HomeMy WebLinkAboutStaff Report 14800
City of Palo Alto (ID # 14800)
City Council Staff Report
Meeting Date: 12/19/2022 Report Type: Action Items
City of Palo Alto Page 1
Title: The Utilities Advisory Commission (UAC) and Staff Recommend the City
Council Approve Option 2 for Building Fiber-to-the-Premises (FTTP) Under a
Phased Approach
From: City Manager
Lead Department: Utilities
Recommendation
The Utilities Advisory Commission (UAC) and staff recommend City Council approve option 2 for
building Fiber-to-the-Premises (FTTP) under a phased approach:
Option 1. Build fiber backbone and FTTP within five years with $98M Revenue Bond.
Construction costs of $127.9 million to build the fiber backbone and FTTP
distribution network based on the business models developed, plus $15
million to provide City-owned internet service provider (ISP) services totaling
$142.9 million for the entire project. Fiber fund balance at $34 million,
leaving a funding gap of $98 million, which may be covered with a revenue
bond.
► Option 2. Build fiber backbone and FTTP under a phased approach without Revenue
Bond. Allocate $34M from the Fiber fund and $13M from the Electric fund to
build the fiber backbone and build phase one of the FTTP distribution
network under a phased approach.
Option 3. Build fiber backbone, pause City-owned ISP plans, and collaborate with
private ISPs. Build the fiber backbone to increase reliability and capacity for
dark fiber licensing and support some “smart city” initiatives. Collaborate
with ISP providers to improve broadband in Palo Alto. Pause FTTP efforts and
potentially redistribute some fiber surplus to support other City initiatives
such as grid modernization or electrification.
20
Packet Pg. 268
City of Palo Alto Page 2
Executive Summary
This report is a culmination of research and findings for building fiber backbone and citywide
FTTP, including follow-up information from the joint session held on September 19, 2022 (Staff
Report #13956). This meeting will provide Council an opportunity to review the findings and
information accumulated so far, weigh the options for moving forward with FTTP, consider staff
and UAC’s recommendation, and provide staff with feedback on how to proceed. This report
contains follow-up information requested from the joint session and options to build FTTP:
1. Comparisons to other California municipalities providing fiber service
2. Use of Micro-trenching to reduce construction costs and accelerate FTTP buildout
3. Use of Fiber Fund surplus for FTTP and other City services
4. Options to build FTTP and enhance broadband in Palo Alto
The UAC unanimously (6-0, Bowie absent) recommended Option 2 – build fiber backbone and
FTTP under a phased approach with existing funds ($34 million from Fiber and $13 million from
Electric). The UAC did not recommend bond financing phase one of FTTP deployment due to
uncertainty of customer take rate and high interest borrowing rates. UAC also suggested staff
provide periodic updates (i.e. construction, installation, financials) to UAC and Council during
phase one to determine whether the City should accelerate the FTTP buildout.
Background
On November 2, 2022 staff reviewed with UAC the accumulated research and findings for
building the fiber backbone and several options for building FTTP shortly following the joint
session held with Council and UAC on September 19, 2022. Plans to build the fiber backbone
were reviewed and the feasibility of expanding the City’s current licensing of dark fiber to end
users by building out the network with a citywide FTTP distribution network were evaluated
during the joint study session. Building out the network to connect to homes and businesses is
commonly known as building out the “last mile” in a network. During the meeting with UAC,
options for building the last mile to provide citywide FTTP were discussed in further detail.
There have been multiple discussions over the decade regarding the tradeoffs for offering
municipal-owned FTTP to the community as a service, investments needed to build the “last
mile,” benefits and risks of becoming a new internet service provider (ISP), and various financial
models and organizational structure. The City’s broadband consultant, Magellan, presented
information demonstrating how building the last mile in the City’s fiber network infrastructure
supports the City’s initiatives, provides the community with more choice among broadband
providers, and becomes a valuable telecommunication investment for the future.
However, it was acknowledged that while the City operates its own utilities and the fiber
network has a successful dark fiber business model, the competitive landscape for FTTP would
be very different from managing other City-owned utilities and dark fiber licensing. The City
would not only have to build a reliable FTTP distribution network capable of delivering ultra-
high-speed Internet options, but also capture market share, provide responsive customer
service and support, implement and install FTTP, and respond to competitors’ efforts. In
response to some concerns about take rate viability, Magellan presented maps based on survey
20
Packet Pg. 269
City of Palo Alto Page 3
results that could be used to infer which areas of the community had higher demand for a City-
owned ISP, less competition from existing incumbents, or even where internet speed may be
slower. This report includes some additional maps based on UAC’s feedback.
Several options for building FTTP came into focus after the joint study session and were
summarily discussed with the UAC. During the meeting with the UAC, concerns were raised
regarding uncertainties surrounding the deployment of FTTP, such as rising interest rates,
which can impact cost estimates, or incumbent response and other market factors which can
impact anticipated take rates. The discussion section contains some ideas to mitigate these
risks weighed alongside the various options presented to the UAC for further discussion with
Council.
Discussion
The following discussion topics include follow-up information requested from the joint session
and options for building FTTP.
Comparisons to other Municipalities with Fiber
According to the Institute for Local Self-Reliance Community Broadband Map, in the United
States there are 83 municipal networks serving 148 communities with a publicly owned FTTP
citywide network. More broadly, there are 315 networks communities in 31 states with some
form of a publicly owned network offering at least 1 gigabit services and more than 30
communities in 10 states with a municipal network delivering 10 gigabit services.1
With limited resources to verify the business models of each municipal network, research was
focused on high-profile municipal networks for those which are out-of-state, and municipal
networks in California since unique legal conditions in each state substantially impact the
deployment of broadband networks state to state. Staff focused on well-known municipal
networks or municipal networks with easily accessible public information to search for those
which contained one or more of the following traits comparable to what the City of Palo Alto
has been exploring:
9 City-owned and built
9 City-ISP (City is the ISP provider)
9 City-wide (100% full buildout to all businesses and residents)
The resulting list in Attachment A is intended to provide the public a general idea of how other
municipalities are providing fiber broadband and would require the information to be further
validated by the individual municipal network listed for accuracy. In general, many
municipalities in California are either in the process of exploring fiber broadband, providing
limited fiber broadband, or partnering with a third-party as the ISP provider. Several cities are
1 Institute for Local Self-Reliance Community Broadband map: https://muninetworks.org/communitymap
20
Packet Pg. 270
City of Palo Alto Page 4
offering commercial services, and some are providing residential services under an incremental
approach. Below are some examples:
Masterplan to explore fiber broadband
1. Developing a master plan
Fremont
Glendale
2. Completed a master plan Vallejo
San Leandro
Limited, Non City-wide fiber broadband
3. Providing only dark fiber and
commercial ISP (no residential)
Burbank
Los Angeles
4. Providing limited (instead of city-wide
fiber broadband)
Loma Linda (homes built after 2004)
Rancho Cucamonga (greenfield developments)
Santa Monica (select multi-family developments)
San Bruno (hybrid fiber/coax; incumbent cable television operator)
Non City-ISP provided fiber broadband
5. Non-city ISP provider Culver City (CLEC Onward)
Ontario (CLEC Onward)
Vallejo (CLEC Onward)
Santa Cruz (Cruzio Internet)
Brentwood licenses conduit to Sonic* in select areas
Micro-trenching as an Alternative Construction Method
In response to Council’s question on micro-trenching as a method to reduce construction costs,
Utilities, Public Works and Magellan evaluated the benefits and disadvantages of applying
micro-trenching in Palo Alto’s FTTP buildout.
In recent years, “micro-trenching” has emerged as a new construction method for deploying
fiber infrastructure. Whereas traditional standards call for fiber to be buried at least 24-36
inches below grade either with directional boring or trenching, micro-trenching uses thinner,
shallower cuts averaging 8-16 inches in depth. Typically, these cuts are made either in the
pavement, sidewalk or the joint between the pavement and guttering. Proponents of micro-
trenching note the shallower placement reduces construction cost by 40%-50% and can
accelerate construction schedules by three to four times of the typical timeline. Street closures
can also be reduced. Thus far, only a few cities (e.g., Austin, Texas) have successfully allowed
fiber network builders to use micro-trenching as the preferred construction methodology to
deploy broadband and other telecommunication services. As of the writing of this report, the
City has not received any inquiries from the incumbent ISPs (wired and wireless) to use micro-
trenching to deploy fiber.
The biggest downside to micro-trenching is the shallow depth and how this impacts future
access to underground facilities for maintenance or repairs. While micro-trenching employs
cuts averaging 8-16 inches deep, Palo Alto streets are resurfaced at 6 to 18 inches deep, and
20
Packet Pg. 271
City of Palo Alto Page 5
routine cuts to access underground facilities such as electric underground lines are 24 to 36
inches deep. Additionally, to minimize damage to trees in areas where such precaution may be
necessary, new techniques in micro-trenching may need to be deployed which may significantly
impact construction. Finally, streets and sidewalks are routinely cut by contractors for
underground work who would have to exercise greater precision and control to avoid cutting
shallow fiber. The shallow depths not only impact how underground facilities are accessed in
the future, but this also increases the vulnerability of micro-trenched fiber to dig ups. Although
there are benefits for implementing micro-trenching, the tradeoffs in future access, ease of
maintenance, risk of outages, and the time and resources to mitigate these issues would need
to be considered as well.
Effective January 2022, state law requires the City to allow micro-trenching for the installation
of underground fiber if the installation in the micro-trench is limited to fiber, unless the City
makes a written finding that allowing micro-trenching for a fiber installation would have a
specific, adverse impact on the public health or safety. The law also requires the City to adopt
or amend its existing policies, ordinances, codes, or construction rules to allow for micro-
trenching for fiber. The Public Works Department in collaboration with the City Attorney’s
Office, Utilities and Urban Forestry, is developing a micro-trenching standard for the City which
will further inform on how this method may be implemented in the future.
As a part of ongoing efforts to explore potential cost savings for building FTTP, staff evaluated
micro-trenching and found although there may be construction cost savings of $10 – $20
million (high-level estimate) in the undergrounded “last mile” portions of a FTTP network, these
savings are unlikely to be fully realized due to cost factors which would offset a significant
portion of potential savings. These cost factors include but are not limited to future outages,
tree root protection, and impacts to street conditions. For a more thorough calculation of cost
savings applying micro-trench, Magellan may need up to four months to conduct a field survey
in the underground areas and redesign the architecture of the FTTP network.
The cost savings and efficiency gained by using micro trenching is achieved when it is used in
large volume. To realize the full savings of micro-trench, the City would have to utilize micro-
trenching for largest part of underground construction. One of the advantages that allows
micro trenching to be less costly is the depth placement of conduit, which is also a disadvantage
to operating and maintaining the network. The cost to mobilize equipment and crew to micro-
trench small segments and/or service laterals would potentially outweigh construction cost
savings combined with high risk of future outages.
Although micro-trenching has the potential to result in reduced construction costs and faster
implementation time in high density and rural areas, staff’s research suggests there are long-
term risks in Palo Alto such as network reliability, impact to the City’s street pavement
condition index score, and the effect on tree canopies – all of which may significantly outweigh
the construction benefits. Based on these concerns, staff does not recommend using micro-
trenching to build the FTTP network in underground districts.
20
Packet Pg. 272
City of Palo Alto Page 6
Use of Fiber surplus for FTTP and other City services
To date, the City has accumulated a surplus of approximately $34 million through its dark fiber
licensing revenues. This surplus may be used for capital expenditures related to the fiber
network, including the FTTP buildout, or as unrestricted funds. However, potential uses for
future surpluses—particularly from a FTTP line of business—are less clear. The legal landscape
for local government revenues, especially revenues contemplated for expenditure on general
municipal services, is complex and frequently changing either through ballot initiatives or
litigation over the meaning and application of existing laws.
For present purposes, the most significant State law is Prop 262, the 2010 ballot initiative that
amended the California Constitution to expand the general definition of a tax to include
"any…charge…of any kind imposed by a local government." If Prop 26 applies to a local
government fee or charge, the charge must either fall within one of seven enumerated
exceptions or be imposed with voter approval. The stated intent behind Prop 26 was to close a
"loophole" under which the state and local governments "disguised new taxes as 'fees' in order
to extract even more revenue from California taxpayers without having to abide by these
constitutional voting requirements."
In August of this year, the California Supreme Court issued a decision in Zolly vs. City of Oakland
that clarified a previously unresolved question about the meaning of the phrase "imposed by a
local government" in the context of franchise fees paid by the City of Oakland's solid waste and
recycling service providers. The City of Oakland argued the franchise fees are not “imposed by a
local government” because they were a product of voluntary contractual negotiations and
therefore Prop 26 does not apply to such fees. The Supreme Court rejected this argument,
finding that the ordinary meaning of the word “imposed” is “enacted” and it is sufficient that
2 The full text of the Prop 26 provisions discussed in this staff report is as follows:
(e) As used in this article, “tax” means any levy, charge, or exaction of any kind imposed by a local government,
except the following:
(1) A charge imposed for a specific benefit conferred or privilege granted directly to the payor that is not provided to
those not charged, and which does not exceed the reasonable costs to the local government of conferring the benefit
or granting the privilege.
(2) A charge imposed for a specific government service or product provided directly to the payor that is not provided
to those not charged, and which does not exceed the reasonable costs to the local government of providing the
service or product.
(3) A charge imposed for the reasonable regulatory costs to a local government for issuing licenses and permits,
performing investigations, inspections, and audits, enforcing agricultural marketing orders, and the administrative
enforcement and adjudication thereof.
(4) A charge imposed for entrance to or use of local government property, or the purchase, rental, or lease of local
government property.
(5) A fine, penalty, or other monetary charge imposed by the judicial branch of government or a local government,
as a result of a violation of law.
(6) A charge imposed as a condition of property development.
(7) Assessments and property-related fees imposed in accordance with the provisions of Article XIII D (Prop 218).
Article XIIC, Section 1(e) of the California Constitution.
20
Packet Pg. 273
City of Palo Alto Page 7
Oakland exercised its legal authority to execute the two franchise agreements and then
enacted those charges into law by ordinance.
A court could very possibly reach a similar conclusion regarding the applicability of Prop 26 to
the City's fiber utility's charges. However, the full impact of the Zolly decision on the fiber utility
rates is still unclear because another Prop 26 case (Howard Jarvis Taxpayers Association vs. Bay
Area Toll Authority) currently pending before the California Supreme Court will provide
additional guidance on how the various exceptions to Prop 26 might apply in the context of the
fiber utility. In addition, a statewide initiative that would impose even stricter limitations on
local government fees and charges3 appears likely to qualify for the ballot in the 2024 election.
Staff will continue to monitor these developments and their implications for the fiber utility's
revenue potential.
Options to build FTTP and enhance broadband service in Palo Alto
Upon review of the findings from the joint session and follow-up information incorporated into
this report, options for providing FTTP and enhancing broadband service were narrowed down
to the following three:
Option 1. Build backbone and citywide FTTP within five years with $98M of Revenue Bond
financing
Based on the current business models and anticipated construction costs for the fiber backbone
and FTTP distribution network, the project team estimates a funding gap of approximately $98
million to be allocated in both the Fiber Fund and Electric Fund. The allocation for the
construction costs and the bond financing structure would still need to be finalized based on
actual plans.
Table 1. Cost estimates as of August 2022
Costs Current Estimates 2022
Fiber Backbone $25.6 M
Fiber-to-the-Premise $102.3 M
Working Capital Set Aside $15.0 M
Total Costs $142.9 M
Funding Current Estimates 2022
Cost Savings if Built Jointly ($10.9 M)
Existing Fiber Fund ($34.0 M)
Total New Funding Required $98.0 M
3 Taxpayer Protection and Government Accountability Act
20
Packet Pg. 274
City of Palo Alto Page 8
New Funding Allocation* Fiber $80 M - $85 M
Electric $13 M - $18 M
Utility Revenue Bonds are a cost-effective option for long-term capital improvement projects
with a large funding gap such as the $98M in this project. There are various revenue bonds
structures and potential bond rating assumptions which could be applied for the fiber backbone
and FTTP distribution network. The bonds are anticipated to be issued by both the Fiber Optics
Fund for the FTTP construction, and the Electric Fund for the building the fiber backbone. Given
the financial strength of the reserves, assets, and lack of current debt by the Electric utility, the
new revenue bonds for the backbone and FTTP has little to no impact on potential future
Electric bond issuance(s). The Electric utility is the City’s financially strongest utility, and the
other City utilities with outstanding bonds have a triple A credit ratings from Standard and
Poor’s therefore, staff anticipates the Electric utility will also receive the highest credit rating as
well. The credit rating for the Fiber utility is less certain, as is the credit rating for a combined
(Fiber and Electric) bonds issuance. The allocation of construction costs between the two funds
and the bond financing structure is still being finalized, and the potential bond rating will be
determined after a rating presentation (usually occurs a month prior to the bond issuance).
Typically, bond issuance takes three to four months to process. Sometimes it takes longer when
associated to a major construction contract Request for Proposal (RFP) being issued, receiving
proposals, and evaluating bids to determine the lowest responsible bidder. The RFP process
needs to be substantially completed to determine the final amount of the bond issuance since
this result can substantially differ from the Engineer project cost estimate.
The table below shows the preliminary potential bond structures estimated on a 30-year $98
million (par) bond issuance needed to fund the project gap for the following scenarios:
Table 2. Debt Schedule as of August 2022
Scenario Rating Capitalized
Interest
All-In True
Interest
Cost
Annual
Average Debt
Service
Total Debt
Service
(net of Capl)
1* AA+ 18 months 4.42% $6.47 M $186.82 M
2** AA+ None 4.42% $5.96 M $179.45 M
3* AA 18 months 4.49% $6.52 M $188.34 M
4** AA None 4.49% $6.01 M $180.80 M
*Scenarios 1 and 3: 18 months Capitalized Interest. First three (interest only) semi-annual debt
service payments during project construction are paid by bond funds. This amount is added to
the principal bond issuance.
**Scenarios 2 and 4: No capitalized interest. First debt service would be due six months after the
bond issuance.
20
Packet Pg. 275
City of Palo Alto Page 9
The advantages of option 1 is it seeks to provide every resident and business in Palo Alto with
equal access to Palo Alto Fiber within five years so construction bid prices may be more
favorable and competitive with a larger construction project and economies of scale, and the
customer take rates would be higher due to speed-to-market and favorable momentum for
Palo Alto Fiber. In addition to providing fiber broadband high-speed internet access to all
residents and businesses, the new citywide FTTP network will provide more dark fiber licensing
opportunities, support “smart city” initiatives for all departments, and serve as future
telecommunication infrastructure for City or third-party needs throughout the city. There may
also be construction savings of approximately 25% in areas where the City must replace poles
or trench to upgrade its electrical distribution system for the grid modernization project. Staff
will evaluate total FTTP and grid modernization joint savings after the construction buildout
areas are determined.
The disadvantages of option 1 is the reliance on customer take rate to be financially
sustainable. Given the uncertainties with take rate, and how existing ISP incumbents may
respond in an adversarial manner to further impact take rates, there is a risk if Palo Alto Fiber
cannot attain a minimum take rate of 25% after the build out, the City will have to adjust the
construction schedule, and/or identify other sources (i.e. General Fund, partnership) of funding
to repay the outstanding debt amount.
Option 2. Build fiber backbone and FTTP with existing fiber reserves under a phased
approach.
The phased deployment approach will decrease the City’s financial risk because no bond
financing is required. Construction can be phased to start in areas with higher potential take
rates and lower accessibility to fiber internet, ramping up instead of committing to a full
deployment. With a ramp up approach, the City could build up revenues from dark fiber
licensing and incremental FTTP deployments to reinvest into building out more of the fiber
network in subsequent areas instead of taking on new debt. Depending on the viability of the
City ISP business and take rate percentage, Council has the option to accelerate the FTTP
citywide buildout with or without bond financing in future years. If Council decides to bond
finance the remainder of citywide FTTP, it will take about two months to complete financing
after Council’s approval. Staff will provide key financial information (i.e. number of subscribers,
revenue, expenses) on a regular basis to inform Council of the new FTTP business and decide
whether the City should 1) continue the phased approach; 2) seek revenue bond financing to
accelerate the buildout; 3) pause FTTP expansion.
Under the phased approach, and assuming a $20 million capital investment in the first phase of
FTTP, the City can provide access to 20% - 30% of homes and businesses within 2-3 years.
Magellan developed three potential phase 1 deployment models for FTTP with the following
initial build out scenarios (Attachment B):
1. Areas with the lowest construction cost and highest density
2. Areas with the highest demand based on deposits
20
Packet Pg. 276
City of Palo Alto Page 10
3. Areas without AT&T Fiber, with the goal of increasing take rates
Areas of Lowest
Cost, Highest
Density
Areas of Highest
Demand
Areas without
AT&T Fiber
Total Cost $20 M $20 M $20 M
Single-Family with Access 4,231 3,540 3,186
Multi-Family with Access 6,213 4,712 4,594
Total Homes with Access 10,444 8,252 7,780
Total Businesses with
Access 1,064 879 912
Total Homes and
Businesses with Access 11,508 9,131 8,692
In response to the UAC request, Magellan created a citywide heatmap of broadband speed
from 25 Mbp to 1Gbp or greater based on 1,700 self-reported speed tests from the internet
survey (Attachment C). Based on the heatmap, it is inconclusive whether there are
neighborhoods with access to only low speed internet since there are adjacent neighborhoods
with moderate to high speed access. Therefore, the speed test heatmaps are not a conclusive
indicator of areas where speeds may be lower, and resultantly where the City should
potentially focus its initial deployment.
For option 2, Magellan created a construction map for phase 1 incorporating the various
scenarios of lowest cost, highest density, least competition, and highest demand. (Attachment
D).
The advantage of option 2 is no initial bond financing required with $34 million of Fiber
Reserves and $13 million contributions from the Electric Fund. The City can build a dedicated
fiber backbone for the Electric utility to enhance reliability, security, redundancy, and future
electric-related initiatives such as automated SCADA sensors. The City will also be able to
provide internet access to approximately 20% - 30% of homes and residents who prefer to
switch to City-owned ISP. Council can decide whether to accelerate or decelerate the FTTP
expansion plan in one or two years based on how the deployment of Phase 1 goes. Under
Option 2, the City will evaluate the feasibility of integrating FTTP expansion into future capital
improvement projects such as electric grid modernization, electrification and undergrounding.
Similar to option 1, there may be construction savings of approximately 25% in areas where the
City must replace poles or trench to upgrade its electrical distribution system for the grid
modernization project. Staff will evaluate total FTTP and grid modernization joint savings after
phase 1 construction buildout areas are determined.
20
Packet Pg. 277
City of Palo Alto Page 11
The disadvantages of Option 2 are primarily the longer implementation time, which would likely
impact take rates and weaken the City’s competitive position. Additionally, the phased
approach could impact equity as not all parts of the community would be built out at the same
time.
Option 3. Build fiber backbone, Pause City-owned ISP plans, and Collaborate with private ISPs
The City’s fiber optic backbone network was planned, designed and constructed in the mid to
late 1990s. The estimated useful life of fiber plant is typically 30 to 40 years. The Utilities
Department began to license dark fiber service connections for commercial purposes in the late
1990s. Since then, several sections of the fiber network have reached capacity, which limits the
City's ability to effectively serve its existing customers and acquire new customers. Building a
separate fiber backbone is essential to maintaining and improving network reliability and
security, increasing network coverage and capacity for commercial dark fiber customers, and
supporting some “smart city” initiatives for City departments such as Information Technology,
Office of Emergency Services, Public Safety, Public Works, Transportation and Utilities.
In light of future network and service upgrade plans by incumbent ISPs and the development of
emerging technologies such as mobile and fixed 5G wireless, instead of building out the last
mile the same time, the City may postpone its FTTP development. The City could, instead,
identify resources to improve coordination of City policies, processes, and access to
communication infrastructure in the PROW to facilitate network upgrades. These efforts could
enhance transparency and predictability for ISPs planning network upgrades and/or building
new networks. In general, municipal strategies for advancing broadband deployment without
building their own FTTP are typically to:
1. Facilitate access to key assets such as fiber, communication conduit and utility poles in
the PROW in addition to access to City-owned real property;
2. Provide frequently used information such as construction standards and pole
requirements to potential broadband service providers; and
3. Streamline and publicize local construction, PROW and permitting processes.
The advantages of option 3 are primarily proceeding with building the fiber backbone to
provide a more robust network and increase capacity, possibly creating new opportunities for
dark fiber licensing in fiber congested areas such as Stanford Research Park. Dark fiber reserve
could be redistributed to support other City initiatives such as grid modernization or
electrification. The City could focus on working with existing ISPs to enhance broadband service
throughout the City with substantially less financial investment and risk for the City as delivery
of broadband services would be completely controlled by the existing providers.
The disadvantages of option 3 are postponing FTTP. If FTTP is resumed at a future date, labor
and materials costs may be higher, other providers may be further entrenched, and the City’s
share of the market would be limited or more difficult to compete in, dwindling any chances for
the community to invest in a community-owned asset. New technologies such as wireless could
20
Packet Pg. 278
City of Palo Alto Page 12
evolve and become widely available to residents and businesses, further reducing market
share. The net impact to the community for postponing FTTP is a loss in the opportunity for the
City to maintain local control of how broadband is delivered in Palo Alto citizens and
businesses, in terms of pricing, internet speeds, net neutrality, and privacy.
UAC/Staff Feedback
The UAC noted there are portions of the City which present an opportunity for more parity
based on the map provided from Magellan. The map indicates there are areas with higher
customer demand and no fiber broadband options which the City can serve. With rising interest
rates and uncertain take rates, UAC recommended option 2 which provides a path forward that
helps mitigate some of these risks. Staff acknowledged the phased approach was previously not
recommended as it stretched the implementation timeline to 10-20 years. However, the
phased approach provides the City opportunities to check in and evaluate if take rates justify
continued build out at a faster pace. Construction of the first phase may take 12 – 24 months;
however, community interest may accelerate the construction of subsequent phases.
Furthermore, the phased construction approach may also be more efficient and cost-effective
to overlap with the electric grid modernization work.
Resource Impact
Option 1
↓
Option 2 Option 3
Description
Fiber Backbone +
FTTP in 5 Years
Fiber Backbone +
$20M FTTP Phase 1
Fiber Backbone +
Pause City FTTP
Expenses:
Fiber Backbone $25.6 M $25.6 M $25.6 M
Fiber-to-the-Premises $102.3 M $20.0 M $ 0.0
Working Capital $15.0 M $3.0M $ 0.0
Total Costs $142.9 M $48.6 M $25.6 M
Funding:
Cost Savings if Built
Jointly ($10.9 M) ($1.7 M) -
Fiber Fund ($34.0 M) ($34.0 M) ($12.8 M)
Electric Fund $0.00 ($12.9 M) ($12.8 M)
Total Funding/Savings ($57.9M) ($48.6 M) ($25.6 M)
New Funding Required $98.0M* $ 0.0 $ 0.0
* Allocation of revenue bond between Electric and Fiber will be determined based on actual
construction costs.
20
Packet Pg. 279
City of Palo Alto Page 13
The staffing resources needed to implement, support, and manage options 1 and 2 are included
in the working capital estimates in the table above. However, staff will need to bring forward
new job classifications if the City Council chooses to proceed with options 1 or 2.
Stakeholder Engagement
On September 19, 2022, City Council and UAC held a public joint session to discuss the City’s
fiber expansion plan and specifically FTTP. The joint session included information for Council
and UAC as they consider next steps for FTTP. The information included FTTP engineering
design details, construction cost estimates, market analysis results, financial models, and
organizational structure options.
On November 2, 2022, the UAC unanimously (6-0, Bowie absent) recommended Option 2 –
build fiber backbone and FTTP under a phased approach with existing funds ($34 million from
Fiber and $13 million from Electric) (Staff Report #14845). UAC expressed the goal of FTTP is to
provide ubiquitous or citywide high-speed internet access to all residents and businesses in
Palo Alto. Option 2 can become a springboard to Option 1 which is a citywide FTTP deployment
within five years if phase 1 of Palo Alto Fiber is deemed to be successful and financially self-
sustaining. Other topics raised during the meeting was partnering with an ISP service since the
City does not have experience in this area which is under consideration with the outsourcing
model. Multi-dwelling unit (MDU) is a different market segment than residential or commercial
customers because most property owners have an exclusive multi-year term contractual
agreement with an ISP. If Council approves Option 1 or 2, the City will begin a MDU marketing
campaign in the beginning of 2023 to gauge MDU interest in negotiating access agreement.
Environmental Review
Council action on this item is not a project as defined by CEQA because the Council direction regarding
the fiber utility is a general policy making activity. CEQA Guidelines section 15378(b)(2)."
Attachments:
x Attachment20.a: Attachment A: Comparison to Other Municipalities
20
Packet Pg. 280
ƚ
ƚ
Ă
Đ
Ś
ŵ
Ğ
Ŷ
ƚ
͘
Ž
ŵ
Ɖ
Ă
ƌ
ŝ
Ɛ
Ž
Ŷ
ƚ
Ž
K
ƚ
Ś
Ğ
ƌ
D
Ƶ
Ŷ
Đ
ŝ
Ɖ
Ă
ů
ŝ
ƚ
ŝ
Ğ
Ɛ
ǁ
ŝ
ƚ
Ś
&
ŝ
ď
Ğ
ƌ
dž
Ă
ŵ
Ɖ
ů
Ğ
Ɛ
Ž
Ĩ
D
Ƶ
Ŷ
ŝ
Đ
ŝ
Ɖ
Ă
ů
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
WƵ
ď
ů
ŝ
Đ
Ő
Ğ
Ŷ
Đ
LJ
^
ƚ
Ă
ƚ
Ğ
d
LJ
Ɖ
Ğ
h
ƚ
ŝ
ů
ŝ
ƚ
ŝ
Ğ
Ɛ
K
Ĩ
Ĩ
Ğ
ƌ
Ğ
Ě
/Ŷ
Đ
Ƶ
ŵ
ď
Ğ
Ŷ
ƚ
d
Ğ
ů
Ğ
Đ
Ž
ŵ
Wƌ
Ž
ǀ
ŝ
Ě
Ğ
ƌ
Ɛ
WƵ
ď
ů
ŝ
Đ
ů
LJ
Ž
ǁ
Ŷ
Ğ
Ě
ĂŶ
Ě
ď
Ƶ
ŝ
ů
ƚ
WƵ
ď
ů
ŝ
Đ
Ő
Ğ
Ŷ
Đ
LJ
ĂƐ
ƚ
Ś
Ğ
/
^
W
ϭϬ
Ϭ
й
ďƵ
ŝ
ů
Ě
Ž
Ƶ
ƚ
Ě
Ě
ŝ
ƚ
ŝ
Ž
Ŷ
Ă
ů
Đ
Ž
ŵ
ŵ
Ğ
Ŷ
ƚ
Ɛ
ͬ
Ŷ
Ž
ƚ
Ğ
Ɛ
ůĂ
ŵ
Ğ
Ě
Ă
͕
ŝ
ƚ
LJ
Ž
Ĩ
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
z
Ğ
Ɛ
z
Ğ
Ɛ
z
Ğ
Ɛ
tŚ
Ğ
Ŷ
ŝ
Ŷ
ŝ
ƚ
ŝ
Ă
ů
ů
LJ
ď
Ƶ
ŝ
ů
ƚ
͕
ƚ
Ś
Ğ
ŝ
ƚ
LJ
Ś
ŝ
ƌ
Ğ
Ě
Ă
Đ
Ž
Ŷ
ƚ
ƌ
Ă
Đ
ƚ
Ž
ƌ
ƚ
Ž
Ɖ
ƌ
Ž
ǀ
ŝ
Ě
Ğ
/
^
W
Ɛ
Ƶ
Ɖ
Ɖ
Ž
ƌ
ƚ
͘
,
LJ
ď
ƌ
ŝ
Ě
&
ŝ
ď
Ğ
ƌ
Ž
Ă
dž
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
Ɛ
Ž
ů
Ě
ƚŽ
Ž
ŵ
Đ
Ă
Ɛ
ƚ
ŝ
Ŷ
Ϯ
Ϭ
Ϭ
ϴ
͘
Ŷ
Ă
Ś
Ğ
ŝ
ŵ
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
h
Ŷ
ŬŶ
Ž
ǁ
Ŷ
z
Ğ
Ɛ
ƚ
Ś
Ğ
ƌ
ƚ
Ž
Ŷ
͕
d
Ž
ǁ
Ŷ
Ž
Ĩ
E
Ž
Ŷ
Ğ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
h
Ŷ
Ŭ
Ŷ
Ž
ǁ
Ŷ
t
/
W
ƚ
Ś
Ğ
ƌ
ƚ
Ž
Ŷ
&
ŝ
ď
Ğ
ƌ
ŝƐ
Ŷ
Ž
ƚ
Ă
ŵ
Ƶ
Ŷ
ŝ
Đ
ŝ
Ɖ
Ă
ů
ů
LJ
Ͳ
Ž
ǁ
Ŷ
Ğ
Ě
Ğ
Ŷ
ƚ
Ğ
ƌ
Ɖ
ƌ
ŝ
Ɛ
Ğ
͘
ŝ
ƚ
LJ
Ă
ů
ů
Ž
ǁ
Ğ
Ě
W
Z
K
t
Ă
Đ
Đ
Ğ
Ɛ
Ɛ
͘
Ğ
ǀ
Ğ
ƌ
ů
LJ
,
ŝ
ů
ů
Ɛ
͕
ŝ
ƚ
LJ
Ž
Ĩ
^
Ž
ů
ŝ
Ě
t
Ă
Ɛ
ƚ
Ğ
͕
^
ƚ
Ž
ƌ
ŵ
t
Ă
ƚ
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
t
Ă
ƚ
Ğ
ƌ
͕
tĂ
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
z
Ğ
Ɛ
h
Ŷ
Ŭ
Ŷ
Ž
ǁ
Ŷ
z
Ğ
Ɛ
ŝ
ƚ
LJ
ǁ
ŝ
Ě
Ğ
&
d
d
W
Ɖ
ƌ
Ž
ũ
Ğ
Đ
ƚ
Ś
Ă
ů
ƚ
Ğ
Ě
Ă
Ŷ
Ěƌ
Ğ
Ɛ
Ƶ
ŵ
Ğ
Ě
͕
Đ
Ƶ
ƌ
ƌ
Ğ
Ŷ
ƚ
ů
LJ
ƚ
Ś
Ğ
Ɖ
ƌ
Ž
ũ
Ğ
Đ
ƚ
ŝ
Ɛ
Ƶ
Ŷ
Ě
Ğ
ƌ
Ğ
ǀ
Ă
ů
Ƶ
Ă
ƚ
ŝ
Ž
Ŷ
͘
ƌ
Ğ
Ŷ
ƚ
ǁ
Ž
Ž
Ě
͕
ŝ
ƚ
LJ
Ž
Ĩ
'
Ă
ƌ
ď
Ă
Ő
Ğ
͕
^
Ğ
ǁ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
ŝ
ƚ
LJ
ů
Ğ
Ă
Ɛ
Ğ
Ɛ
Đ
Ž
Ŷ
Ě
Ƶ
ŝ
ƚ
ƚ
Ž
^
Ž
Ŷ
ŝ
Đ
ŝ
Ŷ
Ɛ
Ğ
ů
Ğ
Đ
ƚ
Ă
ƌ
Ğ
Ă
Ɛ
Ž
Ĩ
ƚ
Ś
Ğ
Đ
ŝ
ƚ
LJ
ǁ
Ś
Ğ
ƌ
Ğ
ƚŚ
Ğ
Đ
ŝ
ƚ
LJ
ŝ
Ŷ
Ɛ
ƚ
Ă
ů
ů
Ğ
Ě
Ğ
ŵ
Ɖ
ƚ
LJ
Đ
Ž
ŵ
ŵ
Ƶ
Ŷ
ŝ
Đ
Ă
ƚ
ŝ
Ž
Ŷ
ĐŽ
Ŷ
Ě
Ƶ
ŝ
ƚ
Ĩ
Ž
ƌ
Ŷ
Ğ
ǁ
Ś
Ž
Ƶ
Ɛ
ŝ
Ŷ
ŐĚ
Ğ
ǀ
Ğ
ů
Ž
ƉŵĞ
Ŷ
ƚ
Ɛ
͘
Ƶ
ƌ
ď
Ă
Ŷ
Ŭ
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
ϭ͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
z
Ğ
Ɛ
E
Ž
Ž
ŵ
ŵ
Ğ
ƌ
Đ
ŝ
Ă
ů
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
Ă
Ŷ
Ě
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ^
Ğ
ƌ
ǀ
ŝ
Đ
Ğ
Ɛ
͘
Ś
Ă
ƚ
ƚ
Ă
Ŷ
Ž
Ž
Ő
Ă
W
d
E
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
&
ŝ
ď
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
z
Ğ
Ɛ
z
Ğ
Ɛ
z
Ğ
Ɛ
Ś
Ă
ƚ
ƚ
Ă
Ŷ
Ž
Ž
ŐĂ
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
W
Ž
ǁ
Ğ
ƌ
Ž
Ă
ƌ
Ě
&
d
d
W
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
Ž
Ĩ
Ĩ
Ğ
ƌ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
͕
Ă
ď
ů
Ğ
d
s
Ă
Ŷ
Ě
W
Ś
Ž
Ŷ
Ğ
Ƶ
ů
ǀ
Ğ
ƌ
ŝ
ƚ
LJ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
E
Ž
E
Ž
W
Ă
ƌ
ƚŶ
Ğ
ƌ
Ɛ
Ś
ŝ
Ɖ
ǁ
ŝ
ƚ
Ś
>
K
Ŷ
ǁ
Ă
ƌ
Ě
Ͳ
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
͘
&Ž
ƌ
ƚ
Ž
ů
ů
ŝ
Ŷ
Ɛ
͕
ŝ
ƚ
LJ
Ž
Ĩ
K
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
͕
^
ƚ
Ž
ƌ
ŵ
ǁ
Ă
ƚ
Ğ
ƌ
͕
&
ŝ
ď
Ğ
ƌ
Ğ
Ŷ
ƚ
Ƶ
ƌ
LJ
>
ŝ
Ŷ
Ŭ
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
z
Ğ
Ɛ
z
Ğ
Ɛ
t
/
W
&Ž
ƌ
ƚ
Ž
ů
ů
ŝ
Ŷ
Ɛ
h
ƚ
ŝ
ů
ŝ
ƚ
ŝ
Ğ
Ɛ
ŝ
Ɛ
Ă
ŵ
Ğ
ŵ
ď
Ğ
ƌ
Ž
Ĩ
ƚ
Ś
Ğ
W
ů
Ă
ƚ
ƚ
Ğ
Z
ŝ
ǀ
Ğ
ƌ
W
Ž
ǁ
Ğ
ƌ
Ƶ
ƚŚ
Ž
ƌ
ŝ
ƚ
LJ
͘
&
d
d
W
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
ď
ƌ
Ă
Ŷ
Ě
Ŷ
Ă
ŵ
Ğ
ŝ
Ɛ
Ž
Ŷ
Ŷ
Ğ
dž
ŝ
Ž
Ŷ
Ͳ
Ž
Ĩ
Ĩ
Ğ
ƌ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
͕d
s
Ă
Ŷ
Ě
W
Ś
Ž
Ŷ
Ğ
&ƌ
Ğ
Ɛ
Ŷ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
^
Ğ
ǁ
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
'ů
Ğ
Ŷ
Ě
Ă
ů
Ğ
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
W
ůĂ
Ŷ
Ŷ
ŝ
Ŷ
Ő
W
ů
Ă
Ŷ
Ŷ
ŝ
Ŷ
Ő
Ƶ
ƌ
Ğ
Ŷ
ƚ
ů
LJ
ǁ
Ž
ƌ
Ŭ
ŝ
Ŷ
Ő
ǁ
ŝ
ƚ
Ś
D
Ă
Ő
Ğ
ů
ů
Ă
Ŷ
Ž
Ŷ
&
ŝ
ď
Ğ
ƌ
K
Ɖ
ƚ
ŝ
Đ
Ƶ
ƐŝŶ
Ğ
Ɛ
Ɛ
W
ů
Ă
Ŷ
Ĩ
Ž
ƌ
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
Ğ
dž
Ɖ
Ă
Ŷ
Ɛ
ŝ
Ž
Ŷ
,ŝ
ů
ů
Ɛ
ď
Ž
ƌ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
K
Z
t
Ă
ƚ
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
Ğ
Ŷ
ƚ
Ƶ
ƌ
LJ
>
ŝ
Ŷ
Ŭ
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
z
Ğ
Ɛ
z
Ğ
Ɛ
t/
W
,
ŝ
ů
ů
Ɛ
ď
Ž
ƌ
Ž
&
d
d
W
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
ď
ƌ
Ă
Ŷ
Ě
Ŷ
Ă
ŵ
Ğ
ŝ
Ɛ
,ŝ
Ő
Ś
>
ŝ
Ő
Ś
ƚ
Ͳ
ŽĨ
Ĩ
Ğ
ƌ
Ɛ
,
ŝ
Ő
Ś
Ͳ
^
Ɖ
Ğ
Ğ
Ě
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
Ă
Ŷ
Ě
ŝ
Ő
ŝ
ƚ
Ă
ů
s
Ž
ŝ
Đ
Ğ
,Ƶ
Ŷ
ƚ
ŝ
Ŷ
Ő
ƚ
Ž
Ŷ
Ğ
Ă
Đ
Ś
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
ϭ͕
^
Ğ
ǁ
Ğ
ƌ
͕
d
ƌ
Ă
Ɛ
Ś
͕
t
Ă
ƚ
Ğ
ƌ
^Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
͖
&
ƌ
Ž
Ŷ
ƚ
ŝ
Ğ
ƌ
E
Ž
E
Ž
E
Ž
>Ă
Ĩ
Ă
LJ
Ğ
ƚ
ƚ
Ğ
h
ƚ
ŝ
ů
ŝ
ƚ
ŝ
Ğ
Ɛ
^
LJ
Ɛ
ƚ
Ğ
ŵ
>
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
͕
&
ŝ
ď
Ğ
ƌ
d
Θ
d
͖
Ž
dž
z
Ğ
Ɛ
z
Ğ
Ɛ
z
Ğ
Ɛ
>
h
^
&
/
Z
&
d
d
W
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
Ž
Ĩ
Ĩ
Ğ
ƌ
Ɛ
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
/
Ŷ
ƚ
Ğ
ƌ
ŶĞ
ƚ
͕
Ă
ď
ů
Ğ
d
s
Ă
Ŷ
Ě
W
Ś
Ž
Ŷ
Ğ
>Ž
ŵ
Ă
>
ŝ
Ŷ
Ě
Ă
͕
ŝ
ƚ
LJ
Ž
Ĩ
Z
Ğ
Ĩ
Ƶ
Ɛ
Ğ
͕
^
Ğ
ǁ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
zĞ
Ɛ
ϭ
zĞ
Ɛ
E
Ž
ŝ
ƚ
LJ
&
d
d
W
ŝ
Ɛ
ů
ŝ
ŵ
ŝ
ƚ
Ğ
Ě
ƚ
Ž
Ɛ
Ğ
ů
Ğ
Đ
ƚ
Ă
ƌ
Ğ
Ă
Ɛ
Ă
Ŷ
Ě
Ɖ
ƌ
Ž
ǀ
ŝ
Ě
Ğ
Ě
ƚ
Ž
Ś
Ž
ŵ
Ğ
Ɛ
ď
Ƶ
ŝ
ů
ƚ
Ă
Ĩ
ƚ
Ğ
ƌ
Ϯ
Ϭ
Ϭ
ϰ
͘
>Ž
Ŷ
Ő
Ğ
Ă
Đ
Ś
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
ϭ͕
'
Ă
Ɛ
͕
t
Ă
ƚ
Ğ
ƌ
^Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
͖
&
ƌ
Ž
Ŷ
ƚ
ŝ
Ğ
ƌ
E
Ž
E
Ž
E
Ž
>Ž
Ŷ
Ő
ŵ
Ž
Ŷ
ƚ
͕
ŝ
ƚ
LJ
Ž
Ĩ
K
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
&
ŝ
ď
Ğ
ƌ
Ğ
Ŷ
ƚ
Ƶ
ƌ
LJ
>
ŝ
Ŷ
Ŭ
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
z
Ğ
Ɛ
z
Ğ
Ɛ
z
Ğ
Ɛ
>Ž
Ŷ
Ő
ŵ
Ž
Ŷ
ƚ
W
Ž
ǁ
Ğ
ƌ
Θ
Ž
ŵ
ŵ
Ƶ
Ŷ
ŝ
Đ
Ă
ƚ
ŝ
Ž
Ŷ
Ɛ
ŝ
Ɛ
Ă
ŵ
Ğ
ŵ
ď
Ğ
ƌ
Ž
Ĩ
ƚ
Ś
Ğ
W
ů
Ă
ƚ
ƚ
Ğ
Z
ŝ
ǀ
Ğ
ƌ
W
Ž
ǁ
Ğ
ƌ
Ƶ
ƚ
Ś
Ž
ƌ
ŝ
ƚ
LJ
͘
&
d
d
W
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
ďƌ
Ă
Ŷ
Ě
Ŷ
Ă
ŵ
Ğ
ŝ
Ɛ
EĞ
dž
ƚ
>
ŝ
ŐŚƚ
>Ž
Ɛ
Ŷ
Ő
Ğ
ů
Ğ
Ɛ
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
ϭ͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
E
Ž
E
Ž
Ž
ŵ
ŵ
Ğ
ƌ
Đ
ŝ
Ă
ů
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
Ă
Ŷ
Ě
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
^
Ğ
ƌ
ǀ
ŝ
Đ
Ğ
Ɛ
͘
>Ž
ǀ
Ğ
ů
Ă
Ŷ
Ě
͕
ŝ
ƚ
LJ
Ž
Ĩ
K
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͖
t
Ă
ƚ
Ğ
ƌ
͕
&
ŝ
ď
Ğ
ƌ
Ğ
Ŷ
ƚ
Ƶ
ƌ
LJ
>
ŝ
Ŷ
Ŭ
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
z
Ğ
Ɛ
z
Ğ
Ɛ
t
/
W
>Ž
ǀ
Ğ
ů
Ă
Ŷ
Ě
t
Ă
ƚ
Ğ
ƌ
Θ
W
Ž
ǁ
Ğ
ƌ
ŝ
Ɛ
Ă
ŵ
Ğ
ŵ
ď
Ğ
ƌ
Ž
Ĩ
ƚ
Ś
Ğ
W
ů
Ă
ƚ
ƚ
Ğ
Z
ŝ
ǀ
Ğ
ƌ
W
Ž
ǁ
Ğ
ƌ
Ƶ
ƚŚ
Ž
ƌ
ŝ
ƚ
LJ
͘
&
d
d
W
Ŷ
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
ď
ƌ
Ă
Ŷ
Ě
E
Ă
ŵ
Ğ
ŝƐ
WƵ
ů
Ɛ
Ğ
Ͳ
Ž
Ĩ
Ĩ
Ğ
ƌ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
d
s
Ă
Ŷ
Ě
W
Ś
Ž
Ŷ
Ğ
KŶ
ƚ
Ă
ƌ
ŝ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
͖
&
ƌ
Ž
Ŷ
ƚ
ŝ
Ğƌ
E
Ž
E
Ž
t
/
W
W
Ă
ƌ
ƚ
Ŷ
Ğ
ƌ
Ɛ
Ś
ŝ
Ɖ
ǁ
ŝ
ƚ
Ś
>
K
Ŷ
ǁ
Ă
ƌ
Ě
Ͳ
Z
Ğ
Ɛ
ŝ
Ě
Ğ
Ŷ
ƚ
ŝ
Ă
ů
Ă
Ŷ
Ě
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
͘
WĂ
ů
Ž
ů
ƚ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
'
Ă
Ɛ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
d
WĂ
Ɛ
Ă
Ě
Ğ
Ŷ
Ă
͕
ŝ
ƚ
LJ
Ž
Ĩ
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
E
Ž
E
Ž
Ž
ŵ
ŵ
Ğ
ƌ
Đ
ŝ
Ă
ů
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
Ă
Ŷ
Ě
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
^
Ğ
ƌ
ǀ
ŝ
Đ
Ğ
Ɛ
͘
WůƵ
ŵ
Ă
Ɛ
Ͳ
^
ŝ
Ğ
ƌ
ƌ
Ă
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
Ž
Ͳ
Ž
Ɖ
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
&
ŝ
ď
Ğ
ƌ
ͬ
Ž
Ă
dž
ͬ
t
ŝ
ƌ
Ğ
ů
Ğ
Ɛ
Ɛ
ͬ
^
Ă
ƚ
Ğů
ů
ŝ
ƚ
Ğ
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
d
Θ
d
z
Ğ
Ɛ
z
Ğ
Ɛ
E
Ž
W
ů
Ƶ
ŵ
Ă
Ɛ
Ͳ
^
ŝ
Ğ
ƌ
ƌ
Ă
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
Ž
Ͳ
Ž
Ɖ
Ɛ
Ğ
ƌ
ǀ
Ğ
Ɛ
W
ů
Ƶ
ŵ
Ă
Ɛ
͕
^
ŝ
Ğ
ƌ
ƌ
Ă
͕
>
Ă
Ɛ
Ɛ
Ğ
Ŷ
Θ
t
Ă
Ɛ
Ś
Ž
Ğ
Ž
Ƶ
Ŷ
ƚ
ŝ
Ğ
Ɛ
ZĂ
Ŷ
Đ
Ś
Ž
Ƶ
Đ
Ă
ŵ
Ž
Ŷ
Ő
Ă
͕
ŝ
ƚ
LJ
Ž
Ĩ
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚƌ
Ƶ
ŵ
E
Ž
E
Ž
t
/
W
ŝ
ƚ
LJ
D
Ƶ
Ŷ
ŝ
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
Ɖ
Ă
ƌ
ƚ
Ŷ
Ğ
ƌ
Ɛ
ǁ
ŝ
ƚ
Ś
>
K
Ŷ
ǁ
Ă
ƌ
Ě
Ͳ
Z
Ğ
Ɛ
ŝ
Ě
Ğ
Ŷ
ƚ
ŝ
Ă
ů
Θ
ƵƐ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
͘
ŝ
ƚ
LJ
Ĩ
Ƶ
Ŷ
Ě
Ğ
Ě
ƚ
Ś
Ğ
Ĩŝď
Ğ
ƌ
͘
Zŝǀ
Ğ
ƌ
Ɛ
ŝ
Ě
Ğ
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
E
Ž
EŽ
W
ů
Ă
Ŷ
Ŷ
ŝ
Ŷ
Ő
Ž
Ƶ
Ŷ
Đ
ŝ
ů
ƌ
Ğ
Đ
Ğ
Ŷ
ƚ
ů
LJ
Ă
Ɖ
Ɖ
ƌ
Ž
ǀ
Ğ
Ě
Ă
Ő
ƌ
Ğ
Ğ
ŵ
Ğ
Ŷ
ƚ
ǁ
ŝ
ƚ
Ś
^
ŝ
&
ŝ
ƚ
Ž
ď
Ƶ
ŝ
ů
Ě&
d
d
W
;
Ψ
ϯ
Ϭ
Ϭ
D
Ϳ
Ͳ
K
Ɖ
Ğ
Ŷ
Đ
Đ
Ğ
Ɛ
Ɛ
D
Ž
Ě
Ğ
ů
͘
^Ă
Ŷ
ƌ
Ƶ
Ŷ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
&
ŝ
ď
Ğ
ƌ
͕
^
Ž
ů
ŝ
Ě
t
Ă
Ɛ
ƚ
Ğ
͕
t
Ă
Ɛ
ƚ
Ğ
t
Ă
ƚ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ă
Ŷ
ƌ
Ƶ
Ŷ
Ž
ŝ
ƚ
LJ
E
Ğ
ƚ
zĞ
Ɛ
Ϯ
zĞ
Ɛ
zĞ
Ɛ
Ϯ
ŝ
ƚ
LJ
Ͳ
Ž
ǁ
Ŷ
Ğ
Ě
^
Ă
Ŷ
ƌ
Ƶ
Ŷ
Ž
ŝ
ƚ
LJ
E
Ğ
ƚ
^
Ğ
ƌ
ǀ
ŝ
Đ
Ğ
Ɛ
ŝ
Ɛ
ƚ
Ś
Ğ
ŝ
Ŷ
Đ
Ƶ
ŵ
ď
Ğ
Ŷ
ƚ
Đ
Ă
ď
ů
Ğ
d
s
Ă
Ŷ
Ě
ď
ƌ
Ž
Ă
Ě
ď
Ă
Ŷ
Ě
Ɖ
ƌ
Ž
ǀ
ŝ
Ě
Ğ
ƌ
͘
&
Ž
ƌ
ŵ
Ğ
ƌ
ŶĂ
ŵ
Ğ
^
Ă
Ŷ
ƌ
Ƶ
Ŷ
Ž
Ă
ď
ů
Ğ
d
s
͘
^Ă
Ŷ
&
ƌ
Ă
Ŷ
Đ
ŝ
Ɛ
Đ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
t
Ă
Ɛ
ƚ
Ğ
ǁ
Ă
ƚ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
^Ă
Ŷ
>
Ğ
Ă
Ŷ
Ě
ƌ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
E
Ž
Ŷ
Ğ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
ŝ
ƚ
LJ
Ɖ
ƌ
Ž
ǀ
ŝ
Ě
Ğ
Ɛ
Đ
Ž
Ŷ
Ě
Ƶ
ŝ
ƚ
ǁ
Ś
ŝ
ů
Ğ
Ɖ
ƌ
ŝ
ǀ
Ă
ƚ
Ğ
Ɖ
Ă
ƌ
ƚ
Ŷ
Ğ
ƌ
Ɖ
ƌ
Ž
ǀ
ŝ
Ě
Ğ
Ɛ
ƚ
Ś
Ğ
Ĩ
ŝ
ď
Ğ
ƌ
ĂŶ
Ě
Ɛ
Ğ
ƌ
ǀ
ŝ
Đ
Ğ
Ɛ
ƚ
Ž
ď
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
Ğ
Ɛ
Ĩ
Ž
ƌ
Η
>
ŝ
ƚ
^
Ă
Ŷ
>Ğ
Ă
Ŷ
Ě
ƌ
Ž
Η
͘
^Ă
Ŷ
ƚ
Ă
ů
Ă
ƌ
Ă
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
'
Ă
ƌ
ď
Ă
Ő
Ğ
͕
^
Ğ
ǁ
Ğ
ƌ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
^
Ă
Ŷ
ƚ
Ă
ů
Ă
ƌ
Ă
Ğ
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
Ƶ
ƚ
ŝ
ů
ŝ
ƚ
LJ
ŝ
Ɛ
Đ
Ă
ů
ů
Ğ
Ě
^
ŝ
ů
ŝ
Đ
ŽŶ
s
Ă
ů
ů
Ğ
LJ
W
Ž
ǁ
Ğ
ƌ
͘
^Ă
Ŷ
ƚ
Ă
ƌ
Ƶ
nj
͕
ŝ
ƚ
LJ
Ž
Ĩ
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
ϮϬ
ϭ
ϲ
͗
ƌ
Ž
Ɖ
Ğ
Ĩ
Ĩ
Ž
ƌ
ƚ
Ɛ
Ĩ
Ž
ƌ
ŝ
ƚ
LJ
Ă
Ŷ
Ě
ů
Ž
Đ
Ă
ů
/
^
W
Ƶ
ƌ
nj
ŝ
Ž
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
ƚ
Ž
ď
Ƶ
ŝůĚ
&
d
d
W
Ƶ
Ŷ
Ě
Ğ
ƌ
Ă
Ɖ
Ƶ
ď
ů
ŝ
Đ
Ͳ
Ɖ
ƌ
ŝ
ǀ
Ă
ƚ
Ğ
ƉĂƌ
ƚ
Ŷ
Ğ
ƌ
Ɛ
Ś
ŝ
ƉĂ
ŐƌĞ
Ğ
ŵ
Ğ
Ŷ
ƚ
ď
ƌ
Ă
Ŷ
Ě
Ğ
Ě
^
Ă
Ŷ
ƚ
Ă
ƌ
Ƶ
nj
&
ŝ
ď
Ğ
ƌ
͘
ŝ
ƚ
LJŝƐ
Ŷ
Ž
ƚ
ŝ
Ŷ
ǀ
Ž
ů
ǀ
Ğ
Ě
ǁ
ŝ
ƚ
Ś
ƚ
Ś
Ğ
ů
ŝ
ŵ
ŝ
ƚ
Ğ
Ě
Ɛ
Ğ
ů
Ĩ
Ͳ
Ĩ
Ƶ
Ŷ
Ě
Ğ
Ě
^Ă
Ŷ
ƚ
Ă
D
Ž
Ŷ
ŝ
Đ
Ă
͕
ŝ
ƚ
LJ
Ž
Ĩ
Ă
ƌ
Ŭ
&
ŝ
ď
Ğ
ƌ
͕
Z
Ğ
Ɛ
Ž
Ƶ
ƌ
Đ
Ğ
Z
Ğ
Đ
Ž
ǀ
Ğ
ƌ
LJ
͕
^
Ğ
ǁ
Ğ
ƌ
͕
t
Ă
ƚĞ
ƌ
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
͖
&
ƌ
Ž
Ŷ
ƚ
ŝ
Ğ
ƌ
E
Ž
E
Ž
E
Ž
Ž
ŵ
ŵ
Ğ
ƌ
ŝ
Đ
Ă
ů
Ě
Ă
ƌ
Ŭ
Ĩ
ŝ
ď
Ğ
ƌ
Ă
Ŷ
Ě
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
^
Ğ
ƌ
ǀ
ŝ
Đ
Ğ
Ɛ
͘
^Ś
Ă
Ĩ
ƚ
Ğ
ƌ
͕
ŝ
ƚ
LJ
Ž
Ĩ
&
ŝ
ď
Ğ
ƌ
͕
^
Ğ
ǁ
Ğ
ƌ
͕
d
ƌ
Ă
Ɛ
Ś
͕
t
Ă
ƚ
Ğ
ƌ
K
Ɖ
ƚ
ŝ
ŵ
Ƶ
ŵ
͖
d
Θ
d
z
Ğ
Ɛ
E
Ž
z
ĞƐ
Ϯ
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
Ă
Ŷ
Ě
ƌ
Ğ
Ɛ
ŝ
Ě
Ğ
Ŷ
ƚ
ŝ
Ă
ů
&
d
d
W
Ͳ
/
^
W
Ɛ
ŝ
Ŷ
Đ
ů
Ƶ
Ě
Ğ
>
Ğ
ǀ
Ğ
ů
ϯ
Ă
Ŷ
Ě
s
Ă
Ɛ
ƚE
Ğ
ƚ
ǁ
Ž
ƌ
Ŭ
Ɛ
͘
sĂ
ů
ů
Ğ
ũ
Ž
͕
ŝ
ƚ
LJ
Ž
Ĩ
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
y
Ĩ
ŝ
Ŷ
ŝ
ƚ
LJ
E
Ž
E
Ž
E
Ž
W
Ă
ƌ
ƚ
Ŷ
Ğ
ƌ
Ɛ
Ś
ŝ
Ɖ
ǁ
ŝ
ƚ
Ś
>
K
Ŷ
ǁ
Ă
ƌ
Ě
Ͳ
Z
Ğ
Ɛ
ŝ
Ě
Ğ
Ŷ
ƚ
ŝ
Ă
ů
Ă
Ŷ
Ě
Ƶ
Ɛ
ŝ
Ŷ
Ğ
Ɛ
Ɛ
/
Ŷ
ƚ
Ğ
ƌ
Ŷ
Ğ
ƚ
͘
sĞ
ƌ
Ŷ
Ž
Ŷ
͕
ŝ
ƚ
LJ
Ž
Ĩ
&
ŝ
ď
Ğ
ƌ
͕
'
Ă
Ɛ
͕
ů
Ğ
Đ
ƚ
ƌ
ŝ
Đ
͕
t
Ă
ƚ
Ğ
ƌ
d
Θ
d
͖
^
Ɖ
Ğ
Đ
ƚ
ƌ
Ƶ
ŵ
z
Ğ
Ɛ
z
Ğ
Ɛ
z
Ğ
Ɛ
W
ƌ
ŝ
ŵ
Ă
ƌ
ŝ
ů
LJ
Ă
Ŷ
ŝ
Ŷ
Ě
Ƶ
Ɛ
ƚ
ƌ
ŝ
Ă
ů
Đ
Ž
ŵ
ŵ
Ƶ
Ŷ
ŝ
ƚ
LJ
ǁ
ŝ
ƚ
Ś
Ă
Ɖ
Ɖ
ƌ
Ž
dž
͘
ϭ
͕
ϴ
Ϭ
Ϭ
ď
Ƶ
Ɛ
ŝ
Ŷ
Ğ
ƐƐ
Ğ
Ɛ
͖
Ɖ
Ž
Ɖ
Ƶ
ů
Ă
ƚ
ŝ
Ž
Ŷ
Ϯ
Ϭ
Ϯ
Ϭ
Đ
Ğ
Ŷ
Ɛ
Ƶ
Ɛ
͗
Ϯ
Ϯ
Ϯ
ϭ͘
>
ŝ
ŵ
ŝ
ƚ
Ğ
Ě
ƚ
Ž
ƌ
Ğ
Ő
ŝ
Ž
Ŷ
Ɛ
ͬ
ŵ
Ă
ƌ
Ŭ
Ğ
ƚ
Ɛ
Ğ
Ő
ŵ
Ğ
Ŷ
ƚ
Ɛ
Ϯ͘
,
LJ
ď
ƌ
ŝ
Ě
ŝ
Ŷ
Ɛ
ƚ
Ğ
Ă
Ě
Ž
Ĩ
ũ
Ƶ
Ɛ
ƚ
&
ŝ
ď
Ğ
ƌ
͕
Ĩ
Ž
ƌ
Ğ
dž
Ă
ŵ
Ɖ
ů
Ğ
&
ŝ
ď
Ğ
ƌ
Ž
Ă
dž
20
.
a
Pa
c
k
e
t
P
g
.
2
8
1
Ar
e
a
s
w
h
e
r
e
s
u
r
v
e
y
r
e
s
p
o
n
d
e
n
t
s
re
p
o
r
t
e
d
A
T
&
T
F
i
b
e
r
s
e
r
v
i
c
e
Ar
e
a
s
w
i
t
h
l
o
w
e
s
t
c
o
s
t
t
o
b
u
i
l
d
a
n
d
hi
g
h
e
s
t
d
e
n
s
i
t
y
Ar
e
a
s
w
i
t
h
l
o
w
e
s
t
c
o
s
t
t
o
b
u
i
l
d
b
u
t
re
p
o
r
t
e
d
A
T
&
T
F
i
b
e
r
s
e
r
v
i
c
e
Ar
e
a
s
w
i
t
h
h
i
g
h
e
s
t
n
u
m
b
e
r
o
f
d
e
p
o
s
i
t
s
fr
o
m
h
o
u
s
e
h
o
l
d
s
Fi
b
e
r
Z
o
n
e
s
o
f
D
i
f
f
e
r
e
n
t
S
c
e
n
a
r
i
o
s
At
t
a
c
h
m
e
n
t
A
/
͗
ϭ
ϰ
ϴ
Ϭ
Ϭ
^ƚ
Ă
Ĩ
Ĩ
͗
Ă
ǀ
Ğ
z
Ƶ
Ă
Ŷ
Ž
Ŷ
Ɛ
Ƶ
ů
ƚ
Ă
Ŷ
ƚ
D
Ă
Ő
Ğ
ů
ů
Ă
Ŷ
͗
:
Ž
Ś
Ŷ
,
Ž
Ŷ
Ŭ
Ğ
ƌ
&2
8
1
&
,
/
0
(
(
7
,
1
*
5H
F
H
L
Y
H
G
%
H
I
R
U
H
0
H
H
W
L
Q
J
20
12
/
1
9
/
2
0
2
2
✔
Hi
g
h
e
s
t
S
p
e
e
d
(>
1
G
i
g
)
Lo
w
e
s
t
S
p
e
e
d
(>
1
0
0
M
e
g
)
Mo
d
e
r
a
t
e
S
p
e
e
d
(5
0
0
M
e
g
)
Ci
t
y
w
i
d
e
H
e
a
t
m
a
p
o
f
B
r
o
a
d
b
a
n
d
S
p
e
e
d
At
t
a
c
h
m
e
n
t
B
/
͗
ϭ
ϰ
ϴ
Ϭ
Ϭ
^ƚ
Ă
Ĩ
Ĩ
͗
Ă
ǀ
Ğ
z
Ƶ
Ă
Ŷ
Ž
Ŷ
Ɛ
Ƶ
ů
ƚ
Ă
Ŷ
ƚ
D
Ă
Ő
Ğ
ů
ů
Ă
Ŷ
͗
:
Ž
Ś
Ŷ
,
Ž
Ŷ
Ŭ
Ğ
ƌ
0(
(
7
,
1
*
5H
F
H
L
Y
H
G
%
H
I
R
U
H
0
H
H
W
L
Q
J
CO
U
NCIL
De
c
e
m
b
e
r
1
9
,
2
0
2
2
✔
20
Be
s
t
c
o
m
b
i
n
e
d
a
r
e
a
s
t
o
b
u
i
l
d
u
n
d
e
r
ph
a
s
e
1
:
l
o
w
e
s
t
c
o
s
t
,
h
i
g
h
e
s
t
d
e
n
s
i
t
y
,
le
a
s
t
c
o
m
p
e
t
i
t
i
o
n
,
h
i
g
h
e
s
t
d
e
m
a
n
d
At
t
a
c
h
m
e
n
t
C
Ph
a
s
e
1
–
F
T
T
P
C
o
n
s
t
r
u
c
t
i
o
n
M
a
p
/
͗
ϭ
ϰ
ϴ
Ϭ
Ϭ
^ƚ
Ă
Ĩ
Ĩ
͗
Ă
ǀ
Ğ
z
Ƶ
Ă
Ŷ
Ž
Ŷ
Ɛ
Ƶ
ů
ƚ
Ă
Ŷ
ƚ
D
Ă
Ő
Ğ
ů
ů
Ă
Ŷ
͗
:
Ž
Ś
Ŷ
,
Ž
Ŷ
Ŭ
Ğ
ƌ
0(
(
7
,
1
*
5H
F
H
L
Y
H
G
%
H
I
R
U
H
0
H
H
W
L
Q
J
CO
U
NCIL
De
c
e
m
b
e
r
1
9
,
2
0
2
2
✔
20