Doc #127 — NZ Telecommunications Maintenance Under Isolation

---

RECOMMENDED ACTIONS

These actions are recommended within the first weeks and months. They are ordered by genuine urgency — based on the cost of delay, not rhetorical priority.

First 72 hours

1. Test NZ–Australia submarine cable connectivity. Determine whether internet connectivity to Australia survives. This single assessment determines whether NZ has high-bandwidth digital communication with its most important partner or is limited to HF radio. Cost of delay: Low — the cable either works or it doesn’t, and testing it a few days later doesn’t change the answer. But the information shapes all subsequent communication planning.

2. Mirror critical international data to NZ-hosted servers. If international internet connectivity is available (even briefly), government agencies, universities, libraries, and businesses should immediately begin downloading and locally caching critical international content. Wikipedia, medical databases, engineering references, open-access journals. Cost of delay: Potentially very high — international connectivity may fail at any time. Every hour of delay is lost download time.

3. Secure cable landing station sites. Physical security and power continuity for the submarine cable terminal equipment at Whenuapai and Takapuna (Auckland, Southern Cross and Southern Cross NEXT) and Mangawhai Heads (Hawaiki). Identify and contact the technical staff responsible. Cost of delay: Low in the first days (the sites are secure under normal conditions), but establishing protection protocol early prevents problems later.

First 30 days

4. Issue government directive: halt copper network decommissioning. Any planned copper withdrawal should stop immediately. Copper infrastructure that still exists should be preserved. Cost of delay: Every day of continued decommissioning permanently removes copper infrastructure that cannot be replaced.

5. Begin inventory of critical telecom equipment. Through the national skills census (Doc #8): fusion splicers, OTDRs, fibre cleavers, spare OLT cards, spare radio units, spare ONTs, battery stocks, cable stocks (fibre and copper), test equipment. This inventory shapes the entire maintenance strategy. Cost of delay: Moderate — the equipment doesn’t disappear in 30 days, but without inventory, maintenance is ad hoc rather than strategic.

6. Designate telecommunications maintenance as an essential service. Ensure telecom workers are not drawn into other labour mobilisation. Provide priority resource allocation. Cost of delay: If not done early, skilled workers may be reassigned. Recalling them later is harder than retaining them.

7. Begin network consolidation planning. The three mobile operators, Chorus, and LFCs should begin coordinating a unified maintenance approach. This requires government facilitation and probably emergency powers (Doc #144). Full consolidation takes months, but planning should begin immediately. Cost of delay: Low — consolidation is a months-long process, and a few weeks of delay in starting doesn’t significantly change outcomes.

Months 1–6

8. Execute network consolidation. Select the primary mobile network per region. Begin decommissioning secondary networks and recovering equipment.

9. Establish fusion splicer and test equipment depot. Central (Wellington or Christchurch) with regional sub-depots. All fusion splicers not in active use should be stored in controlled conditions. Allocate splicers to regions based on fibre cable length and damage risk.

10. Begin telecom-specific skills training programme. Priority: fibre splicing (the most vulnerable specialised skill), cell site power systems, core network management. Integrate with the broader trade training framework (Doc #157).

11. Deploy additional battery backup and solar supplementation at Tier 1 sites. Using batteries recovered from decommissioned sites and solar panels from existing NZ stock.

12. Document all network configurations, cable routes, and equipment inventories. Offline copies stored at multiple locations.

13. Establish contact with Australian telecommunications counterparts (via submarine cable or HF radio) to coordinate mutual maintenance of the trans-Tasman cable link.

---

ECONOMIC JUSTIFICATION

Person-years required for telecom maintenance

Maintaining NZ’s telecommunications infrastructure under isolation is a labour-intensive programme. The following table estimates the workforce required by function, distinguishing Year 1 (network assessment, consolidation, and inventory) from ongoing steady-state maintenance.

Function	Year 1 (person-years)	Ongoing per year	Notes
Network engineers — core network, routing, OLT management	200–400	150–300	Drawn from Spark, One NZ, 2degrees, Chorus existing workforce; consolidated under unified structure
Fibre cable technicians and fusion splicers	100–200	80–160	Covers Chorus, LFC, and contractor workforce; repair demand falls as network contracts
Cell site RF engineers and technicians	150–300	100–200	Reduced progressively as site count falls through consolidation
Tower and antenna maintenance crews	50–100	40–80	Structural inspection, antenna hardware, feeder replacement; some overlap with electrical trades
Power systems and battery technicians	60–120	50–100	Battery replacement, solar installation, generator maintenance; shared with broader electrical workforce
Exchange and data centre operations	40–80	30–60	Environmental controls, physical security, power management
Equipment repair workshops (ONT refurbishment, board-level repair)	20–40	20–40	Electronics technicians; can be trained from existing pool
Cable landing station operations (submarine cable SLTE)	10–20	10–20	Highly specialised; small team, high strategic value
Network management, software, and configuration	50–100	40–80	Ongoing configuration, software maintenance, fault diagnosis
Inventory, logistics, and depot management	20–40	20–40	Equipment recovery from decommissioned sites, parts allocation
Training and skills transfer	20–40	15–30	Declining over time as workforce stabilises
Total	~720–1,440	~555–1,110	Wide ranges reflect genuine uncertainty in network state and consolidation pace

Notes on the estimates: These figures are rough approximations. The Year 1 figure is higher because consolidation and inventory work is additional to routine maintenance. The ongoing figure assumes network consolidation has reduced the number of active sites by approximately one-third to one-half. The existing telecommunications workforce (Section 7.1) is approximately 2,000–4,000 engineers and technicians — the programme requires retaining a significant fraction of this existing workforce in their current roles, not creating a new workforce from scratch. This is a retention and prioritisation problem, not a training problem in Year 1.

The binding constraint is not the total workforce number but the availability of highly specialised skills: submarine cable SLTE technicians, core network routing engineers, and trained fibre splicers. Loss of any of these small cohorts cannot be quickly compensated by training replacements.

What is lost without telecommunications maintenance

The comparison is not “telecommunication maintenance workforce vs. that workforce doing something else.” The comparison is “network-coordinated recovery vs. uncoordinated recovery.”

Coordination multiplier effect. Every recovery programme described in the Recovery Library depends on coordination between geographically dispersed actors: food distribution routing (Doc #1), medical consultation between regional centres and Wellington, emergency response, government administration, agricultural supply chains. A functioning telecommunications network does not add to the output of these programmes — it multiplies it. Without telecom, the same number of food distribution workers move less food because they cannot coordinate routing, cannot report blockages, cannot redirect vehicles. The coordination tax on every recovery activity rises sharply as communication degrades.

Estimated impacts of communication loss:

• Medical: Remote clinical consultation via video link eliminates approximately 70–80% of the logistical cost of providing specialist medical advice to rural communities. Without telecom, this requires physical travel of specialists or patients — in an isolation scenario, this is severely constrained by fuel. Many remote consultations that could be handled via telephone or video will result in untreated conditions or preventable deaths. ³
• Food distribution: Real-time communication between distribution centres, transport operators, and regional coordinators underpins efficient routing. Without it, food distribution falls back on fixed schedules established days or weeks in advance, with no ability to redirect supply in response to blockages, spoilage, or demand shifts. Studies of disaster-affected supply chains estimate coordination losses of 20–40% of logistics efficiency in communication-degraded environments.⁴
• Government administration: Inter-island and inter-regional communication between Wellington and Civil Defence regions, between hospitals, between emergency services. Without reliable telecommunications, government administrative capacity falls to what can be achieved by courier (days of latency) or HF radio (limited bandwidth). Policy decisions that depend on current information will be made on stale data.
• Emergency response: The 111 emergency system and Civil Defence communication networks are telecommunications-dependent. Degraded telecom directly translates to slower emergency response and more preventable deaths in the first years, before HF radio backup systems (Doc #128) are fully established.

Breakeven analysis — telecommunications as force multiplier

The question of whether to commit 720–1,440 person-years in Year 1 to telecom maintenance must be evaluated against the productivity gain across the entire recovery programme.

Conservative estimate of coordination benefit: If maintaining telecommunications improves the efficiency of other recovery programmes by 5–15% on average — a conservative range given the 20–40% logistics efficiency losses cited above for fully communication-degraded environments, discounted because telecom loss is gradual rather than total — and if the total recovery workforce engaged in food production, medical care, infrastructure maintenance, and other activities is 500,000–1,000,000 people, the efficiency gain represents 25,000–150,000 person-year equivalents of additional productive output. The lower bound assumes a small workforce with modest coordination benefit; the upper bound assumes a large workforce with significant coordination benefit. The break-even ratio of telecom maintenance labour invested to coordinated output gained is approximately 17:1 to 210:1.

This is not a rigorous economic calculation — it is an order-of-magnitude argument that the labour cost of telecom maintenance is small relative to the coordination benefit it enables. The argument holds even at the bottom of the range, and even if the actual efficiency benefit is only 1–2%.

The force multiplier property. Telecommunications is one of the few investments that simultaneously increases the output of every other recovery sector. Additional workers in food production increase food output proportionally. Additional telecom workers increase the output of every other recovery programme, weighted by how much coordination that programme requires. This asymmetry justifies disproportionate priority for telecom maintenance relative to its share of the workforce.

Breakeven timeline. The network does not require maintenance for the full degradation timeline to justify its cost. Active maintenance is estimated to extend useful network life by 3–10 years beyond what an unmaintained network would achieve — the lower bound assumes poor equipment condition and limited spares yield from consolidation; the upper bound assumes good initial equipment condition and effective cannibalisation. Even at the low end, 3 additional years of coordinated communications capability across the entire recovery programme pays back the maintenance investment many times over. The marginal cost of each additional year of network life is declining (fewer sites, less equipment) while the value of coordination is stable or rising as the recovery matures.

Opportunity cost

The opportunity cost of employing 555–1,110 people per year in telecom maintenance is the agricultural, medical, and general labour those people would otherwise provide. This cost is real and should not be dismissed.

However, several factors reduce the effective opportunity cost:

• Existing specialists. The telecommunications workforce has skills not readily substitutable in other sectors. A fibre splicer or core network engineer reassigned to farming provides approximately one unit of unskilled agricultural labour — not an upgrade for the farming sector, and a permanent loss of a scarce technical capability. The opportunity cost of redeploying specialists to general labour is low; the cost of losing the specialist role is high.
• Declining labour demand over time. As the network contracts, fewer technicians are needed. The telecom workforce can be reduced progressively as sites are decommissioned, releasing people for other sectors without abrupt disruption.
• Complementarity with electrical trades. Many telecom maintenance skills (power systems, battery maintenance, electrical installation) are shared with the general electrical trades sector. Telecom power technicians can contribute to grid maintenance, solar installation, and electrical infrastructure more broadly — the opportunity cost of retaining them in a combined telecom/electrical role is lower than retaining pure specialists.

The real opportunity cost is the 20–40 specialist person-years in roles with no alternative: submarine cable SLTE technicians, core routing engineers, and trained fusion splicers. These people cannot be usefully redeployed elsewhere. Releasing them saves their wages but destroys a capability that cannot be rebuilt on any near-term timescale.

---

1. NZ TELECOMMUNICATIONS INFRASTRUCTURE — CURRENT STATE

1.1 Network operators

NZ’s telecommunications network is operated by multiple companies, each owning different layers of infrastructure:⁵

Chorus Limited — The national fibre and copper access network provider. Chorus owns approximately 31,000–33,000 km of fibre-optic cable and the residual copper local loop connecting premises to local exchanges.⁶ Chorus does not provide retail services — it is a wholesale-only infrastructure company created by the structural separation of Telecom NZ in 2011. Chorus is responsible for the “last mile” fibre to approximately 87% of NZ addresses under the Ultra-Fast Broadband (UFB) programme, with local fibre companies (LFCs) covering additional areas.⁷

Spark New Zealand — NZ’s largest telecommunications company. Operates a nationwide mobile network (approximately 1,600–2,000 cell sites), a core IP network, international connectivity, and retail services.⁸ Spark also operates data centres and cloud services infrastructure.

One NZ (formerly Vodafone NZ) — Second-largest mobile operator. Operates approximately 1,800–2,200 cell sites and a core network. Sold by Vodafone Group to a consortium of Infratil and Brookfield Asset Management in 2019 and rebranded to One NZ in 2023.⁹

2degrees — Third mobile operator, now merged with Vocus NZ. Operates approximately 1,500–1,800 cell sites. Has grown through network sharing arrangements and its own infrastructure build.¹⁰

Local Fibre Companies (LFCs) — Northpower Fibre (Northland), Enable (Christchurch), Tuatahi First Fibre (formerly Ultrafast Fibre, covering Hamilton, Tauranga, and other centres). These companies own fibre access networks in their regions under the UFB programme.¹¹

Rural Connectivity Group (RCG) — A joint venture of Spark, One NZ, and 2degrees established under the Rural Broadband Initiative (RBI) and Mobile Black Spot Fund (MBSF) programmes to extend mobile and broadband coverage to rural NZ.¹² RCG cell sites often serve areas with no alternative connectivity.

Submarine cable operators — Southern Cross Cable Network (owned by Spark, Singtel, Verizon, and Telstra) operates the Southern Cross NEXT cable system connecting NZ to Australia and the United States.¹³ Hawaiki Cable connects NZ to Australia and the US west coast.¹⁴ These provide NZ’s international internet connectivity.

1.2 Network layers

Understanding which parts of the network to prioritise requires understanding its layered structure:

Layer 1 — Physical infrastructure (cables, towers, ducts):

• Fibre-optic cable in ducts or aerial deployments
• Copper cable (where it remains)
• Cell tower structures (steel or concrete poles, rooftop installations, guyed masts)
• Underground ducting and manholes
• Exchange buildings and equipment rooms
• International submarine cables

This layer is the most durable. Fibre-optic glass lasts indefinitely if undisturbed. Copper cable lasts decades. Tower structures last 25–50+ years. Buildings and ducts last longer. Physical infrastructure is not the primary vulnerability.

Layer 2 — Active network equipment (the vulnerability):

• Optical line terminals (OLTs) in exchanges — convert between fibre and network protocols
• Optical network terminals (ONTs) at customer premises — the fibre modem
• Cell site equipment: radio units (RUs), baseband units (BBUs), power amplifiers, filters, antennas
• Core network routers and switches
• Transmission equipment (DWDM multiplexers on fibre backbone)
• Power supplies and battery backup at all sites
• Network management and monitoring systems

This layer is where failure will occur. Every piece of active equipment contains electronic components — semiconductors, electrolytic capacitors, power transistors, oscillators, connectors — with finite operational lifespans. Without imported replacements, this equipment fails piece by piece over years to decades.

Layer 3 — Software and management systems:

• Network operating software on routers and switches
• Element management systems
• Billing, provisioning, and customer management systems
• Network monitoring and fault management

Software does not wear out, but it runs on hardware that does. Software is also the layer where NZ has the most domestic capability — NZ has a significant software engineering workforce that can maintain and adapt software systems as long as the hardware runs.

1.3 The fibre network in detail

NZ’s fibre network is one of the most extensive per capita in the world, a result of the government-funded UFB programme that ran from 2011 to approximately 2022.¹⁵

Architecture: The UFB network uses Gigabit Passive Optical Network (GPON) technology. From each exchange or fibre aggregation node, a single fibre runs to a passive optical splitter (typically a 1:32 or 1:64 split ratio), which splits the optical signal to serve 32 or 64 premises from a single fibre port at the exchange.¹⁶ The “passive” part is important — the splitters contain no electronics and no power. They are glass prisms in a weatherproof enclosure. They do not fail.

What fails in the fibre network:

• OLTs (Optical Line Terminals) in exchanges — These are the exchange-end equipment that transmits and receives optical signals. They are active electronics manufactured by companies like Nokia, Huawei, Calix, or similar.¹⁷ OLTs contain laser transmitters, optical receivers, processing electronics, and power supplies. Lifespan under normal conditions: 10–15 years for the electronics, though the lasers may last longer. These are the critical bottleneck — when an OLT card fails, every customer served by that card (typically 32–64 premises) loses service.
• ONTs (Optical Network Terminals) at premises — The customer-side fibre modem. Each premises has one. There are approximately 1.2–1.5 million ONTs deployed across NZ.¹⁸ They fail individually — one customer’s ONT failing does not affect others. Lifespan: 7–15 years typically, with electrolytic capacitor aging as the primary failure mode. Failed ONTs can be replaced from the stock of ONTs at premises that have been abandoned or where service is no longer needed.
• Fibre cable breaks — Physical damage from construction, storms, earthquakes, or rodents. Fibre cable breaks are repaired by fusion splicing — heating the broken fibre ends in a precision arc fusion splicer to fuse the glass. This requires a fusion splicer (an imported precision instrument costing $5,000–$30,000) and optical test equipment (OTDR — Optical Time Domain Reflectometer) to locate the break and verify the repair.¹⁹ NZ has an inventory of fusion splicers held by Chorus, LFCs, and contractors — this inventory must be identified, preserved, and maintained. Fusion splicers are irreplaceable with current NZ manufacturing capability. When they fail, fibre repair becomes extremely difficult.

What does not fail:

• Fibre cable itself (silica glass, indefinite life if undamaged)
• Passive optical splitters (no electronics, no moving parts)
• Underground ducts and manholes
• Exchange buildings

1.4 The cellular network in detail

NZ’s mobile network provides coverage to approximately 98–99% of the population, though geographic coverage is substantially less (NZ is mountainous with large areas of low population density).²⁰

Cell site types:

• Macro sites — Full-sized cell towers serving large areas. Steel monopoles, lattice towers, or rooftop installations. Typically 3 sectors (three sets of antennas pointed in different directions to provide 360° coverage). Each sector has radio units, baseband processing, power amplifiers, and antennas. These serve the backbone of cellular coverage.
• Small cells and micro sites — Smaller installations providing infill coverage in dense urban areas or capacity hotspots. Lower power, simpler equipment, shorter range.
• Rural sites (including RCG) — Often single-sector or two-sector sites serving small communities. May use satellite backhaul rather than fibre, which introduces an additional dependency.

What fails at cell sites:

• Radio units (RUs) and power amplifiers — These are the most stressed components. Power amplifier transistors (GaN or LDMOS types) degrade under sustained operation, particularly in hot conditions. Typical lifespan: 7–15 years, depending on loading and environmental conditions.²¹ When a radio unit fails, that sector loses service.
• Baseband units (BBUs) — The digital processing equipment that handles the cellular protocol. Complex electronics with multiple processor chips, memory, and FPGA/ASIC components. Lifespan: 10–15 years under normal conditions. BBU failure takes the entire site offline.
• Battery backup systems — Most cell sites have battery backup (typically 4–8 hours, sometimes longer for critical sites). Lead-acid or lithium-ion batteries degrade over time — lead-acid batteries last 3–7 years, lithium-ion 5–10 years.²² When batteries fail and grid power is interrupted (planned outages, faults), the site goes down during the outage.
• Backhaul links — The connection from the cell site back to the core network. Most sites use fibre backhaul; some rural sites use microwave links or satellite. Microwave equipment has similar failure profiles to cell site electronics. Satellite backhaul depends on the satellite constellation continuing to function.
• Air conditioning and environmental control — Cell site equipment generates heat and requires cooling. In NZ’s temperate climate, this is less critical than in tropical locations, but equipment rooms still need ventilation or cooling, particularly for dense urban sites. Air conditioning units fail and require maintenance.

Antenna hardware — The passive antenna panels and feeders at cell sites are relatively durable (20+ years), similar to HF radio antennas. They are not the primary vulnerability.

1.5 The copper network

NZ’s copper network — the traditional Public Switched Telephone Network (PSTN) — is being progressively retired as fibre replaces it. Chorus has been withdrawing copper services from areas where fibre is available.²³

What remains valuable about copper:

• Copper voice telephony is the most maintainable telecommunications technology. A copper line from a premises to an exchange, powered from the exchange battery (which is grid-charged), provides basic voice communication with no complex electronics at the customer end — only a telephone handset, which contains a speaker, a microphone, and a switch.
• Exchange equipment for copper (though increasingly decommissioned) is conceptually simpler than fibre or cellular equipment, though modern exchanges are still computerised.
• Copper cable, while less capable than fibre, is physically robust and can be spliced with basic tools (soldering iron, heat-shrink tubing) — no fusion splicer required.

The copper dilemma: Under normal conditions, NZ was rationally retiring copper in favour of fibre — fibre is faster, cheaper to maintain at scale, and more future-proof. Under isolation conditions, copper’s simplicity becomes an advantage. The recommendation is: stop copper decommissioning immediately. Any copper infrastructure that still exists should be preserved. Reactivation of copper voice service to areas where it has been withdrawn but the cable remains in place should be assessed.

Assumption: This recommendation assumes the copper cable is still physically in place — Chorus’s withdrawal of copper service does not necessarily mean immediate cable removal, but cables that have been physically removed cannot be restored. The actual state of copper infrastructure requires a field assessment through the skills census (Doc #8).

1.6 Submarine cables and international connectivity

NZ’s international internet connectivity depends on submarine fibre-optic cables:

Southern Cross NEXT — Completed in 2022. A modern cable system with landing points at Takapuna Beach (Auckland) and Coogee Beach (Sydney), continuing to the United States (Los Angeles).²⁴ Capacity: approximately 72 Tbps design capacity. This is NZ’s primary international cable.

Hawaiki Cable — Completed in 2018. Landing points at Mangawhai Heads (north of Auckland) and connects to Sydney (Australia) and Oregon (United States).²⁵

Southern Cross Cable (original) — Older system, also connecting NZ to Australia and the US.

The international connectivity picture under isolation:

The Northern Hemisphere cable endpoints (Los Angeles, Oregon) connect to US telecommunications infrastructure that is likely destroyed or severely degraded. These connections are severed.

The Australian cable endpoints (Sydney) connect to Australian telecommunications infrastructure. If Australian infrastructure survives — which is plausible, as Australia is in the Southern Hemisphere and distant from the conflict — then NZ–Australia internet connectivity may continue as long as the cable terminal equipment at both ends functions.

This is one of the most important early assessments. Within the first days, NZ telecommunications operators should test NZ–Australia cable connectivity. If it works, NZ retains high-bandwidth digital communication with its most important partner — enabling file transfer, email, video conferencing, and database synchronisation that are functionally impossible over HF radio voice links (Doc #128), which are limited to approximately 100–300 words per minute of voice communication or very low-speed data.

If NZ–Australia cable connectivity survives, maintaining it becomes an extremely high priority. The cable terminal equipment (submarine line terminal equipment, or SLTE) at the NZ landing stations is specialised and irreplaceable. Protecting it, ensuring its power supply, and having technicians available who understand it should be an immediate action.

Estimate: NZ–Australia submarine cable connectivity, if initially functional, could last 10–20+ years — the cables are new, the terminal equipment is modern, and the failure modes are the same gradual electronics aging that affects other network equipment. This estimate is uncertain and depends heavily on conditions at both ends.

---

2. DEGRADATION TIMELINE

2.1 What drives failure

Telecommunications equipment fails through several mechanisms, all of which accelerate without imported replacement parts:

Electrolytic capacitor aging. Electrolytic capacitors are used in virtually every piece of electronics for power filtering and energy storage. They contain a liquid or gel electrolyte that slowly dries out, causing capacitance to drop and equivalent series resistance (ESR) to rise. This is the single most common failure mode in aging electronics. Typical lifespan: 5,000–15,000 hours at rated temperature, which translates to roughly 7–25 years under normal operating conditions depending on quality and ambient temperature.²⁶ Higher temperatures dramatically shorten life — every 10°C increase roughly halves capacitor lifespan.

Semiconductor degradation. Transistors, integrated circuits, and power devices degrade through mechanisms including electromigration, hot-carrier injection, and gate oxide breakdown. For most digital electronics operating within ratings, semiconductor life exceeds 20 years. For power amplifiers (in cell site radio units and power supplies), which operate near their thermal limits, life is shorter — typically 7–15 years.²⁷

Connector and solder joint degradation. Repeated thermal cycling (day/night temperature changes) stresses solder joints. Connectors corrode, particularly in humid or coastal environments — relevant for many NZ cell sites, especially those along the west coast of both islands (high rainfall, salt-laden wind) and in Northland and the Hauraki Gulf coast. This is usually slow but accumulates over decades.

Software issues. Clock rollover bugs, certificate expiry, and date-related issues may affect network equipment on timelines of years to decades. Some equipment may cease functioning at specific future dates due to hardcoded assumptions. This is unpredictable but should be anticipated — NZ’s software engineering workforce can potentially address these issues for equipment where the software is accessible.

Mechanical wear. Fans in equipment enclosures, air conditioning units, and backup generators. Bearings wear, motors fail, filters clog. More readily maintainable than electronics — NZ has the mechanical capability to repair or fabricate replacement fans, filters, and basic mechanical components.

2.2 Estimated degradation timeline

The following timeline is an estimate based on typical equipment lifespans and represents the most likely trajectory under the baseline scenario (grid power continues, competent maintenance, equipment operated within ratings). Actual outcomes depend heavily on maintenance quality and on whether specific critical components fail early.

Timeframe	Expected State	Key Drivers
Months 0–12	Network fully functional. No perceptible degradation.	All equipment recently maintained. Spares from existing stock cover any failures.
Years 1–3	Network substantially functional. Isolated equipment failures, repaired from spares. Battery backup degradation begins at some cell sites.	Electrolytic capacitor failures begin in oldest equipment. Battery replacement stock depletes.
Years 3–7	Noticeable contraction begins. 5–15% of cell sites may be offline due to equipment failure. Some ONT failures at customer premises — replaced from recovered stock. Core network still functional.	Accelerating component failures. Spares cannibalised from decommissioned sites. Some rural/remote sites abandoned to preserve parts for higher-priority sites.
Years 7–15	Significant network contraction. Cellular coverage reduced to major population centres. Fibre network contracts as OLT equipment fails — maintained in priority areas by cannibalising from low-priority exchanges. Copper voice service, where preserved, may be the most reliable remaining technology.	Equipment now well past design life. Power amplifiers, capacitors, and processing boards failing at increasing rates. Fusion splicer failures begin limiting fibre repair capability.
Years 15–30	Residual modern telecom serving only the most critical links. Majority of communication transitions to HF radio (Doc #128), VHF/UHF radio, and possibly locally maintained copper voice. Some fibre backbone links may survive with careful husbandry.	Modern equipment fleet largely exhausted. Only cannibalised and refurbished units remain.
Years 30+	Modern telecom equipment fully exhausted. Communication systems are those NZ can build: copper wire telephony, HF/VHF radio, and eventually locally manufactured electronics (Doc #135).	Complete dependency on locally producible technology.

Important caveat: This timeline is an estimate with substantial uncertainty. It could be significantly better or worse depending on:

• The actual condition and age distribution of NZ’s installed equipment
• How effectively spares are managed and cannibalisation is organised
• Whether specific critical components (e.g., a core router at a major exchange) fail early
• The quality of maintenance — competent maintenance extends equipment life; poor maintenance shortens it
• Environmental conditions (NZ’s temperate climate is favourable compared to tropical environments)

2.3 The cascade risk

The degradation spiral described in the about page applies directly to telecommunications. A cell site fails. The technician who would repair it has been reassigned to farming. The spare part needed is held at a depot that has been repurposed. The vehicle to transport the part has no fuel allocation. Each failure makes the next failure harder to address.

Preventing this cascade requires institutional commitment to telecom maintenance as a national priority — not something that happens automatically when the labour market tightens and food production competes for every available worker.

---

3. PRIORITISATION — WHAT TO MAINTAIN, WHAT TO LET GO

3.1 Prioritisation principles

Not all parts of the network are equally important. Under isolation, the government must decide what to maintain and what to allow to fail. The principles:

1. Core before periphery. The backbone fibre network connecting major centres is more important than the last-mile connection to individual premises. A functioning backbone with patchy local access is far more useful than perfect local access in a few areas with no backbone connecting them.

2. Population density determines priority. A cell site serving 10,000 people in Hamilton is more important than a cell site serving 50 people in a remote valley — assuming the 50 people have alternative communication (HF radio, copper phone).

3. Interconnection matters. Links between Civil Defence regions, between government agencies, and between NZ and Australia (if submarine cable survives) are higher priority than consumer broadband.

4. Simplicity aids longevity. Simpler technologies (copper voice, basic data) outlast complex ones (5G, high-capacity DWDM). Where a choice must be made, preserve the simpler system.

5. Redundancy has diminishing value under scarcity. Three mobile networks covering the same area made sense when competition drove investment. Under isolation, one working network in each area is sufficient. Consolidation releases equipment for spare parts.

3.2 Priority tiers

Tier 1 — Maintain at all costs:

• Inter-city fibre backbone links (connecting Auckland, Hamilton, Wellington, Christchurch, Dunedin, and intermediate cities)
• NZ–Australia submarine cable terminal equipment (if link is functional)
• Core network routers and switches at major exchanges (Auckland, Wellington, Christchurch as minimum)
• Government communication links (Civil Defence, Parliament, key ministries)
• Key exchange OLT equipment serving major population centres

Tier 2 — Maintain as long as practical:

• Cellular coverage in major urban areas (Auckland, Wellington, Christchurch, Hamilton, Tauranga, Dunedin) — consolidated to one operator’s network per area
• Fibre local access in major urban areas
• Copper voice service where it still exists
• Inter-island fibre links (Cook Strait crossings)

Tier 3 — Maintain where resources allow:

• Cellular coverage in smaller towns
• Fibre local access in smaller centres
• Microwave transmission links to areas without fibre

Tier 4 — Allow to fail (with alternative provision):

• Cellular coverage in remote rural areas (replaced by HF radio — Doc #128)
• Consumer broadband in areas where basic voice is sufficient
• Duplicate mobile network infrastructure (cannibalise for parts)
• Satellite-backhauled rural sites (dependent on satellite constellation survival)

3.3 Network consolidation

NZ currently has three mobile networks covering substantially the same geography. Under isolation, this redundancy becomes wasteful — three sets of equipment aging simultaneously when one set could serve the population.

Recommendation: Within the first 6–12 months, the government should negotiate (or mandate under emergency powers — Doc #144) network consolidation. This means:

• Selecting one mobile network per geographic area (likely based on which operator has the best equipment condition and coverage in that area)
• Decommissioning the other operators’ cell sites in that area
• Recovering all usable equipment (radio units, BBUs, power supplies, antennas, batteries, cables) from decommissioned sites for use as spares
• Consolidating the technical workforce — all mobile network engineers from all three operators working together on maintaining the selected network

This is politically and commercially difficult under normal conditions. Under isolation, the commercial considerations are irrelevant — there is no shareholder value to protect, no competitive market to maintain. The question is purely: how do we make the available equipment last as long as possible?

Estimate: Consolidating three mobile networks into one approximately triples the available spare parts for the surviving network, potentially extending the cellular network’s functional life by 5–10 years beyond what any single operator’s network would achieve alone. This estimate is rough — it depends on the degree of equipment compatibility between operators (different vendors complicate component sharing) and on the actual spares yield from decommissioned sites.

Equipment compatibility challenge: NZ’s three mobile operators use equipment from different vendors (Ericsson, Nokia, Huawei, Samsung, and others). Equipment from one vendor is generally not interchangeable with another’s at the board level. However, some components are common across vendors: power supplies, batteries, cables, connectors, antenna hardware, air conditioning units, and some standard electronic components. Network consolidation should prioritise keeping the network with the largest installed base of compatible equipment, or the network whose vendor’s equipment is most common across all three operators.

3.4 The cellular-to-copper transition

As cellular coverage contracts, some areas will lose mobile service. Where copper infrastructure remains, basic voice telephony can continue — copper line service requires only:

• Intact copper cable from premises to exchange
• A working exchange (or bypass equipment that connects copper lines directly)
• Power at the exchange (grid or battery/solar)
• A basic telephone handset at the customer end

This is a technology NZ can maintain indefinitely. The copper cable lasts for decades. Exchange equipment is complex but simpler than cellular. Telephone handsets are about as simple as electrical devices get.

The transition path: As cellular sites fail in a given area, revert that area to copper voice if the copper infrastructure exists. This requires advance planning — identifying which areas have surviving copper, testing it, and ensuring exchange equipment can support it.

---

4. FIBRE NETWORK MAINTENANCE

4.1 Fibre cable maintenance

Fibre-optic cable requires minimal ongoing maintenance — the glass itself does not degrade. The maintenance challenge is repairing cable breaks when they occur.

Causes of fibre cable breaks:

• Excavation damage (the most common cause under normal conditions — digging equipment striking buried cable)
• Storm damage (aerial cable, or flooding causing ground movement)
• Earthquake damage (ground movement, liquefaction)
• Rodent damage (in some environments)
• Vandalism or theft (copper theft sometimes results in fibre damage when cables are co-located)

Under isolation, excavation damage may decrease (less construction activity) but earthquake risk remains — particularly along the Alpine Fault (South Island), the Hikurangi subduction zone (eastern North Island), and the Wellington fault system — and storm damage may increase if tree management around aerial cable routes is reduced. NZ’s seismic risk is a significant ongoing threat to buried fibre, especially where routes cross fault lines or traverse liquefaction-prone soils (Canterbury, Hawke’s Bay, parts of Wellington).

Fibre splice repair process:

1. Locate the break using an OTDR (Optical Time Domain Reflectometer) — this instrument sends a pulse of light down the fibre and analyses the reflection to determine the break location, typically to within 1–2 metres.²⁸
2. Access the cable at the break location (excavation for buried cable; access for aerial)
3. Expose the individual fibre strands (a cable typically contains 12–288 individual fibres)
4. Prepare each fibre end: strip the protective coating, clean, cleave the fibre to a clean perpendicular face using a precision fibre cleaver
5. Fusion splice: place both fibre ends in the fusion splicer, which automatically aligns the fibre cores (typically 9 microns diameter for single-mode fibre) using a camera system, then fuses them with an electric arc. A good splice has less than 0.1 dB loss — essentially invisible to the network.²⁹
6. Protect the splice with a heat-shrink splice protector
7. Place the splice in a splice closure and restore the cable

Critical equipment:

• Fusion splicer — NZ probably has 50–200 fusion splicers across Chorus, LFCs, and contractor companies.³⁰ These are precision instruments manufactured in Japan (Fujikura, Sumitomo) or China. They contain precision motors, cameras, image processing electronics, and high-voltage arc generators. They cannot be manufactured in NZ. Preserving the national fusion splicer inventory is a high-priority immediate action.
• Fibre cleaver — Precision tool for cutting fibre. Simpler than a splicer but still a precision instrument. Also not NZ-manufacturable.
• OTDR — Test instrument. Essential for efficient fault location but not for the actual repair. Faults can be located by visual inspection or trial-and-error without an OTDR, though the process takes significantly longer — potentially hours or days per fault instead of minutes.
• Splice closures and splice protectors — Consumable items. NZ should have significant stock, and some could potentially be fabricated locally from appropriate materials.
• Fibre cable stock — Spare fibre cable for replacing damaged sections. NZ likely has some stock held by Chorus and contractors. Domestic manufacture of fibre-optic cable is not feasible — it requires high-purity silica, precision drawing towers, and UV-curable coating materials that NZ does not produce.³¹

Fusion splicer lifespan: With careful use and maintenance (keeping electrodes clean, protecting the alignment camera, storing in proper cases), fusion splicers can last 15–25 years.³² The electrodes (which create the arc for splicing) wear out after 2,000–5,000 splices and need replacement — electrode stock should be inventoried and preserved.³³ Some splicer failures (broken cameras, failed motors) may be repairable by skilled electronics technicians even without manufacturer support.

Mechanical splicing as fallback: If fusion splicers become unavailable, fibre can be joined using mechanical splices — devices that align and clamp the fibre ends together without fusing them. Mechanical splices have higher loss (typically 0.2–0.5 dB) and lower reliability than fusion splices, but they work and require no powered equipment.³⁴ Mechanical splice stock should be identified and preserved. In the long term, it may be possible to fabricate crude mechanical splicing devices from local materials, though alignment precision is a significant challenge.

4.2 OLT and ONT maintenance

OLT maintenance: OLTs are rack-mounted equipment in exchanges. They are modular — individual line cards serving 32–64 customers can be swapped without replacing the entire unit. When an OLT line card fails, that group of customers loses fibre service.

Maintenance strategy:

• Maintain a stock of spare OLT line cards, recovered from decommissioned exchanges
• As customer numbers in an exchange area decrease (population movement, service no longer needed), consolidate customers onto fewer line cards and recover the rest as spares
• Protect OLT equipment from power surges (lightning, grid faults) — surge protection is simple and extends equipment life
• Maintain clean power supply to exchanges — power conditioning equipment should be preserved

ONT maintenance: ONTs are individually small and numerous. The strategy is straightforward in concept though labour-intensive in execution: maintain a large pool of spare ONTs recovered from premises where service is no longer needed. With approximately 1.2–1.5 million deployed, even a 5–10% failure rate per year generates 60,000–150,000 failed ONTs annually — a large number, but also a large number of potential donor units.

ONT refurbishment: Common failures (power supply capacitors, blown rectifier diodes) can be repaired by competent electronics technicians with soldering equipment and salvaged components. This is well within NZ’s capability — the electronics are not complex by modern standards. Establishing ONT repair workshops at regional centres would extend the useful ONT fleet significantly.

4.3 Exchange and data centre maintenance

Exchanges and data centres house the core network equipment — routers, switches, OLTs, power systems, and environmental controls. They require:

• Reliable power: Grid power, backed by battery banks (typically large lead-acid or lithium-ion systems providing 4–24 hours of backup) and standby generators. Battery banks degrade and need replacement — lead-acid batteries last 5–10 years in float service.³⁵ Generator fuel is a constrained resource (Doc #1). Exchange power systems should be a priority for solar and micro-hydro supplementation.
• Environmental control: Temperature and humidity management. Equipment generates heat; excessive temperature shortens component life. NZ’s temperate climate helps — natural ventilation may suffice for many sites if equipment density is reduced through consolidation. For sites where active cooling is needed, maintaining air conditioning units is important but within NZ’s mechanical engineering capability.
• Physical security: Protecting against water ingress (flooding, roof leaks), rodent damage, and dust. Routine building maintenance.
• Fire suppression: Exchanges typically have fire suppression systems. These should be maintained — a fire destroying an exchange full of irreplaceable equipment would be a severe loss.

---

5. CELLULAR NETWORK MAINTENANCE

5.1 Cell site maintenance priorities

Each cell site requires periodic maintenance:

• Equipment monitoring: Under normal conditions, cell sites are monitored remotely via the network management system. This monitoring should continue as long as the NMS and its communication links function. When remote monitoring is no longer possible, scheduled physical inspections replace it.
• Battery replacement: The most frequent maintenance need. Battery degradation is predictable and can be monitored. Replacing batteries with recovered units from decommissioned sites extends site life.
• Power amplifier replacement: When a radio unit’s power amplifier degrades, that sector’s output power drops, reducing coverage. Eventually the amplifier fails entirely. Replacement from spare stock (including units from decommissioned sites) is the primary strategy.
• Antenna and feeder maintenance: Connectors corrode (particularly in coastal environments), cables degrade from UV exposure. Antenna hardware is relatively maintainable — connectors can be cleaned or replaced, cables can be fabricated from available coaxial cable stock.
• Software updates: Cell site software may need adjustment as the network configuration changes (new neighbour lists when adjacent sites are decommissioned, parameter changes to extend coverage from remaining sites). NZ has the software engineering capability for this.

5.2 Extending coverage from fewer sites

As sites are decommissioned or fail, the remaining sites need to cover larger areas. This can be partially achieved through:

• Increasing transmit power on remaining sites (within equipment limits). Most sites operate below maximum power under normal conditions.
• Antenna tilting — adjusting the electrical or mechanical downtilt of antennas to cast coverage further. This is a standard network optimisation technique that NZ mobile engineers understand well.
• Adding external amplifiers where available, to boost output power beyond the radio unit’s native capability.
• Relocating antennas to higher positions (taller towers or higher buildings) to extend line-of-sight coverage.

These measures cannot fully compensate for lost sites — cellular coverage is fundamentally limited by the physics of radio propagation. But they can reduce the impact of losing peripheral sites.

5.3 Cell site power resilience

Cell sites depend on power. Under the baseline scenario, grid power continues, but localised outages (storms, faults, equipment failure) will occur. Resilient cell sites need:

• Battery backup: The first line of defence. Currently, most NZ cell sites have 4–8 hours of battery backup.³⁶ This should be extended to 24–48 hours at Tier 1 and Tier 2 sites by deploying additional battery banks recovered from decommissioned sites.
• Solar supplementation: A 500W–1kW solar panel installation at a cell site can extend backup duration meaningfully. Cell site equipment draws roughly 500–2,000W depending on configuration (a single-sector rural site at the low end; a three-sector urban macro site at the high end).³⁷ Solar panels cannot fully replace grid power for a busy cell site but can reduce grid dependence and extend battery life during outages. NZ has existing solar panel stock and the electrical skills to install it.
• Wind supplementation: In appropriate locations (hilltop sites, rural areas with good wind exposure), small wind turbines can supplement power. Many NZ cell sites are on hilltops or ridgelines — often sites with good wind exposure, particularly in the Cook Strait corridor, the Canterbury Plains, the Manawatū, and Southland.
• Generator backup: Priority sites may warrant standby generators. Fuel allocation is the constraint (Doc #1); wood gasifier generators (Doc #56) are a renewable alternative for sites accessible by road.

---

6. DOMESTIC INTERNET UNDER ISOLATION

6.1 What continues to work

The domestic internet — communication between NZ-based servers, websites, and users — continues to function as long as the core network equipment operates. This includes:

• NZ-hosted websites and services: Government websites, NZ news sites, NZ-hosted email, local business services. Any content physically hosted on servers in NZ remains accessible.
• Internal corporate networks: Businesses and government agencies with NZ-hosted infrastructure continue to operate.
• Domestic email: Email between NZ addresses, routed through NZ servers.
• Domestic social media and messaging: To the extent that NZ-hosted platforms or locally cached content exists. Major social media platforms (Facebook, Instagram, TikTok) are hosted internationally and will become inaccessible when international links fail. Locally developed or self-hosted alternatives could fill this gap if there is demand.
• NZ DNS resolution: The .nz domain zone is maintained by InternetNZ (now Internet New Zealand) with name servers located in NZ.³⁸ Domestic DNS resolution continues as long as these servers function. Resolution of international domains (.com, .org, etc.) will fail when international connectivity is lost, unless NZ caches the relevant zone data.

6.2 What ceases

• All international websites and services: Google, Facebook, Amazon, international news, streaming services, international email — anything hosted outside NZ (or outside Australia, if the NZ–Australia cable survives).
• Cloud services hosted internationally: Many NZ organisations use AWS, Azure, or Google Cloud for hosting. Any data or services on international cloud infrastructure become inaccessible. This is a significant early vulnerability — organisations should immediately mirror critical data and services to NZ-hosted infrastructure in the first hours/days.
• Software updates and security patches: No more updates for operating systems, applications, or network equipment firmware. This creates a slowly growing security vulnerability, though in an isolation scenario, the threat model changes — external cyberattacks become less relevant, while the risk of local system failures from unpatched bugs increases.
• Content delivery networks: Much web content is served through international CDNs (Cloudflare, Akamai). NZ-based caches may continue serving cached content for some time, but fresh international content ceases.

6.3 The NZ–Australia internet link

If the submarine cable connection to Australia survives, the useful internet expands significantly:

• Communication with Australian government and institutions via standard internet tools (email, video conferencing, file transfer) — vastly more capable than HF radio
• Access to Australian-hosted content and services — Australia hosts significant data centre infrastructure
• Relay potential — If Australia has connectivity to other surviving regions, NZ might access a broader internet through Australia

Maintaining this link is extremely high priority. The actions required:

• Test NZ–Australia cable connectivity immediately (within first hours)
• If functional, protect the cable terminal equipment at NZ landing stations (Whenuapai and Takapuna for Southern Cross/Southern Cross NEXT, Mangawhai Heads for Hawaiki)
• Ensure continuous power to terminal equipment
• Identify and protect the technical staff who understand the SLTE systems
• Coordinate with Australian counterparts on mutual maintenance
• Establish backup communication (HF radio — Doc #128) to coordinate cable maintenance with Australia if the cable itself fails temporarily

6.4 Data preservation

NZ has a window — while modern computing and storage infrastructure is functional — to preserve digital knowledge that will be needed later. This is addressed in detail in other documents (Doc #5 for printing, Doc #134 for computing), but the telecommunications implications are:

• Mirror critical international content to NZ servers while international links are available (if they are available at all — they may already be down). Wikipedia, technical reference databases, open-access scientific papers, agricultural databases, medical references, engineering standards.
• Ensure NZ-hosted services have local backup — NZ organisations relying on international cloud services need to pull their data home immediately.
• Preserve network configuration data — All network configurations, equipment inventories, cable route maps, exchange layouts, and maintenance procedures should be documented on local (NZ-hosted) systems with offline backup.

---

7. WORKFORCE AND SKILLS

7.1 The telecommunications workforce

NZ’s telecommunications workforce is a critical asset. The sector employs approximately 20,000–25,000 people directly, though estimates vary depending on how the sector is defined (the Information Media and Telecommunications industry was estimated at 40,000–50,000 in the broader classification).³⁹ This includes:

• Network engineers — Design, build, and maintain fibre, cellular, and core networks. Spark, One NZ, 2degrees, Chorus, and LFCs each employ engineering teams. Total across all operators: perhaps 2,000–4,000 engineers and technicians with hands-on network skills.
• Fibre splicers and cable technicians — Chorus and its contractors employ trained fibre splicers. This is a specialised skill — fusion splicing requires training and practice. Estimated 200–500 trained fibre splicers in NZ.⁴⁰
• RF engineers — Cellular network radio planning and optimisation specialists. Perhaps 100–300 across NZ.⁴¹
• Software and systems engineers — Network management, software development, systems administration. NZ has a large software sector; the subset with specific telecom network management skills is smaller but substantial.
• Electrical technicians — Power systems, battery maintenance, generator operation. Shared with the broader electrical trades workforce.

7.2 Skills retention

The telecommunications workforce will be under pressure from competing demands — food production, general infrastructure maintenance, and other essential industries all need skilled workers. The government must explicitly designate telecommunications maintenance as an essential service and allocate sufficient workforce to it.

Key retention actions:

• Designate telecom maintenance personnel as essential workers — exempted from general labour mobilisation, with priority food and resource allocations (Doc #1)
• Consolidate the workforce — Under network consolidation (Section 3.3), engineers from all three mobile operators and Chorus work as a single team. Eliminate competitive barriers to knowledge sharing.
• Cross-train — Ensure fibre splice skills, RF engineering knowledge, and core network management skills are held by multiple people in each region. No single point of failure for critical knowledge.
• Document everything — Equipment-specific maintenance procedures, cable route maps, exchange configurations, equipment inventory. This knowledge currently exists partly in corporate systems and partly in the heads of experienced technicians. It must be captured and preserved in locally accessible formats.
• Train replacements — Begin training new telecom technicians immediately (Doc #157 — Trade Training). Basic fibre splicing can be taught in weeks; cellular RF engineering takes longer. Prioritise the most critical skills (fibre splicing, power systems, core network management).

7.3 Knowledge capture urgency

Some telecommunications knowledge is held by a small number of specialists:

• Submarine cable terminal equipment (SLTE) maintenance — Operated by a very small team at each cable landing station. If these people are unavailable, operating and maintaining the SLTE becomes extremely difficult. Their knowledge must be documented immediately and additional staff trained.
• Core network routing — NZ’s internet routing depends on configuration maintained by a small number of core network engineers at each major operator. The routing tables, peering arrangements, and network architecture must be documented so that others can maintain it.
• Legacy equipment — Older equipment (copper exchange switches, early-generation fibre equipment, microwave link systems) is understood by older technicians who may retire or become unavailable. Their knowledge of how to maintain, troubleshoot, and repair this equipment should be captured before it is lost.

---

8. BATTERY BACKUP SYSTEMS

8.1 The battery challenge

Every telecommunications site depends on battery backup for continuity during power outages. The battery population across NZ’s telecom network is large:

• Cell sites: Approximately 5,500–7,000 sites, each with battery banks (typically 48V lead-acid or lithium-ion systems providing 4–8 hours backup)⁴²
• Exchanges: Each of several hundred exchanges has substantial battery banks (48V systems, often providing 8–24 hours backup for the entire exchange)
• Cabinet and pillar sites: Smaller battery systems at street-level aggregation points

Lead-acid batteries in float service (continuously trickle-charged from mains power, as in telecom applications) typically last 5–10 years. Lithium-ion batteries last 7–15 years.⁴³ This means the existing battery fleet begins failing within the first 3–7 years.

8.2 Battery management strategy

• Inventory all telecom batteries through the skills census (Doc #8). Knowing what exists and its age/condition is the first step.
• Recover batteries from decommissioned sites for redeployment at priority sites. Network consolidation (Section 3.3) releases batteries from decommissioned cell sites.
• Reduce battery load at sites by lowering equipment power consumption where possible — turning off unnecessary equipment, reducing transmit power during low-traffic periods (night), and supplementing with solar/wind.
• Maintain batteries properly — Correct float voltage, temperature compensation, regular inspection, cleaning of terminals. Poor maintenance accelerates degradation. Telecom battery maintenance is a well-understood discipline.
• Transition to locally produced lead-acid batteries — NZ can produce lead-acid batteries from recycled lead and locally manufactured sulfuric acid (Doc #113). The dependency chain: recycled lead must be smelted (requiring furnace capability and fuel — Doc #91), cast into plates, and paired with sulfuric acid electrolyte (which itself requires sulfur or pyrite, a roasting furnace, and an absorption process — Doc #113). Separators can be fabricated from locally available materials (PVC or rubber sheeting, or treated wood veneers as a lower-performance substitute). Battery cases require acid-resistant containers (repurposed polypropylene or locally produced stoneware). The performance of locally produced batteries will be lower than commercial telecom batteries — expect 50–70% of rated capacity and 40–60% of cycle life — but functional for backup power applications. See Doc #35 (Battery Management and Lead-Acid Production).

---

9. LONG-TERM TRANSITION

9.1 The transition to simpler communication

Modern telecommunications — 4G/5G cellular, GPON fibre, IP networking — is extraordinarily complex. It works because global supply chains provide a continuous flow of precisely manufactured components. Under isolation, this complexity becomes a liability.

The long-term trajectory (Phase 4 and beyond) is a transition to communication technologies that NZ can sustain:

• Copper wire telephony — The most maintainable electrical communication technology. Requires copper wire (drawn from recycled copper stock using wire-drawing equipment — Doc #70), basic exchange switching, and telephone handsets. NZ has extensive copper cable still in the ground and the electrical knowledge to maintain it. New copper wire production requires: copper smelting from recycled stock (furnace, charcoal or coke fuel, flux), wire drawing through a series of progressively smaller dies (requiring hardened steel or tungsten carbide dies — a precision manufacturing step), and insulation (enamel coating or fabric wrapping). The exchange switching can be simplified from modern digital systems to basic analogue or manual switching if needed — manual switchboard operation is a low-skill task that can be trained in days.
• HF and VHF radio — Documented in Doc #128. Radio transmitters and receivers can eventually be built from local materials (Doc #131), including locally manufactured vacuum tubes. Performance is vastly below modern telecom: voice-only communication (or very low-speed data), subject to atmospheric propagation conditions, requiring skilled operators, and limited to one conversation per frequency channel at a time. Adequate for coordination and emergency communication; not a substitute for the volume and convenience of modern telecom.
• Locally manufactured electronics — The long-term path to restoring electronic communication involves rebuilding basic electronic component manufacturing (Doc #135). This is a decades-long project.

9.2 What NZ can build

Some telecommunications components can be produced or substituted with NZ materials and skills:

Component	NZ Capability	Timeline	Notes
Copper wire and cable	Yes — wire drawing from recycled copper	Phase 2–3	Adequate for voice telephony over distances up to 5–10 km without amplification. Locally drawn wire will have higher resistance and less consistent gauge than commercial cable, limiting maximum line length. Not suitable for data transmission beyond very low speeds. See Doc #70.
Basic telephone handsets	Yes — carbon microphone, magnet-and-coil speaker, basic wiring	Phase 2–3	19th-century technology. Requires: carbon granules (from charcoal), thin metal diaphragms (sheet brass or tin), permanent magnets (salvaged), fine copper wire for coils (Doc #70), and a housing (wood, bakelite, or salvaged plastic). Audio quality substantially below modern handsets.
Battery banks (lead-acid)	Yes — from recycled lead and local sulfuric acid	Phase 2–3	See Doc #35, Doc #113. Expect 50–70% of commercial battery capacity and 40–60% of cycle life due to less uniform plate casting and lower electrolyte purity. Adequate for telecom backup if sized accordingly.
Antenna hardware	Yes — wire, timber masts, basic metalwork	Phase 1+	See Doc #128.
Fibre-optic cable	No — requires high-purity silica preform fabrication at ~2,000°C, precision drawing tower, UV-curable acrylate coatings (petrochemical-derived)	Not foreseeable	Dependency chain includes: high-purity synthetic silica (not natural sand), germanium or fluorine dopants, precision temperature control, clean-room conditions. Maintain existing stock.
Semiconductors (transistors, ICs)	No — decades of industrial development required	Phase 6–7	See Doc #135.
Fusion splicers	No — precision optics and electronics	Not foreseeable	Preserve existing stock absolutely.
Radio transmitters (valve-based)	Yes — with locally produced vacuum tubes	Phase 3–4	Power output 10–100W typical vs. modern solid-state 500W+. Higher power consumption, shorter component life (1,000–5,000 hours per valve vs. decades for transistors). See Doc #131.
Radio receivers	Yes — simpler designs achievable	Phase 3–4	Sensitivity and selectivity well below modern digital receivers. Adequate for voice communication on HF/VHF; unsuitable for high-data-rate applications. See Doc #131.
Coaxial cable	Possible — copper tube with insulation	Phase 2–3	Requires: copper tube or wire (Doc #70), dielectric insulator (polyethylene from recycled stock, or wax-impregnated cotton as lower-performance substitute), and outer shielding (copper braid or foil). Locally produced coaxial cable will have higher signal loss and lower bandwidth than commercial cable — adequate for HF/VHF radio connections, marginal for higher-frequency applications.
Solar panels	No — semiconductor manufacturing required	Not foreseeable	Maintain existing stock. See Doc #73.

9.3 The postal service as communication backbone

As electronic communication contracts, physical mail becomes increasingly important for non-urgent communication. NZ Post’s infrastructure (sorting facilities, delivery routes, post offices) can provide nationwide message delivery as long as transport is available. Paper communication is slow but requires no electronics.

The Recovery Library’s documents on printing (Doc #5) and paper production address the material supply chain for physical communication. In the longer term, a robust postal service may carry more non-urgent communication volume than the surviving electronic network.

9.4 Last-resort communication: runners, signal fires, and oral relay

These methods become relevant when electronic communication has failed entirely — they are not substitutes for functioning telecom infrastructure but are the last-resort communication layer.

When all electronic communication has failed at a given location — no cellular, no copper telephone, no HF radio — communication reverts to pre-electronic methods. Pre-European Māori communication across NZ’s dispersed geography was sophisticated, reliable, and maintained over centuries. These methods represent validated solutions to communication under exactly the conditions that emerge at the end of the degradation timeline: no electronics, dispersed communities, and physical geography as both barrier and medium.

Runners and human messenger networks (kaikarere):

The pre-European kaikarere (messenger runner) system maintained communication between hapū and iwi across NZ’s terrain. Runners carrying verbal messages could cover 60–100 km per day on established tracks.⁴⁴ Messages were memorised precisely — transmission fidelity was a trained skill, not an approximation. In some cases, messages were encoded on tally sticks or similar physical objects as memory aids.

Recovery value: Kaikarere networks can carry non-urgent messages between communities at the pace of foot travel — typically 1–4 days between most NZ community pairs. This is slow but orders of magnitude more reliable than no communication at all. The primary cost is runner time; the knowledge exists within communities connected to traditional practices. Establishing kaikarere routes between communities that lose electronic connectivity should be explicitly planned, not improvised.

Signal fires and visual communication:

Pū (fire-and-smoke signals), tōtara bark torch signals, and hilltop visual communication were used to transmit simple, pre-agreed messages (warnings, assembly calls, harvest signals) across long distances.⁴⁵ NZ’s geography — numerous high points with clear sightlines across valleys and between headlands — is well suited to line-of-sight visual signalling.

Recovery value: Visual signals are not a general-purpose communication medium — they can convey pre-agreed simple signals, not arbitrary messages. But for emergency alerts, assembly calls, and status signals between known nodes, they have real utility. Establishing a set of pre-agreed visual signals between marae, Civil Defence nodes, and community hubs requires negligible resources (agreed protocols and a hillside) and provides a zero-technology fallback for emergencies.

Oral communication networks and whakapapa-based relay:

Māori oral communication has always used social networks — whakapapa connections between hapū mean that a message can travel through a chain of kin relationships, each link adding a local relay. This is distinct from kaikarere (which uses dedicated runners) — it is the normal process by which information spreads through social networks. Under isolation, when people travel between communities (for trade, welfare checks, agricultural exchange), they carry information.

Recovery value: Oral relay is slow, subject to distortion (unlike kaikarere, which involves trained transmission), and limited by travel frequency between nodes. But in areas where all electronic communication has failed and dedicated runners are not organised, social oral networks are the default that operates automatically. Planning recognition: Civil Defence communication planning should acknowledge oral relay as a real (if imperfect) channel and design welfare check-in systems that function through it — for instance, community representatives who travel to market or exchange points also carry welfare status information back to Civil Defence nodes.

These traditional communication methods are most valuable as the last layer of redundancy — when electronic communication has contracted, HF radio is not available at a given location, and nothing else works. They should be documented and planned as part of the regional Civil Defence communication architecture, not left as an unacknowledged informal fallback. Doc #160 (§4.5–4.7) provides the governance framework for integrating traditional knowledge into recovery planning.

---

10. RURAL AND COMMUNITY COMMUNICATION ACCESS

This section addresses the practical communication needs of rural and remote communities — particularly Māori communities — under network degradation, the role of marae as communication infrastructure, and Treaty obligations for equitable access. These are not supplementary considerations — in rural and remote areas, Māori communities represent a significant share of the population in communication-stressed zones (Northland, East Cape, the Waikato, Taranaki hinterland, Marlborough Sounds, and Southland). Planning that does not account for their specific connectivity situation will fail them and will undercount the actual communication gap the network must close.

Cross-reference: Doc #160 (Heritage Skills Preservation, §4.5–4.7) provides the broader governance framework. Doc #150 (Community Health and Rural Medical Access) addresses the medical communication dependencies that make telecom access survival-critical in remote Māori communities.

10.1 Rural and remote Māori community connectivity needs

Māori communities are disproportionately located in rural and remote areas — a legacy of land alienation that concentrated Māori landholding in areas that were agriculturally marginal or geographically isolated. Many of the areas at greatest risk of early cellular network contraction under the degradation timeline (Section 2.2) are predominantly or substantially Māori: Te Tai Tokerau (Northland), the East Cape and Tairāwhiti, the upper King Country, and parts of Northland, Whanganui, and Southland.

Specific connectivity dependencies:

• Medical consultation. Remote Māori communities frequently lack resident general practitioners. Telehealth via video link or telephone is the primary means of specialist medical access for many whānau in these areas. Doc #150 documents this dependency in detail. Loss of telecommunications connectivity means medical consultation reverts to physical travel — in an isolation scenario, constrained by fuel — or goes without. The health burden falls disproportionately on the most isolated communities.
• Educational continuity. Distance education for children in remote rohe, and adult education programmes operated through kohanga reo and kura kaupapa Māori networks, depend on telecommunications. These have both humanitarian and economic value — an educated workforce is a recovery asset.
• Agricultural and resource management coordination. Many iwi and hapū manage land, fisheries (Doc #150), and forestry (Doc #160) collectively across large rohe. Coordination of these activities depends on communication between far-flung whanau and hapū nodes.
• Whānau welfare monitoring. In the first phases of isolation, knowing who needs help and where is a primary welfare function. Remote Māori communities that lose connectivity lose visibility to Civil Defence and welfare networks — a serious gap in the recovery’s ability to reach people who need assistance.

The network contraction asymmetry. Under the prioritisation framework in Section 3.2, rural and remote communities are lower-priority for cellular maintenance than major urban centres. This is defensible at the network level but means that Māori communities in rural rohe will lose cellular service earlier than urban communities. The network contraction plan must explicitly address what replaces cellular in these areas before the sites fail, not after. Providing HF radio (Doc #128) capability to key nodes in each affected rohe, before cellular goes down, is the intervention that prevents these communities from becoming communication-dark.

10.2 Marae as communication hubs

Marae are the pre-existing community infrastructure most suited to serving as communication nodes when the general telecommunications network contracts. This is an extension of their established role as community gathering, coordination, and welfare centres.

The practical case for marae as communication nodes:

• Marae are geographically distributed across rural and remote NZ in patterns that reflect historic Māori settlement — they are often located where no other community facility exists.
• Marae have established governance structures (hapū or iwi management) capable of operating shared community resources.
• Marae already function as hubs during emergencies — Civil Defence has existing relationships with marae in many regions as welfare and evacuation centres.
• Physical marae buildings provide housed, secure locations for communication equipment (HF radio, battery banks, solar panels) that might not be viable in open community locations.

Connectivity provision for marae as communication hubs:

Priority should be given to maintaining or establishing connectivity at marae serving communities that would otherwise have no alternative. In the network prioritisation framework (Section 3.2), marae in rural areas should be treated as equivalent to other critical community infrastructure nodes — not as residential premises subject to residential service contraction.

Practical provision: - Where fibre or copper service reaches the marae site, that connection should be maintained as long as technically feasible. - Where cellular is the primary access technology and the site is in a Tier 3 or Tier 4 coverage area under the contraction plan, deploy HF radio capability at the marae before cellular ceases. The transition should be planned and managed — not abrupt. - Marae communication nodes should receive priority power supplementation (solar panels, battery backup) to ensure they remain operational during grid outages. The skills census (Doc #8) should identify which marae currently have solar or independent power. - Equipment to install: a basic HF transceiver (Doc #128), a computer terminal for data communication (where network connectivity survives), a notice board function (physical, in the absence of electronic alternatives), and a charge point for community members’ devices.

Governance of marae communication nodes. Iwi and hapū governance of marae means that the operational model for marae communication hubs must be developed with, not imposed on, marae communities. Civil Defence and government communications agencies should engage iwi and hapū as partners in establishing the hub model, training operators, and integrating marae nodes into the regional communication network. The skills census (Doc #8) should specifically identify people within each marae community who hold or could be trained in radio operation.

10.3 Te Reo Māori content and language equity in the communications network

As the telecommunications network contracts and communication becomes scarcer and more mediated, the language in which communication services operate becomes practically significant.

Te reo Māori in operational communication:

• Civil Defence emergency broadcasts, welfare information, and health advisories should be delivered in te reo Māori as well as English in areas with significant Māori populations. This is not only a cultural matter — comprehension affects compliance and response. Emergency instructions not understood are not followed. Communities where te reo is the primary language of elderly members (who may be at greatest health risk) must receive critical communications in te reo.
• Network operators and hub coordinators at marae communication nodes in te reo-speaking communities should be capable of working in te reo. A communication hub that cannot serve the language of its community does not function as a community resource.
• Training materials for HF radio operation, basic telecommunications maintenance, and emergency communication protocols should be available in te reo Māori. The language of instruction affects who can be trained.

Te reo Māori content delivery through the network:

The telecommunications network carries te reo Māori language media, education, and cultural content — radio stations (Irirangi), Māori Television via internet streaming, iwi-operated media, and te reo distance education. As the network contracts, these content streams narrow. The cultural stakes are real: te reo Māori revitalisation is an ongoing project whose continuity depends partly on electronic delivery infrastructure. The Recovery Library takes no position on ranking cultural outcomes against other recovery priorities, but the loss of te reo broadcast and education infrastructure should be acknowledged as a consequence of network contraction, not treated as invisible.

Within the degradation timeline, Māori Television and iwi radio stations are likely to be affected in similar ways to other broadcasters — see Doc #2 (Public Communication) for the broadcast infrastructure analysis. The practical recommendation here is narrower: where marae communication nodes include data access to surviving domestic internet, te reo content distribution should be included in the content prioritisation decisions for NZ-hosted servers (Section 6.4).

10.4 Treaty obligations and equitable communication access

The Treaty of Waitangi and its principles are relevant to how the telecommunications network is maintained and contracted under isolation. Specifically:

Article 2 (tino rangatiratanga) and resource governance. Iwi and hapū have the right to manage resources within their rohe. Telecommunications infrastructure within a rohe — particularly infrastructure serving predominantly Māori communities — should not be decommissioned without engagement with the relevant iwi or hapū. Where decommissioning is unavoidable, the alternative provision (HF radio, copper fallback) must be agreed with the community, not unilaterally decided by government agencies.

Article 3 (ōritetanga / equity) and communications access. The principle of equity requires that Māori communities do not receive systematically worse communication services than comparable non-Māori communities. The network contraction plan will, by its nature, reduce rural coverage earlier than urban coverage. If rural areas are disproportionately Māori, this asymmetry requires explicit justification and mitigating action. Mitigating action does not necessarily mean maintaining rural cellular coverage that is not technically sustainable — it means ensuring that when cellular fails, a genuine alternative is in place, not a gap.

Practical Treaty compliance framework for network contraction decisions:

1. For any proposed decommissioning of cellular or fibre service in areas with significant Māori population, require iwi and hapū engagement before decommissioning proceeds.
2. For Tier 4 (allow to fail) sites in predominantly Māori areas, require that alternative provision (HF radio or copper) is confirmed operational before the site is decommissioned.
3. Include iwi and hapū representatives in regional Civil Defence communication planning committees.
4. In the skills census (Doc #8), specifically identify Māori community members with existing telecommunications or radio skills — they are assets for the community communication hub model.

These requirements add modest process overhead to the network contraction programme. They are not presented as constraints that can override technical necessity — if a cell site fails catastrophically, it cannot be un-failed on Treaty grounds. They are requirements for the planned, managed elements of network contraction.

---

11. CRITICAL UNCERTAINTIES

Uncertainty	Why It Matters	How to Resolve	Impact if Adverse
NZ–Australia submarine cable survival	Determines whether NZ has high-bandwidth or HF-only international communication	Test within first hours	If cables fail, all international coordination drops to HF radio bandwidth — orders of magnitude less capable
Actual equipment age distribution across NZ telecom network	Determines degradation timeline accuracy	Inventory through skills census (Doc #8)	If equipment is older than assumed, degradation timeline compresses
Fusion splicer inventory size and condition	Determines how long fibre cable repair is possible	Immediate inventory of all splicers nationally	If fewer than expected or in poor condition, fibre network lifespan shortens significantly
Equipment compatibility across mobile operators	Determines how effective network consolidation is	Technical assessment by combined engineering team	If equipment is highly incompatible across vendors, cannibalisation yields fewer usable spares
Battery fleet condition	Determines backup power timeline	Condition assessment at all sites	If batteries are already aged, backup failures begin sooner
Nuclear winter effects on outdoor electronics	Temperature cycling and moisture may affect equipment differently under altered climate	Empirical monitoring	Uncertain — could be better (cooler temperatures reduce thermal stress) or worse (increased condensation)
Government capacity to execute network consolidation	Requires overriding commercial interests and coordinating across companies	Political assessment; emergency powers framework (Doc #144)	If consolidation fails, each operator’s network degrades independently — shorter total network life
Copper network actual surviving footprint	Determines the copper fallback option’s viability	Field survey by Chorus	If most copper has been physically removed, the copper fallback is limited

---

12. CROSS-REFERENCES

Document	Relevance to This Document
Doc #1 — National Emergency Stockpile Strategy	Framework for managing finite equipment stocks; fuel allocation for generator backup
Doc #2 — Public Communication	Telecommunications as delivery channel for public communication
Doc #5 — Printing and Knowledge Distribution	Physical alternatives as electronic communication contracts
Doc #156 — Skills Census	Inventory of telecom equipment, workforce, and skills
Doc #135 — Computer Construction	Long-term path to locally manufactured electronics
Doc #35 — Battery Management and Lead-Acid Production	Battery replacement for telecom backup power
Doc #56 — Wood Gasification	Generator fuel for critical sites
Doc #65 — Hydroelectric Maintenance	Grid power continuity — the foundation of telecom operation
Doc #70 — Copper Wire Production	Long-term copper cable replacement
Doc #73 — Solar Panel and Inverter Maintenance	Solar supplementation at cell sites and exchanges
Doc #91 — Machine Shop Operations	Fabrication of mechanical components for telecom infrastructure
Doc #113 — Sulfuric Acid Production	Input for lead-acid battery production
Doc #128 — HF Radio Network	Backup and eventual successor communication system
Doc #131 — Radio Equipment Fabrication	Long-term path to locally manufactured radio equipment
Doc #144 — Emergency Powers	Legal framework for network consolidation and resource requisition
Doc #157 — Trade Training	Training pipeline for telecom maintenance workforce
Doc #160 — Heritage Skills Preservation	Governance framework for integrating iwi and hapū into recovery planning (§4.5–4.7); referenced in Sections 9.4 and 10
Doc #150 — Community Health and Rural Medical Access	Medical communication dependencies in remote Māori communities; rural telehealth vulnerability

---

13. WHAT THIS DOCUMENT DOES NOT COVER

• Broadcast radio and television — Related but distinct infrastructure. Broadcast transmitters are fewer, more powerful, and follow a different maintenance trajectory. See Doc #2.
• Satellite communication — Inmarsat, Iridium, and Starlink constellations may remain partially functional for years but depend on constellation maintenance that is unlikely to continue indefinitely. Their lifespan is uncertain. A separate assessment would be valuable.
• HF radio network establishment — Covered in detail in Doc #128. This document focuses on the conventional telecom network; Doc #128 addresses the backup and long-term successor.
• Cybersecurity under isolation — Without international threats, the cybersecurity landscape changes. Residual concerns (insider threats, system failures from unpatched vulnerabilities) deserve separate analysis.
• Detailed equipment-specific maintenance procedures — This document provides strategic guidance. Specific procedures for Ericsson, Nokia, Huawei, and other vendors’ equipment are in manufacturers’ documentation, which should be preserved locally.
• Radio equipment fabrication — Long-term manufacturing of radio transmitters and receivers from NZ materials is covered in Doc #131.

---

APPENDIX A: PRIORITY ACTIONS CHECKLIST

For government decision-makers and telecommunications sector leaders:

Week 1:

• Test NZ–Australia submarine cable connectivity
• Begin mirroring critical international content to NZ servers
• Secure cable landing stations (power, physical security, staff)
• Contact all three mobile operators and Chorus — establish unified coordination
• Halt copper network decommissioning

Month 1:

• Designate telecommunications as essential service; protect workforce
• Begin national inventory of fusion splicers, test equipment, and critical spares
• Begin network consolidation planning
• Document all network configurations and cable routes (offline copies)
• Assess copper network surviving footprint

Months 1–6:

• Execute mobile network consolidation per region
• Establish fusion splicer and equipment depot
• Deploy additional battery backup and solar at Tier 1 sites
• Begin telecom-specific skills training (fibre splicing priority)
• Coordinate trans-Tasman cable maintenance with Australian counterparts

Year 1+:

• Ongoing: track equipment failure rates and adjust maintenance strategy
• Ongoing: recover equipment from decommissioned sites for parts
• Ongoing: train replacement telecom technicians
• Begin ONT and equipment refurbishment workshops
• Plan for copper voice service restoration in areas losing cellular coverage

---

APPENDIX B: GLOSSARY

Term	Definition
BBU	Baseband Unit — digital processing equipment at a cell site
DWDM	Dense Wavelength Division Multiplexing — technology for sending multiple data streams on a single fibre using different light wavelengths
GPON	Gigabit Passive Optical Network — fibre-to-premises technology used in NZ’s UFB network
LFC	Local Fibre Company — regional fibre network operators under the UFB programme
NMS	Network Management System — software for monitoring and managing network equipment
NVIS	Near-Vertical Incidence Skywave — HF radio technique for medium-distance communication (see Doc #128)
OLT	Optical Line Terminal — exchange-side fibre equipment
ONT	Optical Network Terminal — customer-side fibre equipment (the “fibre modem”)
OTDR	Optical Time Domain Reflectometer — test instrument for locating fibre cable faults
PSTN	Public Switched Telephone Network — traditional copper-based telephone network
RCG	Rural Connectivity Group — joint venture for rural mobile/broadband coverage
RU	Radio Unit — the radio transmitter/receiver at a cell site
SLTE	Submarine Line Terminal Equipment — shore-end equipment for submarine cables
UFB	Ultra-Fast Broadband — NZ government programme for nationwide fibre deployment

---

This document was prepared by Recoverable Foundation using AI-assisted research and drafting. It has not been reviewed by NZ telecommunications engineers or operators. The equipment counts, lifespan estimates, and degradation timelines are estimates based on publicly available information and general telecommunications engineering knowledge. They require verification from Chorus, Spark, One NZ, 2degrees, and submarine cable operators. The strategic recommendations — particularly network consolidation and copper preservation — involve significant commercial and political considerations that this document can identify but not resolve.

---

---

1. NZ cell site numbers are estimates. The Commerce Commission’s annual telecommunications monitoring reports provide some data on network scale, but exact site counts vary by source and definition. Spark, One NZ (formerly Vodafone NZ), and 2degrees collectively operate approximately 5,500–7,000 macro and micro cell sites across NZ. See Commerce Commission, “Annual Telecommunications Monitoring Report,” various years. https://comcom.govt.nz/regulated-industries/telecommunica...↩︎

2. NZ international submarine cable infrastructure: Southern Cross Cable Network (https://www.southerncrosscables.com/) and Hawaiki Cable (https://www.hawaikicable.co.nz/). Both systems have landing points in NZ, Australia, and the United States.↩︎

3. Telehealth in rural NZ: Ministry of Health and Health New Zealand data on telehealth uptake in rural and remote communities, accelerated during the COVID-19 pandemic. Studies of rural telehealth indicate that telephone and video consultations handle approximately 60–80% of the consultation volume that would otherwise require physical specialist access. In an isolation scenario with constrained fuel, the constraint on physical travel is more binding than in normal conditions. Exact figures for NZ rural Māori communities require verification from Health New Zealand regional data.↩︎

4. Supply chain coordination losses from communication degradation: Based on disaster logistics research. See Tomasini, R. and Van Wassenhove, L., “Humanitarian Logistics,” INSEAD, 2009; and UN OCHA humanitarian logistics guidance. The 20–40% logistics efficiency loss figure is an estimate for communication-degraded environments; exact NZ-specific figures would require modelling. The estimate is conservative — severely communication-degraded environments (e.g., post-earthquake situations with no functioning telecommunications) show larger coordination losses.↩︎

5. NZ telecommunications market structure is documented in Commerce Commission reports and the Telecommunications Act 2001 (as amended). The structural separation of Telecom NZ into Spark and Chorus in 2011 is a key structural feature. See https://comcom.govt.nz/regulated-industries/telecommunica...↩︎

6. Chorus network scale: Chorus annual reports and investor presentations describe the network as including approximately 31,000+ km of fibre cable and extensive copper infrastructure. See https://www.chorus.co.nz/investor-centre. Exact current figures require verification.↩︎

7. Ultra-Fast Broadband (UFB) programme: A NZ government initiative launched in 2009 (Phase 1) and extended through subsequent phases, aiming to provide fibre broadband to approximately 87% of NZ premises. Chorus is the primary provider; local fibre companies cover additional areas. See Crown Infrastructure Partners (now Crown Infrastructure). The target was substantially achieved by the early 2020s. https://www.crowninfrastructure.govt.nz/↩︎

8. Spark NZ network scale: Based on Spark investor reports and Commerce Commission monitoring data. Spark operates NZ’s largest mobile and fixed-line network. Cell site counts are approximate — operators do not always publish exact figures, and counts change as the network is expanded and reconfigured. https://www.spark.co.nz/↩︎

9. One NZ (formerly Vodafone NZ): Sold by Vodafone Group to a consortium of Infratil and Brookfield Asset Management in 2019. Rebranded to One NZ in 2023. Cell site count is an estimate based on publicly available information and regulatory filings. https://one.nz/↩︎

10. 2degrees network: NZ’s third mobile operator, now merged with Vocus NZ. Cell site count is estimated. 2degrees has grown significantly since launching in 2009 and participates in the Rural Connectivity Group. https://www.2degrees.nz/↩︎

11. Local Fibre Companies under the UFB programme: Northpower Fibre (Northland), Enable Services Ltd (Christchurch), and Tuatahi First Fibre (formerly Ultrafast Fibre, covering Hamilton, Tauranga, and other centres). See Crown Infrastructure Partners documentation.↩︎

12. Rural Connectivity Group (RCG): A joint venture of Spark, One NZ, and 2degrees, established to deliver rural broadband and mobile coverage under government-funded programmes (Rural Broadband Initiative, Mobile Black Spot Fund). https://www.rcg.co.nz/↩︎

13. Southern Cross Cable Network: Originally deployed in 2000 (Southern Cross Cable), with the Southern Cross NEXT cable completed in 2022. Owned by a consortium including Spark, Singtel, Verizon, and Telstra. Connects Auckland to Sydney and Los Angeles. https://www.southerncrosscables.com/↩︎

14. Hawaiki Cable: Completed in 2018. Connects Mangawhai Heads (NZ) to Sydney (Australia) and the US west coast (Oregon). https://www.hawaikicable.co.nz/↩︎

15. Ultra-Fast Broadband (UFB) programme: A NZ government initiative launched in 2009 (Phase 1) and extended through subsequent phases, aiming to provide fibre broadband to approximately 87% of NZ premises. Chorus is the primary provider; local fibre companies cover additional areas. See Crown Infrastructure Partners (now Crown Infrastructure). The target was substantially achieved by the early 2020s. https://www.crowninfrastructure.govt.nz/↩︎

16. GPON (Gigabit Passive Optical Network) architecture: ITU-T G.984 series standards. GPON uses a single fibre from the OLT, split passively to serve multiple premises (typically 1:32 or 1:64 split ratios in NZ deployments). The passive splitters contain no active components. This is well-established telecommunications engineering; see any modern fibre-optic network reference.↩︎

17. OLT equipment vendors in NZ: NZ’s UFB networks use OLT equipment from various vendors. Specific vendor selections vary by network operator and region. Major GPON OLT vendors globally include Nokia (Alcatel-Lucent), Huawei, Calix, ZTE, and others. The specific equipment deployed in NZ requires verification from Chorus and LFCs.↩︎

18. ONT deployment numbers are estimated from the number of fibre connections reported in Commerce Commission monitoring data. As of recent reports, fibre connections number approximately 1.2–1.5 million in NZ. Each connection requires one ONT at the customer premises. See Commerce Commission annual telecommunications monitoring reports.↩︎

19. Fusion splicing and OTDR technology: Standard fibre-optic maintenance references. Fusion splicers align and fuse optical fibres using an electric arc. OTDRs locate faults by analysing reflected light pulses. Both are precision instruments manufactured by companies like Fujikura, Sumitomo (splicers) and EXFO, Viavi Solutions (OTDRs). Cost and capability data based on general industry knowledge.↩︎

20. NZ mobile population coverage: Commerce Commission monitoring reports indicate population coverage above 98% for the combined mobile networks. Geographic coverage is substantially lower due to NZ’s terrain. Rural areas served by the RCG programme have expanded coverage in recent years but gaps remain.↩︎

21. Radio unit and power amplifier lifespan: Based on general telecommunications equipment engineering data. GaN (gallium nitride) and LDMOS (laterally diffused metal oxide semiconductor) power amplifiers are the dominant technologies in modern cell site radio equipment. Lifespan depends on operating temperature, loading, and duty cycle. 7–15 years is an estimate for typical cell site operation; some units last longer, some fail sooner.↩︎

22. Cell site battery backup duration: Industry standard for NZ cell sites is typically 4–8 hours of battery backup, though this varies by operator and site classification. Critical sites may have longer backup. Battery type is predominantly valve-regulated lead-acid (VRLA) with increasing deployment of lithium-ion. Based on general NZ telecommunications industry knowledge; exact figures by site require operator verification.↩︎

23. Chorus copper withdrawal: Chorus has been progressively withdrawing copper services from areas where fibre is available, as permitted under regulatory frameworks. The Commerce Commission has overseen this process. See Commerce Commission telecommunications determinations and Chorus public communications regarding copper withdrawal. https://www.chorus.co.nz/↩︎

24. Southern Cross Cable Network: Originally deployed in 2000 (Southern Cross Cable), with the Southern Cross NEXT cable completed in 2022. Owned by a consortium including Spark, Singtel, Verizon, and Telstra. Connects Auckland to Sydney and Los Angeles. https://www.southerncrosscables.com/↩︎

25. Hawaiki Cable: Completed in 2018. Connects Mangawhai Heads (NZ) to Sydney (Australia) and the US west coast (Oregon). https://www.hawaikicable.co.nz/↩︎

26. Electrolytic capacitor aging: Well-established electronics engineering knowledge. Capacitor lifespan models (Arrhenius equation for temperature dependence) are standard. The “10°C rule” (lifespan halves for every 10°C temperature increase above rated temperature) is widely used as a first approximation. See any electronics reliability engineering reference.↩︎

27. Semiconductor degradation mechanisms: Electromigration, hot-carrier injection, gate oxide breakdown, and related mechanisms are well-studied in semiconductor reliability engineering. Digital ICs operating well within ratings typically have projected lifespans exceeding 20 years. Power devices under thermal stress have shorter lifespans. See standard references on semiconductor reliability, e.g., Srinivasan, J. et al., “The Impact of Technology Scaling on Lifetime Reliability,” IEEE DSN, 2004.↩︎

28. Fusion splicing and OTDR technology: Standard fibre-optic maintenance references. Fusion splicers align and fuse optical fibres using an electric arc. OTDRs locate faults by analysing reflected light pulses. Both are precision instruments manufactured by companies like Fujikura, Sumitomo (splicers) and EXFO, Viavi Solutions (OTDRs). Cost and capability data based on general industry knowledge.↩︎

29. Fusion splice loss: A well-executed fusion splice on single-mode fibre typically achieves less than 0.05–0.1 dB loss. This is standard fibre-optic splicing performance documented in ITU-T recommendations and manufacturer specifications.↩︎

30. NZ fusion splicer inventory: This figure is an estimate. Chorus, local fibre companies, and fibre contracting companies (e.g., Downer, Visionstream, UCG, and others) hold fusion splicers. The exact national inventory has not been publicly documented and requires verification. The number 50–200 is a rough estimate based on the scale of NZ’s fibre industry and typical equipment-to-technician ratios.↩︎

31. Fibre-optic cable manufacturing requires drawing glass fibre from high-purity synthetic silica preforms at approximately 2,000°C, applying UV-curable acrylate coatings at high speed, and cabling multiple fibres into protected structures. This is a highly specialised industrial process concentrated in a small number of factories globally (Corning, Prysmian, Fujikura, YOFC). NZ does not have this manufacturing capability and cannot develop it without extensive industrial prerequisites.↩︎

32. Fusion splicer lifespan: Based on manufacturer specifications (Fujikura, Sumitomo) and general industry experience. Splicer lifespan depends heavily on usage frequency, storage conditions, and maintenance. Units used infrequently and stored properly may exceed 25 years; heavily used field units in poor conditions may fail sooner. The primary failure modes are electrode degradation, camera sensor failure, and alignment motor wear.↩︎

33. Fusion splicer electrode life: Manufacturer specifications typically rate electrode life at 2,000–5,000 arcs (splices) before replacement is recommended. Electrode performance degrades gradually — splice quality deteriorates before total failure. See Fujikura and Sumitomo Electric splicer maintenance documentation.↩︎

34. Mechanical fibre splicing: An alternative to fusion splicing that uses precision V-grooves or other alignment mechanisms to hold fibre ends together with index-matching gel. Higher loss than fusion splicing (typically 0.2–0.5 dB per splice) but requires no power or complex equipment. Mechanical splice products include 3M Fibrlok and Corning UniCam. NZ stock of mechanical splices should be inventoried.↩︎

35. Telecommunications battery lifespan: VRLA (valve-regulated lead-acid) batteries in float service (continuously charged, typical telecom application) have published design lives of 5–12 years, depending on construction quality and operating temperature. Actual achieved life in NZ conditions is typically 5–10 years. Lithium-ion batteries are increasingly deployed and have longer float life (7–15 years). Based on battery manufacturer specifications and general telecom industry experience.↩︎

36. Cell site battery backup duration: Industry standard for NZ cell sites is typically 4–8 hours of battery backup, though this varies by operator and site classification. Critical sites may have longer backup. Battery type is predominantly valve-regulated lead-acid (VRLA) with increasing deployment of lithium-ion. Based on general NZ telecommunications industry knowledge; exact figures by site require operator verification.↩︎

37. Cell site power consumption: Based on general telecommunications engineering data. A typical modern macro cell site with three sectors draws 1,000–2,000W including radio units, baseband processing, and ancillary equipment. Rural single-sector sites draw 300–800W. Older equipment generations tend toward the higher end. These figures are approximate and vary by vendor and configuration. See GSMA, “Green Power for Mobile” programme documentation for cell site power consumption benchmarks.↩︎

38. InternetNZ (now Internet New Zealand) administers the .nz country code top-level domain. DNS infrastructure for .nz includes name servers located in NZ. See https://internetnz.nz/. The .nz DNS infrastructure operates independently of international DNS root servers for .nz zone queries, though resolving international domains requires root server access.↩︎

39. NZ telecommunications workforce: Statistics NZ Business Demography Statistics and Linked Employer-Employee Data (LEED) provide employment data by industry. The Information Media and Telecommunications sector (ANZSIC Division J) employed approximately 40,000–50,000 people, but this includes media, publishing, and other non-telecom subsectors. The telecommunications-specific workforce (ANZSIC subdivision J58) is a subset. The figure of 20,000–25,000 for telecommunications specifically is an estimate. See Doc #156 for broader workforce data.↩︎

40. Trained fibre splicer numbers: This is an estimate based on the scale of NZ’s fibre deployment workforce. During the peak UFB build (2015–2022), NZ had a substantial fibre construction workforce. Post-build, the active splicing workforce has contracted but remains significant for maintenance. The exact number of currently trained and active splicers requires verification from Chorus, LFCs, and contracting companies.↩︎

41. NZ RF engineering workforce: This is an estimate based on the number of mobile operators (three), their typical engineering team sizes for network planning and optimisation, and the scale of NZ’s mobile network. Exact figures require verification from Spark, One NZ, and 2degrees. The broader Engineering NZ register and IPENZ workforce data may provide additional data.↩︎

42. Cell site battery backup duration: Industry standard for NZ cell sites is typically 4–8 hours of battery backup, though this varies by operator and site classification. Critical sites may have longer backup. Battery type is predominantly valve-regulated lead-acid (VRLA) with increasing deployment of lithium-ion. Based on general NZ telecommunications industry knowledge; exact figures by site require operator verification.↩︎

43. Telecommunications battery lifespan: VRLA (valve-regulated lead-acid) batteries in float service (continuously charged, typical telecom application) have published design lives of 5–12 years, depending on construction quality and operating temperature. Actual achieved life in NZ conditions is typically 5–10 years. Lithium-ion batteries are increasingly deployed and have longer float life (7–15 years). Based on battery manufacturer specifications and general telecom industry experience.↩︎

44. Pre-European Māori messenger runner (kaikarere) systems: Well documented in ethnographic and historical sources. See Best, E., “Māori and Moriori,” Journal of the Polynesian Society, 1924; Salmond, A., “Hui: A Study of Māori Ceremonial Gatherings,” 1975. Runner networks maintained inter-hapū and inter-iwi communication across NZ’s terrain. Distances of 60–100 km per day for fit runners on established tracks are consistent with historical accounts and comparable pre-European communication systems globally.↩︎

45. Pre-European Māori visual and fire signalling: See Best, E., “The Māori as He Was,” 1924; and Colenso, W., historical accounts of Māori hill-to-hill communication. Pū (fire signals) and torch signals were used for military alerts and community assembly calls, particularly in areas with clear sightlines between hilltops. The specific signals used were locally variable and agreed within communities — there was no single national code.↩︎