Doc #48 — Water Treatment Without Imports

---

RECOMMENDED ACTIONS (BY ACTUAL URGENCY)

First week:

1. Inventory all water treatment chemical stocks nationally — chlorine (gas and hypochlorite), alum, PAC, fluoride compounds, lime, soda ash, all polymers. Contact Watercare (Auckland), Wellington Water, all district council water operators. This feeds into the national asset census (Doc #8).
2. Reduce chemical dosing to minimum effective levels at all plants. Many plants operate with safety margins above minimum required doses; these margins can be narrowed to extend stocks.
3. Confirm grid power status for all water treatment plants and pumping stations. Water treatment is an absolute priority load for the electrical grid (Doc #65).

First month:

4. Commission assessment of every registered water supply for source water quality — which sources could operate safely with reduced or no coagulation? Artesian and deep bore sources are the immediate candidates.
5. Begin planning domestic sodium hypochlorite production. Identify salt stocks, electrolysis equipment (existing industrial electrochlorination units, pool chlorination systems, and laboratory equipment), and potential fabrication sites.
6. Inventory all UV disinfection equipment in NZ — installed systems, spare lamps, uninstalled units at suppliers. UV lamps are a finite consumable; the stock sets the effective life of UV as a primary disinfection method.
7. Issue public guidance on household water treatment: boiling protocols, rainwater collection best practice, recognition of unsafe water signs.

First 3 months:

8. Begin construction of pilot slow sand filtration units at sites where coagulation chemicals are expected to run out earliest. Slow sand filters take 2–6 weeks to mature biologically before they reach full effectiveness.⁴
9. Establish salt supply chain from Lake Grassmere and/or Dominion Salt to support chlorine/hypochlorite production. Coordinate with Doc #103 (salt production).
10. Begin fabrication of electrolysis cells for sodium hypochlorite generation. These can range from simple two-electrode brine cells to more sophisticated membrane cells; start with what can be built immediately and improve over time.
11. Assess feasibility of local alum production from NZ clay and acid sources. This is a longer-term project (see Section 6) but assessment should begin early.

First year:

12. Commission first municipal-scale domestic hypochlorite production facility.
13. Convert highest-risk surface water treatment plants to slow sand filtration where source water quality permits.
14. Develop training program for water treatment operators in new methods — many operators have only ever worked with chemical coagulation and chlorine dosing from manufactured stocks.
15. Begin catchment protection programs to reduce raw water contamination — fencing waterways, managing stock access, protecting riparian zones. Clean source water reduces treatment requirements.

Ongoing:

16. Expand domestic chlorine production capacity to match national requirements.
17. Develop coagulant alternatives (see Section 6) as chemistry and mineral processing capabilities develop.
18. Coordinate with wastewater treatment adaptation (Doc #49) to prevent upstream contamination of drinking water sources.

---

ECONOMIC JUSTIFICATION

Water treatment is a prerequisite for urban habitation. The economic justification is therefore which approach to pursue.

Cost of failure. Waterborne disease outbreaks under breakdown of treatment would produce mass casualties in concentrated urban populations. The Havelock North campylobacteriosis outbreak of 2016, which affected an estimated 5,500 people from a contaminated bore supply serving ~14,000 people, illustrates the scale: roughly 40% of the served population became ill, at least 45 were hospitalised, and 3–4 deaths were attributed to the outbreak.⁵ Scale that infection rate to Auckland’s 1.7 million people and the consequences would be severe: widespread illness reducing workforce productivity by a significant fraction for weeks, overwhelming an already strained health system, and undermining public confidence in governance.

Person-year estimates for the recommended approach:

Investment	Estimated person-years	Timeline
Inventory and assessment (Actions 1–4)	5–10	Month 1
Domestic hypochlorite production (pilot)	10–20	Months 2–6
Slow sand filtration (first installations)	20–50 per plant	Months 3–12
Operator retraining	5–10	Months 3–12
Municipal-scale hypochlorite production	30–50	Year 1–2
Catchment protection (ongoing)	50–100/year	Ongoing

These are rough estimates based on analogous infrastructure construction projects and water treatment engineering practice.⁶ The person-year cost of slow sand filtration construction depends heavily on the scale of the plant — a village-scale filter serving 500 people requires a few person-weeks of construction; a municipal plant serving 50,000 requires major earthworks.

Comparison with alternatives:

• Do nothing: Waterborne disease outbreaks within weeks of chemical depletion. Based on the Havelock North infection rate of ~40% of the served population,⁷ untreated urban water supplies could produce hundreds to thousands of deaths nationally, depending on the pathogen and speed of population response. Urban population dispersal likely. Effectively not an option.
• Boiling only: Technically effective for disinfection, but the energy cost of boiling all municipal water is prohibitive. Auckland consumes approximately 350 million litres per day.⁸ Boiling this volume would require energy equivalent to approximately 30–35 MW of continuous thermal input — feasible with grid electricity but representing a significant grid load, and impractical for the many households and institutions that would need to boil at point of use.
• Recommended approach: Domestic chemical production plus alternative treatment methods. Higher upfront investment but sustainable indefinitely with NZ resources.

The recommended approach pays for itself immediately relative to the “do nothing” alternative, and within months relative to boiling-only, in terms of reduced energy consumption and health system load.

---

1. CURRENT NZ WATER TREATMENT INFRASTRUCTURE

1.1 The national picture

New Zealand has approximately 830 registered drinking water supplies serving populations of more than 25 people.⁹ These range from major metropolitan systems serving hundreds of thousands to small rural supplies serving a few dozen. Approximately 85% of the population receives water from a registered supply; the remainder — predominantly rural — use private supplies (rainwater tanks, private bores, springs).

The major metropolitan systems are:

Auckland (Watercare Services Ltd): Serves approximately 1.7 million people. Sources include the Hunua Ranges dams (surface water), Waitakere Ranges dams (surface water), and the Waikato River (surface water). All surface water sources require full conventional treatment: coagulation, flocculation, sedimentation, filtration, and disinfection. Auckland is NZ’s most treatment-dependent city.¹⁰

Wellington (Wellington Water Ltd): Serves approximately 415,000 people across the Wellington metropolitan area. Sources include the Hutt River (surface water), Wainuiomata River (surface water), and artesian bores in the Hutt Valley. Surface sources require full treatment; artesian sources require minimal treatment.¹¹

Christchurch (Christchurch City Council): Serves approximately 390,000 people. Sourced almost entirely from deep artesian aquifers fed by the Waimakariri River system percolating through Canterbury Plains gravel. This water is naturally filtered through tens of metres of gravel and emerges at consistent quality. Historically, Christchurch did not chlorinate or treat its water at all — it was one of the largest unchlorinated urban supplies in the developed world. UV disinfection and residual chlorination were introduced from 2018 following the Havelock North inquiry recommendations, but the underlying source quality remains excellent.¹² Christchurch is NZ’s most resilient major city for water treatment under import cutoff.

Hamilton, Tauranga, Dunedin, and other centres: A mix of surface water and groundwater sources with varying treatment requirements. Hamilton draws from the Waikato River (full treatment required). Tauranga uses a combination of bores and surface water. Dunedin uses surface reservoirs with conventional treatment.

1.2 Treatment processes and their chemical dependencies

A conventional surface water treatment plant in NZ follows this process chain:

1. Coagulation/flocculation: Aluminium sulfate (alum) or polyaluminium chloride (PAC) is added to raw water to destabilise suspended particles. These particles aggregate into larger “flocs.” Sometimes a polymer flocculant aid is also added. All coagulant chemicals are imported.
2. Sedimentation: Flocs settle out in large basins. No chemical inputs required.
3. Filtration: Water passes through sand and/or activated carbon filters. Sand is locally available; activated carbon is imported but the filters function (less effectively for taste and organic removal) with sand alone.
4. Disinfection: Chlorine gas or sodium hypochlorite is added to kill remaining pathogens. Some plants use UV disinfection as a primary barrier with chlorine for residual disinfection in the distribution network. Chlorine gas is imported. Sodium hypochlorite is manufactured domestically by some suppliers (Ixom, Ravensdown) from imported or domestic precursors, but the manufacturing depends on chlor-alkali chemistry that uses imported components.¹³
5. pH adjustment: Lime (calcium hydroxide) or soda ash (sodium carbonate) adjusts pH after chlorination. Lime can be produced domestically from NZ limestone. Soda ash is imported.
6. Fluoridation: Hydrofluorosilicic acid or sodium fluoride is added at some plants. Entirely imported. Fluoridation is a public health measure for dental health, not a water safety requirement — it can be discontinued immediately without risk to water potability.

1.3 Groundwater vs. surface water: the treatment gap

The critical distinction for post-import planning is between surface water sources (rivers, lakes, dams) and protected groundwater sources (deep bores, artesian aquifers):

• Protected groundwater is naturally filtered through geological strata and typically has low turbidity, low organic content, and low or zero pathogen load. It may need only disinfection (and sometimes not even that) to be safe.
• Surface water is exposed to contamination from agriculture, wildlife, human activity, and weather events. It typically has higher turbidity, organic loading, and pathogen risk, and requires the full coagulation-filtration-disinfection treatment chain.

Under import cutoff, communities on protected groundwater are in a fundamentally different position from communities on surface water. This difference should drive prioritisation of chemical stocks and infrastructure investment.

Traditional Maori practice prioritised spring and seepage sources over river water where possible, and maintained strict controls on activities upstream of water collection points. These source protection principles align directly with the technical need to protect raw water quality and reduce treatment requirements – particularly under import cutoff, where reducing source contamination translates directly into reduced chemical demand.

---

2. IMPORTED CHEMICAL DEPENDENCIES AND DEPLETION TIMELINES

2.1 Chlorine and sodium hypochlorite

Current use: NZ’s water treatment plants use either chlorine gas (Cl2) or sodium hypochlorite (NaOCl) solution for disinfection. Chlorine gas is used at larger plants; sodium hypochlorite at smaller ones. Combined national consumption for water treatment is estimated at 3,000–5,000 tonnes of chlorine equivalent per year.¹⁴ This is a rough estimate — the national asset census (Doc #8) should establish the precise figure.

Supply chain: Chlorine gas is manufactured by the chlor-alkali industry (electrolysis of brine). NZ has limited domestic chlor-alkali production. Ixom (formerly Orica) has operated chlor-alkali facilities in NZ, but the scale and current status of domestic production versus importation is uncertain and should be verified.¹⁵ Much of NZ’s chlorine supply is imported from Australia or manufactured from imported precursors.

Sodium hypochlorite degrades over time — a 12.5% solution loses approximately 0.5–1% concentration per month at 25degC, faster in heat and light.¹⁶ Stockpiled hypochlorite solution therefore has a shelf life of months, not years. Chlorine gas is more stable but requires specialised storage (pressurised cylinders).

Estimated depletion timeline: Assuming stocks at treatment plants represent 1–3 months of normal consumption (typical for a just-in-time supply chain), and a further 1–3 months of stock in the supply chain (warehouses, distributors, ports), total available stock is estimated at 2–6 months of normal use. With reduced dosing and prioritisation, this could stretch to 6–12 months. This is one of the first critical depletion points for water treatment.

2.2 Coagulants (alum and PAC)

Current use: Aluminium sulfate (Al2(SO4)3) and polyaluminium chloride (PAC) are the primary coagulants used in NZ water treatment. Watercare Auckland alone uses approximately 15,000–20,000 tonnes of liquid alum per year.¹⁷ National consumption across all surface water treatment plants is estimated at 25,000–40,000 tonnes of liquid alum equivalent per year. These figures are estimates; actual consumption varies with source water quality and seasonal turbidity.

Supply chain: Alum is manufactured by reacting aluminium hydroxide (derived from bauxite) with sulfuric acid. NZ has no bauxite deposits and no alumina refinery. PAC is manufactured from aluminium chloride, also derived from imported alumina or bauxite. Both coagulants are entirely dependent on imported raw materials.

Estimated depletion timeline: Similar to chlorine — stocks of 2–6 months at normal consumption, stretchable to perhaps 12–18 months with reduced dosing and prioritisation of high-turbidity sources. Coagulant depletion is a serious problem because there is no straightforward domestic substitute at municipal scale (see Section 6).

2.3 pH adjustment chemicals

Lime (calcium hydroxide / calcium oxide): Produced by calcining (heating) limestone or shell deposits. NZ has extensive limestone deposits — major quarries at Otorohanga, Oamaru, and elsewhere.¹⁸ Lime production requires only limestone and a kiln heated to approximately 900–1000degC, well within NZ’s capability.¹⁹ Lime is one of the few water treatment chemicals that NZ can produce domestically at any required scale. Depletion is not a concern for lime.

Soda ash (sodium carbonate): Imported. Can be produced domestically from the Solvay process (requiring salt, limestone, and ammonia) or by trona mining (NZ has no known trona deposits). The Solvay process is chemically straightforward but industrially complex. Soda ash is used in smaller quantities than lime for pH adjustment; lime alone can serve most pH adjustment needs.

2.4 Fluoride compounds

Hydrofluorosilicic acid and sodium fluoride: Entirely imported. Fluoridation is not a water safety measure — it is a dental public health intervention. Under import cutoff, fluoridation should be discontinued immediately to conserve stocks of all chemicals for essential treatment functions. There is no water safety consequence to stopping fluoridation.²⁰

2.5 Polymer flocculant aids

Polyacrylamide and other synthetic polymers: Imported. Used in small quantities to improve floc formation and settling. Not essential — treatment plants can operate without polymer aids, though coagulation efficiency is reduced. Stocks will outlast the primary coagulants.

2.6 Summary depletion table

Chemical	NZ production possible?	Estimated stock duration	Criticality
Chlorine / hypochlorite	Yes — from salt electrolysis	6–12 months (stretched)	Critical — must be replaced
Alum / PAC	Not readily — no bauxite	12–18 months (stretched)	Critical — hardest to replace
Lime	Yes — from limestone	Indefinite	Low concern
Soda ash	Possible but complex	12–24 months	Moderate — lime substitutes
Fluoride compounds	No	N/A — discontinue	None for safety
Polymer aids	No	12–24 months	Low — not essential

---

3. DOMESTIC CHLORINE AND HYPOCHLORITE PRODUCTION

3.1 The chemistry

Chlorine production from salt is one of the oldest and most established electrochemical processes. The chlor-alkali process electrolyses brine (concentrated salt water) to produce chlorine gas, sodium hydroxide (caustic soda), and hydrogen gas:

2NaCl + 2H2O -> Cl2 + 2NaOH + H2

For water treatment purposes, the most practical product is sodium hypochlorite, produced either by:

• Dissolving chlorine gas in sodium hydroxide solution: Cl2 + 2NaOH -> NaOCl + NaCl + H2O
• Or direct electrochemical generation, where brine is electrolysed in an undivided cell to produce hypochlorite solution directly: NaCl + H2O -> NaOCl + H2

The direct electrochemical route is simpler and produces a dilute hypochlorite solution (0.5–1% available chlorine) suitable for water treatment dosing. This is the technology used in existing on-site generation (OSG) systems at swimming pools, food processing plants, and some water treatment facilities worldwide.²¹

3.2 NZ’s salt supply

NZ produces salt domestically at two main sites:

Dominion Salt, Lake Grassmere (Marlborough): Solar evaporation of seawater in large salt pans. Production capacity is approximately 60,000–80,000 tonnes per year, depending on weather conditions (solar evaporation requires dry, sunny weather — which nuclear winter would significantly reduce).²² This is NZ’s largest salt production facility.

Dominion Salt, Mount Maunganui: Processes imported salt and domestically produced salt. Under import cutoff, this facility’s output depends on domestic supply.

Nuclear winter impact on salt production: Lake Grassmere relies on solar evaporation during the Marlborough summer. Under 5–8degC cooling and reduced sunlight, evaporation rates would decline substantially — perhaps 30–60% reduction in annual output, though this is an estimate that depends on the specific climate conditions in Marlborough. The salt pans would still function; they would produce less. Supplementary heat-assisted evaporation using waste heat or wood fuel could partially compensate.²³

Salt quantity required for water treatment: To produce 3,000–5,000 tonnes of chlorine equivalent per year (the estimated range for NZ’s water treatment requirement — see Section 2.1), the electrolysis process requires approximately 5,100–9,000 tonnes of salt, based on the stoichiometry of the chlor-alkali reaction (1 mole Cl2 requires 2 moles NaCl: 2 x 58.4g NaCl per 71g Cl2, giving a theoretical ratio of 1.64 kg salt per kg chlorine, or approximately 1.7–1.8 kg/kg in practice due to process inefficiencies).²⁴ At the upper end, this is roughly 12–15% of Lake Grassmere’s reduced output under nuclear winter — a manageable fraction, especially given that water treatment is a non-negotiable priority allocation.

3.3 Electrolysis equipment

Existing equipment: NZ already has some on-site electrochlorination systems installed at swimming pools, food processing facilities, and possibly some smaller water treatment plants. The national asset census (Doc #8) should identify these. Each is a working example of the required technology.

Fabrication requirements: A basic electrochlorination cell requires:

• Electrodes: The anode must resist chlorine corrosion. Commercially, dimensionally stable anodes (DSA) — titanium coated with mixed metal oxides (iridium, ruthenium) — are used. These are imported items. Alternatives for domestic fabrication include graphite anodes (available from pencil/electrode stock, or fabricable from coke/pitch) and platinum-coated titanium (if platinum is available from laboratory or jewellery stocks). Graphite anodes work but erode over time, requiring periodic replacement.²⁵
• Cathode: Stainless steel or mild steel. Readily available in NZ.
• Cell body: Any non-conductive, chlorine-resistant container — PVC, polyethylene, fiberglass, or even wooden tanks lined with plastic sheet.
• Power supply: DC power at low voltage (3–5V) and high current (hundreds of amps for municipal-scale production). Rectifiers converting grid AC to DC are required. These are standard electrical engineering components within NZ fabrication capability.
• Brine supply system: Tanks and pumps for dissolving salt and feeding brine to the cells. Standard plumbing and chemical handling equipment.

The dependency chain: Salt -> brine preparation -> electrolysis cell (requiring electrodes, power supply, containment) -> hypochlorite solution -> dosing system. Each step has sub-dependencies, but none require materials or skills unavailable in NZ. The weakest link is electrode fabrication — graphite electrodes are adequate but consumable (eroding at approximately 1–3 kg per tonne of chlorine produced²⁶), and superior mixed-metal-oxide electrodes require imported specialty metals. Overall feasibility for domestic hypochlorite production: [B] Feasible.

Scale-up pathway:

• Immediate (months 1–6): Repurpose existing electrochlorination units. Build simple graphite-electrode cells at workshop scale. Produce hypochlorite for highest-priority treatment plants as chemical stocks deplete.
• Short-term (months 6–18): Build larger electrolysis facilities at or near major treatment plants. Improve electrode quality as fabrication experience develops.
• Medium-term (years 2–5): Establish full-scale domestic chlor-alkali production, potentially co-producing caustic soda (useful for many industrial processes) and hydrogen (usable as fuel or chemical feedstock).

3.4 Energy requirements

Electrochlorination consumes approximately 3.5–4.5 kWh of electricity per kilogram of chlorine produced.²⁷ To produce NZ’s estimated water treatment chlorine requirement of approximately 3,000–5,000 tonnes/year (see Section 2.1), using a mid-range estimate of 4,000 tonnes/year:

4,000,000 kg x 3.5–4.5 kWh/kg = 14,000,000–18,000,000 kWh/year = approximately 1.6–2.1 MW continuous

This is approximately 0.02% of NZ’s grid capacity of approximately 9,000–9,500 MW of installed generation (predominantly hydro and geothermal).²⁸ Power availability is not a constraint for domestic chlorine production.

---

4. ALTERNATIVE DISINFECTION METHODS

Where domestic chlorine production is insufficient or not yet established, several alternative disinfection methods are available.

4.1 UV disinfection

How it works: Ultraviolet light at 254 nm wavelength damages pathogen DNA, preventing replication. UV disinfection is highly effective against bacteria, viruses, and protozoa (including Cryptosporidium and Giardia, which are resistant to chlorination at normal doses).²⁹

NZ installation base: Many NZ water treatment plants already have UV disinfection systems, often as a primary barrier with chlorine providing residual disinfection in the distribution network. UV is also widely used in rural NZ on private supplies (domestic UV units are common at hardware stores and rural suppliers). The total installed base is significant but unquantified — the national asset census (Doc #8) should establish this.

Advantages: No chemical inputs required — only electricity and functioning UV lamps. Highly effective. No taste or odour effects on water.

Limitations: - UV lamps are consumable. Low-pressure mercury vapour lamps (the standard type) have rated lifetimes of 8,000–12,000 hours (approximately 1–1.5 years of continuous operation).³⁰ The total stock of replacement lamps in NZ at the time of the event determines how long UV disinfection can continue. Once lamps are exhausted, UV capacity declines unless lamps can be manufactured domestically — which requires mercury, quartz glass tube fabrication, and vacuum sealing. This is possible but not straightforward. - Turbidity interference. UV is less effective in turbid water because particles shield pathogens from UV exposure. Water must be filtered to below 1 NTU (nephelometric turbidity unit) before UV treatment is reliable.³¹ This means UV alone does not solve the coagulation problem for turbid surface water. - No residual disinfection. UV kills pathogens at the point of treatment but provides no ongoing protection in the distribution network. In long pipe networks (as in cities), recontamination is possible between the treatment plant and the tap. This is why many plants use UV plus chlorine — UV for primary kill, chlorine for residual protection.

Recommendation: UV disinfection is an excellent complement to domestic chlorine production. It should be maintained and expanded wherever possible. Lamp stockpiling is a high priority. Domestic UV lamp fabrication is a Phase 3–4 project requiring mercury sourcing (from existing NZ stocks of mercury thermometers, switches, and laboratory supplies), quartz glass tube fabrication (requiring high-purity silica sand and furnaces capable of 1700–2000degC), and vacuum sealing capability. Feasibility for domestic lamp fabrication: [C] Difficult — requires precursor industries that do not currently exist in NZ at the necessary scale.

4.2 Boiling

Effectiveness: Boiling water for one minute (at NZ altitudes) kills all common waterborne pathogens — bacteria, viruses, protozoa, and helminths. This is the oldest and most reliable disinfection method.³²

Limitations: - Energy cost at scale. Raising 1 litre of water from 15degC to 100degC requires approximately 0.1 kWh of energy. For a household using 200 litres/day of drinking and cooking water, this is 20 kWh/day — a substantial household energy load, though manageable with grid electricity. For a city, it is impractical as a centralised method (see Section on economic justification above). - Practicality. Boiling is a household-level method. It cannot substitute for municipal treatment but is an essential backup during treatment disruptions. - Does not remove turbidity or chemical contaminants. Boiling disinfects; it does not clarify or purify in the broader sense.

Recommendation: Boiling should be the standard public guidance for household water treatment whenever municipal supply safety is uncertain. Every household should be capable of boiling water — which, with grid electricity, requires only an electric jug. Without grid power, wood fuel serves but is slower and less convenient. Public guidance on boiling should be issued in the first week.

4.3 Solar disinfection (SODIS)

How it works: Clear water in transparent plastic (PET) bottles is exposed to direct sunlight for 6+ hours. UV-A radiation and thermal heating inactivate pathogens.³³

Limitations under nuclear winter: SODIS depends on strong sunlight. Under nuclear winter conditions with substantially reduced solar radiation, SODIS effectiveness would be severely compromised. This method is not reliable as a primary treatment strategy under the scenario, though it retains value as a household backup during clearer periods and in Phase 3+ as the atmosphere clears.

4.4 Household-scale ceramic and biosand filters

Ceramic filters: Locally fabricable from clay, sawdust (as burnout material to create pores), and colloidal silver (as a disinfectant coating). NZ has suitable clays and pottery infrastructure. Ceramic filters remove bacteria and protozoa but are less effective against viruses. Flow rates are low — typically 1–3 litres per hour — making them suitable for household use only.³⁴

Biosand filters: A household-scale adaptation of slow sand filtration. A concrete or plastic container filled with layers of sand and gravel. As water percolates through, a biological layer (the “schmutzdecke”) develops on the sand surface and removes pathogens. Effective against bacteria, protozoa, and to some extent viruses. Flow rate approximately 12–18 litres per hour.³⁵

Both are useful household-level technologies that can be manufactured and distributed without imported materials. They should be promoted for household and community use, especially in areas where municipal treatment is uncertain.

---

5. SLOW SAND FILTRATION: THE COAGULATION BYPASS

5.1 Why slow sand filtration matters

The central problem of water treatment without imports is not disinfection (chlorine can be produced locally) but coagulation (alum and PAC cannot be readily replaced). Slow sand filtration offers a way to bypass the coagulation step entirely for many water sources.

5.2 How it works

Slow sand filtration passes water slowly (0.1–0.4 metres per hour) through a bed of fine sand approximately 0.8–1.2 metres deep.³⁶ Over 2–6 weeks of operation, a biological community (the “schmutzdecke” — German for “dirt layer”) develops on the surface of the sand. This biofilm, composed of bacteria, algae, protozoa, and invertebrates, actively consumes and traps pathogens, organic matter, and fine particles in the raw water.

The result is a combined biological and physical filtration process that achieves: - 90–99% removal of bacteria - 99–99.99% removal of protozoa (Giardia, Cryptosporidium) - 90–99% removal of viruses (through biological predation and adsorption) - Significant turbidity reduction (typically to below 1 NTU from moderate raw water turbidity) - Partial removal of dissolved organic matter and colour

These removal rates are achieved without any chemical inputs — only sand, gravel, water, and time.³⁷ For comparison, conventional coagulation-flocculation-rapid filtration typically achieves >99.9% removal of bacteria and >99.99% of protozoa — so SSF matches or approaches conventional performance for protozoa but provides somewhat lower bacterial and viral removal, making post-filtration disinfection (chlorine or UV) important as a secondary barrier.

5.3 Construction requirements

A slow sand filter requires:

• Sand: Clean, uniform sand with effective size 0.15–0.35 mm and uniformity coefficient below 3. NZ has abundant sand deposits, but not all sand is suitable — beach sand is often too coarse and too variable, and river sand may need washing and grading. Purpose-prepared filter sand can be produced by washing and sieving local sand deposits.³⁸
• Gravel: A support layer of graded gravel beneath the sand bed. Locally available throughout NZ.
• Containment: The filter bed must be contained in a watertight basin. Concrete is the standard material for municipal-scale filters; NZ has cement production (Golden Bay Cement, now Holcim NZ, at Portland, Whangarei and elsewhere) and aggregate.³⁹ Smaller filters can use lined earthen basins, masonry, or ferrocement.
• Inlet and outlet controls: Piping to maintain constant water level above the sand bed and to collect filtered water from the underdrain. Standard plumbing.

What is NOT required: No electricity (gravity-fed). No chemical inputs. No sophisticated instrumentation — the filter is monitored by observing flow rate and testing output water quality. No imported materials if concrete, sand, and gravel are locally sourced.

5.4 Performance limitations and honest assessment

Slow sand filtration is not a universal solution. Its limitations must be understood:

• Turbidity ceiling. SSF works well with raw water turbidity below approximately 10–20 NTU. Above this, the schmutzdecke clogs rapidly and the filter requires frequent scraping (cleaning), reducing capacity. For highly turbid water (common during storm events on NZ rivers), pre-treatment is needed — either roughing filters (coarse gravel filters that remove gross turbidity) or sedimentation basins. Neither requires imported chemicals, but both require additional infrastructure.⁴⁰
• Cold temperature effects. Biological activity in the schmutzdecke declines with temperature. Under nuclear winter cooling, filter effectiveness may be reduced, particularly during the coldest months. This is a genuine concern but not fatal — slow sand filters have operated successfully in northern European and Canadian climates with water temperatures near 0degC, though with reduced pathogen removal rates requiring longer contact times or post-filtration disinfection.⁴¹
• Land area. SSF requires large filter areas because the filtration rate is slow. A plant treating 10,000 cubic metres per day (serving approximately 50,000 people) requires approximately 1,000–2,500 square metres of filter area. For Auckland’s 350,000+ cubic metres per day, the filter area would need to be 35,000–90,000 square metres (3.5–9 hectares). This is a substantial land requirement, though not impractical — it is comparable to a few sports fields.⁴²
• Maturation time. A new filter takes 2–6 weeks to develop an effective schmutzdecke. Filters should be built and matured before chemical stocks run out, not after.
• Not effective against all contaminants. SSF does not remove dissolved chemicals (pesticides, heavy metals, some industrial pollutants). Source water protection is essential.

5.5 Roughing filtration as pre-treatment

For high-turbidity source water, horizontal or vertical roughing filters — large beds of progressively finer gravel through which water flows — can reduce turbidity from hundreds of NTU to below 10–20 NTU, making the water suitable for slow sand filtration. Roughing filters require no chemicals, only gravel of various sizes and containment structures. They have been used extensively in developing countries for exactly this purpose.⁴³

The combination of roughing filtration + slow sand filtration + chlorine disinfection provides a complete treatment chain for surface water without any imported chemicals. This is the recommended long-term treatment solution for NZ surface water supplies under import cutoff. Feasibility for this combined approach: [A] Achievable with existing NZ capability and materials, though construction at municipal scale requires significant labour and lead time.

5.6 Historical precedent

London’s water supply relied on slow sand filtration from the 1830s through much of the 20th century. The technology was proven to dramatically reduce cholera and other waterborne diseases decades before the germ theory of disease was understood. Many European cities used SSF as their sole water treatment for over a century. It remains in use today in various countries, including in the UK and the Netherlands. This is not an experimental technology — it is one of the most thoroughly proven water treatment methods in history.⁴⁴

---

6. THE COAGULANT PROBLEM: LOCAL ALTERNATIVES

6.1 Why coagulation matters

Coagulation removes turbidity, dissolved organic matter (which forms harmful disinfection byproducts when chlorinated), and colour from water. It also improves the effectiveness of downstream filtration and disinfection. For surface water with significant organic loading (common in NZ’s bush-fed catchments, where humic acids produce brown-coloured water), coagulation is the step that makes the difference between aesthetically acceptable clear water and unpleasantly brown, organically-laden water that is difficult to disinfect safely.

Slow sand filtration (Section 5) addresses turbidity and pathogen removal. It partially addresses organic matter and colour, but for highly coloured water (common in NZ peat-influenced and bush catchments), coagulation provides substantially better removal of dissolved organics.

6.2 Aluminium sulfate (alum) from NZ materials

Alum is produced by reacting aluminium hydroxide with sulfuric acid:

Al(OH)3 + 3H2SO4 -> Al2(SO4)3 + 6H2O

Aluminium source: NZ has no bauxite deposits (the standard commercial aluminium source). However, aluminium is the third most abundant element in the Earth’s crust and is present in many NZ clays and minerals. The challenge is not the presence of aluminium but its extraction — dissolving aluminium from clay requires strong acid or alkali at high temperatures, and yields are lower and less pure than from bauxite. NZ does have the NZ Aluminium Smelter at Tiwai Point (Bluff), which holds substantial stocks of alumina and aluminium metal, but these are finite.⁴⁵

Sulfuric acid source: Sulfuric acid production in NZ is possible from elemental sulfur (available from geothermal sources and petroleum refining residues) or from pyrite (iron sulfide, which occurs in some NZ mineral deposits). Ballance Agri-Nutrients operates a sulfuric acid plant at Mount Maunganui that processes imported elemental sulfur, but the technology and infrastructure exist.⁴⁶ The question is whether sulfur supply can be maintained from domestic geothermal sources — this is plausible but would need to be established (see also Doc #113 on sulfuric acid if it exists in the catalogue).

Feasibility assessment: Producing alum from NZ clay and domestically-produced sulfuric acid is chemically possible but industrially challenging. The quality would be lower than commercial alum (more impurities, less consistent), and the cost in energy and labour would be higher. This is a Phase 3–5 project — not available in the critical first years, but worth developing as industrial chemistry capability grows. Rating: [C] Difficult.

6.3 Ferric coagulants

Iron-based coagulants (ferric chloride, ferric sulfate) are alternatives to aluminium coagulants. NZ has abundant iron ore — ironsand deposits on the west coast of the North Island (exploited by NZ Steel at Glenbrook).⁴⁷ Ferric sulfate can be produced by dissolving iron in sulfuric acid and oxidising the result:

2Fe + 3H2SO4 + oxidation -> Fe2(SO4)3

This is simpler than producing alum from clay, because metallic iron dissolves readily in acid. The Glenbrook steel mill produces iron and steel; scrap iron and steel offcuts are available throughout NZ. Sulfuric acid remains the limiting reagent.

Feasibility: Ferric coagulant production from local iron and locally-produced sulfuric acid is more feasible than alum production from clay. It should be prioritised as the replacement coagulant pathway. Rating: [B] Feasible.

6.4 Natural coagulants

Various plant-based and mineral coagulants have been used or studied:

Moringa oleifera seeds: The most studied natural coagulant. Crushed seeds contain proteins that act as cationic polyelectrolytes, effectively coagulating turbid water. Widely used in sub-Saharan Africa. However, Moringa is a tropical tree — it does not grow outdoors in NZ’s climate, particularly not under nuclear winter. Greenhouse cultivation would be necessary, competing for limited heated growing space with food crops. Not recommended as a primary strategy.⁴⁸

NZ-native plant coagulants: Research into coagulant properties of NZ native plant materials is limited. Some research has investigated the coagulant properties of various seed extracts and plant tannins globally, but no NZ-native species has been established as an effective municipal-scale coagulant. Tannin-based coagulants (from bark extracts) have shown some promise internationally, and NZ has tannin-rich bark sources (e.g., radiata pine bark), but these remain at the research stage for water treatment applications.⁴⁹

Bentonite and clay coagulants: Some NZ bentonite deposits exist (notably in the Northland region). Bentonite can assist with flocculation of certain water types but is not a substitute for primary chemical coagulation.

Honest assessment: Natural coagulants are not a viable replacement for alum/PAC at municipal scale with current knowledge. They may supplement chemical coagulants or serve at household/community scale, but they do not solve the municipal coagulation problem. The recommended pathway is ferric coagulants produced from domestic iron and sulfuric acid, combined with slow sand filtration to reduce the need for chemical coagulation.

---

7. SOURCE-SPECIFIC STRATEGIES

7.1 The Christchurch advantage

Christchurch is NZ’s best-positioned major city for water treatment under import cutoff. Its deep artesian aquifer supply is naturally filtered to a quality that does not require coagulation. Disinfection — the one treatment step applied — can be maintained with domestic hypochlorite production or UV (as long as lamps last). Christchurch should require minimal adaptation and no significant infrastructure investment for water treatment continuity.⁵⁰

Risk: The Canterbury Plains aquifer recharge depends on the Waimakariri River and rainfall. Under nuclear winter, reduced precipitation could affect aquifer levels over years. Monitoring of aquifer levels is important. The 2010–2011 earthquakes also demonstrated that seismic events can affect bore quality — some bores showed increased turbidity or bacteriological contamination post-earthquake.⁵¹ Emergency treatment capability should be maintained even for Christchurch.

7.2 Auckland: the hardest case

Auckland is NZ’s most treatment-dependent major city. Its surface water sources (Hunua Ranges dams, Waitakere dams, Waikato River) all require full conventional treatment. The Waikato River in particular carries agricultural and urban contamination requiring robust treatment.

Recommended approach for Auckland:

1. Stretch existing chemical stocks through reduced dosing and optimised use.
2. Build domestic hypochlorite production capacity early — Auckland should be the priority site.
3. Construct roughing filters and slow sand filters as pre-treatment to reduce reliance on chemical coagulation.
4. Investigate source switching: are there groundwater sources in the Auckland region that could supplement surface water and reduce treatment requirements? The Onehunga aquifer has historically supplied some Auckland water; other bores may be developable.⁵²
5. Implement aggressive catchment protection for dam catchments to maintain raw water quality.

Auckland’s transition will be the most complex and resource-intensive in the country. It should receive proportional investment.

7.3 Wellington

Wellington’s mix of surface water (Hutt and Wainuiomata rivers) and artesian bores (Hutt Valley) gives it a middle position. Prioritise the artesian sources, extend treatment chemical use to the surface water sources, and build slow sand filtration for the surface sources over time.

7.4 Smaller towns and rural supplies

Many smaller NZ towns draw from relatively clean catchment sources or bores. For these communities, domestic hypochlorite production at local scale (a single electrochlorination cell serving the town) plus basic filtration may be sufficient. Some communities may be able to operate with UV and filtration alone, particularly if source water quality is high.

7.5 Rural private supplies

Approximately 15% of NZ’s population uses private water supplies — predominantly rainwater tanks, private bores, and springs.⁵³ Many of these already operate with no chemical treatment (UV units, or no treatment at all). For this population, import cutoff has minimal impact on water supply, though guidance on maintaining water quality (roof and tank hygiene, UV unit maintenance, bore protection) should be distributed.

Rainwater collection may actually become more important under nuclear winter: while precipitation patterns may change, rainwater is inherently low in turbidity and pathogens (if collection systems are clean) and requires only disinfection. Communities that can expand rainwater harvesting reduce their dependence on surface water treatment.

Concern: Nuclear fallout does not affect NZ directly under the scenario (the exchange is in the Northern Hemisphere and NZ is physically unscathed), but atmospheric particulate from the nuclear winter itself — primarily soot — could contaminate rainwater in the early months. First-flush diversion (discarding the first few litres of roof runoff after dry periods) mitigates this. After the first year or two, as atmospheric particulates settle, rainwater quality should return to normal.⁵⁴

---

8. WASTEWATER TREATMENT INTERFACE

Safe drinking water cannot be considered separately from wastewater management. If wastewater treatment degrades, contamination of drinking water sources follows — particularly for surface water supplies downstream of urban areas.

NZ’s wastewater treatment plants also depend on imported chemicals — chlorine for effluent disinfection, polymers for sludge management, and various process chemicals. However, the core biological treatment processes (activated sludge, trickling filters, oxidation ponds) do not require chemical inputs beyond what the wastewater itself provides. The main risk is not treatment failure but reduced effluent quality — specifically, higher pathogen loads in treated effluent discharged to rivers and coastal waters.

Key coordination points:

• Wastewater plants discharging upstream of drinking water intakes must maintain disinfection. Domestic hypochlorite production should supply both drinking water and critical wastewater treatment plants.
• Where wastewater treatment degrades, drinking water treatment must compensate with more robust disinfection.
• Land-based wastewater disposal (irrigation) becomes more attractive as it keeps pathogens out of waterways. This requires land area and careful management but avoids the river contamination problem.

Detailed wastewater treatment adaptation is covered in Doc #49 (Wastewater Treatment). The two documents should be read together.

---

9. SCALING FROM HOUSEHOLD TO MUNICIPAL

A useful conceptual framework for water treatment without imports is to recognise three scales of operation, each with different appropriate technologies:

9.1 Household scale (1–10 people)

• Boiling (with grid electricity: electric jug; without: wood fuel)
• Household UV units (while lamps last)
• Biosand filters (locally constructable, no imports)
• Ceramic filters (locally fabricable)
• Rainwater collection (with first-flush diversion)

These methods are available immediately and require no municipal infrastructure. Public education and distribution of biosand filter construction guides should be an early action.

9.2 Community scale (10–5,000 people)

• Small electrochlorination units (workshop-fabricated, salt + electricity)
• Community slow sand filters (concrete or lined earth construction)
• Gravity-fed systems where topography allows
• Protected bore or spring supplies with minimal treatment

Many rural NZ communities already operate at this scale. The model is well understood and does not require chemical imports.

Where laboratory testing capacity is limited, field-observable water quality indicators supplement formal testing: water clarity, taste, the presence of certain healthy aquatic species (freshwater invertebrates indicating good oxygen levels), and the absence of algal blooms or surface films. These observational methods – drawn in part from traditional Maori knowledge of waterway health – do not replace chemical and microbiological testing but provide useful early warning of source quality changes between formal sampling events.

Marae-based water systems provide a working template for community-scale water management outside the municipal framework. Many marae maintain their own water supply, and where iwi have developed environmental monitoring and management programmes for their rohe, this capability should be integrated into the regional water management response.

9.3 Municipal scale (5,000–1,700,000 people)

• Industrial electrochlorination/chlor-alkali production
• Municipal slow sand filtration (large land area required)
• Roughing filtration for pre-treatment
• Ferric coagulant production (from local iron and sulfuric acid) for sources requiring chemical coagulation
• UV disinfection (while lamp stocks last)
• Source protection and catchment management

This is the most challenging scale and requires the most investment in new infrastructure. The transition period — from depletion of imported chemicals to full domestic capability — is the risk window.

---

CRITICAL UNCERTAINTIES

1. Actual chemical stock levels. The depletion timelines in this document are estimates based on typical supply chain inventory levels. The actual stocks at the time of the event could be significantly higher or lower. Establishing actual stocks (Action 1) is the single most important immediate step.

2. Nuclear winter impact on Lake Grassmere salt production. The 30–60% reduction estimate for solar evaporation is based on general climate impact reasoning, not specific modelling. The actual impact depends on Marlborough’s specific climate conditions under nuclear winter, which are uncertain.

3. Slow sand filtration performance under nuclear winter temperatures. Biological filtration efficiency declines with temperature, but the magnitude of this decline for NZ water types under nuclear winter conditions has not been studied. Pilot testing under local conditions is needed.

4. UV lamp stock. The total NZ stock of UV lamp replacements is unknown and sets the effective life of UV disinfection as a primary barrier. This is a critical inventory item.

5. Coagulant replacement pathway timeline. The feasibility and timeline for ferric coagulant production from domestic iron and sulfuric acid depends on sulfuric acid production from geothermal sulfur — a capability that exists in NZ but would need to be redirected and possibly expanded. The timeline for this is uncertain.

6. Seismic disruption. NZ is seismically active. A major earthquake during the post-import period could damage water treatment infrastructure, contaminate groundwater sources (as observed in Canterbury 2010–2011), and disrupt distribution networks. The baseline scenario does not assume major seismic events, but this is a significant contingency.

7. Population movement. If internal migration concentrates population in centres with better water infrastructure (e.g., Christchurch), this could either ease or strain water treatment capacity depending on the centre. Planning should account for possible population redistribution.

8. Institutional capacity. This document assumes that water treatment operators, council engineers, and relevant government agencies continue functioning. Under the extreme stress of the scenario, institutional capacity is a real uncertainty — not the baseline assumption, but a risk to monitor.

---

CROSS-REFERENCES

• Doc #8 — National Asset and Skills Census: Establishes actual chemical stocks, UV lamp inventory, existing electrochlorination equipment, water treatment workforce numbers.
• Doc #49 — Wastewater Treatment: Companion document covering the other half of the water cycle. Degraded wastewater treatment directly impacts drinking water source quality.
• Doc #65 — Hydro Grid Maintenance: Water treatment depends on grid electricity for pumping, UV systems, and electrolysis. Grid reliability is a prerequisite for most treatment methods described here.
• Doc #103 — Salt Production: Salt supply is the critical feedstock for domestic chlorine production. Coordination between water treatment chlorine demand and other salt uses (food preservation, industrial) is essential.
• Doc #1 — Resource Requisition Framework: Treatment chemical stocks may need to be requisitioned and centrally allocated if private or industrial stocks exist outside the municipal system.
• Doc #74 — Pastoral Farming Under Nuclear Winter: Agricultural land management affects catchment water quality. Stock exclusion from waterways reduces treatment requirements.
• Doc #7 — Fertiliser: Agricultural fertiliser use affects nitrogen and phosphorus loading in waterways, which affects treatment difficulty. Reduced fertiliser availability under import cutoff may actually improve some water source quality over time.

---

APPENDIX A: HOUSEHOLD WATER TREATMENT QUICK REFERENCE

If your municipal supply is still functioning: Continue using tap water as normal. Authorities will advise if treatment is compromised.

If you are advised to treat water at home:

1. Boiling: Bring water to a rolling boil for 1 minute. Allow to cool. Effective against all pathogens. Uses approximately 0.1 kWh per litre (electric jug).
2. If you have a UV unit: Ensure lamp is functioning (check indicator light). Replace lamp at rated interval. UV-treated water should be used promptly — no residual protection.
3. If you have household bleach (sodium hypochlorite): Add 2 drops of 5% bleach (or 1 drop of 10% bleach) per litre of clear water. Wait 30 minutes before drinking. Water should have a slight chlorine smell. If it does not, add another drop and wait 15 minutes. Note: bleach degrades over time — old bleach may be less effective.⁵⁵
4. Rainwater: Ensure roof and gutters are clean. Use first-flush diverter if available. Rainwater from a clean catchment is generally safe if stored in a covered tank. Consider disinfection (boiling or bleach) if immunocompromised people are using the water.

Long-term household options:

• Biosand filter construction: A concrete or bucket-based filter with layered sand and gravel can be built at home. Contact local council or civil defence for construction guides. Takes 2–4 weeks to mature before full effectiveness.

---

APPENDIX B: SODIUM HYPOCHLORITE PRODUCTION — SIMPLIFIED METHOD

For community or workshop-scale production. This method produces a dilute (~0.5%) hypochlorite solution suitable for water treatment.

Materials required: - Salt (sodium chloride) — food grade or coarse salt - Clean water - Two electrodes — graphite rods (from large batteries or industrial supply) or stainless steel plates - DC power supply — 12V battery, solar panel with charge controller, or grid-powered rectifier/battery charger - Non-metallic container (plastic bucket, glass jar) - Connecting wires

Procedure: 1. Dissolve salt in water at approximately 30g per litre (about 2 tablespoons per litre). Stir until dissolved. 2. Place both electrodes in the salt solution, spaced approximately 2–5 cm apart. Ensure they do not touch. 3. Connect electrodes to DC power supply. The electrode connected to the positive terminal (anode) will produce chlorine; the negative terminal (cathode) will produce hydrogen gas bubbles. 4. Run electrolysis for 30–60 minutes per litre of solution. The solution will develop a chlorine smell — this indicates hypochlorite formation. 5. Remove electrodes. The solution now contains sodium hypochlorite. 6. Test concentration if possible (pool test kits measure free chlorine). Target approximately 0.5–1% available chlorine for a stock solution. 7. Dosing: Add approximately 1–2 ml of this stock solution per litre of water to be treated (adjust based on testing — target 0.5 mg/L free chlorine residual after 30 minutes contact time).

Safety warnings: - Hydrogen gas is produced at the cathode. Ensure adequate ventilation — hydrogen is flammable and explosive in enclosed spaces. - Chlorine gas may be released from the anode, particularly at higher concentrations. Work in a ventilated area. - The solution is a mild oxidiser. Avoid contact with eyes. Store in a cool, dark place — hypochlorite degrades in heat and light. - Graphite electrodes gradually erode. Inspect and replace when significantly reduced in size. - This is a simplified method producing a dilute solution. Municipal-scale production requires purpose-built equipment with proper safety controls.

---

---

1. Taumata Arowai (NZ’s dedicated water services regulator, established 2021). National drinking water statistics. Approximately 82–85% of NZ’s population is served by registered drinking water supplies. https://www.taumataarowai.govt.nz/ — Exact figures depend on the definition of “registered supply” and population estimates used.↩︎

2. Dominion Salt Ltd. Lake Grassmere solar salt works, Marlborough. Production capacity figures are approximate based on industry reporting. https://www.dominionsalt.co.nz/ — Production is weather-dependent and varies year to year. The 60,000–80,000 tonne range represents typical good-weather production.↩︎

3. Christchurch City Council, Water Supply. Christchurch’s water is sourced from over 50 deep wells drawing from artesian aquifers in the Canterbury Plains. UV treatment and chlorination were introduced progressively from 2018 following the Havelock North Inquiry recommendations, though the underlying source quality has not changed. https://ccc.govt.nz/services/water-and-drainage/water-sup...↩︎

4. Slow sand filtration maturation times and operating parameters from: Huisman, L. and Wood, W.E., “Slow Sand Filtration,” WHO, 1974; also Visscher, J.T. et al., “Slow Sand Filtration: Planning, Design, Construction, Operation and Maintenance,” IRC International Water and Sanitation Centre, 1987. The 2–6 week range reflects typical conditions; colder temperatures extend maturation time.↩︎

5. Government Inquiry into Havelock North Drinking Water, Report of the Havelock North Drinking Water Inquiry: Stage 1, May 2017. https://www.dia.govt.nz/Government-Inquiry-into-Havelock-... — The inquiry found that approximately 5,500 people were estimated to have become ill. At least 3 deaths were directly attributed; a fourth probable. 45 hospitalisations confirmed. The contamination was caused by sheep faecal matter entering a bore water supply following heavy rainfall.↩︎

6. Person-year estimates for water treatment infrastructure are order-of-magnitude figures based on general civil engineering practice for concrete structures, earthworks, and mechanical/electrical installation. Actual labour requirements depend on site conditions, equipment availability, and the skill level of the available workforce. These figures should be refined with input from NZ water treatment engineers during the review process.↩︎

7. Government Inquiry into Havelock North Drinking Water, Report of the Havelock North Drinking Water Inquiry: Stage 1, May 2017. https://www.dia.govt.nz/Government-Inquiry-into-Havelock-... — The inquiry found that approximately 5,500 people were estimated to have become ill. At least 3 deaths were directly attributed; a fourth probable. 45 hospitalisations confirmed. The contamination was caused by sheep faecal matter entering a bore water supply following heavy rainfall.↩︎

8. Watercare Services Limited, Annual Report and operational data. Auckland’s average daily water demand is approximately 340–380 million litres per day, serving approximately 1.7 million people. https://www.watercare.co.nz/ — Demand varies seasonally.↩︎

9. Ministry of Health / Taumata Arowai, Register of Drinking Water Supplies. The exact number of registered supplies fluctuates; ~830 is an approximate figure for supplies serving 25+ people. Many smaller private supplies are not registered.↩︎

10. Watercare Services Limited, Annual Report and operational data. Auckland’s average daily water demand is approximately 340–380 million litres per day, serving approximately 1.7 million people. https://www.watercare.co.nz/ — Demand varies seasonally.↩︎

11. Wellington Water Ltd, Water Supply Information. https://www.wellingtonwater.co.nz/ — Wellington’s water sources include the Hutt River (Te Marua and Waterloo treatment plants), Wainuiomata River (Wainuiomata treatment plant), and artesian sources in the Hutt Valley (Waterloo wellfield).↩︎

12. Christchurch City Council, Water Supply. Christchurch’s water is sourced from over 50 deep wells drawing from artesian aquifers in the Canterbury Plains. UV treatment and chlorination were introduced progressively from 2018 following the Havelock North Inquiry recommendations, though the underlying source quality has not changed. https://ccc.govt.nz/services/water-and-drainage/water-sup...↩︎

13. Ixom (formerly Orica Watercare) is a major supplier of water treatment chemicals in NZ and Australia. Domestic production vs. import proportions for chlorine and hypochlorite in NZ are not publicly detailed and should be verified through the national asset census. https://www.ixom.com/↩︎

14. National chlorine consumption for water treatment is an estimate based on typical dosing rates (1–5 mg/L depending on source water quality) applied to total treated water volumes. This is an order-of-magnitude estimate; the national asset census should establish the actual figure. Assumes average ~2 mg/L across all treated supplies, with total treated water volume of approximately 1.5–2 billion litres per day nationally.↩︎

15. Ixom (formerly Orica Watercare) is a major supplier of water treatment chemicals in NZ and Australia. Domestic production vs. import proportions for chlorine and hypochlorite in NZ are not publicly detailed and should be verified through the national asset census. https://www.ixom.com/↩︎

16. Sodium hypochlorite degradation rates are well-documented in water treatment literature. See: Black & Veatch, “White’s Handbook of Chlorination and Alternative Disinfectants,” 5th edition, 2010. Degradation is temperature-dependent and accelerated by UV light, heat, and metal contamination.↩︎

17. Watercare Auckland coagulant consumption is estimated based on typical alum dosing rates for NZ surface water (20–80 mg/L depending on turbidity) applied to Auckland’s ~350 ML/day treatment volume. Exact figures should be verified with Watercare.↩︎

18. NZ limestone resources are described in: GNS Science, Mineral Resources of New Zealand. Major limestone deposits exist at Otorohanga (Waikato), Oamaru (Canterbury/Otago), Golden Bay, Whangarei, and numerous other locations. Lime production by calcination is an established NZ industry.↩︎

19. Lime (calcium oxide) is produced by calcining limestone (calcium carbonate) at 900–1000degC. The lower end of this range initiates decomposition; practical kiln operation typically targets 1000degC for complete conversion. Standard industrial chemistry; see: Boynton, R.S., “Chemistry and Technology of Lime and Limestone,” Wiley, 1980.↩︎

20. Fluoridation is a dental public health measure with no bearing on microbiological water safety. The Ministry of Health recommends fluoridation where practicable, but discontinuation under emergency conditions has no water safety consequence. See: Ministry of Health, “Guidelines for Drinking-water Quality Management for New Zealand.”↩︎

21. On-site generation (OSG) of sodium hypochlorite is a well-established technology. See: USEPA, “Alternative Disinfectants and Oxidants Guidance Manual,” EPA 815-R-99-014. Commercial OSG systems are manufactured by companies including MIOX, De Nora, and Prominent.↩︎

22. Dominion Salt Ltd. Lake Grassmere solar salt works, Marlborough. Production capacity figures are approximate based on industry reporting. https://www.dominionsalt.co.nz/ — Production is weather-dependent and varies year to year. The 60,000–80,000 tonne range represents typical good-weather production.↩︎

23. The nuclear winter impact on Marlborough’s climate is estimated based on the general 5–8degC cooling and reduced solar radiation parameters of the scenario. Marlborough’s current climate is NZ’s driest and sunniest region, which is why solar salt production is located there. The estimate of 30–60% production reduction is an assumption based on reduced evaporation rates; actual impact would depend on specific conditions. See also Doc #103 on salt production.↩︎

24. Stoichiometric calculation: the chlor-alkali reaction 2NaCl + 2H2O -> Cl2 + 2NaOH + H2 requires 2 moles NaCl (116.8g) to produce 1 mole Cl2 (71g). This gives a theoretical ratio of 1.64 kg NaCl per kg Cl2. Practical consumption is slightly higher due to inefficiencies: approximately 1.7–1.8 kg NaCl per kg Cl2 in industrial practice. Standard electrochemistry; see any chemical engineering reference.↩︎

25. Electrode materials for chlor-alkali cells: commercial DSA (dimensionally stable anodes) use titanium substrates with iridium oxide and/or ruthenium oxide coatings. These are specialty imported items. Graphite has been used as an anode material in chlor-alkali cells since the 19th century but erodes at approximately 1–3 kg per tonne of chlorine produced. See: O’Brien, T.F., Bommaraju, T.V., and Hine, F., “Handbook of Chlor-Alkali Technology,” Springer, 2005.↩︎

26. Electrode materials for chlor-alkali cells: commercial DSA (dimensionally stable anodes) use titanium substrates with iridium oxide and/or ruthenium oxide coatings. These are specialty imported items. Graphite has been used as an anode material in chlor-alkali cells since the 19th century but erodes at approximately 1–3 kg per tonne of chlorine produced. See: O’Brien, T.F., Bommaraju, T.V., and Hine, F., “Handbook of Chlor-Alkali Technology,” Springer, 2005.↩︎

27. Energy consumption for chlorine production by electrolysis: 3.0–3.5 kWh/kg Cl2 for modern membrane cells; up to 4.5 kWh/kg for less efficient diaphragm or mercury cells. Simple undivided cells for direct hypochlorite production have varying efficiency but are typically in the 4–8 kWh/kg range for available chlorine produced. See: Euro Chlor, “Chlor-Alkali Industry Review.”↩︎

28. NZ electricity generation capacity: approximately 9,500 MW installed capacity as of 2023, with approximately 82% from renewable sources (hydro ~57%, geothermal ~18%, wind ~7%). Source: MBIE, “Energy in New Zealand.” https://www.mbie.govt.nz/building-and-energy/energy-and-n...↩︎

29. UV disinfection effectiveness: well-established in water treatment literature. UV dose of 40 mJ/cm2 achieves >4-log (99.99%) inactivation of most bacteria and viruses, and >3-log inactivation of Cryptosporidium and Giardia. See: USEPA, “Ultraviolet Disinfection Guidance Manual for the Final Long Term 2 Enhanced Surface Water Treatment Rule,” 2006.↩︎

30. UV lamp lifetimes: standard low-pressure mercury vapour lamps are rated for 8,000–12,000 hours. Medium-pressure lamps have shorter lifetimes (4,000–8,000 hours) but higher intensity. Lamp output declines over time; the rated lifetime typically represents the point at which output drops below 80% of initial intensity. Manufacturer data from Trojan UV, Xylem, and others.↩︎

31. Turbidity interference with UV disinfection: particles shield pathogens from UV exposure and can harbor pathogens within their structure. Most UV system validations assume influent turbidity below 1 NTU. See: USEPA UV Guidance Manual (note 21).↩︎

32. WHO, “Guidelines for Drinking-water Quality,” 4th edition, 2011. Boiling for 1 minute at altitudes below 2,000m is sufficient to inactivate all common waterborne pathogens. NZ’s maximum altitude for inhabited areas is well below this threshold.↩︎

33. SODIS (Solar Water Disinfection): developed by EAWAG/SANDEC. Requires at least 6 hours of strong sunlight (or 2 days under cloudy conditions). UV-A radiation at 320–400 nm and thermal inactivation above 50degC are the active mechanisms. See: Wegelin, M. et al., “Solar Water Disinfection: Scope of the Process and Analysis of Radiation Experiments,” Journal of Water SRT — Aqua, 1994.↩︎

34. Ceramic water filters: see Lantagne, D., “Investigation of the Potters for Peace Colloidal Silver Impregnated Ceramic Filter,” USAID, 2001. Flow rates typically 1–3 L/hr. Bacterial removal >99%, protozoa removal >99.9%, virus removal variable (60–99% depending on design and silver content).↩︎

35. Biosand filters: see CAWST (Centre for Affordable Water and Sanitation Technology), “Biosand Filter Manual,” 2012. Well-documented community water treatment technology used globally. Bacterial removal 85–99%, protozoa removal >99%, virus removal 70–99% (improves with filter maturity).↩︎

36. Slow sand filtration maturation times and operating parameters from: Huisman, L. and Wood, W.E., “Slow Sand Filtration,” WHO, 1974; also Visscher, J.T. et al., “Slow Sand Filtration: Planning, Design, Construction, Operation and Maintenance,” IRC International Water and Sanitation Centre, 1987. The 2–6 week range reflects typical conditions; colder temperatures extend maturation time.↩︎

37. Slow sand filtration pathogen removal performance: summarised in Logsdon, G.S. et al., “Slow Sand Filtration for Small Water Systems,” Journal AWWA, 2002; and Hijnen, W.A.M. et al., “Elimination of micro-organisms by drinking water treatment processes,” IWA Publishing, 2004. Performance ranges given are typical for mature filters operating within design parameters.↩︎

38. Filter sand specifications for slow sand filtration: effective size 0.15–0.35 mm, uniformity coefficient <3 (ideally <2). See Huisman and Wood (note 4). NZ has extensive sand deposits but most natural sand requires washing and grading to meet these specifications. Construction aggregate suppliers typically have sieving capability.↩︎

39. Holcim (NZ) Ltd (formerly Golden Bay Cement) operates cement manufacturing at Portland (Whangarei). NZ cement production capacity is approximately 1.3–1.5 million tonnes per year. This is sufficient for domestic construction needs including water infrastructure. https://www.holcim.co.nz/↩︎

40. Turbidity limits for slow sand filtration: filter clogging becomes problematic above approximately 10–20 NTU. Some sources cite 50 NTU as an absolute maximum with frequent scraping. Roughing filtration pre-treatment extends the usable range. See: Galvis, G. et al., “Multi-Stage Filtration: An Innovative Water Treatment Technology,” IRC, 2006.↩︎

41. Cold-weather slow sand filtration performance: documented in: Logsdon, G.S., “Slow Sand Filtration,” ASCE, 1991; and operational experience from northern European and Canadian installations. Pathogen removal can decline by 0.5–1.0 log units at near-zero temperatures. Post-filtration disinfection compensates.↩︎

42. Watercare Services Limited, Annual Report and operational data. Auckland’s average daily water demand is approximately 340–380 million litres per day, serving approximately 1.7 million people. https://www.watercare.co.nz/ — Demand varies seasonally.↩︎

43. Roughing filtration: see Wegelin, M., “Surface Water Treatment by Roughing Filters,” SANDEC/EAWAG, 1996. Horizontal roughing filters using graded gravel (20mm to 4mm) can reduce turbidity from several hundred NTU to below 10 NTU with no chemical inputs.↩︎

44. Historical slow sand filtration: the Chelsea Water Company’s slow sand filter at Chelsea, London, began operation in 1829. John Snow’s 1854 investigation of the Broad Street cholera outbreak demonstrated the link between contaminated water and cholera. By the 1850s, London mandated filtration of all Thames water, decades before the germ theory was established. See: Huisman and Wood (note 4), and Baker, M.N., “The Quest for Pure Water,” AWWA, 1948.↩︎

45. NZ Aluminium Smelters Ltd (NZAS), Tiwai Point, Bluff. Operated by Rio Tinto. The smelter’s alumina stocks at any given time are uncertain but represent weeks to months of production input. Aluminium metal in stock and in the NZ economy (vehicles, building materials, etc.) represents a secondary aluminium source, but recycling aluminium into aluminium sulfate for water treatment is chemically possible but circuitous.↩︎

46. Ballance Agri-Nutrients, Mount Maunganui. Operates a sulfuric acid plant for fertiliser production. The plant processes imported elemental sulfur. Domestic sulfur sourcing from geothermal fields is discussed in: GNS Science, “Geothermal Energy in New Zealand.” Geothermal sulfur recovery is technically possible from Rotorua/Taupo geothermal areas but would require dedicated extraction infrastructure.↩︎

47. NZ Steel, Glenbrook. Produces approximately 600,000–650,000 tonnes of steel per year from Waikato ironsands (titanomagnetite). The ironsand resource is vast — billions of tonnes along the west coast of the North Island. Iron availability for ferric coagulant production is not a material constraint. https://www.nzsteel.co.nz/↩︎

48. Moringa oleifera as a natural coagulant: extensively studied. See Ndabigengesere, A. and Narasiah, K.S., “Quality of Water Treated by Coagulation Using Moringa oleifera Seeds,” Water Research, 1998. Effective for turbidity removal in tropical settings. However, Moringa is a tropical species requiring temperatures above 15degC (ideally 25–35degC) and is not frost-tolerant. Greenhouse cultivation in NZ under nuclear winter would compete with food production for limited heated growing space.↩︎

49. Natural coagulant research: see Yin, C.Y., “Emerging usage of plant-based coagulants for water and wastewater treatment,” Process Biochemistry, 2010. Tannin-based coagulants (from bark extracts) show some promise. NZ-specific research on native plant coagulants is limited; this is identified as a research gap.↩︎

50. Christchurch City Council, Water Supply. Christchurch’s water is sourced from over 50 deep wells drawing from artesian aquifers in the Canterbury Plains. UV treatment and chlorination were introduced progressively from 2018 following the Havelock North Inquiry recommendations, though the underlying source quality has not changed. https://ccc.govt.nz/services/water-and-drainage/water-sup...↩︎

51. Canterbury earthquake impacts on water supply: documented in Giovinazzi, S. et al., “Performance of the Water Supply System during the Canterbury Earthquake Sequence,” NZSEE Conference, 2012. Some bores showed increased turbidity and E. coli detection post-earthquake due to aquifer disturbance and bore casing damage.↩︎

52. Auckland groundwater: the Onehunga Springs source has historically supplied water to parts of Auckland. Other potential groundwater sources in the Auckland region exist but are generally small relative to the city’s demand. See: Auckland Council, “Auckland Water Strategy.”↩︎

53. ESR (Institute of Environmental Science and Research), “Drinking-water Assessment.” The ~15% figure for private supplies is an estimate; the actual proportion varies by region and is higher in rural areas. Some rural regions have >50% of the population on private supplies.↩︎

54. Atmospheric particulate from nuclear winter: the primary particles are soot (black carbon) from fires ignited by nuclear detonations. These particles are lofted into the stratosphere and descend over 1–5 years. Contamination of rainwater by soot particulate is plausible in the early period but has not been specifically modelled for NZ. First-flush diversion is standard rainwater harvesting practice for removing initial roof contaminants and would address most deposited particulate. See: Robock, A. et al. (note in Doc #74, [^5]).↩︎

55. Household bleach disinfection: WHO and CDC guidelines recommend 2 drops of 5–6% sodium hypochlorite per litre for clear water, doubled for cloudy water. Wait 30 minutes before consumption. Bleach should have a chlorine smell; if not, the bleach may be degraded. Household bleach is typically 3–6% sodium hypochlorite but concentration varies by brand and age. See: CDC, “Making Water Safe in an Emergency.”↩︎