decree, was considered adequate by
New Jersey and the other Parties.. The drought of the 1960’s changed this basic criterion for both
systems, yet even 15 years after the 1960’s drought and nearly 60 years after the design of the
Catskill System, the Good Faith negotiations disregarded the fact that the safe yield
reassessments of the Hudson reservoirs were still conducted with 25 percent reserve storage in
the Hudson reservoirs. Thus the intent of designing reserve storage, that is, to provide flexibility
in safe yield in the event of a drought worse than the design drought to enable revising
operations accordingly, was never realized insofar as the Delaware Basin is concerned. That is
not to say NYC and NYS do not independently operate to optimize their safe yield by reducing
reserve storage, only that the Delaware Basin derives no benefit if such action is taken.
Continuing to withhold 25 percent reserve storage in calculating the Hudson reservoirs’ safe
yields nominally reduces the safe yields of those reservoirs and results in passing most of the
drought risk onto the Delaware Basin by reducing Montague flows and over-drafting the
Delaware reservoirs.

Isolating the Delaware reservoirs from the total system when assessing system safe yield likely
results in a total yield less than that obtained by conjunctive use analysis, thus also resulting in
under-estimating the ERQ. The three Delaware Basin reservoirs are simulated under a “worst
case” over-draft scenario, without any consideration of operating alternatives, demand effects or
conjunctive use impacts. Under such assumptions, the safe yield of the Delaware reservoirs is
only 480 mgd. However, it is not hydraulically or hydrologically realistic to assume that during
droughts the Delaware Basin reservoirs are not affected by total system use and design, including

14
releases, reserve storage, critical duration, demand patterns, distribution reservoir capacity and
conjunctive use optimization.

The 1983 Good Faith Agreement did not quantify the sustainable, effective safe yield of NYC’s
water supply system relative to the tiered drafts and Montague flow targets, and there was no
attempt to develop a conjunctive model of NYC’s total system yields; only two parallel studies
partitioning the Delaware reservoirs from NYC’s total system were done, both between 1967 and
1968. The 1968 study was produced by the USGS under contract to the DRBC and the 1967
study was sponsored by the New York State Department of Health (NYSDH) and produced by
Metcalf and Eddy, Malcolm Pirnie and Hazen and Sawyer, consultant engineers. Assuming a
minimum Montague flow of 1,750 cfs is being maintained, the USGS study produced a safe
yield of 480 mgd and the NYSDH sponsored study produced a safe yield of 510 mgd, the
difference due to slight methodological variations. Both studies retained the minimum
conservation releases, equivalent to about 41 cfs (26.5 mgd), on average that had been applied at
the request of NYS after the 1954 Decree.

The 1979 Task Group Report for Docket D-77-20 listed as one of its goals (under Section 2,
Authorization for Investigation, Findings, A. Drought Emergency Criteria and Conservation
Measures, 3.), “The Task Group should explore the operations of the total New York City system
to assure under emergency conditions the maximum possible conservation of waters stored in the
Delaware System reservoirs and balanced use of supplies available throughout the City system.”
This was never done and the safe yields obtained by partitioning the Delaware reservoirs was
assumed to be the minimum yield possible when maintaining Montague flow at 1,750 cfs. The
modeling needed to resolve issues of realistic, equitable and optimal reapportionment of the safe
yields was never performed and the 1983 Good Faith Agreement was finalized without the
benefit of an accurate yield analysis available to the lower basin Parties.

Because of the conjunctive use of the Hudson reservoirs and because curtailments from storage
are offset by reductions of Montague flow targets, the diversion curtailments provide little
impact to NYC’s water supply needs, while the Montague flow reductions help restore NYC's
storage and seriously impact the lower Basin. Further, because NYC operates to retain more
storage than it can use, its reservoirs are prone to be near-full, full or surcharged more frequently
and for longer durations, which unnecessarily increases the likelihood of high spills during
precipitation events. This phenomenon may “telegraph” to the Catskill and Croton Systems of
the Hudson Basin because, in order to over-draft the Delaware System, NYC must under-utilize
the Catskill and Croton Systems. With Hudson Basin reservoirs’ reserve storage at 25 %, these
reservoirs should also be prone to be near-full, full or surcharged more frequently and for longer
duration because the effect of NYC’s normalized demand pattern is to reduce its needed yields
and increase the probability of unusable storage held in both basins.

The partitioning of the Delaware Basin for yield analysis may have been facilitated by the
erroneous assumption that, because the Hudson Basin reservoirs are not subject to DRBC
jurisdiction, the lower basin Parties could only analyze the Delaware Basin reservoirs. In fact, no
permission is needed to do an effective conjunctive use analysis and the Task Group report of
1979 clearly shows the intent to do so. While the information needed in the late 1970’s may have
been difficult to obtain without NYC’s cooperation, today all data needed are in the public

15
domain. In any event, the necessary realistic total system safe yield assessment was never
performed by or provided to the lower basin. The Special Master’s Report documents that the
engineers representing the Parties during the Decree negotiations considered the inter-basin
water issues, including issues of combined total safe yield, system capacity, total build-out
population and equitable apportionment. After the drought of the 1960’s, realistic needs and
demands, diversion options, reserve storage options, release options and equitable use of both the
Hudson and Delaware systems needed to be transparently examined and open to periodic
verification and reassessment in order to make the Decree meaningful. If the DRBC were to
appropriately develop a conjunctive use model it could regulate NYC’s Delaware Basin
reservoirs so that they had an appropriate conjunctive effect on the Hudson Basin reservoirs
instead of NYC demanding the impunity to operate so as to transfer an inequitable amount of
risk to the Delaware Basin.

The USGS and the ACE, each of which have delivered to the DRBC past partitioned analyses
and models, could also assess the conjunctive safe yield. The prior partitioned analyses have
thwarted the lower basin states understanding of how conjunctive use works and how NYC has
obtained a greater safe yield than it has acknowledged. NYC experiences virtually no stress on
meeting its current demands and most likely will easily meet its ultimate demands, shown later in
this paper. Meanwhile, the lower basin is severely impacted in adjusting to the reduced
Montague flows, reduced lower basin storage, reduced lower basin water supply, economic
instability and rising salinity, fishery losses and the collateral effect of increasing flood risks as
NYC’s demand continues to stabilize.

In addition to minimizing the calculation of the safe yields, partitioning the two basins obscures
the uneven use between the Delaware and Hudson basins. Until recently, NYC claimed that it
used the Delaware and Hudson Basin reservoirs equitably, since they were used approximately
equally; that the Delaware reservoirs are actually used slightly less. However, as is indicated in
the Special Masters Report, “…equal is not necessarily equitable…” in water resources
apportionment. The safe yield NYC used in the Good Faith Agreement negotiations for its
combined total Hudson reservoirs (includes Rondout) is 810 mgd, which excludes about 65 to 66
billion gallons of reserve storage in the calculation and is therefore under-estimated. Although
recent River Master Reports indicate NYC uses a combined total of less than 1,200 mgd, for
illustrative purposes, 1,250 mgd will be used to demonstrate the inequity of use between the
basins relative to safe yield. According to NYC’s published reports, about 55 % of its use comes
from the Hudson reservoirs and 45 % from the Delaware reservoirs. This means currently about
0.55*1,250= 687 mgd comes from the Hudson reservoirs and 0.45*1,250=563 mgd is derived
from the Delaware Basin reservoirs. However, 687 mgd is 85 % of the Hudson reservoirs’ total
safe yield of 810 mgd, whereas 563 mgd is 117 % of the 480 mgd used in calculating the
FFMP’s IERQ and the safe yield that would assure meeting Montague flows. Thus NYC
maximizes the potential for Montague flow reductions and minimizes stressing the Catskill and
Croton Systems. There is then, an inherent inequity in the over-draft operating plan since NYC
relies on greater depletion of the safe yield of the Delaware reservoirs while under-utilizing the
Hudson system reservoirs .Additionally, in the Delaware Basin reservoirs, zero reserve storage
is assumed for the calculation of safe yield while 65 to 66 BG is held back in the Hudson
reservoirs.. This is an inequitable and unrealistic amount of reserve storage and results in
minimizing the ERQ. Clearly a reassessment must include an acceptable conjunctive use analysis

16
based on a reasonable reapportionment of reserve storage, perhaps optimally redistributing
reserve storage between the basins.

Assessing the safe yield of several reservoir systems individually is also problematic due to the
differences in the critical drought duration between the basins. The safe yield based operations of
the Hudson System resulted in two lesser drought periods from 1961 to 1964 when the Catskill
and Croton Systems refilled and from 1964 to 1967. The maximum Hudson Systems’ critical
duration was 948 days starting in May, 1964 and ending in December, 1966. Using the over-draft
operating program of Rev 1, makes it appear as though the Delaware Basin reservoirs would be
in drought stress from July, 1961 to November, 1965 when they would begin to recover; a
critical duration of about 1,600 days. However, under a different and equitable safe yield based
program it may be possible to reduce the critical duration of the Delaware System to two lesser
drought periods. Since the two drought periods do not coincide, it is difficult to determine
NYC’s safe yield relative to the 1960’s drought without conjunctively modeling both the
Delaware and Hudson Systems for the overlapping critical durations, which should optimize and
clarify the combined systems’ effective, conjunctive use safe yields. (Figure 1 demonstrates how
different the critical duration is under different safe yield based diversion and release
assumptions.) A summation of the individual worst case yields of each system’s reservoirs will
tend to be unreliable compared to the safe yield when appropriately optimized for conjunctive
use over the longest combined critical duration, from June, 1961 to December 1966 or about
1,640 days. Currently, improved computer tools and data are available to perform a reliable,
independent conjunctive analysis. As stated previously, all the necessary information for a
conjunctive use safe yield analysis is available via public records and through public domain
internet sources.

Nominal Safe Yields for the Delaware and Hudson Systems, Past and Present

Many of the estimates for Hudson and Delaware Reservoirs’ safe yields cited herein are nominal
figures taken from NYC publications and reports that represent a minimized safe yield. No safe
yield assessment has ever been independently conducted using modern accepted standards for
water supply engineering and applying consistent, realistic criteria.

It is essential to use safe yield as the basis to evaluate the adequacy of available sources.
However, a reasonable and equitable methodology should be applied so that past nominal safe
yields are not arbitrarily continued without verification of their appropriateness. Table 1 is a
compilation of Hudson and Delaware Systems’ safe yields for the droughts of the 1930’s and the
1960’s, and demonstrates the range of yields possible under varying criteria, such as reserve
storage, that can skew the total value. Some of the assumptions of 1954 appear obsolete or
unrealistic in relation to the longer flow records available, the demand patterns that should now
be used and the severity of the Montague flow reductions. The yields are compiled from past
NYC Chief Engineer’s reports and also supplemented by the NJDEP with conservative
reassessments using techniques explained in the paragraphs following Table 1. When including
Rondout Reservoir and Croton pumping, the combined safe yields totaled 1,805 mgd in 1954,
excluding 10 mgd from Kensico and 5 mgd from pre-existing wells that have since been
abandoned, and without considering the extra 65 BG of reserve storage in the Hudson reservoirs.
In 1954 the Chief Engineer for NYC indicated a total of 850 mgd for the Hudson reservoirs,

17
including Rondout, after reducing the Croton System to only its gravity yield (555 mgd for the
Catskill System, 175 mgd for the Croton System, plus 120 mgd for Rondout).The Decree
therefore appears to have apportioned 865 mgd to the Catskill/Croton Systems with Rondout
Reservoir, including 10 mgd from Kensico Reservoir and 800 mgd (50 mgd less than the
calculated safe yield at that time) for the Delaware Basin reservoirs for a total of 1,665 mgd.
Although this did not represent NYC’s maximum safe yield under equivalent reserve storage
criteria and did not include augmentation pumping to the Croton System, the Special Master’s
Report and Decree testimony shows the lower basin Parties had no objection as long as NYC’s
reservoir design and operations were predicated on its maintaining 1,750 cfs at Montague.

After the Decree, NYS requested minimum conservation releases that essentially used an
additional 50 mgd of safe yield that had, as an extra safety factor not been apportioned (since the
1954 operating plan had as yet not been tested). Theoretically, these additional releases were to
have no affect on the 1954 apportionment. However, after the drought of the 1960’s, all
conservation releases have been subsidized by reductions in Montague flow, not adjustments of
NYC’s safe yield.
COMBINED
Hudson Systems SUB-TOTAL
DELAWARE SYSTEM SUB-TOTAL Delaware &
Reference or Source Catskill System Croton System Catskill & Hudson Basin Delaware Basin Reservoirs Delaware Hudson
Schoharie Ashoken (12 Res's) Croton Rondout Neversink Pepacton Cannonsvile System Systems
1
Hochlerner (1945) 500 315 815 120 50 375 360 905 1,720
(25 % Reserve in Hudson Reservoirs)
Combined Total Kennison, 1954
2
Clark (1950) 555 330 885 120 50 375 360 905 1,790
(25 % Reserve in Hudson Reservoirs) Combined Total Kennison,1954
3 175 w/o 4
Kennison (1954) 215 340 730 120 115 375 310 by 1954 Decree 920 1,650
(25% Reserve in Hudson Croton NOTE, Combined Neversink & 1,805 w/Croton
Reservoirs) Pumping Pepacton Safe Yield of 490 Pumping;
{330 was apportioned at 440 by
w/Croton 1931 Decree (Actual Del {360 = Actual 1930's
Reservoirs Safe Yield in 1954 1,855 w/Croton
Pumping} Drought Safe Yield} Pumping & 850 Del
= 850)
reservoirs

Good Faith 470 240 710 100 480 580 1,290

Agreement (1983)
( 25% Reserve in Hudson Reservoirs)

NJDEP Estimated Actual 465 245 710 100 625 725 1,435
Safe Yield Under Rev 1 (25 NJDEP Min. Est. under FFMP Over-draft Plan
%R i H d R i t
NJDEP Estimated Actual 489 254 743 100 625 725 1,468
Safe Yield Under Rev 1 (10
% Reserve in Hudson Reservoirs, except NJDEP Min. Est. under FFMP Over-draft Plan
Schoharie)

NJDEP Estimated Actual 502 264 766 100 620 720 1,486
Safe Yield Under Rev 1 (ZERO
Reserve in Hudson Reservoirs, except NJDEP Min. Est. under FFMP Over-draft Plan
Rondout held at 25 % reserve)

Recent NYCDEP Estimate 470 175 645 97 480 577 1,222

Under Rev 1 (Approximated by w/o Pumping
NJDEP w/25 % Reserve in Hudson NJDEP Min. Est. under FFMP Over-draft Plan
Reservoirs, except Schoharie )

1. Chief Engineer, NYC Dept of Water Supply, Gas & Electricity, 1945. 2. Chief Engineer, NYC Bureau of Water Supply, Dept of Water Sup, Gas & Electricity, 1950.
3. Chief Engineer, NYC Board of Water Supply, 1954. 4. Total Safe Yield in 1954 included Kensico Reservoir (10 mgd) and 5 mgd from wells for a total of 1,665 mgd. NOTE: All safe yields exclude
10 mgd of Kensico Reservoir safe yield and 5 mgd wells.

Table 1. Compilation of safe yields of NYC's reservoirs obtained under various design presumptions and
operating criteria. (Add 10 mgd for Kensico Reservoir to all totals for estimate of total combined safe yield)

NYC’s Chief Engineer, Karl R. Kennison, prepared safe yield assessments in 1954 showing the
safe yield is 1,820 mgd, excluding 10 mgd from Kensico Reservoir and 65 BG of reserve storage
in the Hudson reservoirs, but including the 155 mgd of safe yield augmented by pumping to
Croton System storage later excluded in the Decree. (Table 1 excludes 15 mgd of wells and other

18
minor systems that were taken out of service shortly after 1954 so that the total safe yield of
1,820 by Kennison in 1954 is adjusted to 1,805 mgd.) It is important to note that the safe yields
in 1954 were limited by aqueduct capacities and nominal design criteria, such as reserve storage,
and not hydrology. After the drought of the 1960’s, the safe yields were limited by a more
severe drought than the design drought of the 1930’s. Once the ability to divert more than the
hydrologic yield became possible, that is to over-draft the safe yield, it was imperative that the
USGS and the Parties understand how demand patterns, conjunctive use and operations affect
safe yield and the safe yield need to protect the Delaware Basin and preserve the principles of
equitable apportionment. The NJDEP conservatively estimates that the actual 1954 safe yield,
assuming all reservoirs have no reserve storage, including Kensico Reservoir’s safe yield and
Croton System augmentation pumping, and with the Delaware reservoirs held at 800 mgd, to be
about 1,890 mgd. If the full 850 mgd of the Delaware reservoirs is accounted for, the safe yield
is 1,940 mgd. (Note: In the Decree, Cannonsville had been assigned a safe yield of 310 mgd
instead of its total safe yield of 360 mgd, as calculated against the drought of the 1930’s. The
Special Master’s Report and Decree testimony also indicate that the full 850 mgd safe yield of
the Delaware reservoirs was not apportioned because the operating plan had not yet been tested.
When preserving the 25 % reserve capacity in the Hudson Basin reservoirs, but using the full
850 mgd safe yield of the Delaware Reservoirs, NYC’s safe yield was estimated to be 1,855
mgd. Today, with Montague flow subject to reduction, Croton pumping should always be
included in the total safe yield determination since NYC operates the Croton System to maintain
that system’s safe yield regardless of Montague flows or Delaware Basin operations and
hydrologic conditions. It is highly likely that significant electrical cost savings have accrued to
NYC due to the reduced need of augmentation pumping to the Croton System because over-
drafting the Delaware Basin allows under-utilization of the Croton System, thus keeping it filled.
In addition, the Croton System’s West Branch Reservoir is also part of the Delaware System and
receives water from NYC’s Delaware and Rondout reservoirs which also reduces the need for
augmentation pumping.

Safe yields for the Catskill/Croton/Hudson Reservoirs can also be approximated by using the
storage obtained from the product of the critical duration and the nominal safe yield provided by
NYC for the Good Faith Agreement negotiations. The Catskills System’s critical duration
occurred from about May 3, 1964 to November 1, 1966 (911 days) and the Croton System’s
critical duration occurred from May 23, 1964 to December 27, 1966 (948 days). These critical
durations were noted and applied in the Task Group Report of 1979 (taken from “The
Coordinating Committee for the Reappraisal of Water Supply Resources of the Delaware River
Basin and Service Area” study completed in 1969) and were calculated from the start of
drawdown to 75 % depletion of total storage (25% reserve storage). After the product of the
critical duration and nominal safe yield is obtained, all or some percentage of the reserve storage
can be added back and the total divided by the critical duration to give a conservative estimate of
the safe yield relative to the chosen amount of reserve storage. Using 948 days for both the
Catskill and Croton Systems is simple and, again, conservative. The Catskill/Croton Systems,
which NYC currently rates at a safe yield of 710 mgd (with about 52.6 BG of reserve storage), is
reassessed herein at varying percents of reserve storage. The 710 mgd safe yield is broken down
as 470 mgd in the Catskill System and 240 mgd in the Croton System, as derived from past
NYCDEP publications.

19
Table 2 shows the method and results. The calculations all show extrapolations of the
Catskill/Croton Systems’ safe yields but retain Rondout’s safe yield at 100 mgd; that is, the 25 %
reserve storage in Rondout is conservatively maintained. Thus, Rondout’s 12.4 BG reserve
storage is unchanged and conservatively allows for even more flexibility and yield than any of
the NJDEP preliminary modeling shows.

Current Assumed Safe Yields Hudson Reservoirs Safe Yield Estimate w/948
Reserve Storage Critical Duration Days (Rondout
Condition Reservoir safe yield unchanged w/25 % (12.4
Catskill Croton Rondout
BG) reserve storage maintained)
System System Reservoir
With Zero Reserve ((470 mgd + 240 mgd)*948 days+52,600 mg)/948 days +100
Storage mgd (Rondout) = 866
(Hold 0, add back 470 mgd 240 mgd 100 mgd mgd
52,600 mg)

With 10 % Reserve ((470 mgd + 240 mgd)*948 days+31,560 mg)/948 days +100
Storage mgd (Rondout) = 843
(Hold 21,040 mg, add 470 mgd 240 mgd 100 mgd mgd
back 31,560 mg)

With 15% Reserve ((470 mgd + 240 mgd)*948 days+21,040 mg)/948 days +
Storage 100 mgd (Rondout) = 832
(Hold 31,560 mg, add 470 mgd 240 mgd 100 mgd mgd
back 21,040)

Note: The above estimates include Croton System’s augmentation pumping as it is logical to do so since
NYC maintains that safe yield regardless of Delaware Basin uses and operations. Also, since Montague
flow reductions benefit NYC, NYC should include the Croton pumped safe yield augmentation in
reciprocity.

Table 2. Estimation of NYC’s Catskill and Croton Systems’ safe yields at various reserve storages. Rondout
Reservoir is held at 100 mgd, which includes 25 % (12.4 BG) reserve storage.

A final note on total system safe yield; Table 1 also shows a recent NYC estimate of 1,222 mgd,
for which no supporting information has yet been provided. However, the same approximation
can be obtained by simply summing the nominal safe yields used in 1983 but eliminating Croton
System yield from augmentation pumping or may also be obtained by modeling the variable
demand pattern under the current operating rule curves which may result in reducing the
Delaware reservoirs’ safe yield. Although 1,222 mgd may appear to be a significantly lower safe
yield, such an operating scenario is also likely to virtually eliminate any drought stress on NYC
and also ensure the 1,750 cfs flow is maintained at Montague.

Demand and Demand Patterns

Currently NYC’s Delaware System’s peak-to-average use ratio ranges between 1.2 and 1.48 to 1,
depending on precipitation levels, with an average of about 1.33 to 1 normal precipitation years
(Above average rainfall years enable NYC to use at greater peak to average ratios). The use ratio
is reflective of NYC’s normalized pattern due to effective implementation of universal metering,
leakage/loss control and repair, installation of water conserving fixtures and modernized
plumbing code programs initiated in the 1990’s. Until the universal metering program, NYC had
little control of use or accurate information on use characteristics. The normalized pattern for a
public water supply system typically varies between the highest summer peak months (July and
August) and the rest of the year’s off-peak seasonal use. The ratio will likely continue to
increase with ongoing improvements in leakage and loss control, and other conservation
programs that normalize NYC’s demand pattern. Therefore, the demand pattern is not a

20
temporary phenomenon; it is a typical permanent expression of a contemporary public water
supply system implementing common conservation practices.

It is efficient and advantageous for all Parties to consider the demand pattern in the development
of a realistic, equitable, optimized safe yield-based operating plan. The ACE’s Engineering
Manual (EM 1110-2-1420: Engineering and Design - Hydrologic Engineering Requirements for
Reservoirs) indicates:

4-1. Introduction
a. Mathematical modeling. It is usually necessary to simulate the detailed sequential
operation of a system, representing the manner in which each element in the system will
function under realistic conditions of inputs and requirements on the system. The
simulation can be based on the results from the optimization of system outputs or
repeated simulations. Successively refining the physical characteristics and operational
rules can be applied to find the optimum output.

Further the Manual describes:

12-15. General Study Procedures: a. Water supply. (4) Determine seasonal patterns of
demands and total annual requirements for all project purposes, if applicable, as a
function of future time. Synthesize stochastic variations in demands, if significant.

The NJDEP found NYC’s estimated peak-month to average annual demand had not been
accounted for in any lower basin model or operating plan. Because NYC’s peak-month to
average annual demand when the 1983 Good Faith Agreement operating plan was first adopted
was close to 1 to 1, due to a combination of substantial system leakage, high consumption and
lack of metering, the assumption of a constant, unvarying diversion could be considered
appropriate. The implementation of conservation practices such as universal metering has
fundamentally changed NYC’s demand pattern and the reservoirs’ operating plan and rule curves
no longer are representative of the NYC’s use. Further these demand changes significantly alter
the NYC’s current and ultimate needs such that the ultimate build-out population could be fully
served by much less yield than 800 mgd. Modeling efforts (e.g. OASIS) that do not reflect these
demand patterns are inconsistent with standard engineering practice and the results do not
accurately reflect realistic operations or predict future effects. Consequently, the OASIS
assumptions result in an over-estimation of the number of “drought days” and under-estimates
the robustness of NYC’s water supply.

In 1989, Hazen & Sawyer (H&S) as consulting engineers for NYC, produced a study (“Study of
Water Demands on New York City System Final Report,” Nov., 1989.) recommending a water
conservation program (e.g. installation of water meters, leakage and loss programs, public
education, etc.) designed to enable NYC to serve a year 2035 build-out population of about 11
million people. This was to be accomplished with a safe yield of 1,470 mgd, at a per capita use of
approximately 133 gpcd. With conservation, the predicted demand extrapolated from the H&S
study for 2010 is about 1,481 mgd, including communities served in counties outside NYC.
Projected demand without conservation for 2010 was about 1,816 mgd. The original 1954 Decree
design population, as per the Special Masters Report, was 12 million people for design year 2000,

21
which now is conservatively high since NYC’s population is about 9 million. Since NYC’s 2010
demand will likely be about 1,200 mgd or less, the H&S study under-estimated the effect of
conservation and also did not address the potential effect of conservation on NYC’s demand
pattern and the subsequent effect of that pattern on operations, reservoir levels, releases and
usable safe yield. The H&S study referred to the minimized nominal safe yields established in
public documents and the Good Faith Agreement and thus avoided the potentially contentious
issue of the NYC’s realistic, effective safe yield under various operating scenarios such as over-
drafting, conjunctive use and alternative reserve storage volumes. The possibility that NYC’s
reservoirs would tend to be near-full, full or surcharged more frequently and for longer durations
and their potential impact on downstream flooding was also not anticipated.

According to published reports, (“Summary of the Hydraulic Aspects of the Catskill/Delaware

Aqueduct System,” NYCDEP, March, 1994) the Catskill/Croton Systems’ combined aqueduct
capacity is 860 mgd. In the 1983 Good Faith Agreement negotiation records and in publications
since 1983, NYC attributes 710 mgd of its estimated combined total safe yield of 1,290 mgd to its
Catskill/Croton Systems, excluding about 53 BG of reserve storage or about 25 % of the
combined Catskill/Croton storage (excluding the 17.6 BG Schoharie Reservoir) and about 12 BG
of the Rondout Reservoir in the Delaware System (for a total of about 65 BG). The
Catskill/Croton Systems now operate at just less than 1.2 to 1 peak-to-average ratio. Since the
peak carrying capacity of the Catskill and Croton Aqueducts is less than the Delaware Aqueduct,
the Catskill/Croton Systems are used at a lesser peak to average variation, in effect operated to
provide a year round ”base flow” to NYC’s system. An alternate description is that the
Catskill/Croton Systems’ peak capacity, 860 mgd, is closer to their genuine combined safe yields.
Therefore, they are optimally used at a lower peak to average ratio. (Estimated peak capacity to
safe yield ratio ranges from: 860/766 = 1.12 to 860/710 = 1.21 since the combined
Catskill/Croton Systems safe yield with 0 reserve storage is about 766 and with 25 % reserve
storage, about 710 mgd).

The demand pattern is not only affected by use and need, but by system distribution capacity at
the delivery end. Distribution reservoirs, such as Kensico (30 BG) and Hillview, (0.9 BG) accept
water from the Delaware and Hudson Systems and use only the upper several feet of storage to
maintain energy head. This limits the time that the aqueducts can divert and, in turn, affects how
much and when water can be drafted from the Hudson and Delaware Systems; this affects the
accuracy and usefulness of both safe yield and flood risk assessments. Modeling without
considering the conjunctive effects would tend to over-estimate drought risk and under-estimate
flood risk for the Delaware Basin. Internal system constraints affect reservoir operations and
conflict with the assumption of a continuous over-draft of 800 mgd from the Delaware reservoirs
as well as a continuous diversion from the Hudson Basin reservoirs.

By examining demand patterns in both Delaware and Hudson systems and those systems’
operating constraints, including realistic equitable use, need, demand pattern and the ERQ, the
safe yield can be more clearly defined and understood and a realistic, optimized operating plan
can be developed.

22
NYC’s Historic Reservoir Operating Rules Criteria

The current operations of NYC’s reservoirs are derived from the 1930’s, “New York Rules.” The
New York Rules were originally published in 1950 by the NYC’s Bureau of Water Supply Chief
Engineer Edward J. Clark in an AWWA article titled, “New York Control Curve.” The article
explains that the New York Rules maximize the probability of spills thereby minimizing
expected storage shortages. By designing the probability of spill to be equally maximized for
each reservoir at the end of every refill season, it follows that physical spill is minimized and
system refill maximized. This is a result of a distribution of the probability of spill,
proportionally between parallel reservoirs, which reduces the likelihood of wasting real spillage
at any one reservoir and, therefore, for the total system.

The New York Rules would remain a valid method if; 1, they were derived from a safe yield-
based operating plan developed for the 1960's drought; 2, they entailed a well-defined
downstream flow objective, and: 3, a constant diversion was applied. The 1983 Good Faith
Agreement over-draft policy eliminated the first criterion, and the tiered Montague flow and
diversion schedules eliminated the second and third criteria. The FFMP, coupled with NYC’s
current normalized demand pattern, results in maximizing reservoir refill thereby increasing the
probability of surcharged reservoirs. The increased frequency, duration and intensity of
surcharged reservoirs results in extended extreme daily flow averages and therefore adds to
downstream flood potential.

The FFMP’s discharge mitigation program provides no measurable and reliable reduction of
severe flood risks. A more applicable operating plan design method might be the “Space Rule,”
(See, Lund, Jay R., “Some Derived Operating Rules for Reservoirs in Series or in Parallel”,
Journal of Water Resources Planning & Management, Vol. 125, No. 3 pp.143-153, May/June
1999). The Space Rule is a special case of the New York Rules designed to increase the
probability of more space in reservoirs where greater inflows are expected. According to Lund,
the Space Rule seeks to minimize the total volume of spill. None the less, neither the Space Rule
nor any other appropriate operating program can be effective without application of legitimate,
safe yield-based operating criteria.

OASIS Model and Daily Flow Model (DFM)

Reservoir operations and the impacts on flows and flooding have been assessed using the OASIS
modeling tool. The current OASIS operating plan is a revised and updated “desktop” version of
the original Daily Flow Model (DFM), a Fortran based program developed in 1981 by Camp,
Dresser and McKee, Inc., for the Army Corps of Engineers (ACE) in cooperation with the
Pennsylvania Department of Environmental Protection (PADEP), the New York State
Department of Environmental Conservation (NYSDEC), the U.S. Geological Survey (USGS)
and the Delaware River Basin Commission (DRBC). The ACE indicated that the purpose of the
regulated flow model in the DMF was to determine the model’s ability to match the desired flow
conditions within the constraints of the reservoir operating rules. Therefore, the DFM was never
designed to develop a safe yield based operating plan derived from conjunctive effects of NYC’s
Hudson reservoirs, if even only the Rondout Reservoir, or to evaluate total system optimization
or reapportionment; the DFM and later the OASIS Model became task specific models for

23
generating flows and droughts given the various unrealistic operational presumptions of the over-
draft policy. The DFM essentially uses the rule curve from Model # 2 provided in Task Group
Report for DRBC Docket # D-77-20-CP, and attempts to maintain 800 mgd to NYC, which
results in an operating plan that essentially replicates all the emergency conditions of the drought
of the 1960’s rather than developing a new, sustainable safe yield based program. In 2002, via a
DRBC contract, Hydrologics, Incorporated derived the more user-friendly OASIS model from
the DFM. The OASIS Model is essentially the DFM model with some minor updated estimates
of lower basin demand and consumptive use projections. Although based on obsolete operating
assumptions, OASIS is available to all the Parties, the public and NYC.

After completing the OASIS Model, Hydrologics, Inc. developed another, separate OASIS
model (OASIS-W2) for use by NYC and NYS which has not been made available to the DRBC
or the Parties for review because it is considered confidential and proprietary by NYC and
Hydrologics. The existence of two models, one publicly vetted but flawed and the other having
no public vetting at all, undermines the transparency necessary to assure equity of operations
between the basins. The Decree clearly extends the right to any Party to obtain the necessary
information to assure such equity. To foster verification of appropriate reservoir use and safe
yield as provided by the Decree, the OASIS-W2 should be made fully available for review by the
lower basin Parties. The same is true of any other NYC models, such as its proposed Operations
Support Tool (OST) model. Such models should be developed with the DRBC and the Decree
Parties, not autonomously as they affect the entire lower basin. The changes in NYC’s demand,
accurate representation of the effective safe yields of the Hudson or Delaware Systems and their
conjunctive use, equity of use, realistic effects between the systems, reasonable reserve storage
levels, and appropriate reservoir operating curves should all be under the lower basin Parties’
purview, especially if agreeing to lower Montague flow targets.

The results produced by the DFM and OASIS Model are not reflective of the system in part or in
whole because Delaware basin reservoirs are analyzed without regard for the effects of Rondout
and the assumption of a constant 800 mgd (or 765 mgd used in the FFMP) diversion. A realistic
operating scenario representing both current and ultimate demands has never been applied.
Because the OASIS Model’s operating criteria are based on several obsolete or unrealistic
assumptions, the modeled results are not useful in developing a sound and equitable operating
plan or flood mitigation program. The key concerns with the current operating plan are:

a) Omits the conjunctive use effect of Rondout and other Hudson basin reservoirs from the
Delaware reservoirs safe yield and flood analyses.
b) Assumes NYC is compelled to continuously draft 800 mgd.
c) Does not apply NYC’s normalized demand pattern.
d) Includes operating rules instead of drought rules that cause droughts and restrictions for
the Delaware Basin while enabling NYC to meet its needs.
e) Incorporates the lower basin’s incorrect assumption of NYC’s safe yield based on a
“worst year” average draft under an over-draft of safe yield policy.
f) Establishes an unrealistic, counter-productive and flood risky Hudson Basin reserve
storage volume (65 BG).

24
Not considering the hydraulic constraints that Rondout Reservoir and NYC’s distribution
reservoirs impose on diversions from NYC’s Delaware Basin Neversink, Pepacton and
Cannonsville Reservoirs obscures the potential for filling or surcharging these reservoirs when
all or part of the modeled diversions cannot be made simply because the receiving and
distribution reservoirs are themselves too full to accept the modeled diversions. Although
omission of Rondout Reservoir’s conjunctive use effect inhibits an accurate assessment of the
Delaware reservoirs’ effective safe yield, it particularly under-predicts flood risk analyses by
simulating voids in the Delaware reservoirs that are exaggerated or non-existent. Therefore, the
current OASIS model over-estimates the probability of void space in the Delaware reservoirs and
under-estimates flood risk. Further, it should not be assumed that the conjunctive effect of
Hudson Systems’ storage usage does not also affect the Delaware System flood risk and safe
yield, with the converse also being true.

As noted above, the current OASIS model, like the original DFM model, assumes NYC will
attempt to over-draft at 800 mgd, that is, attempt to withdraw at a greater rate than the reservoirs
can sustain in a repeat of the Drought of Record, year round. The assumption of an 800 mgd
over-draft leads to the need to reduce the Montague minimum flow goal to 1,100 cfs. During
such times, NYC’s diversion is also curtailed to as low as 520 mgd. Each of these curtailments
improves NYC’s storage recharge and therefore increases NYC’s actual safe yield. The safe
yield assumed by FMTAC under the Rev 1 over-draft plan in 1983 was 480 mgd, taken from the
worst year of the 5-year critical duration of the 1960’s drought, 1965. This minimized 480 mgd
yield has been used by NYC in all its public reports of safe yield despite the fact that 480 mgd is
the same safe yield NYC would have if it maintained Montague flows at 1,750 cfs at all times,
which it does not. The reduced Montague flow enables NYC to increase its safe yield to some
point greater than 480 mgd, but less than 800 mgd. Without an appropriate conjunctive use
model, it is difficult to firmly calculate what NYC’s safe yield is under the seasonal peak varying
demand pattern; such a pattern tends to reduce safe yield since peak demand coincides with
minimum inflow. However, conjunctive use between multiple in-series and parallel reservoir
systems can compensate for such a potential reduction by optimizing diversions in relation to the
demand pattern. The NJDEP has preliminarily estimated that NYC can apply such conjunctive
use operating alternatives, independently of what is assumed in the FFMP plan to optimize the
effective safe yield of its total system to serve its entire built out population of between 11 and
12 million people for the ultimate design year of 2060. The 1983 Good Faith Agreement and the
modeling done at that time appears not to have grasped some key fundamental characteristics of
safe yield assessment, conjunctive use, and the effects of storage curtailments versus run-of-
stream curtailments.

Reservoir Operating Rule Curves

The NJDEP’s preliminary estimates indicate that, whether the Rev 1 or the FFMP operating plan
is applied, NYC’s Delaware reservoirs’ contribute to a combined total safe yield of at least 580
mgd, perhaps as high as 625 mgd or greater. A safe yield significantly greater than 480 mgd is
deliverable within the envelope of the operating rule curves, release schedules and diversion
rules of the FFMP or Rev 1. Often during transitory low flow periods, particularly in early
summer, the existing operating plan responds to the increase in seasonal peak over-drafting of

25
storage by triggering pre-mature Montague flow release reductions, ultimately resulting in an
increase of NYC’s effective safe yield but less water for the lower basin. The hydrology of the
1960’s drought will support neither a constant draft of 800 mgd nor the 765 mgd used in the
FFMP, yet, the storage curves are set on that premise. In short, by NYC taking less than 800
mgd, on average (even though it may average 800 mgd starting in June 1st and through the
summer peak season), it secures more than the 480 mgd NYC represents as its safe yield. The
result is a substantial over-estimation of the number of “drought days.” Further, any assumption
that NYC’s over-draft reduction to 765 mgd in the FFMP is equivalent to its having given up 35
mgd for lower basin fishery maintenance is fallacious since 765 mgd also significantly exceeds
the safe yield NYC actually obtains under the over-draft policy; it exceeds it by more than 100
mgd. In reality, most fishery improvements are still coming from reduced Montague flows, not
NYC storage. These increased releases are only available when most likely to be unneeded,
during average to above average flow and storage conditions. The increased fishery releases
efficiently hasten reservoir drawdown to trigger their severe reduction or cessation and thus
enhance NYC’s conservation of storage; New York ceded essentially nothing in the FFMP and
gained approximately 20 mgd of safe yield and removal of the vernier releases that help offset
salinity encroachment at the estuary intakes.

1983’s Lower Basin Parties Assessment of the Good Faith Agreement Operating Plan (Rev 1)

During the 1983 Good Faith Agreement negotiations, the Compact States formed a Flow
Management Technical Advisory Committee (FMTAC) to assess the appropriateness of the
original, Docket D-77-20-CP, Revision 1 operating plan. In June of 1983, the FMTAC submitted
to the DRBC the “Interim Report of the Technical Responses to “Good Faith” Recommendations
#3 and #4. The report was issued after Rev 1 had already been signed, in January of 1983, and it
was inferred from this report that NYC’s safe yield remained at 480 mgd even though the
minimum Montague flow had been reduced. The 1983 Good Faith Agreement operating plan
essentially was modeled to replicate the same crisis that occurred in the 1960’s drought as if
NYC were compelled to attempt taking 800 mgd until it was no longer sustainable. However,
because of reduced release requirements in relation to reservoir levels, as well as NYC’s
prerogative to take less than 800 mgd, NYC’s safe yield would actually increase to significantly
more than 480 mgd. However, the FMTAC report did not recognize the additional yield that
alternative diversions could afford NYC under the Rev 1 operating “envelope” derived from the
assumption of NYC’s attempt to draft 800 mgd, continuously.

Further, the FMTAC report did not include the effect of conjunctive use of Rondout, West
Branch Reservoirs (West Branch Reservoir in the Croton System effectively captures an
additional 10 billion gallons of Delaware Basin water) or other Hudson Basin reservoirs in the
event NYC could not meet its Montague flow obligations when attempting to over-draft the
Delaware basin reservoirs. No analysis of any pragmatic and equitable reapportionment between
the Hudson and Delaware basins prior to agreeing to Montague flow reductions was ever
performed and a unanimous agreement was reached by the 1983 Good Faith Agreement
negotiators to change key Decree operating conditions that had originally been appropriately
derived from the fundamental tenets of safe yield.

26
Since 1983, the conjunctive use of Rondout, West Branch and other Hudson Basin reservoirs
allows NYC to under-utilize its Catskill and Croton Systems and meet its peak demands by over-
drafting the Delaware System with little total system stress while passing the larger risk of
drought impact to the lower Basin. By examining only the Delaware basin reservoirs, the
FMTAC report found that a constant 800 mgd diversion results in the apparent necessity of
severe curtailments in the latter years of a five-year drought period; during the worst year, 1965,
an average diversion of only 480 mgd results. The FMTAC report therefore reinforced the
misperception that 480 mgd was NYC’s permanent, fixed safe yield because it was the same safe
yield that the United States Geological Survey (USGS) estimated for NYC’s Delaware
reservoirs, when partitioned from the Hudson reservoirs effect, in a 1968 study with the
minimum Montague flow maintained at 1,750 cfs. Ironically, these drought impacts are now
only artifacts of an unrealistic operating plan and a flawed model. Since NYC’s demand has
become normalized and its needed yield closer to 480 mgd than 800 mgd, actual drought risks
are thousands of drought days less than the OASIS Model indicates; however flood risk has
conversely increased.

Often, by the end of the peak demand/low inflow season, summer, NYC may induce lower basin
drought conditions by beginning each June 1 at the 800 mgd over-draft. By the end of the peak
demand season storage levels may have dropped to trigger Montague flow reductions which
serve to recharge NYC’s reservoirs. Even if NYC is curtailed to 520 mgd during peak demand
periods, by conjunctive use of its Rondout and other Hudson Basin reservoirs, NYC can still
serve its peak season needs with little or no conservation. With off-peak season droughts NYC’
“curtailment” to 520 mgd exceeds its need and Montague flow reductions continue to maximize
recharge. The 1983 Good Faith Agreement allows and promotes such an over-draft scenario
resulting in many unnecessary drought warnings, while NYC is curtailed to a level that exceeds
its current annual average need from the Delaware reservoirs.

Effects of Safe Yield Determination on the ERQ and IERQ

Both the ERQ and the IERQ include an 83 percent reduction of the difference between expected
year’s demand and the safe yield, less 7.25 BG for a sudden growth spurt. Why an 83 percent
reduction was used has been the subject of various testimonies. NYC’s Chief Engineer’s
testimony from 1954 indicated it is the result of the expectation that for two months of every
year the reservoirs were expected to spill, for which NYC did not think it should counted against
its unused water; therefore, at that time it was agreed that 10 of the 12 months of controlled
storage, or 83 %, would be applied to the combined total water supply systems’ unused volume.
Elsewhere in NYC testimony, a critical duration of about 2.5 years over the total 3-year 1930’s
drought period is used, which also equates to 83 percent. Neither of these explanations is
particularly hydrologically sound relative to either the 1930’s or 1960’s drought. In reviewing
the 1954 testimony, it appears the 83 percent reduction, as well as allowance for an expected
growth spurt equivalent to an additional use of 7.25 BG, functioned more as an agreeable ‘rule of
thumb’ to justify creating a reasonable cushion against an as yet untested operating plan. A cap
of 70 BG for the ERQ was based on the most recent demand in 1952-53, about 1,125 mgd in
total. When adding the apportioned safe yield of the then proposed Cannonsville Reservoir, 310
mgd, the total is 1,435 mgd. This total was used to set the resultant cap of 70 BG when

27
calculating the ERQ with 1,665 mgd. This cap is no longer relevant either compared to the
current demands, safe yields and design drought of record, the drought of the 1960’s.

The Delaware reservoirs’ effective safe yield as a result of Montague flow reductions and
conjunctive use of Rondout, West Branch and other Hudson Basin reservoirs has never been
determined and the 480 mgd found in the original 1968 USGS yield analysis with the basins
partitioned has been used in the FFMP. To be clear, the 480 mgd is the same minimized yield
obtained by the partitioning the Delaware Basin from the Hudson in the safe yield analysis done
by the USGS in 1968 if NYC were to maintain its minimum Montague flow obligation of 1,750
cfs. Diversion records show NYC has not designed its operations to try meeting 1,750 cfs but to
attempt over-drafting at 800 mgd, continuously; its diversions are designed, therefore to trigger
the Montague flow reductions in the event of a developing dry period at the start of each peak
demand/low inflow season, beginning June 1. The 480 mgd used in the IERQ calculation
coupled with the minimized safe yield of the Hudson reservoirs, results in a significant
underestimation of any bank. In 1977, an ERQ of 31 BG was set in Docket D-77-20. Although it
was also calculated inappropriately, it provided an ERQ 3 times greater than the IERQ used now
(2010) in the FFMP because even though demand and need have decreased, NYC and NYS have
applied yield estimates that have outpaced this decreased need so as to minimize its actual
optimal safe yield. The FFMP’s IERQ is approximately only 10 BG as a result of using NYC’s
minimized combined total safe yield of 1,290 mgd compared with a projected demand of 1,257
mgd and a reduction factor of 0.83. If the D-77-20-CP, Rev 1 Docket were to be reinstated, a
new ERQ would be required and it could be done relative to a realistic expected demand, about
1,200 mgd, relative to a realistic, equitable and effective total safe yield, yet to be determined.
The IERQ is unrepresentative of actual hydraulic conditions, hydrology and system operating
characteristics such as projected demand; the FFMP has and will continue to result in shortfalls
of storage available for Montague flow maintenance as well as in the IERQ bank. A preliminary
conjunctive use analysis by the NJDEP indicates that the operating constraints of Rev 1 resulted
in about 35 BG of unusable storage in the Delaware reservoirs in a repeat of the drought of
record and under the FFMP this storage is made accessible resulting in an increase of 20 mgd or
more of safe yield to NYC.

About 1,250 mgd is a reasonably high, “next year’s” demand, as per the intent of the Decree
formula. The Decree formula is based on accepted protocols for projecting a growing use trend
and not “fixed” to any highest year of demand as has been incorrectly applied in past ERQ’s.
Given declining demand rates due to normalization and use trends, 1,250 mgd or less is currently
a reasonable projection of any forthcoming year’s use estimate in an ERQ calculation. For
example, in calculating the ERQ, the total systems’ actual safe yield should be used and
compared to the expected coming year’s demand. The Decree is clear, (III. B. 1. c.)

“If, at any time after the completion of the Cannonsville reservoir and prior to the year
1993, the continuous net safe yield for water supply of all of NYC’s sources of water
supply, obtainable without pumping, is increased by the development of additional
sources, such greater safe yield shall be used in determining the excess releases.”

Since 1954, there has been only one significant new water supply “source” developed by NYC;
the Montague flow reductions. The 480 mgd value is the safe yield without the Montague flow

28
reductions. Further, it is clear that the figure to be used for estimating the expected consumption
should be the anticipated demand of the coming year, not the highest demand on record,
especially since the record high no longer typifies NYC’s demand. Flow probability and demand
(including releases) are the foundations of an appropriate safe yield analysis and the ERQ is
based on the assumption that NYC’s safe yields, growth rates, and demand characteristics are
accurately projected.

For example, a conservatively low estimate of a total minimum equitable safe yield obtainable is
about 1,350 mgd (520 mgd Delaware Reservoirs + 830 mgd Catskill/Croton System’s w/10 %
reserve storage) A sample ERQ calculation is shown and is based on an expected use of 1,200
mgd:

Using 1,200 mgd for 2010 and 7.25 BG/yr for a sudden growth spurt:
ERQ= 0.83*(365days*(1.350 BGD-1.200 BGD)-7.25 BG) = 39.4 BG.

The sample ERQ is significantly greater than the 10 BG IERQ of the FFMP’s and is based on a
more appropriate method for calculating it. A safe yield range that is equitable and would serve
NYC’s peak needs of 800 mgd from the Delaware reservoirs under a variable demand pattern,
through the drought of record and provide far greater minimum Montague flow is possible. If an
appropriate safe yield based operating plan were designed, no cutbacks would be theoretically
necessary, although prudent drought conservation curves should be superimposed to reduce
diversions in the event of a drought worse than the drought of record, which is the normal,
accepted standard applied to water supply reservoir operations. Such conservation curves would
curtail NYC’s drafts by a modest percentage of the expected demand relative to the season such
that, for example, during peak periods when NYC’s demand may be 800 mgd, under appropriate
drought emergency conservation curves NYC may be curtailed by 10 %; that is to about 720
mgd; during a winter drought emergency the 10 % reduction may mean NYC’s winter draft need
of about 300 mgd would be reduced to 270 mgd. This would eliminate the practice of curtailing
NYC to 520 mgd when its needs are far more or far less.

Reserve Storage Quantity

Table 2 shows a reasonably conservative extrapolation of safe yields for NYC’s Catskill and
Croton Systems’ using alternate reserve storage volumes for comparative purposes. For the
Catskill and Croton Systems, NYC still applies the principle of safe yield but also still reserves
25 % storage from its safe yield calculation. Continuing to withhold 25 % reserve storage in
calculating these systems’ safe yields lowers the safe yields and results in passing more of the
1960’s drought risk onto the Delaware Basin by applying a theoretical over-draft policy, that is
by applying the presumption of a daily draft of 800 mgd to the Delaware reservoirs that
consequently triggers reduced Montague flows, in accordance with the 1983 Good Faith
Agreement; this policy is continued in the FFMP. Reducing the disproportionably high reserve
storage, (compared to zero in the Delaware reservoirs) in the Hudson reservoirs, (about 65 billion
gallons, total, including about 12 billion gallons in Rondout Reservoir and about 53 billion
gallons in the Catskill/Croton Systems) and redistributing it such that perhaps a 10 % void could
be reserved in the Delaware reservoirs would enable a much more flexible, realistic and optimal
operating plan between the basins. This may create opportunities to increase the probability of

29
meaningful void spaces for flood mitigation. The partitioned operating plans, with the
Catskill/Croton Systems based on safe yield and the Delaware System on over-draft, appear to
increase flood risk to both basins because of the resultant tendency for the reservoirs to be
frequently full or surcharged. A realistic, optimized conjunctive use plan should help reduce such
risks. Reserve storage values used in safe yield determinations and reservoir operation plans
should be consistent with modern standards (in the range of 10-15 percent) of water supply
engineering given the longer, now more reliable, period of flow record since the design of the
Hudson reservoirs in the 1920’s and 1930’s when only about 10 to 20 years of flow records were
available among far fewer gage stations.

The drought of the 1960’s changed the basic design drought criteria for both systems, yet NYC
still held 25 % reserve storage in the Hudson reservoirs when calculating its reduced Hudson
Systems safe yield. Thus the intent of designing reserve storage, that is, to provide flexibility in
safe yield in the event of a drought worse than the design drought and enable revising operations,
accordingly was never realized and the Delaware Basin was over-drafted to supplement the
Hudson Basin safe yields. This was made possible by reducing Montague flows.

D. NJDEP Preliminary Findings on Safe Yield Based Operations of Delaware Reservoirs

Compared to Over-Draft Operations under Current Good Faith Agreement Assumptions.

Safe Yield, Delaware Reservoirs

A conservative estimate of a realistic and sustainable safe yield from the Delaware System, that
is, the safe yield NYC can sustain within the unavoidable constraints of equitable use, hydrologic
sustainability and the systems’ hydraulic limits can most transparently be assessed by beginning
with peak demand needs. NYC’s combined total peak-day aqueduct capacity is about 1,810 mgd.
A peak operating capacity of 950 mgd is available from the Delaware Aqueduct and about 860
mgd from the Catskill and Croton Aqueducts (about 585 mgd Catskill Aqueduct and 275 mgd
Croton). These figures are derived from NYSDEC reports such as, “Summary of the Hydraulic
Aspects of the Catskill/Delaware Aqueduct System,” by Alec Y. Fu, NYCDEP, March, 1994,
and represent actual practical aqueduct and tunnel capacities, not the nominal capacities, such as
the 1,000 mgd often applied to the Delaware Aqueduct. The estimated peak month capacity of
the Delaware Aqueduct in a 1989 Hazen and Sawyer study was 890 mgd. Since 2002, NYC has
been using the Delaware Aqueduct at a peak month capacity of less than 860 mgd as presented in
Figure 2 (page 36) using data obtained from NYC diversion records. NYC has applied this peak
rate, apparently as a result of a Delaware System operating plan that is independent of the 1983
Good Faith Agreement operating assumptions and indicates NYC now operates the Delaware
System on the basis of maximizing its safe yield. The diversion pattern is an indication of the
effect of the normalized demand pattern which constrains NYC’s operations to the resultant
diversion pattern. As will be shown, by analyzing the effect of NYC’s normalized demand and
diversion pattern against NYC’s peak need, a safe yield of 520 mgd from the Delaware
reservoirs alone would enable NYC to serve its peak needs through a repeat of the drought of
record. However, a 520 mgd need would also enable NYC to support its Montague flow
obligation much more effectively; since NYC does not support the Montague flows as much as it
could, its reservoirs tend to be near-full, full or surcharged more frequently.

30
For approximating practical operating limits, a peak month’s average use can be conservatively
estimated by using 90 % of the total combined peak carrying capacity of 1,810 mgd. The total
combined systems’ peak-month maximum use would then be 0.9*1,810= 1,630 mgd. The peak
capacity of the Catskill/Croton aqueducts alone is 860 mgd and 90 % of this rate is about 775
mgd. This leaves a peak month delivery need of about 855 mgd from the Delaware System:
1,630-775 = 855 mgd. That is, the peak-month need from the Delaware System, 855 mgd, is less
than its peak capacity of 950 mgd; however 950 mgd would still be available to serve occasional
peak days over any peak month. The last eight years of diversion records verify that NYC’s
peak-month diversions from the Delaware System average around 855 mgd. The diversion
record also demonstrates that prior to 2002 NYC rarely exceeded 855 mgd from the Delaware
System and when it did so it appears to be because of the skewing effect the over-draft policy has
on use of the Delaware reservoirs during above average rainfall periods.

The peak month to annual average use rate of NYC’s total system is about 1.33 to 1 but, because
the Catskill/Croton Systems have less peak carrying capacity, they are used at about 1.2 to 1 and
the Delaware System at about 1.4 to 1. At 1.4 to 1 the needed safe yield from the Delaware
System, on an annual average basis, is therefore 855/1.4 = 610 mgd. The safe yield of Rondout
Reservoir alone is at least 100 mgd without using the 12 BG of reserve storage. This leaves a
safe yield need from the Delaware reservoirs of about 610 – 100 = 510 mgd or 10 mgd less than
the drought emergency diversion curtailment of 520 mgd. Recent diversion records from the
Delaware reservoirs indeed show NYC averages close to 510 mgd. This is a characteristic of
NYC’s yields, system hydraulics, demand patterns and the over-draft policy. In fact, NYC
should need even less than 510 mgd under an appropriate safe yield based operating program.

A safe yield as low as 480 mgd might meet NYC’s ultimate needs if a safe yield based operating
plan were applied and operating rules and rule curves were developed that correlated to realistic
demand patterns, hydrology and hydraulic limits. Further, if less reserve storage was withheld in
NYC’s Hudson Reservoirs, the conjunctive uses between the basins could improve both yield
and flood mitigation opportunities. However, even without reducing NYC’s estimated 65 BG
reserve storage volume in the Hudson reservoirs, the minimum drought curtailment rate of
diversion of 520 mgd should enable NYC to ultimately serve up to 12 million people (the build-
out design population used in the Decree negotiations; see Special Masters Report) when NYC’s
expected use is extrapolated with a very conservative gradual increase in conservation efficiency
to achieve a per capita use of about 111 gallons per day, a reduction of about 16 % of its current
per capita use of about 133 gpcd. Such a reduction should be confidently expected given the
growing efficiencies in conservation technology and growing public support for all forms of
sustaining the nation’s resources, including water. Further, past growth and demand projections
have been overly conservative and never corrected. The 1989 Hazen and Sawyer water supply
study for NYC indicates the ultimate build-out population is more likely to be 11 million. Given
NYC’s past growth and per-capita use rates, it should take over 100 years, possibly 150 years for
NYC’s service population to reach 12 million people if it is ever reaches that number at all.
NYC has justified its high projections of need based on the drought of 1981, a relatively minor
hydrological drought that never the less caused severe storage draw-downs. A 1985 American
Meteorological Society report by Robert A. Weisman of the Department of Atmospheric

31
Science, State University of New York at Albany titled, “The 1980-81 Drought in Southeastern
New York” found that,

“…while New York City did experience a hot dry summer in 1980, it was within normal
climatological variation and, with a reservoir system at 100 % capacity to start off, the system
should have been able to weather this relatively dry period without such a serious situation
developing.”

The Weisman report attributed NYC’s storage deficit problem to a combination of increasing per
capita consumption and a lack of effective conservation. With NYC’s normalized demand, a safe
yield of 520 mgd appears to be sufficient to meet its current and ultimate build-out needs.
However, NYC continues to present over-estimations of its future needs to justify over-drafting
the Delaware Basin even though an 800 mgd over-draft is not sustainable and not deliverable,
continuously, during drought. The Hazen and Sawyer report severely over-estimated the
expected year 2000 demand and under-estimated the effect of conservation (see the “Demand
and Demand Patterns” section of this report). These projections should be updated and
appropriate demand curves used in reassessing a conjunctive, realistic, sustainable safe yield
based operating plan.

Example: Minimum Montague Flow Maintainable at 520 mgd Safe Yield Based Plan

At 520 mgd, NYC could maintain a minimum flow of approximately 1,650 cfs at Montague, as
indicated by the 1968 USGS Hardison study. However, maintaining Montague at a flat minimum
rate may not be the optimal operating option. An optimized and realistic safe yield-based
operating plan could continue a yearly, tiered, Montague flow reduction, for example a minimum
flow of 1,450 cfs, with the volumetric difference between maintaining 1,450 cfs and 1,650 cfs to
be banked and always available for release even through the drought of record for providing the
most effective augmentation of environmental flows, as needed in the tributaries in combination
with Montague. A program of multiple flow goals that include the tributary branches should
optimize the sustainable quality of the stream flows even if minimum quantities may be lower
than 1,650 cfs during drought. Once tributary flow goals are included, the volumetric difference
between maintaining 1,650 cfs and 1,450 cfs may not equate to 73 BG; it would likely be less but
it should be substantial and it should be always available, year on year, under a safe yield based
operating plan. The result of such an operating plan would likely mean that, during a repeat of
the drought of record, Montague flow would be maintained in a range between 1,450 cfs and
1,650 cfs with an average of about 1,550 cfs, but never less than 1,450 cfs. This is far greater
than the 1,100 cfs worse case scenario in the 1983 Good Faith Agreement and the FFMP. The
1,550 mgd is coincidentally very close to the original recommendation of the 1979 Task Group
that looked into a new operating plan and is equivalent to the Montague minimum accepted after
the 1931 Supreme Court Decree, before Cannonsville’s construction. The 1,750 cfs target would
be permanently changed so as not to wastefully release water to maintain unneeded high flows at
Montague in order to meet reliable conservation flows in the tributaries to Montague. This does
not mean that at times the Montague flow would not be raised to 1,750 cfs or more due to
releases by NYC; it's simply that Montague flow would include other more useful flow

32
precedents, such as temperature control, rather than a flow goal at Montague alone, that may
facilitate a lower but more effective flow.

NYC Per Capita Need, Now and for Ultimate Build-out Population-(Design Year 2110)

If the reserve capacity in the Catskill/Croton Systems were reduced to 10 %, a percentage more
consistent with current standards for water supply engineering, the Catskill/Croton & Systems’
combined safe yield may increase to at least 733 mgd. (Under an appropriate conjunctive use
reassessment, the effective combined safe yield is likely to be significantly more than 733 mgd.)
With the redesign of the Delaware reservoirs’ operations to obtain a sustainable safe yield of 520
mgd and without reducing the Rondout Reservoir’s 25 % reserve storage (about 12 BG), a
sustainable safe yield of 620 mgd from the Delaware System would bring the combined total
safe yield to a minimum of, 1,353 mgd. If the Hudson reservoirs’ reserve storage remains
unused, the combined total sustainable safe yield is approximately 1,330 mgd. At the time of
projected ultimate build-out in 2110 using the Decree design population of 12 million, NYC will
have a per capita water availability of approximately 1,353 mgd/12 million=113 gpcd or
1,330/12 million = 111 gpcd. Given the advancement in water conservation technology and the
growing focus on conservation and energy efficiency in future construction, it is far more
probable that, in less than 50 years, NYC will have reduced its per capita needs from 133 gpcd to
111 or 113 gpcd than there would be additional storage constructed; additional storage appears
not to be needed now or ever with effective conservation. NYC’s 1954 estimates of year 2000
use were about 750 mgd higher than actual. The 1989 H&S report overestimates 2010 demand
by about 280 mgd even with conservation and by about 600 mgd without it. At NYC’s current
per capita use, 133 gpcd, it could now serve 1,330/133 =10 million people or about 1 million
people more than it currently serves. At NYC’s past rate of growth, population growth of an
additional 1 million is unlikely to occur until 2060, if at all. Meanwhile the unusable yield
garnered from the over-draft policy adds to the reservoirs flood risk and is withheld from
sustaining Delaware basin flows.

Regardless, should the ultimate demand exceed the established safe yield of the NYC reservoir
systems, it is NYC’s responsibility to meet this demand and not that of the lower basin states
through reductions in releases, loss of reservoir void space and/or development of additional
storage.

Use Exceeding a Sustainable Safe Yield

If one were to interpret NYC’s average yearly use in the recent past as an indication of its future
need, NYC may at times be able use more than 520 mgd from the Delaware Basin reservoirs.
This is an operating phenomenon promoted by the over-draft plan and an artifact of diversions
during wet years wherein a reservoir may be used without danger of drought at a rate exceeding
its nominal safe yield, in this case 480 mgd, or the actual safe yield, under the over-draft policy,
which has yet to be fully assessed. An unacceptable characteristic of the over-draft policy is that
during average or dry years such an operating criterion promotes NYC’s over-drafting of the safe
yield of the Delaware reservoirs, at least if one assumes NYC will try to obtain 800 mgd (or 765

33
mgd) constantly, in order to draw the Delaware reservoirs low and trigger Montague flow release
reductions during times when the basin needs it most and NYC’s demand is at off-season levels.

Estimated Safe Yield Under Rev 1 and FFMP Over-Draft Operating Plans

Properly assessing the safe yield from the Delaware reservoirs cannot be performed
independently of Rondout Reservoir’s safe yield since Rondout is operated in series with the
Delaware basin reservoirs and acts as a governor on diversions to it. In fact, West Branch
Reservoir in the Croton System and Kensico and Hillview reservoirs also have such effects.
Since the Kensico and Hillview reservoirs are also supplied by the Catskill/Croton Systems, the
Delaware Reservoirs are also affected by the Catskill/Croton Systems’ conjunctive use. The
estimate of the safe yield of the Delaware reservoirs should not be isolated from or jeopardized
in deference to the Catskill/Croton systems; these systems are not operated by over-drafting but
are dependent on over-drafting the Delaware System and reductions of Montague flow. By
parsing the results of the OASIS Model output under the over-draft policy, one can reverse
calculate the equivalent releases needed to maintain Montague flows at adjusted storage levels
resulting from alternate diversion rates and patterns. This is an inexact method and a proper
assessment needs to be performed using a range of Montague flow objectives, diversion patterns
and the conjunctive effect of Rondout Reservoir and other Hudson Basin reservoirs. This is not
simply an issue of equitable reapportionment but of effective hydrological modeling and design,
especially relative to appropriate flood analyses. For example, Rondout reservoir, its demand, its
inflow, and its reserve storage, determines the ability for diversions to be made from Pepacton,
Cannonsville and Neversink because the rate of water entering Rondout from the Delaware
reservoirs is limited by the rate of water exiting Rondout via the Delaware Aqueduct and the
void space in Rondout as well as distribution reservoirs and West Branch Reservoir. Below is a
brief synopsis of how NYC operates its water supply systems based on the data reviewed.

• Typical NYC Water Year Operations, June 1st to May 31st, under Existing Rules:
Starting each June 1st, NYC attempts to divert water within an annual running average of
800 mgd, but remains significantly below 800 mgd on an annual average basis. Under the
rules of the Decree, NYC is allowed to, and often does, divert well over 800 mgd for several
weeks during peak demand periods while still remaining under the annual average of 800
mgd. By examining the impact of NYC’s normalized demand and other realistic hydraulic,
hydrologic, and equity based effects, NYC cannot exceed a running yearly average of 800
mgd; it is for all practical purposes, impossible to do so without under-utilization of the
Hudson reservoirs. Since an 800 mgd running average is applied starting June 1st, the last 29
days of May are typically under-used to enable averaging NYC’s initial diversions such that
NYC can exceed 800 mgd when averaged with the lower May drafts, which rarely exceed
about 400 mgd or whatever is necessary to keep Rondout Reservoir and 10 BG in the West
Branch Reservoir of the Croton System “topped off” while also supplying NYC’s mid-spring
water supply needs. By this periodic fill and draw technique, Rondout and West Branch
Reservoirs can be sufficiently charged for the peak demand season even if the NYC draft is
reduced to only 520 mgd. This can be done via conjunctive use of the Catskill/Croton
Systems, which are under-utilized (especially Croton) to conserve water to supplement the
peak demand period needs for short periods (120 days of summer) during Delaware
reservoirs’ curtailments.

34
• Current Off-Peak Season Operations: The Rev 1 and FFMP operating plans do not model
the probability that off-peak season demand reduction will enable more effective refill; that
is, NYC is able to drastically reduce diversions at such times, especially under its current
demand pattern. NYC’s normalized demand enables substantial off-peak season diversion
cutbacks of the Delaware System without needing to increase use of the Catskill/Croton
Systems. Under NYC’s normalized demand the Catskill/Croton Systems are also used less
and their off-peak seasonal use and total use are much lower than was originally anticipated
in 1983. The off-peak season’s probability of recharge is far greater than the OASIS model
shows and the reservoirs of both the Hudson and Delaware systems are often full and spilling
which exacerbates flood conditions. The assumption of a constant 800 mgd (or 765 mgd)
diversion results in underestimating the storage available in off-peak season and the
probability of refill and therefore underestimates the actual safe yield; it also results in over-
estimating void spaces and drought days and therefore is not conducive for flood mitigation
operating alternatives analysis.

• Example: Worst-Case Delaware System Operations: The following is a simplified safe

yield analysis to illustrate the increases of safe yield relative to conjunctive use and
operations and why the over-draft plan is no longer a practical representation of operating
conditions. Beginning each water year (June 1), the Rev 1 and FFMP operating plans enable
NYC to attempt a draft of 800 mgd, an over-draft of 320 mgd of the minimum safe yield
obtained by partitioning the basins, 480 mgd, when Montague is maintained at 1,750 cfs and
there is no consideration of the conjunctive use effect between the Hudson and Delaware
reservoirs. If a rainfall deficit exists and persists, a drought condition may be triggered by the
over-draft. At such times, Montague releases are reduced along with reservoir drafts and the
reservoirs begin to recover since they are drawn down at a lesser rate with both releases and
diversions curtailed. In a worst-case scenario, during peak periods of demand, having pre-
filled Rondout by each June 1st enables NYC to meet peak demand with only the 520 mgd
provided by the Delaware reservoirs as base inflow to Rondout Reservoir. This is because the
Delaware reservoirs are conjunctively used with Rondout Reservoir’s own 100 mgd
minimum safe yield, 10 BG of West Branch Reservoir in the Croton System, and under-
utilization of the Catskill Reservoirs and the remaining Croton System reservoirs. The West
Branch Reservoir effectively extends Rondout's and the Delaware System's reservoir storage
by 10 BG. Each June 1st, NYC also draws down West Branch Reservoir simultaneously with
use of the other Delaware System reservoirs. (West Branch Reservoir is described as part of
the Delaware System in a 1989 Hazen & Sawyer report.) Often, drawing down West Branch
reservoir by 10 BG is virtually the only use NYC puts to the Croton System. Assuming a
worst case average inflow to Rondout Reservoir of 20 mgd (actual June thru Sept average
inflow for 1965 was about 23 mgd) and reserving 12 BG of Rondout’s 49 BG storage and
extending that storage to 59 BG due to the use of West Branch Reservoir, that is, relying on
47 BG of the Rondout-West Branch conjunctive use system, a sustainable peak season draft
via the Delaware Aqueduct from conjunctive use for 120 days can be estimated:

(120days*520 +120 days*20 +47,000 mg)/120 days = 931 mgd.

(This still leaves 12 BG in Rondout Reservoir).

35
 However, a more realistic sustainable peak season draft from conjunctive use is closer to 90
days and at 90 days the peak sustainable draft is about:

(90days*520 mgd +90days*20 +47,000 mg)/90days= 1,062 mgd or about 112 mgd more
than the Delaware Aqueduct (950 mgd practical operating capacity) can handle, in practice.

It is apparent that NYC can amply meet its peak demands with only 520 mgd from the
Delaware reservoirs through the summer peak season and both the Rondout and Delaware
reservoirs would recover during the 8 or 9-month off-peak season when demand and
evapotranspiration wane. One can estimate the demand that can be sustained while also
assuring off-season recharge such that the total combined Delaware and Rondout reservoirs
average use does not exceed their minimum combined safe yield, 620 mgd. The 8-month off-
peak demand that will assure a reliable draft of only 520 mgd from the Delaware reservoirs
and 100 mgd from Rondout reservoir will serve all NYC's needs is about:

12mos*(620 mgd) = 4mos*931mgd+8mos*X; or X= (12*620-4*931)/8= 464.5 mgd.

The 8-month off-seasonal use of the Delaware System alone would need to average no more
than 465 mgd to ensure getting through the drought of record. To sustain a 3-month peak
draft of 950 mgd, a 9-month off-season sustainable use of up to 510 mgd can be gained via
the Delaware Aqueduct. Recent year-round use has averaged around 500 mgd from the
Delaware reservoirs, alone. An off-season average is typically about 65 to 75 % of the yearly
average so that an off-season average of about 350 mgd from the Delaware reservoirs and 70
mgd from Rondout gives a Delaware System expected off-season sustainable draft of 420
mgd. Therefore, the existing worst case scenario exceeds NYC's likely sustainable need. By
inspection of NYC's per capita use data and diversion patterns, its ultimate build-out needs,
will average significantly less than 465 mgd. In this way, NYC’s peak demands can be met
consistently, every year, through a repeat of the drought of record. NYC's peak-month
average drafts via the Delaware Aqueduct can easily range to 860 mgd, which appears to be
greater than or equal to its ultimate and sustainable peak month need. Peak days of 950 mgd
via the Delaware Aqueduct can also be easily accommodated. The fact that NYC has more
storage than it can practically use contributes to frequently unnecessarily full and surcharged
reservoirs that exacerbate flooding.

• Trial Estimate of Safe Yield of FFMP & Previous Operating Plans: The NYS
Department of Health study from 1967, the USGS study of 1968, the ACE study of 1974, the
Hydrologics Incorporated’s OASIS model and the Daily Flow Model simulate operating
programs that are not based on appropriate safe yield-based operating criteria, conjunctive
use effects and boundary conditions applicable to the 1960’s design drought. Previous
analyses each partitioned the basins and assumed Hudson System yields were “off limits”
simply because these reservoirs were not under DRBC jurisdiction. No assessment of the safe
yields of the Hudson reservoirs appears to have ever been done and only the nominal safe
yields provided by NYC have been used, which continued the 25 % reserve storage despite
the lower basin Parties having agreed to lower Montague flow targets. The NJDEP
performed preliminary estimates of NYC’s safe yield under the current over-draft operating
plans, with and without consideration of the effect of Rondout and West Branch reservoirs.

36
 Such analyses are affected by one over-arching operating criterion: whether or not a year-
round, flat rate, demand pattern or a seasonal peak demand pattern is applied. For a flat rate
diversion pattern the lowest range of possible safe yields obtained by the NJDEP exceed the
480 mgd now assumed in the FFMP. An effective minimum safe yield of about 590 to 595
mgd was obtained when attempting to draft starting June 1st 1961, at a 765 mgd flat rate
from the Delaware reservoirs and about 695 mgd was obtained from the Delaware System,
including Rondout and West Branch reservoirs while still maintaining about 12 BG reserve
storage in Rondout. The diversion from Rondout, however, is not limited to the tiered draft
schedule relative to Delaware Basin storage so that, at an annual average safe yield of 690
mgd, NYC can easily serve peak season needs of up to 950 mgd from the Delaware System
even in a repeat of the drought of record. During such times, the Delaware Basin would be in
a much more difficult situation with low Montague flow inducing lower basin storage losses
for make-up releases and salinity control.

E. Other Projects

New Storage

Several alternate storage projects have been discussed, most prominently, the F.E. Walter and
Prompton Reservoir enlargements for Trenton flow maintenance and the Cannonsville Reservoir
increase of 13 BG conjectured in the FFMP. Recently, the owners of the Mongaup Reservoir
system have made overtures to the Decree Parties regarding possible purchase of storage
capacity in their system. The following is a brief synopsis of the NJDEP’s assessment of the
issues related to these projects.

• Cannonsville Reservoir Increase: NYC and NYS have projected a 13 billion gallon storage
increase for Cannonsville Reservoir and incorporated the affect into the FFMP by modeling a
surrogate 35 mgd reduction in the over-draft, reduced from 800 mgd to 765 mgd on a
temporary basis. It is temporary because NYC expects that the 800 mgd over-draft will be
needed if the 13 BG of additional storage is not built. But since NYC does not have a
projected need for this additional storage in the foreseeable future, the projection of future
construction of an additional 13 BG of storage by NYC serves only to obscure the actual
system operating criteria; a 765 mgd over-draft results in virtually no reduction in NYC’s
actual safe yield and has little relevance to their actual operations; 765 mgd is still far greater
than NYC will ever need or be able to divert given its normalized demand and is just as
effective a trigger of Montague flow reductions. In effect, the increased fishery releases are
not coming from NYC’s safe yield or its 800 mgd but from Delaware Basin flow reductions.
In the event of a developing drought, additional fishery releases only enable NYC to draw
down the reservoirs to trigger Montague flow reductions earlier and at a faster rate. If a
drought ensues, the fisheries suffer more and sooner. Certainly it is NYS’ prerogative to
fund storage for fisheries releases, however, such releases could largely, if not fully, be
achieved through simply reverting to a safe yield-based reservoir operating plan, rather than
one based on over-drafting.

37
• F. E. Walter & Prompton Reservoir Projects: If an accurate, objective safe yield-based
operating plan is developed, it is unlikely that the lower basin will need to enlarge F. E.
Walter or Prompton Reservoirs. F. E. Walter is rated for 22.5 BG/year; Prompton for 10
BG/year. This is estimated as providing, during a 120 day drought period each year; (22.5 +
10)/120*1000= 271 mgd or about 420 cfs. This is roughly equivalent to raising the minimum
flow at Trenton by 420 cfs. The NJDEP estimate of what can be maintained at Montague
under a safe yield based operating plan is 1,640 cfs, which is roughly equivalent to raising
Trenton by 540 cfs. As stated previously, NJDEP is amenable to a more effective tiered
Montague flow objective that is lower than 1,640 cfs for part of the year, but significantly
higher than 1100 cfs. However, no matter what the minimum flow would be at Montague
under such a program, it would certainly obviate the need for the F. E. Walter and Prompton
Reservoir projects.

• Mongaup System Purchase of Storage Capacity: Until such time as the Delaware Basin
reservoir system’s safe yield is properly assessed, and an operating plan is developed based
on safe yield, there appears to be little basis for purchasing additional storage by the lower
basin. As described herein, NYC has sufficient capacity in the Delaware Reservoirs to meet
its ultimate water supply demand. However, a purchase of Mongaup System storage by
NYC may facilitate meeting its Montague flow obligations and allow for more effective
flood management. If NYC did not currently have sufficient safe yield it would certainly be
much more efficient and cost effective for NYC to purchase storage from the Mongaup
System. Prior to establishing a safe yield-based operating plan derived from more realistic
operating assumptions, a purchase of Mongaup storage capacity by the lower basin Parties
would result in the lower basin Parties essentially expending funds to re-acquire water
inappropriately reapportioned toward NYC’s water supply system by the Montague flow
reductions of the 1983 Good Faith Agreement.

New Aqueducts and Connection of Delaware-Catskill Aqueducts

NYC’s potential construction of a bypass tunnel around Rondout Reservoir will not provide
additional yield nor will it improve the hydrologic limits of the 1960’s drought and the peak
carrying capacity of the Delaware Aqueduct. It will only improve the ability for NYC to
maintain peak flow in the Delaware Aqueduct for greater duration since the drawing down of
Rondout Reservoir’s hydraulic head will no longer limit NYC’s peak draft period. Improving the
hydraulic head to the Delaware Aqueduct from the three Delaware reservoirs will not increase
the safe yield or NYC's demand. Safe yield is limited by other factors, such as the demand
pattern, distribution reservoir capacity and, ultimately the hydrology, not NYC’s peak carrying
capacity. Even if constructed it is likely to help the lower basin as much or more than NYC
under an appropriate safe yield based operating plan because the more efficiently, reliably and
flexibly NYC can take its peak demand, the better the ability to meet its water supply needs as
well as its release obligations.

Connecting the Delaware Aqueduct to the Catskill Aqueduct will enable NYC to over-draft the
Delaware System more efficiently by using the Delaware Aqueduct to displace combined
diversions during off-peak days, weeks, seasons and months when the Catskill System can be

38
'rested'. This is not necessarily a detriment in an appropriate safe yield based program if modeled
such that an equitable and sustainable safe yield can be maintained. It would allow for NYC to
blend the higher quality Delaware basin waters with the Catskill waters especially during turbid
water periods. Such periods typically occur during major rainfall events when it is likely periodic
peak Delaware basin water use is offset by precipitation. However, the Parties should have a
transparent, total combined systems conjunctive use model to establish what is sustainable and
equitable; this requires an appropriate reassessment study.

Related Issue

Hydroelectric Power

In 2008, an independent hydroelectric power project was proposed to the Federal Energy
Regulatory Commission (FERC) by the Delaware County Cooperative (DCC). DCC proposed to
add to the three NYC Delaware Basin reservoirs siphons with hydroelectric turbines that would
likely enable “high flow skimming” of the reservoirs when they are near full or spilling, and
generate power during release periods. This project presented an opportunity to improve the
release capacity of the three reservoirs, which might have significantly improved the ability to
make pre-flood draw downs, potentially with no construction cost to the NYC. NYC submitted
its own counter-proposal, a hydroelectric project that eliminated the potential for effective
release improvements.

Further, modification to major reservoirs within the jurisdiction of the DRBC is a reviewable
project. The DRBC could require any proposed modifications to NYC’s Delaware Basin
reservoirs to improve or add release facilities to enable effective mitigation of the unnecessary
flood risks associated with the over-draft plan and deny any reservoir modifications until such
time as an appropriate safe yield based operating plan is developed. The NJDEP is unaware if the
DRBC has yet commented on NYC’s proposal or considered the opportunities for improvements
in the release facilities despite having urged the DRBC to do so. In addition to DRBC
reviewability, the proposed hydroelectric project affects the interests of the Decree Parties, and
as such will ultimately require unanimous approval by the parties.

F. Preliminary Supporting Example: Graphs for Delaware System Operations with Over-
Draft Compared to Safe Yield-Based Operations

Figures 1 through 3

The NJDEP modeled operation of the Delaware System including inflow to Rondout Reservoir
from its 95 square mile drainage area while meeting the minimum releases required. Rondout
Reservoir inflow and operating data was obtained from USGS flow records and NYCDEP
drought operation plans. The modeling was derived from deconstructing the releases relative to
diversions generated under the OASIS Model in running the FFMP. When evaluating a lesser
average diversion, increases in releases were made, accordingly, to offset the difference between
releases under the FFMP and those possible under alternate lesser diversions using existing rule

39
curves. While the NJDEP model is not precise, it supports the conclusion that NYC may operate
to attain a safe yield substantially greater than 480 mgd, though remaining less than either 800
mgd or 765 mgd. NYC’s use records indicate that this is their mode of operation and supports the
fact that reducing the theoretical over-draft from 800 mgd to 765 mgd is not a reduction in
NYC’s effective safe yield. The NJDEP graphed the Rondout Reservoir storage levels through
the drought of the 1960’s for both the FFMP and a safe yield based operating plan using 520
mgd from the Delaware reservoirs. Under the FFMP, an effective safe yield of about 595 mgd
could be obtained from the Delaware Reservoirs when considered the conjunctive effect of
Rondout Reservoir. The NJDEP recommends initiating a safe yield based operating plan by
testing a variable demand pattern for an annual average safe yield starting at 415 mgd under
NYC’s normalized demand pattern and increasing the average to up to 520 mgd. This range of
average annual safe yields appears to be the optimal boundary under a varying seasonal demand
pattern and may effectively meet NYC’s peak and off peak seasonal needs immediately, far into
the future and with conservation, until ultimate build-out.

• Figure 1: Comparing a Safe Yield Based Diversion of 520 mgd, for a Total
Conjunctive Safe Yield of 620 mgd, to an Estimated 695 mgd Safe Yield under the
FFMP: The heavy black line shows the storage hydrograph for Rondout Reservoir (the
Delaware System) with the application of a safe yield-based diversion pattern simulating
NYC’s demand as shown in Figure 1a. The inflow from the Delaware reservoirs was set
at 520 mgd, but the drafts from Rondout Reservoir, shown in the black dashed line, were
seasonally varied as per the demand curve between 855 mgd and 484mgd. By
considering Rondout watershed’s inflow and relying on an annual average diversion of
only 520 mgd incoming from the Delaware reservoirs, Rondout Reservoir was modeled
to derive a seasonally varying safe yield via Delaware Aqueduct diversions. This
operating scenario also conservatively preserves the 25 % (12.4 BG) that NYC omits
from Rondout Reservoir’s safe yield assessment. (With conjunctive use, the diversion
average from Rondout is actually closer to 630 mgd because the added efficiency of
multiple reservoirs tends to increase the combined safe yield greater than the sum of the
individual yields.) The peak month diversion averages 855 mgd, and the peak, 3-month
diversion, averages 830 mgd. This diversion pattern and safe yield-based operating plan
exceeds NYC’s need under its current system limits and use characteristics, and it
expected to its ultimate future build-out.

Also in Figure 1, two gray lines represent the NJDEP’s trial estimate of the FFMP’s
effective safe yield. A diversion pattern, shown in a gray dashed line, allowing 695 mgd
from the Delaware System, which given that Rondout Reservoir’s minimum safe yield
with 12 BG of reserve storage is 100 mgd, develops an effective safe yield of, 695 mgd-
100 mgd = 595 mgd from the Delaware reservoirs alone. The heavy gray storage curve
indicates the Delaware System under the FFMP and shows that, in a repeat of the 1960’s
drought, only in one year, 1965 does NYC take Rondout Reservoir to 12 BG of reserve
storage; in the other years the Delaware System recovers to full or nearly full levels.
Under the over-draft policy, NYC’s drought risk is reduced to a critical duration of about
1.67 years, beginning around March, 1964. Under a safe yield based program the critical
duration would be about 4.33 years, beginning about December, 1961. At first glance, the

40
over-draft policy may appear superior if one were to judge its effectiveness in reducing
NYC’s drought duration but it actually shows several years of wasted storage and
spillage. While the black line represents a longer drought duration, it also represents a
more efficient and equitable use of storage for maintaining releases. The difference in
the storage between the curves is the water that should be released to the lower basin if
NYC were to operate under an appropriate safe yield based plan. Therefore the over-draft
policy enables NYC to minimize its risk and maximize the lower basin’s risk of drought
emergencies. Ironically, it appears NYC does not and can not operate to take water as
modeled under the over-draft policy. While the model develops thousands of “drought
days” over the period of record, NYC’s normalized operations would not actually
generate such days. The actual water that would be available for flow and fishery
maintenance goes unrecognized and is therefore unused, when in fact it is available.

Further, since NYC is actually constrained by its hydraulic limits and demand pattern
characteristics to operate at less than 520 mgd, on average, the Delaware reservoirs are
near-full, full or surcharged with greater frequency and for longer duration than the
OASIS Model’s FFMP plan operating criteria would show. Hence, the OASIS program
generates the appearance of void spaces that may no longer be possible and cannot be
used to reliably evaluate flood mitigation opportunities. Under the FFMP, the
consequences of these unrealistic operating assumptions have already been seen with
temperature and fishery releases being inadequate at times even during above average
rainfall years and dismissal of the genuine flood mitigation opportunities indicated by the
results of the recent USGS/ACE/DRBC flood study.

41
 60,000 1000
BLACK SOLID & DASHED LINES : SY Draft = 620 mgd for Del System; 520 mgd for Del Reservoirs, alone.
Peak Mo = 855 mgd; Peak 3-Mo= 830 mgd; Off Peak 9-Mo Avg = 565 mgd; Min Mo = 484 mgd
900
Peak-Mo to Avg Ratio = 1.38 to 1
50,000
800

700
40,000

Diversion (mgd)
600

30,000 500
Volume (mg)

400
20,000
300
Both Methods Reserve 25 % Storage (12.4 B G)
200
10,000 FFMP Over-Draft vs. Safe Yield-Based Plan
Storage under FFMP in Gray: Del Sys SY=695 mgd (595 Del; 100 Rondout) 100
Safe Yield-Based in Black: Del Sys SY =620 mgd (520 Del Res; 100 Rondout)
0 0
J-61

J-62

J-63

J-64

J-65

J-66

J-67

J-68
Ro Drought Watch Ro Drought Warning
Ro Drought Emergency FFMP Over-draft; SY=695 mgd
NJDEP Recomm'd Rondout SY = 620 mgd FFMP Over-draft Diverison; SY=695 mgd
NJDEP Recomm'd Div Pattern for Del Res's=620 mgd DeRes&[Link]

Figure 1 Two Rondout Reservoir storage curves (Delaware System): Heavy gray line is storage
resulting from a Delaware System safe yield of 695 mgd, with 595 mgd provided by the Delaware
reservoirs alone, attainable under the FFMP operating rules. The heavy black line is storage
resulting from a total safe yield of about 620 mgd, which includes adjusting for greater releases to
maintain Montague.

• Figure 1a: A Sample of an Optimized Safe Yield-Based Diversion Pattern: Figure 1a

shows a yearly diversion pattern for the Delaware System (via Rondout Reservoir),
which allows for peak daily diversions of up to 940 mgd throughout the drought of
record. While the running average monthly diversions are restricted to reflect the
appropriate seasonal demand, the demand pattern works to allow more diversion when
needed, during peak season demands. Under this scenario NYC could meet its ultimate
demand more efficiently because the diversion pattern tends to optimize reservoir
storage. Under a lesser annual average safe yield based plan, NYC should be able to
reliably maintain a greater and more purposeful flow regime at Montague to meet the
intents and purposes of the Decree to protect down-basin water resources and uses. The
Figure 1a diversion pattern is based on a truncated Mandelbrot curve and could be
converted to average monthly blocks or some alternate probabilistically-based schedule.
It is not the recommended pattern but only a preliminary example for illustrative
purposes and represents the kind of demand curve that should be used in determining
NYC’s safe yield relative to alternative Montague releases. This curve or any diversion
simulation curve is first created as a dimensionless “Unit Curve” and used to test various

42
 safe yield based diversions by multiplying the annual average, say 620 mgd for the draft
upon Rondout reservoir, by the Unit Curve, to create the daily rate of draft.

1000

900

800
PEAK DIVERSION AT
ANY TIME LIMITED
700 TO 940 MGD

600

500 620 mgd

Optimized Safe Yield Based
400
Diversion Curve
300
(Truncated Mandelbrot Curve)
from Rondout Reservoir via Delaware Aqueduct
MGD

200 (520 mgd from Delaware Reservoirs; 100 from Rondout)

100 Monthly Average (mgd) Annual Constant Avg 620 mgd

Daily Diversion (mgd) Daily Average over Month (mgd)
0
J F M A M J J A S O N D
Figure 1a Diversion pattern for 520 mgd from Delaware reservoirs (620 mgd Delaware System)
that envelopes the 520 mgd draft used in Figure 1. The pattern allows for great flexibility within the
average monthly limits and better simulates actual demand patterns to gage NYC’s current and
ultimate build-out need.

• Figure 2 Ten-Year Record of Rondout Daily Diversions, 1998-2007: Figure 2 shows

10 years of daily diversions from Rondout reservoir in gray and the corresponding 30-day
running average in heavy black.

43
 1000

900

800

700

600

500

400

300

200
Actual Daily Rondout Diversions, 1998 thru 2007,
100 with 30-Day Running Average
mgd

0
Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07

Actual Daily Diversion Record, Rondout Reservoir (mgd)

30-day Running Avg of Daily Diversion Record (mgd)

Figure 2 Ten-year daily diversion record from Rondout Reservoir via Delaware Aqueduct. Gray
line is the daily diversion and black line is the 30-day running average. Note NYC’s reduction of
peak month diversion rate from 940 mgd to about 860 mgd since 2002.

• Figure 3 Ten-Year Record of Rondout Daily Diversions, 1998-2007 w/Model

Diversion Curve Superimposed: Figure 3 shows, in gray, the same 30-day running
average diversion record shown in black in Figure 2 and plotted against a sample
projected demand curve that was used for the NJDEP safe yield estimate of the Delaware
System (shown in a heavy black line). The example safe yield based demand curve is not
the same curve used in Figure 1a but was similarly developed by analysis of the actual
10-year record of Rondout Reservoir diversions, weighed against the effect of current
demand pattern characteristics and system constraints. An allowance for an off-season
diversion spike to recharge Rondout is incorporated in the curve. The example curve
reasonably envelops the actual diversion pattern during critical years such as 1998, 2001,
2002 and 2006.

44
Figure 3 Comparison of the 30-day running average diversion record (gray line) from Rondout
Reservoir (shown in Figure 2 as black line) with modeled 30-day pattern for a Delaware System
safe yield of 620 mgd (black line).

G. Conclusion

If the lower basin states are to continue allowing reduction of Montague flows, a new, safe yield-
based operating plan must be developed to enable optimal, equitable and realistic management of
the Delaware basin reservoirs to meet a set of prioritized objectives of public water supply,
fisheries maintenance and flood risk reduction for both the upper and lower basin, perhaps
including the Hudson Basin, balanced against sustaining NYC’s realistic current and ultimate
water supply needs. The ongoing practice of making reservoir management (e.g. release)
decisions based on assumptions that are not reflective of current or future conditions and
limitations is arbitrary and inconsistent with the intent of the Decree. Preliminary analyses
provided herein indicate that a full reassessment of NYC reservoir systems’ safe yields and
operations has the potential to provide opportunities to substantially improve our ability to meet
multi-objectives in the basin. This would entail effective, transparent and collaborative
engineering analyses, including a conjunctive analysis of NYC’s total combined Delaware and
Hudson basin systems applying the same accepted standards, practices and principles of
professional hydrologic, hydraulic and water supply engineering the Decree is based upon.
Applying these principles in a safe yield based operating plan would reduce the need for the
many short-term fixes and ad-hoc decisions required of the principals to meet constantly
changing conditions. Such an effort is likely to require an independent academic institute
practiced in appropriate reservoir operations design. New Jersey is ready to cooperate in a fully
transparent process to aid NYC in any genuine safe yield based operating alternative that is
equitable, realistic and sustainable.

45
The principle of equitable reapportionment should be applied to establish a safe yield based
operating plan corresponding to adjusted, optimized Montague flows and NYC water supply
needs with a balanced, realistic conjunctive model of NYC’s Hudson System to establish any
appropriate ERQ. The ERQ was designed to prevent NYC from over-drafting the Delaware
reservoirs; the Good Faith Agreement essentially encourages it. Thus the DRBC Water Code,
which established a policy "to discourage the exportation of water from the Delaware River
Basin" needs to be applied. (DRBC Water Code; Importations and Exportations of Water, 2.30.2,
Policy of Protection and Preservation.) An accurate and flexible basin model that includes the
conjunctive affect of Rondout and West Branch reservoirs and the Hudson System on the
Delaware reservoirs’ effective safe yield should be generated to provided a clear understanding
of the risks to lower basin water supplies, the regional economy, natural resources and flooding.
Accurate Rondout Reservoir operations are imperative to include in a conjunctive use analysis
given Rondout’s potential to prevent the creation of meaningful reservoir voids in the Delaware
reservoirs and to prevent surcharge periods from being detected.

From such a model, the Decree Parties may begin to construct an operating plan based on
recognized and established standards for optimizing reservoir operations and gain an accurate
understanding of the affects of such optimizing on drought and flood. A watershed basin that is
largely dependent on the operation of water supply reservoirs needs to develop its protocols from
a properly developed safe yield based plan, not from a series of ad-hoc policies. New Jersey is
not inclined to support a Montague flow target reduction as long as an inequitable, unrealistic,
non safe yield-based operating plan is applied, a plan predicated on over-drafting the Delaware
Basin reservoirs beyond what is sustainable. Analyses pursued in the context of the Delaware
Basin’s role in relation to longer term issues such as salinity repulsion, climate change, long term
flood risk and sea level rise, must incorporate safe-yield based operating scenarios and not serve
as justification to continue having the Delaware Basin provide “overdraft protection” to NYC.

Sources/References/Bibliography

1. Acken, Howard W. and Hardman, A. M., “Use of Delaware River Water through the
Delaware and Raritan Canal,” Proceedings of the American Society of Civil Engineers, N.Y.
N.Y., June, 1954, Vol. 80, Separate No. 452.

2. Archfield, Stacey A. and Vogel, Richard M. “Reliability of Reservoir Firm Yield Determined
from the Historical Drought of Record,” Proceedings of the American Society of Civil
Engineers World Environmental and Water Resources Congress, Honolulu, HI, May, 2008.

3. Austen, Douglas J. PhD., “Proposed Flexible Flow Management Plan and DRBC Water
Code Amendments,” Letter of Executive Director Pennsylvania Fish and Boat Commission,
September 19, 2008.

4. Brooks, Lloyd T. “Flood of September 18-19, 2004 in the Upper Delaware River Basin,
New York,” U.S. Geological Survey Open-File Report 2005-1166, 2005.

46
5. Camp Dresser & McKee, Inc., “Daily Flow Model of the Delaware River Basin,” prepared
for U.S. Army Corps of Engineers in cooperation with PA Dept. of Environmental Resources,
NY Dept. of Environmental Conservation, the U.S. Geological Survey and the DRBC, 1981.

6. Chaves, P. T., Kojiri, T. and Yamashiki, Y., “Optimization of Storage Reservoir Considering
Water Quantity and Quality,” Hydrol Processes, 17(14), 2769–2793. 2003.

7. Chaves, Paulo, Kojiri, Toshiharu and Hori, Tomoharu, “Stochastic Operation for Multi-
purpose Reservoir Using Neuro-Fuzzy Systems,” Annals of Disaster Prevention Research
Institute, Kyoto Univ., No. 48 B, 2005.

8. Chess, C., Hance, B.J. and Gibson, G., “Adaptive Participation in Watershed Management,”
Journal of Soil and Water Conservation, 3rd Qrtr. pp.248-252, 2000.

9. City of New York, Department of Environmental Protection “Drought Management Plan and
Rules,” Dec 29, 1998.

10. City of New York “Mayor’s Intergovernmental Task Force on New York City Water Supply
Needs,” 1992.

11. City of New York, Department of Environmental Protection, “Motion to Intervene,

Competing Application for Preliminary Permit of City of New York under P-13287,”
Application for Federal Energy Regulatory Commission License for new hydroelectric
facilities, October 10, 2008.

12. City of New York, Department of Environmental Protection, “Request for Proposals for
Contract: CAT-330, Catskill Turbidity Control Study (CTS) Design Services for the
Development of an Operations Support Tool (OST),” Published by USGS Delaware River
Master, October 10, 2008.

13. Clark, Edward J. (Chief Engineer, NYC Bureau of Water Supply, Dept of Water Supply, Gas
and Electric), “New York Control Curves,” 1950, Journal of American Water Works
Association, Vol. 42. No. 9, pages 823-827.

14. Clark, Edward J. (Chief Engineer, NYC Bureau of Water Supply, Dept of Water Supply, Gas
and Electric), “Impounding Reservoir,” 1956, Journal of American Water Works Association,
pages 349-354.

15. Committee to Review the New York City Watershed Management Strategy, National
Research Council, “Watershed Management for Potable Water Supply: Assessing the New
York City Strategy,” 2000.

16. Delaware County Electric Cooperative, Federal Energy Regulatory Commission (FERC),
“Notification of Intent and Pre-Application Document, Western Catskills Hydro Project,”
FERC public notice, May 8, 2008.

47
17. Delaware River Basin Commission Flow Management Technical Advisory Committee,
“Interim Report of the Technical Responses to “Good Faith” Recommendations #3 and #4,”
June 10, 1983.

18. Delaware River Basin Commission, “Water Code,” Comprehensive Plan Projects, 2007.

19. Delaware River Basin Commission, “Operational Analysis and Simulation of Integrated
Systems (OASIS) Model” prepared by HydroLogics, Inc., 2002.

20. Delaware, New Jersey, New York City, New York State and Pennsylvania, “Good Faith
Agreement,” 1983.

21. Department of Atmospheric Science, NY State University at Albany, “The 1980-81 Drought
in Southeastern New York,” July, 1985.

22. Dittmann, R., Froehlich F., Pohl, R. and Ostrowski, M., “Optimum Multi-Objective
Reservoir Operation with Emphasis on Flood Control and Ecology,” Nat. Hazards Earth Syst.
Sci., 9, 1973–1980, 2009.

23. Draper, Andrew James, “Implicit Stochastic Optimization with Limited Foresight for
Reservoir Systems,” (B. A. University of Cambridge, UK 1978: [Link]., University of
Southampton, UK), Dissertation submitted in partial satisfaction of the requirements for the
degree of Doctor of Philosophy in Engineering in the Office of Graduate Studies of the
University of California, Davis, 1985.

24. Dysarz, Tomasz, "Control of the Reservoirs System during Flood: Concept of Learning in
Multi-Stage Decision Process," Interim report IR-05-031, International Institute for Applied
Systems Analysis, Austria, June, 2005.

25. Effler, S. W., Gelda, R. K., Johnson, D. L. and Owens, E. M., “Sediment Resuspension in
Cannonsville Reservoir,” Jrnl. of Lake and Reservoir Management, 14 (2-3):225-237, 1998.

26. Fill, Heinz Dieter and Steiner, Alexandre Arns, “Estimating Instantaneous peak Flow from
Mean Daily Flow Data,” Journal of hydrological Engineering ASCE, Nov/Dec, pp. 365-369,
2003.

27. Fu, Alec Y. “Summary of the Hydraulic Aspects of the Catskill/Delaware Aqueduct
System,” New York City Dept of Environmental Protection, Bureau of Water Supply and
Wastewater Collection, Division of Strategic Services, March, 1994.

28. Goldin, Harrison J., Comptroller, “Waste Not, Want Not: Managing New York City’s Water
Supply,” Office of Policy management, Office of the Comptroller, The City of New York, 62
pp. July, 1989.

29. Goode, Daniel J., Koerkle, Edward H., Hoffman, Scott A., Hay, Lauren E. and Markstrom,
Steven L., “Simulation of Runoff and Reservoir Inflow for Use in a Flood-Analysis Model

48
 for the Delaware River, Pennsylvania , New Jersey, and New York, 2004-2006” USGS
Open-File Report 2010-1014, Feb., 2010.

30. Gorokhovich, Yuri, and Goldsmith, Victor, “Water Supply Management of the New York
Metropolitan Area and Future Challenges from Population Growth and Global Climate
Change,” International Assoc of hydrological Sciences Publ. 327, 2009.

31. Hardison, Clayton. H., “Probability Analysis of Allowable Yield of New York City
Reservoirs in the Delaware River Basin,” United States Geological Survey, 1968; sponsored
by the DRBC.

32. Hazen and Sawyer, Engineers “Delaware-Lower Hudson Region Water Resources
Management Study,” prepared for NYSDEC, Sept, 1989.

33. Hazen and Sawyer, Engineers “Study of Water Demands on New York City System Final
Report,” NYSDEC and NYCDEP, Nov, 1989.

34. Hochlerner, Tobias, “New York City’s Water Supply to Outside Communities,” Journal of
the American Water Works Association, August, 1945, pages 754-764.

35. Hogarty, Richard A. “The Delaware River Drought Emergency,” Inter-University Case
Program # 107, Bobbs-Merrill Co. Inc., College Division, 1970.

36. Homa, E. S., Vogel, R. M., Smith, M. P. Apse, C. D., Huber-Lee, A. and Sieber, J., “An
Optimization Approach for Balancing Human and Ecological Flow Needs’” Proceedings of
the EWRI Woprld Water and Environmental Resources Congress, ASCE, Alaska, 2005.

37. Hughes, D. A. and Ziervogel, G., “The Inclusion of Operating Rules in a Daily Reservoir
Simulation Model to Determine Ecological Reserve Releases for River Maintenance,” Water
S.A., 24(4), 293–302, 1998.

38. Kennison, Karl R. (Chief Engineer NYC Board of Water Supply), “Notes of Chief Engineer
Kennison on Provisions of Decree June 7, 1954 and Reports and Memoranda by Witnesses
not called to Testify,” Feb. 10, 1955, Delaware River Basin Library.

39. Kennison, Karl R. (Chief Engineer, NYC Board of Water Supply), “Problems Which New
York City Faces in Expanding its Water Supply Sources,” Journal of the Boston Society of
Civil Engineers, Vol.354, 1954, p 354-375.

40. Kho, F. W. L., Law, P. L., Lai, S. H., Oon, Y. W., Ngu, L. H. and Ting, H. S., “Quantitative
Dam Break Analysis on a Reservoir Earth Dam,” Int. J. Environ. Sci. Tech., 6 (2), 203-210,
Spring, 2009.

41. Kim, Taesoon Heo, Jun-Haeng, Bae, Deg-Hyo and Kim, Jin-Hoon, “Single-reservoir
Operating Rules for a Year Using Multi-objective Genetic Algorithm,” Journal of
Hydroinformatics, 10.2, pp. 163-179, 2008.

49
42. Klipsch, Hurst and Fleming, “Delaware River Basin Flood Analysis Model Reservoir
Operations and Streamflow Routing Component’” Report PR-73, Feb, US Army Corps of
Engineers, Hydrologic Engineering Center, Davis, CA. 2010.

43. Labadie, John W., “Optimal Operation of Multi-reservoir Systems: State-of-the-Art

Review,” Journal of Water Resources Planning and management, Amer. Soc. of Civil Eng.,
Mar/Apr, 2004 pp 93-111.

44. Laedlein, Mark and Mayer, Robert A., “Reservoir System Analysis model Users manual and
Program Documentation,” New York City Dept of Environmental Protection, 1983, revised
1985 and 1993.
45. Leurig, Sharlene, "The Ripple Effect: Water Risk in the Municiple Bond Market" A Ceres
Report prepared by Price Waterhouse Coopers International, Limited, 98 pp., Oct, 2010.

46. Lumia, Richard, “Flood of January 19-29, 1996, in New York State,” USGS Water
Resources Investigations Report No. 97-4252, 61 pp., 1998.

47. Lund, J. R, & Guzman, J., “Some Derived Operating Rules for Reservoirs in Series or in
Parallel,” Journal of Water Resources Planning and Management, Vol. 125, No. 3, pp. 143-
153, May/June 1999.

48. Nishida, S., Palmer, R., Slaughterbeck, C. and Walker, S., “An Expert System for Control
Curve Evaluation During Drought,” Proc. of 17th Annual [Link]., Water Resources
Planning and Management Div., ASCE, Ft. Worth, Texas, pp. 294-297, April, 1990.

49. New York State Department of Health Metcalf & Eddy, Hazen & Sawyer, and Malcolm
Pirney, Engineers, “Comprehensive Public Water Supply Study for the City of New York
and County of Westchester,” 1967.

50. New York State Dept. of Environmental Conservation & New York State Dept. of Health,
“water Resources Management Strategy,” Jan., 1989

51. Office of the City of New York Comptroller, “Report of the Comptroller on New York City
Water Supply,” 1989.

52. Owens, Emmet M., “Observation and Simulation of Surface Waves in Two Water Supply
Reservoirs,” Journal of Hydraulic Engineering, Vol. 135, No. 8, Aug 1, 2009, pp. 663-670.

53. Owens, Emmet M., Gelda, Rakesh K., Effler, Steven W. and Hassett, John M. “Hydrologic
Analysis and Model Development for Cannonsville Reservoir,” Journal of Lake and
Reservoir Management, 14(2-3):140-151, 1998.

54. Palmer, Richard N and Fisher, Selene “Managing Water Supplies during Drought: Triggers
for Operational Responses,” Joint Paper of Kramer, Chin Mayo, Consulting Engineers and
University of Washington, 1997.

50
55. Palmer, Richard M., Kersnar, Joan M. and Choi, Dora “Estimating Demand Variability,”
Integrated Water Resources Planning for the 21st Century, Proceedings of the 22nd Annual
Conference, Cambridge, Massachusetts, May 7-11, 1995. (Editor Michael F. Domenica)

56. Palmer, Richard M. and Snyder, Regina M. “Effects of In-stream Flow Requirements on
Water Supply Reliability,” Water Resources Research, Vol. 21, No 4, pp 439-446, 1985.

57. Panzer, Kurt F. (Special Master, U.S. Supreme Court), “Report of the Special Master” May
27, 1954.

58. Perky, Donald J. Young, Kevin N. and Kreitzberg, Carl W., “The 1980-81 Drought in
Eastern Pennsylvania,” Department of Physics and Atmospheric Science, Drexel University,
Phila, PA., 1983.

59. Quirk, Lawler & Matusky, Engineers “Northeastern United States Water Supply Study:
Hydraulic Analysis of the New York City Water Supply System,” for Northeastern United
States Water Supply Study, sponsored by US Army Corps of Engineers, 1974.

60. Raheem, Enayetur and Sekander, Hayat Kim, Ph.D., “Combining Probability of Emptiness
and Mean First Overflow Time of a Dam to Determine Its Capacity,” Journal of Spatial
Hydrology, Vol. 2, No. 2., 7 pp. Fall, 2002.

61. Riggs, H. C. and Hardison, Clayton H., “Storage Analysis for Water Supply,” Chapter B@,
Techniques of Water-resources Investigations of the U.S. Geological Survey, Book 4, Circa
1973, (Reprinted 1983.)

62. Schumann, A. and Sieber, H.-U., "Expectations and reality about the Role of Dams for Flood
Control-Experience of the Extreme Flood of August, 2002, Saxony," Dams and Reservoirs,
Societies and Environment in the 21st Century: Proceedings of the International Symposium
on Dams in the Societies of the 21st Century, Dams (ICOLD), Barcelona, Spain, 18 June
2006 by Luis Berga, J.M. Buil, E. Bofill, and J.C. De Cea (Hardcover - May 22, 2006)

63. Supreme Court of the United States, “New Jersey v. New York, Decree of 1931.”

64. Supreme Court of the United States, “New Jersey v. New York, Amended Decree of 1954.”

65. Supreme Court of the United States, New Jersey v. New York, 345 US 369 (1953) “On
Motion of the City of Philadelphia for Leave to Intervene,” Denial, April 6, 1953.

66. Suro, Thomas P. and Firda, Gary D. “Flood of April 2–3, 2005, Neversink River Basin, New
York,” USGS Open-File Report # 2006-1319, 98 pp., 2006.

67. Suro, Thomas P., Firda, Gary D and Szabo,Carolyn O. “Flood of June 26-29, 2006, Mohawk,
Delaware, and Susquehanna River Basins, New York 2006,” USGS Open File Report # 2009-
1063, 112 pp. 2009.

51
68. Srdjevic, Bojan, Dantas Pinto Medeiros, Yvonilde and La Laina Porto, Rubem, “Date
Development Analysis of reservoir System Performance,” Computers and Operations
Research, Vol. 32 pp. 3209-3226. 2005.

69. Task Group Report, DRBC Docket No. D-77-20, “Appraisal of Upper Basin Reservoir
Systems, Drought Emergency Criteria and Conservation Measures,” March, 1979.

70. United States Environmental Protection Agency, “Regional Approaches to Efficient Water
Uses: Tales from the Trenches,” Cleaner Water through Conservation (1998).

71. United States Environmental Protection Agency, “Filtration Avoidance Determination,” May
[Link] through 2007.

72. United States Army Corps of Engineer (USACE), “Preliminary Results of 3-Dimensional
Salinity Model, Delaware Estuary,” circa April, 2007.

73. United States Army Corps of Engineer (USACE) “Engineering Manual (EM 1110-2-1420):
Engineering and Design - Hydrologic Engineering Requirements for Reservoirs,” Oct, 1997.

74. United States Army Corps of Engineer (USACE), “Resolving Conflict Over Reservoir
Operation, A Role for Optimization and Simulation Modeling,” June, 1999, by US Army
Corps of Engineers, Hydrologic Engineering Center, 609 Second Street, Davis, CA.

75. United States Environmental Protection Agency, “New York City Filtration Avoidance
Determination,” May, 1997.

76. United States Geological Survey, “Report of the Delaware River Master,” 1955- 2005.

77. Valdes, Juan B. and Marco, Juan B., “Managing Reservoirs for Flood Control,” U.S.-Italy
Research Workshop on the Hydrometeorology, Impacts and Management of Extreme
Floods,” Perugia (Italy), Nov., 1995.

78. Vogel, Richard M., Fennessy, Neil M., and Bolognese, Ralph A., “Storage-Reliability-
Resilience-Yield Relations for Northeastern United States,” Journal of Water resources
Planning and Management, Sep/Oct, 1965, pp. 364-374

79. Vogel, Richard M., Sieber, Jack, Archfield, Stacey A., Smith, Mark P., Apse, Colin D. and
Huber-Lee, Annette, “Relations Among Storage, Yield, and Instream Flow,” Water Resources
Research, Vol. 43, W05403, doi:10.1029/2006WR005226, 2007.

80. Vudhivanich, Varawoot and Rittima, Areeya, “Development of Probability Based Rule
Curves for a Reservoir,” Kasetsart Journal. (Natural Science) 37: 234 - 242 (2003).

81. Waldron, Marcus C. and Archfield, Stacey A., “Factors Affecting Firm Yield and the
Estimation of Firm Yield for Selected Streamflow-Dominated Drinking-Water-Supply

52
 Reservoirs in Massachusetts,” Prepared in cooperation with the Massachusetts Dept. of
Environmental Protection, Scientific Investigations Report 2006–5044, U.S. Department of
the Interior, [Link] Survey.

82. Wang, Zhengang and Bowles, David S., “Dam Breach Simulations with Multiple Breach
Locations under Wind and Wave Actions,” Advances in Water Resources 29, pp.1222–1237,
2006.

83. Weisman, R. A. “1980-81 Drought in Southeastern New York,” New York State University,
Albany, published in the Bulletin of the American Meteorological Society, 1985.

84. Weston, R. Timothy, “Statement of the Commonwealth of Pennsylvania to the Delaware

River Basin Commission,” Hearings Regarding Proposed New York State Department of
Environmental Conservation Regulations, Parts 670 & 671 (Reservoir Regulations) Issued
Under N.Y. Environmental Conservation law 1-0101, 3-0303, and 15-0805 Cch.888, L. 1976)
Trenton, NJ Feb 23, 1977.

85. Whitman, Ezra B., Cassedy, Thomas S. Sherman, Arthur L. and Powell, Sheppard T.,
Consulting Engineeers, New Jersey State, “Delaware Diversion Case Report to State of New
Jersey, Division of Water Policy and Supply on Agreement Between the City of New York
and the State of New Jersey,” March 20, 1953.

86. Yates, David., Purkey, David, Sieber, Jack, Huber-Lee, Annette and Galbraith, Hector,
“WEAP21: A Demand-Priority-and Preference-Driven Water Planning Model, Part 1: Model
Characteristics,” International Water Resources Association, Water International, Vol. 30, No.
4, pp. 487–500, Dec, 2005a.

87. Yates, David., Purkey, David, Sieber, Jack, Huber-Lee, Annette and Galbraith, Hector,
“WEAP21: A Demand-Priority-and Preference-Driven Water Planning Model, Part 2: Aiding
Freshwater Ecosystem Service Evaluation,” International Water Resources Association,
Water International, Vol. 30, No. 4, pp. 501–512, Dec, 2005b.

88. Zoppou, Christopher, “Reverse Routing of Flood Hydrographs Using Level Pool Routing,”
Australia National University, Canbarra, Australia, 15 pp. Aug.11, 1998.

53
Figure A. Delaware Basin from the Philadelphia-Camden area to the Upper Basin
showing New York City’s Delaware basin reservoirs (Pepacton, Neversink &
Cannonsville) and Rondout Reservoir, part of the Delaware System located in the
Hudson Basin. The Delaware Aqueduct is also shown. The Mongaup System and several
other major reservoirs are also shown.

54
 Figure B. Pepacton and Neversink Reservoirs to Rondout and Delaware Aqueduct. (Figure 25 from
“Hydraulic Analysis of the New York City Water Supply System,” US Army Corps of Engineers, 1974.)

Figure C. Cannonsville Reservoir to Rondout. (Figure 22 from, “Hydraulic Analysis of the New
York City Water Supply System”, US Army Corps of Engineers, 1974)

55
Figure D. New York City’s Delaware System. (Taken from Figure 8, “Hydraulic Analysis of the
New York City water Supply,” US Army Corps of Engineers, 1974)

56
Figure E. New York City’s Water Supply System, including the Catskill and Croton Systems which
comprise the Hudson System. (Taken from Figure 1, “Hydraulic Analysis of the New York
City Water Supply System,” US Army Corps of Engineers, 1974)

57