Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 165
9
Power Economics
INTRODUCTION
Power resources occupied an important and somewhat unique position
in the Glen Canyon Environmental Studies (GCES). Electrical output from
Glen Canyon is proportional to water flow through the dam, and the value of
that output varies daily, weekly, and seasonally. As a result, power econ-
omics and water flows are closely related.
Traditionally, Glen Canyon Dam has operated with relatively few con-
straints so as to maximize the value of its electrical output (Chapter 4~. Thus,
operational changes (such as those implemented under the interim flow
regime and those that were under active consideration in the environmental
impact statement (EIS) for Glen Canyon Dam) alter the scheduling and
reduce thevalue of power production. This loss of power resources accounts
for most, if not all, of the costs of altered dam operations.
As can be seen from the previous chapters, operational changes at Glen
Canyon Dam can have beneficial effects on native fishes, beaches, recreation,
and archeological sites. Therefore, a principal focus of the decision making
process concerning operation of the dam is the balance between the value
of production of electricity and other resources. This has several important
implications. First, since changes to dam operations reduce the value of
power, it is those who benefit from this output who will tend to be affected
most adversely and thus to be opposed to such changes. Interests related
to power production are typically well defined, strongly organized, and quite
aggressive in advocating their point of view. By contrast, those who might
benefit from modified dam operations are more diverse and must be mo
165
OCR for page 166
166
River Resource Management in the Grand Canyon
tivated by somewhat more diffuse goals, involving environmental or cultural
resources.
Second, the contrasting position of the different interests is accentuated
by the nature of the resources. Electrical output can be measured readily.
Moreover, it is a good that is bought and sold and that can be assessed a
specific monetary value. Other resources are generally harder to measure,
and it can be difficult or impossible to assign them monetaryvalues (Chapter
7~. Thus, in the EIS for Glen Canyon Dam and elsewhere, the costs of
changes to dam operations are reported in dollars, while most of the benefits
are reported in other units (e.g., number of beaches).
Third, the disjunction between electrical power and other resources is
further accentuated by the contrasting levels of historical analysis. Power
resources have been the subject of decades of analysis. The utilities
potentially affected by changes in operations at Glen Canyon Dam are
commonly viewed as having sufficient data and expertise to estimate the
adverse effects on their interests. Prior to the GCES, natural resources were
subject to much less study. Without the requisite data and expertise, it was
difficult to assess how these resources would be affected by changes in dam
operations. The GCES increased the feasibility of comparisons between
power production and the natural resources of Glen Canyon.
The GCES has also played a major role in advancing the study of Glen
Canyon Dam power economics. It became clear during Phase I of GCES that
the existing analysis did not provide an adequate basis for decision making
regarding altered dam operations. The quality of this analysis has sub-
stantially improved as a result of the extensive power economics studies
undertaken during GCES Phase 11. This process has greatly benefited from
broader public participation and review external to federal agencies and
utilities.
FLOWS AFFECT ELECTRICAL OUTPUT AND COSTS
The Colorado River Storage Project Act directs the Secretary of the
Interior to operate power plants "so as to produce the greatest practicable
amount of power and energy that can be sold at firm power and energy
rates." In a hydra project such as Glen Canyon, water is impounded behind
a dam and discharged at a lower elevation into the river downstream. The
mechanical energyofthefallingwateris used toturn turbines, which generate
electricity.
OCR for page 167
Power Economics
167
Electrical output is measured in two ways. Instantaneous output is re-
ferred to as power and is measured in terms of watts Go. Output over time
is electrical energy and is expressed in terms of watt-hours (Wh). The
quantity of electrical output is typically much larger then a waft, so units such
as kW (thousand watts), MW (million watts), and GW (billion watts) are used.
Similarly, typical units for energy are kWh, MWh, and GWh.
The maximum amount of power that can be produced by a power plant
is known as capacity. For a hydra dam, capacity is a function of the number
and size of turbines. Subsequent to a rewinding and uprating of generators
completed in 1987, Glen Canyon Dam capacity has been 1356 MW (at 33,200
cubic feet per second (cfs), which is the maximum flow through the turbines)
(BOR, 1995~. However, the Bureau of Reclamation (BOR) agreed not to use
this increased capacity pending completion of a comprehensive study of the
effects of historic and current dam operations on environmental resources.
Thus, Glen Canyon Dam has generally been limited to a capacity of 1,300 MW
(at 31,500 cfs of flow) (PRO, 1995~.
The maximum amount of energy that can be produced over time by a
hydro dam is determined by the smaller of two constraints: turbine capacity
and amount of water in the reservoir. If Glen Canyon Dam were to operate
at full capacity continuously for 1 year, it would generate almost 12,000 GWh
and discharge 24 million acre-feet (mad of water. This greatly exceeds the
amount of water entering Lake Powell, even in wet years (Chapter 4~. At the
average annual flow of 10 mat, Glen Canyon Dam can produce about 5,000
GWh annually.
Because turbine capacity generally exceeds water supply, the annual
electrical output of a dam is determined by the amount of water in the re-
servoir. Thus, changes in daily or monthly operations will have no effect on
annual power generation (Chapter 4~. They will, however, affect the sched-
uling of electrical output and thus its value.
The value of electricity varies substantially with time. Demand for elec-
tricity fluctuates daily, weekly, and seasonally. It is higher during the day,
when businesses are open, and lower at night and on weekends. It is greater
during the winter and summer, when required for heating and cooling, than
during the fall or spring. In addition, electricity cannot be cheaply or con-
veniently stored.
The reliabil ty of the electrical power system is also difficult to maintain.
Power plants and customers throughout western North America are inter-
connected, and electricity moves at the speed of light. If generation and
consumption are not continuously and closely matched, the power system
OCR for page 168
168
River Resource Management in the Grand Canyon
becomes unstable and blackouts can result. Thus, utility systems are plan-
ned to have sufficient generating capacity to supply customer requirements
arm provide a reserve for malfunctions and other exigencies.
The cost of producing electricity includes two principal components: (1 )
fixed costs to build power plants and keep them in operable condition and (2)
variable costs associated with operation. For fossil fuel power plants, the
variable costs relate mostly to fuel purchases. For hydra plants, which are
powered by wafer, variable costs are relatively low, but the fixed costs of dam
construction are high.
In a typical large power system, several kinds of generating plants are
used. In general, plants with high fixed costs and low operating costs (such
as coal-fired stations) serve the base load, while plants with low fixed costs
and higher operating costs (such as gas- or oil-fired stations) are used to
meet peak demancl. Overall, base load is cheaper to serve than peak de-
mand because fixed costs can be spread over more hours of output.
One notable advantage of hydra plants is that they can respond quickly
to variations in demand for electricity. Hydro turbines can be turned on and
off almost instantaneously. In contrast, conventional thermal powerplants use
boilers to generate steam for a turbine and can require substantial start-up
time (ranging from minutes to hours) to generate electricity. Thus, hydro
plants are especially well suited for providing peaking power.
With the large turbine capacity at Glen Canyon Dam, traditional op-
erations provided great flexibility to schedule electrical production at times
when it would be most valuable. Changes in dam operations have restricted
the maximum flows available during peak periods. These changes have the
effect of shifting power output from periods when it is more valuable to
periods when it is less valuable, with essentially no change in annual energy
production. With less output during peak periods, additional supply is re-
quired from other sources. Also, a quantity of off-peak power from Glen
Canyon Dam is available for sale to other utilities or to displace the need for
other power supplies. Thus, the cost of altered flows has been the difference
between (1 ) the cost of peak power required to replace the output shifted to
off-peak periods and (2) the value of this incremental off-peak power.
THE INSTITUTIONAL CONTEXT
Glen Canyon Dam is owned and operated by the Bureau of Reclamation.
The Western Area Power Administration (WAPA) markets this electricity on a
OCR for page 169
Power Economics
169
wholesale basis to about 180 preferred customers. As required by federal
law, these preferred customers are municipal and county utilities, rural electric
cooperatives, waterand irrigation districts, U.S. government installations, and
other non profit organizations. Each individual customer serves a designated
area in the region. As shown in Figure 9.1, most are in the six states of Ar-
izona, Colorado, Nevada, New Mexico, Utah, and Wyoming, although some
extend into California, Nebraska, and Texas. These preferred customers, in
turn, serve 1.7 million end-use customers, including residential, commercial,
industrial, and agricultural users.
In part, the distribution of costs associated with altered flows is de-
termined by the contractual arrangements for sale of Glen Canyon Dam
electricity. Under existing contracts, WAPA's costs will increase, because it
is obligated to supply fixed quantities of peaking power, which it may have to
purchase at higher costs from other utilities. As contracts expire and are
renegotiated, however, WAPA could contract to sell less firm peak power and
more firm off-peak power. In this case, both the cost and the value of WAPA's
electricity will diminish, and the need to replace the dam's output will be
shifted to WAPA'S customers. Thus, there is a relationship between these
marketing policy issues and the costs associated with altered flow regime.
A separate EIS has been established to address marketing issues (WAPA,
1 994).
It is important to distinguish between the economic perspective that
measures changes in overall costs to society and the financial perspective
concerning changes in the costs borne by specific entities. I n the short term,
economic costs caused by altered flows will be limited, because the region
currently enjoys a surplus of generating capacity (in excess of the required
reserve margin). Underthe national economic perspective, the fixed costs of
this existing capacity are excluded because they must be paid whether or not
the capacity is used as a replacement for Glen Canyon Dam.
From a financial point of view, however, there will be differences in costs
and benefits among utilities. Utilities that have surplus capacity and can sell
power to replace Glen Canyon Dam stand to gain at the expense of other
utilities that must purchase power to replace Glen Canyon Dam power.
Surpluses eventually will decrease, and prices for peak power can be ex-
pected to rise. Eventually, new supplies will be required, which will entail both
economic costs to society and financial costs to utilities.
Within the restricted six-state marketing area for Glen Canyon Dam, 70
percent of electricity consumers (3.9 million) are served by utilities that do not
receive power from the dam. Absent some new type of cost-sharing mech
OCR for page 170
170
River Resource Management in the Grand Canyon
_
_~
I NEVADA
GLEN CANYON I
DAM |
_ ~
·.3 (
WYOMING
~1 , .
UTAH ./
· COLORADO
·:
.1
ARIZONA
NEW MEXICO
FIGURE 9.1 The Salt Lake City Area/lntegrated Projects markets power to approximately 180
utilities, mostly in six states. SOURCE: Bureau of Reclamation (1995~.
anism, electricity bills for these utilities will be unaffected by changes at Glen
Canyon Dam, or may decrease owing to surplus power sales. Notably, most
of the region is served by investor-owned utilities (e.g., Arizona Public
Service, Nevada Power, Public Service of Coloraclo, Utah Power), which are
foreclosed by law from being WAPA preferred customers. Six relatively large
utilities, which serve 1.3 million end-use customers, receive about half of Glen
Canyon Dam's output. The remainder goes to numerous smaller systems,
which together supply only 400,000 customers (7 percent of the region's
total).
Glen Canyon Dam operations, and the resulting revenues, also relate to
a broader set of issues. Access to the relatively low cost electricity produced
by the dam is restricted; about half goes to only 7 percent of regional el-
ectricity consumers. Recipients of the dam's electricity benefit from the use
of public water resources that were developed with low-cost government
clebt. Under the terms of the Colorado River Storage Project Act, revenue
from Glen Canyon Dam is intended to support reclamation irrigation projects.
OCR for page 171
Power Economics
171
Judgments about the appropriateness of these historical arrangements
influence determinations of how the costs of altered flows should be dis
tributed. If the beneficiaries of Glen Canyon Dam have traditionally been
subsidized at the expense of taxpayers and the environment, it is acceptable
that they bear the costs of altered operations. Not surprisingly, this point of
view is strenuously resisted by the beneficiaries of the dam, notably the
Colorado River Energy Distributors Association (CREDA), which represents
the utilities receiving the bulk of Colorado River hydropower. They assert that
the beneficiaries of Glen Canyon Dam have paid their fair share and should
be shielded from any increased costs (Barrett, 1 992a, 1 992b).
COST ESTIMATES FOR ALTERED FLOW REGIMES
Analyzing the costs of altered flow regimes is complex. The operation of
Glen Canyon Dam must be simulated to provide a realistic estimate of the
time pattern of power production under the various constraints of each flow
regime. These simulations must be carried out over a number of years to
reflect annual variations in hydrology.
Once the output of the dam under different flow regimes has been es-
tablished, the costs of these regimes to power users can be estimated. Such
analyses typically rely on computer models, which are used to simulate the
planning and operation of electric utility systems. These models require
extensive data on existing power plants, new supply options, fuel costs,
required reserve margins, interconnections between utilities, and current and
projected electric ty demand. They use various mathematical techniques to
determine the optimal (least-cost) solution, subject to specific constraints.
Output includes data on the operation of each power plant, the new power
plants and demand-side management (conservation) programsthat are used
to meet requirements for new supply, and capital and operating costs.
Relatively little work regarding power economics was undertaken during
Phase I of GOES. In response to the National Research Council (NRC)
comm ttee's review of GOES Phase I (NRC, 1991), WAPA was requested to
analyze the economic effects of increasing minimum flows to either 5,000 or
8,000 cfs all year. WAPA estimated annual costs of $5.0 and $14.7 million,
respectively, during the 1990s and much larger costs subsequently (NRC,
1991~. These cost estimates were based solely on a WAPA financial
perspective. Moreover, the modeling of Glen Canyon Dam operations did not
provide a realistic estimate of the time pattern of power production under the
OCR for page 172
172
River Resource Management in the Grand Canyon
constraints of each flow regime (NRC, 1991~. Therefore, the loss of peak
period capacity and energy output due to altered flows was overstated.
As part of GCES Phase 11, detailed power resource studies were initiated
in November 1988. These studies were conducted by the Power Resources
Committee (PRC). Three of thefour members have historically controlled and
benefited from Glen Canyon Dam power resources: BOR, WAPA, and
CREDA. In addition, however, the PRC included the Environmental Defense
Fund (EDF), a national environmental group. As the developer of a widely
used utility simulation model (ELFIN, or Electric Utility Financial and
Production Cost Moclel), EDF provided the PRC with substantial technical
expertise regarding utility economics, as well as an alternative point of view.
The PRC's work focused on Glen Canyon Dam. However, during the
early part of GCES Phase 11, individual members of the PRC also produced
some analyses of costs related to experimental and interim flows. WAPA
estimated overall costs of $10.9 million for experimental flows planned for
1990 and 1991 (NRC, 1991~. Once again, these cost estimates were based
solely on a WAPA financial perspective. The modeling of Glen Canyon Dam
operations was an improvement over previous work but still did not provide
a realistic estimate of the time pattern of power production under the
constraints of each flow regime (NRC, 1991~.
EDF used ELFIN to evaluate the interim flow regimes proposed by various
parties, including WAPA (EDF, 1991~. These estimates were based solely on
a national economic perspective, which EDF argued should be given pre-
ference over the financial perspective of affected utilities. Estimated costs for
WAPA's proposal were only $1 million to $2 million annually over the period
1992 to 1995. For the other proposals, the costs were on the order of $9
million for 1992, increasing to $15 million to $16 million for 1995.
For interim flows the BOR's draft environmental assessment noted that
conclusive data were not yet available from the detailed power resource
studies under way as part of GCES Phase 11 (BOR, 1991~. The costs were
expected to be small, however, for the 3-year period of interim flow, because
surplus capacity was likely to be available. From a national economic
perspective, effects were characterized as a minor increase. The draft en-
vironmental assessment stated that the financial effects on WAPA were
estimated to be $22 million in fiscal year 1992. If financial exception criteria
were provided, allowing limited exceedances flow criteria, costs would be
reduced to only $3 million.
Exception criteria were ultimately accepted by the BOR, and the costs of
interim flows have been relatively low. To date, WAPA has avoided the need
OCR for page 173
Power Economics
173
to purchase replacement capacity and has been able to operate with 10 to 15
percent available capacity above peak needs, as opposed to 30 percent
previously (BOR, 1995~.
The efforts of the PRO were first directed at determining the best app-
roach to analyzing power economics. A detailed study (BOR, 1990) re-
commended that the EGEAS (Electric Generation Expansion Analysis
System) and ELFIN simulation models be used and their results compared.
EGEAS has the capability to perform expansion planning (addition of new
resources) as well as production cost estimates (dispatch of a given set of
resources). ELFIN estimates production costs and utility financial models.
For the power resources studies, EGEAS modeling was carried out by
Stone and Webster Consultants, Inc., a contractor retained by the BOR
through HERS, Inc. (the contractor for GOES recreation and nonuse value
economics studies). The expansion plans developed by using EGEAS were
used as inputtothe ELFIN modeling conducted by EDF. The production cost
projections of the two models were similar.
Power system costs were measured over a 50-year period. To allow for
changes at Glen Canyon Dam to influence the need for new supplies, the
most detailed analysis was conducted for an initial 20-year planning period
(1991-2010~. To reflect costs over the lifetime of power plants, a 30-year
extension period (201 1-2040) was modeled. Continued escalation in fuel and
other costs was assumed, but the level of demand and set of supply re-
sources forecast for 2010 were held constant (resources being retired were
assumed to be replaced in kind).
The operation of Glen Canyon Dam also had to be simulated to estimate
the time pattern of power production under each flow regime. The BOR's
Colorado River Simulation Model (CRSM) was the source of projections for
monthly water releases, capacity, and energy. CRSM has no capability for
simulating hourly operational constraints, such as those imposed by
fluctuating flow alternatives. As discussed earlier, two marketing approaches
were modelecl. For the Contract Rate of Delivery (CROD) marketing app-
roach, a simple geometric method was used to determine available capacity.
The peak shaving algorithm from ELFIN was utilized for the hydrology mar-
keting approach.
The foregoing description only begins to capture the intricate array of
analyses undertaken for the power resources studies (see Figure 9.2~. In
large part, this complexity results from efforts to reflect the institutional
context, with its multiplicity of affected parties, viewpoints, and practices. In
particular, much of the effort in the power resources studies was devoted to
OCR for page 174
74
No-Action
Case
Flow
Alternative
~ ECE\S
r I ,r
Purchased Avoided
Power Economic
Costs Supply Costs
1 ' ' 1
· Individual and Total
Utility Economic
Results
(CROD only)
River Resource Management in the Grand Canyon
| CROD
| CROD and Fnancal
Hydrology Economic
Studies h Studies
i
Federal
Repayment and
Rate Study
Utility
Wholesale
Rate Impacts
Small Systems
Retail Rate Impacts
Compute Differences
No-Action Alternatives
1
Total System Before
Transfer Payments
Subtract Transfer Payments
· Between Large and Small
Systems
· Firm Purchased Power
Capacity Cost Transactions
Total Federal
Economic Results
1 1
FIGURE 9.2 Study process to determine power values. SOURCE: Adapted from Power
Resources Committee (1994, Fig. 1-3~.
detailed modeling of the financial and rate effects on individual utilities.
Even for the analysis of the national economic perspective, separate
OCR for page 175
Power Economics
175
simulations were conducted for each of seven large utilities (one of these
utilities has now been absorbed by the others), with some attempt to re-
concile and coordinate the individual analyses. Meanwhile, the small utilities,
which rely on WAPA and other utilities for most or all of their supply, were not
included in the EGEAS modeling. These were assumed to purchase re-
placement powerfrom their alternative suppliers, at costs based on the
EGEAS model of large systems.
It would have been simpler to model all of the utilities as a single in-
tegrated system. This approach was recommended by the NRC committee
and rejected by the PRO (NRC, 1992; Roluti, 1993; Power Resources
Committee, 1993, 1994~. The rationale was that the individual utilities do not
now coordinate the operation of their systems so as to minimize overall costs.
Moreover, existing transmission capacity between utilities is limited. Thus,
results based on a single optimized system could understate the actual costs
of system operations. However, even within the constraints imposed, the
resource plans selected were not completely optimized (Power Resources
Committee, 1993, 1994~.
The focus on individual utilities and financial effects also necessitated
detailed modeling of WAPA's marketing practices. The data on Glen Canyon
Dam's power production had to be combined with estimates of the output
from other hydra projects marketed by WAPA to determine the amount of
power and energy that would be contracted for sale to preferred customers.
This determination was affected by the marketing criteria used by WAPA.
Currently, WAPA uses the approach, CROD (Contract Rate of Delivery)
according to which the amount of firm capacity and the amount of energy are
fixed in advance, and WAPA must purchase electricity to supplement hydro
output in dry years. Alternatively, under the hydrology approach, WAPA
would sell only the capacity and energy available given actual hyciro output.
Customers, rather than WAPA, would be responsible for meeting any
additional needs. The power resources studies modeled both the CROD and
hydrology approaches.
Extensive analyses were performed of the sensitivity of results to changes
in input assumptions or methodology. Sensitivity analyses are standard
practice in power economics studies. In keeping with recenttrends, the Glen
Canyon Dam analyses relied to some extent on sophisticated probabilistic
approaches. The use of sensitivity analyses, however, is not a substitute for
selection of appropriate base-case assumptions. When the PRO was unable
to reach consensus on these assumptions, remaining disagreements were
typically "resolved" by specifying a case for sensitivity analysis. This was also
OCR for page 176
176
River Resource Management in the Grand Canyon
the response to certain concerns expressed by the NRC committee (NRC,
1992; Roluti, 19g3; Power Resources Committee, 1993, 1994~.
The EGEAS model was also used to estimate emissions of atmospheric
pollutants. Emissions rates for each power plant (tons of sulfur dioxide (SO2)
and nitrogen oxides (N Ox) per unit of fuel burned) were provided as input.
The model then combined these data with the results concerning power plant
operations to estimate the total tons of SO2 and NOx.
Review of the power resources studies is further complicated by the
significant changes that occurred over the course of GOES work. The PRO
produced three reports that were provided to the NRC committee: a draft
Phase 11 report (PRC, 1992), a final Phase 11 report (PRO, 1993), and a Phase
111 report (PRC, 1995~. Each report incorporated improvements in metho-
dology and data from the previous one. These improvements addressed
problems in the analysis identified bythe NRC committee and other reviewers
(NRC, 1992; NRC, 1994~.
Unfortunately, the usefulness of GOES Phase 11 and Phase 111 results is
compromised by changes in the flow alternatives that were modeled. The
Phase 11 studies analyzed eight flow.alternatives (including "no actions. Prior
to the release of the draft EIS, beach-building and habitat maintenance flows
were added to the moderate fluctuating flow and seasonally adjusted steady
flow alternatives; they were also included in the new preferred alternative of
the EIS (modified low fluctuating flows). The Phase 11 studies were too ad-
vanced to incorporate these changes, so consideration of them was deferred
to Phase 111.
Because of resource constraints, Phase 111 work was limited to the "no
action" and preferred alternatives. As a result, there are no Phase 111 results
for most flow alternatives, and the final EIS largely relies on the earlier Phase
11 work. For the three flow alternatives that include beach-building and habitat
maintenance flows, cost data for the final EIS were derived from a simple
regression analysis based on the amount of capacity lost and the costs
estimated for Phase 11 flow alternatives. The final EIS does provide the Phase
111 results as supplemental data for the preferred alternative.
Table 9.1 summarizes the most important numerical results from the
power resources studies. In the interest of brevity, data are presented for
only three flow regimes. "No action" serves as the benchmark for measuring
the effects of altered flows. Seasonally adjusted steady flows result in the
greatest power resource costs and potentially the greatest benefits in terms
of protecting other resources. The preferred alternative and seasonally
adjusted steadyflow bracketthe range of operational changes likely to be im
OCR for page 177
Power Economics
177
TABLE 9.1 Summary of the Effects of Three Operating Regimes on the Value of Electrical
Power From Glen Canyon Dam
Preferred Alternative: Seasonally
No Modified Low Adjusted
Action Fluctuating Flow SteadyFlow
Marketable Resource
Annual Energy (GWh) 6,010 6,018 6,123
(+0.1%) (+ 1.9%)
Winter Capacity (MOO, 1,407 965 640
(-31.4%) (-54.5%)
Summer Capacity 1,315 845 498
(-35.2%) (-62.1 %)
Increase in Economic Costs
(relative to No Action)
Arnual (nominal$million) Phase II Phase lil
CROD So $44.2 $34.8 $123.5
Hydrology $0 $15.1 $25.0 $88.3
Present Value (1991 $ million)
CROD So $511.2 $402.0 $1,428.4
Hydrology $0 $174.6 $286.8 $1,021.2
WAPA Wholesale Rate
(FY 1993 ¢/kWh) 1.878 2.316 2.820
(+23.3%) (+50.2%)
Retail Rates (¢/kWh)
Customers
Large systems (23% end No Slight See text Slight
users) change decrease to p.178 decrease to
moderate moderate
increase increase
Small systems (7% end 6.41 7.05 See text 7.58
users) . (+10.0%) p. 178 (+16.3%)
No No change No change
Other regional utilities p0% change to slight to slight
of end users) decrease decrease
OCR for page 178
178
River Resource Management in the Grand Canyon
plemented and that might occur in connection with endangered-fish research.
Costs for fluctuating flow regimes not shown in Table 9.1 are generally
intermediate between those for "no action" and the preferred alternative, while
those for other steady flow regimes are intermediate between the preferred
alternative and seasonally adjusted steady flow.
The preferred alternative reduces WAPA's marketable capacity by
approximately 450 MW. In the power resources studies, the stream of annual
nominal dollar costs is present valued to 1991 using an 8.5 percent nominal
discount rate. Based on the GOES Phase 111 study, the present value of the
associated economic cost is on the order of $300 to $400 million. Levelized
over the 50 year analysis period, this is equivalent to an annual cost of $25 to
$35 million in nominal dollars or $15 to $20 million in 1991 dollars.
The power resources studies report annual costs on a nominal levelized
basis; over the 50-year period being analyzed, the estimated equivalent ann-
ual effects remain constant in nominal terms. In constant (inflation-adjusted)
dollars, they are highest in the first year and decline steadily due to inflation.
The figures for 1991 dollar levelized costs in this report were calculated based
on the rea/ discount rate. Given the 3.8 percent inflation rate assumed (Power
Resources Committee, 1993, p. 111-13), the 8.5 percent nominal discount rate
is equivalent to a 4.5 percent real discount rate. For a 50 year period, this
yields a real annualization factor of 5.08 percent. Annual impacts in 1991
dollars are approximately 41 percent less than those calculated on a nominal
basis.
Costs incurred in any given year are expected to vary substantially over
time. In general, they would be lower in earlier years. The region has a
surplus capacity, which the power resources modeling assumes would be
prolonged as the large utilities implement demand-side management pro-
grams. Costs would rise after 1998 as new capacity is required. Costs will
also be affected by variations in hydrology. The constraints associated with
altered flows will have less effect during wet years.
For the preferred alternative, WAPA's wholesale rates are estimated to
increase by about 0.5¢/per kilowatt-hour, but Glen Canyon Dam power
would still be highly competitive with alternative sources. Most regional
electricitycustomerswould experiencelittle, if any, change intheir retail rates.
For several reasons, however, the effect of changes in operations at Glen
Canyon Dam will tend to be much greater for the small preferred customers.
They typically buy most or all of their power from other utilities and rely
heavily on Glen Canyon Dam power. By contrast, the large preferred cus-
tomers generate much of their own power. They currently have surplus cap
OCR for page 179
Power Economics
179
acity, which can be used to replace lost Glen Canyon Dam power for their
own needs and through expanded sales to the small systems.
The final EIS for Glen Canyon Dam reports that increases in small system
retail rates would range from 4 to 16 percent, averaging 10 percent or about
0.6¢/per kilowatt-hour. The basis of these estimates is unclear. The draft EIS
reported smaller effects on rates, similar to those shown in the GOES Phase
11 study (PRO, 1993). Thefinal EIS cites the Phase 111 study (PRO, 1995), but
the version of the report provided to the NRC committee does not include the
final EIS data. The data that are provided in the Phase 111 report indicate that
effects on rates will vary both across utilities and over time, with a maximum
increase of 9.5 percent for one utility in 1 year. On average over the 1991-
2010 period, rates will rise by 1 percent or less for Rural Electric Admin-
istration member utilities and by 2 to 5 percent for municipal utilities.
Seasonallyadjusted steaclyflow reducesWAPA's marketable capacity by
approximately 800 MW. The present value of the associated economic cost
is estimated to be on the order of $1 billion to $1.4 billion. Levelized over the
50-year analysis period, this is equivalent to an annual cost of about $90 to
$120 million in nominal dollars or about $50 to $70 million in 1991 dollars. As
for the preferred alternative, costs would generally be lower in the earlier
years. However, the greater loss of capacity would advance the need for
construction of new capacity and the associated costs. Cost estimates based
on Phase 111 methodology would likely be lower than these, which are based
on the Phase 11 approach, which does not value off-peak energy correctly.
WAPA's wholesale rates are estimated to increase by about 1 ¢/per
kilowatt-hour. Even so, Glen Canyon Dam power would still be competitive
with all but the lowest-cost alternative sources. Most regional electricity
customers would experience limited, if any, change in their retail rates.
Reported increases in small system retail rates would range from 8 to 33
percent or, on average, 18 percent or about 1.2¢/per kilowatt-hours.
In summary, the cost effects of the preferred alternative are relatively
modest. Costs for seasonally adjusted steady flows are generally two to three
times greater than those for the preferred alternative. Even costs of this
magnitude, however, would have only limited effects on agriculture and no
material impact on the overall regional economy (BOR, 1995; WAPA, 1994;
Flaim et al., 1994~. For any altered flow regime, some small utility customers
may bear a disproportionate share of the costs; however, they also received
a disproportionate share of the benefits of low-cost Glen Canyon Dam
electricity in previous years.
The costs of altered flows may be less than estimated, especially for the
OCR for page 180
180
River Resource Management in the Grand Canyon
small utilities, which account for most of these costs. The GOES power
resources studies did not consider the option of having WAPA maintain the
same marketing commitment and use its transmission system to procure low-
cost replacement capacity on behalf of its customers. Studies in support of
the marketing EIS indicated that this approach could considerably reduce
economic and financial effects (BOR, 1995~. Even without such an approach,
changes in the electricity industry (e.g., the U.S. Energy Policy Act of 1992)
are providing greater access to a wide variety of low cost electricity supply
sources, especially for small utilities, which traditionally have been limited in
their supply options.
The base-case analysis in the power resources studies assumes a rapid
escalation in oil and gas prices (averaging 8.4 percent nominal or4.4 percent
real annually), which now appears highly unlikely. Lower oil and gas prices
would reduce the costs of operating the peaking plants used to replace lost
capacity at Glen Canyon Dam. The power resources studies also did not
explicitly consider the relationship between electricity prices and the amount
of energy consumed (price elasticity). If rates increase because of altered
flows, this will reduce future electricity demand. In turn, this will delay the
need for new capacity and reduce the cost of altered flows. The PRO chose
to deal with these issues through sensitivity analyses, which confirm that the
estimated costs of altered flows will be substantially lower if lower fuel prices
and lower demand materialize in the future.
The costs associated with altered flows must be compared with the
benefits. The preferred alternative reduces regional SO2 and NOX power plant
emissions by almost 1 percent, as fossil fuel generation is shifted from base
load to cleaner peaking plants and construction of new cleaner plants is
advanced. Under the Clean Air Act Amendments of 1990, there is a market
for SO2 emissions allowances. On this basis, the GOES Phase 111 report
estimates that SO2 emissions reductions are worth about $5 million (1991 net
present value) or about 1 to 2 percent of the estimated increase in electricity
costs (PRO, 1995~. Seasonally adjusted steady flow would likely result in
even greater reductions in emissions. The power resources studies did not
consider othertypes of environmental impacts, such es those associated with
fuel supply and transportation. Nonetheless, it is reasonable to assume that
by shifting electricity production to newer and cleaner power plants, altered
flows will generally reduce adverse environmental impacts.
The preferred alternative yields increased recreation values of $43.3
million (1991 present value), equivalent to about 10 to 15 percent of the
estimated increase in electricity costs (BOR, 1995~. Nonuse values are of
OCR for page 181
Power Economics
much greater magnitude than the increase in power costs (Chapter 7~.
RECOMMENDATIONS
181
The GCES power resources studies were impeded by an unfortunate
combination of factors. The process was dominated by the entities that
historically have controlled and benefited from Glen Canyon Dam power
resources, notably BOR, WAPA, and CPtEDA. These entities have a clear
incentive to deter implementation of altered flows, which would reduce the
value of the dam's electrical output. On the other hand, they have the ex-
pertise to perform power resource studies, in light of their familiarity with
these issues and high level of analytical resources. In fact, however, very little
useful information regarding the cost effects of altered flow regimes was
provided by those entities during Phase I of GCES.
During GCES Phase 11 there was great progress in developing the
requisite tools for measuring cost effects. Nonetheless, the end result has not
been wholly satisfactory in terms of providing cost estimates that are
accurate, well documented, and readily reviewable. Unfortunately, the data
in the EIS are principally based on Phase 11 Power Resources studies, rather
than the subsequent Phase 111 analyses, which generally indicate lower costs
for altered flows.
It is difficult, even with hindsight, to make completely definitive judgments
of many aspects of the power resources studies. To some degree, the
multiple analyses that were undertaken reflect the complex institutional
context and the distribution of costs and benefits across different groups.
Given the constraints in terms of budget and schedule, however, the strong
focus on distributional issues has adversely affected the accuracy and
timeliness of the analysis from a national economic perspective. Given that
Glen Canyon Dam is federally owned and affects resources of national (and
international) significance, the national economic perspective should be given
precedence in the future. This approach is in keeping with the principles and
guidelines established for federal water projects (Water Resources Council,
1 983~.
The power resources studies generally assumed that current practices
arid constraints would remain in place throughout the 50-year analysis period.
This is problematic because the electric utility industry is evolving toward a
more competitive future, which should help reduce the cost of altered flow
regimes. The PRO missed a valuable opportunity to inform decision makers
OCR for page 182
182
River Resource Management in the Grand Canyon
concerning the effect of these changes. In particular, analyses could have
been conducted both for the current, less integrated system and for a single
optimized system. The former would have provided an upper bound for
costs, assuming continuation of the status quo. The latter would provide a
lower bound for costs, which could be achieved if the existing constraints on
cost minimization were eliminated.
In part, the problems with the GOES power resources studies stem from
the lack of a continuous open planning process that is accessible to the
public. By contrast a very different set of procedures is in place for ad-
dressing electric power issues in the Pacific Northwest concerning the
Columbia River system. Both the Colorado and Columbia River systems
feature extensive hydroelectric facilities operated byfederal agencies that sell
power to government-owned utilities under environmental constraints. But
due in part to the Northwest Power Planning Act, the Northwest region has
long had a major planning effort that has developed the necessary tools and
institutions required to evaluate the effects of various alternatives. There is a
high level of expertise on the part of federal agencies, the utilities, and other
interested parties such as environmental groups and state governments.
Furthermore, this process has been steadily building in expertise over the
past two decades.
In contrast, the GOES and Glen Canyon EIS were the first time that much
of this type of analysis had been undertaken for the Colorado River system
area. In addition, given the many disparate interests, there were many
procedural issues to resolve, and it has been difficult to obtain the requisite
economic and financial data. Overall, the process has been difficult, time
consuming, and costly. Also, unlike the process for the Columbia River,
which involves a continuing mandate and an established institutional en-
vironment, it is unclear to what extent power studies for Glen Canyon Dam
will continue in the future. Clearly, there is a need to update the power
studies over time for the purposes of adaptive management. Moreover,
without such projections, it will be difficult to plan and operate the regional
power system effectively.
The BOR or Department of the Interior should sponsor the development
of analytical and modeling capabilities that can continue to provide in-
formation concerning the cost of dam operations. This would permit regular
revisions to reflect the rapidly evolving electricity industry and other factors.
Future studies relating to the operation of Glen Canyon (and other hydra
facilities) should explicitly consider how current practices and constraint may
be altered by factors such as the evolution of the utility industry. Continued
OCR for page 183
Power Economics
183
modeling also could facilitate the numerous processes that are now affecting
the Colorado River hydra system. Subsequent to the release of the final
WAPA marketing EIS in the summer of 1995, a commitment level will be
established for firm power and energy to be marketed through 2004. This
process may be delayed, however, by consideration of proposals to privatize
WAPA. As mandated by the Grand Canyon Protection Act, WAPA has
initiated a Replacement Resources Process to study and report on methods
to make up for any reductions in Glen Canyon Dam output. The U.S. Energy
Policy Act of 1992 requires WAPA customers to prepare and implement
Integrated Resource Plans that consider a full range of supply options,
including demand-side management and renewable energy sources.
Continuing study of power economics is important, given the complex
issues being examined and the need for updating. Such capabilities could be
used in optimizing the timing of experimental flows. This could be a major
issue. The endangered fish research included as a common element in the
EIS alternatives would involve monthly release volumes similar to the
seasonally adjusted steady flow alternative. It is unclear how long this re-
search would continue, but the EIS indicates that it could be for as much as
1 0 years.
More generally, an attractive strategy would be to experiment with highly
restrictive flow regimes (e.g., seasonally adjusted steady flow) in the short
term when surplus capacity is available and the cost of such alternatives is
low. Then, ~ the costs of altered flows rise in the future, decisions on whether
to move toward a less restrictive alternative (e.g., modified low fluctuating
flows) could be made based on revised studies concerning the effects of flow
regimes on power and other resources.
Near-term experimentation with highly restrictive flow regimes may also
reduce the need to experiment in later years, when less surplus capacity is
available and costs could be much higher. Thus, a strategy of extensive
short-term experimentation could reduce long-term electricity costs. When
evaluating future dam operations, especially experimentation with highly
restrictive flow regimes, decision makers should consider long-term, as well
as short-term, impacts. Experimentation which has significant long-term
benefits should not be unduly restricted in an attempt to minimize short-term
increases in power costs.
OCR for page 184
184
River Resource Management in the Grand Canyon
REFERENCES
Barrett, C. 1992a. Letter to the Editor, The Washington Post, January 21,
from Clifford Barrett, Executive Director, Colorado River Energy Dis-
tributors Association.
Barrett, C. 1 992b. Letter to Sheila David, Program Officer, NRC Committee
to Review the Glen Canyon Environmental Studies, February 10, from
Clifford Barrett, Executive Director, Colorado River Energy Distributors
Association.
Bureau of Reclamation et al. April 1990. Final Report, Evaluation of Methods
of Estimated Power System Impacts of Potential Changes in Glen Canyon
Powerplant Operations. Bureau of Reclamation, Washington, D.C.
Bureau of Reclamation. 1991. Glen Canyon Dam, Interim Operating Criteria,
Draft Environmental Assessment. Bureau of Reclamation, Upper Col-
orado River Regional Office, Salt Lake City.
Bureau of Reclamation. 1995. Operation of Glen Canyon Dam, Colorado
River Storage Project, Final Environmental Impact Statement. Bureau of
Reclamation, Salt Lake City.
Environmental Defense Fund. 1991. Estimates of Power System Impacts of
Proposed Interim Flow Release Patterns at Glen Canyon Dam. Oakland,
Calif.: Environmental Defense Fund.
Flaim, S.~., R.E. Howitt, and B.K. Edwards. 1994. Impacts on Irrtigated
Agriculture of Changes in Electricity Costs Resulting from Western Area
Power Administration's Power Marketing Alternatives. Draft report,
Argonne National Laboratory, Technical Report No. W-31-109 Eng-38, for
the U.S. Department of Energy, Argonne, 111.
National Research Council. 1991. Colorado River Ecology and Dam
Management. Washington, D.C.: National Academy Press.
National Research Council. 1994. Review of the Draft Environmental Impact
Statement on Operation of Glen Canyon Dam. Washington, D.C.:
National Academy Press.
Power Resources Committee. 1992. Power Systems Impacts of Potential
Changes in Glen Canyon Powerplant Operations. Glen Canyon En-
vironmental Studies Technical Report. Draft, Stone and Webster Man-
agement Consultants, Inc., Englewood, Colo.
Power Resources Committee. 1993. Power Systems Impacts of Potential
Changes in Glen Canyon Powerplant Operations. Glen Canyon En-
vironmental Studies Technical Report, Stone and Webster Management
Consultants, Inc., Englewood, Colo.
OCR for page 185
Power Economics
185
Power Resources Committee. 1995. Power Systems Impacts of Potential
Changes in Glen Canyon Powerplant Operations, Glen Canyon
Environmental Studies Technical Report. Stone and Webster Con-
sultants, Inc., Englewood Colo.
Roluti, M.~. 1993. Letter to William M. Lewis, Chair, NRC Committee to
Review Glen Canyon Environmental Studies, Subject: Power System
Impact of Potential Changes in Glen Canyon Environmental Studies, April
29, from Michael J. Roluti, Chair, Power Resources Committee, Bureau
of Reclamation, Upper Colorado Regional Office.
Water Resources Council. 1983. Economic and Environmental Principles
and Guidelines for Water and Related Lana Resources Implementation
Stuclies. Washington, D.C.: Government Printing Office.
Western Area Power Administration. 1994. Salt Lake City Area Integrated
Projects Electric Power Marketing. Draft Environmental Impact State-
ment, Western Area Power Administration, Salt Lake City.
Representative terms from entire chapter:
canyon dam
