Water, Done Right

Longevity Water · Montevideo · OSE tap

The best drinking water you can build in Uruguay.

A science-first guide, led by toxicology and water chemistry rather than influencers, to filtering OSE Montevideo tap water for a health-obsessed home. What is actually in the water, what removes it, and exactly which system to buy or import, with filters you can restock locally.

Verdict: reverse osmosis plus remineralization, bought as insurance, not because your tap demands it.
1 documented exceedance (THMs) 11-agent research swarm 5-agent adversarial verification 3 tiers · buy local or import
TL;DR

The short answer

If you read nothing else: your tap's only proven problem is a chlorination byproduct that carbon removes. Everything past that is you buying insurance and control, which, given your brief (remove all four contaminant classes, longevity-optimal, no budget cap), is a reasonable thing to buy. Here is the call.

Good · buy local now

Local tank RO + UV

US$200–330 installed, local warranty
  • 7 to 8-stage RO with remineralization from Osmoaqua, Carrasco Import or Green Pure
  • Filters and service bought in Montevideo, zero import friction
  • No third-party certification (generic units)
For "I want clean water handled this month."
My pick for you Better · certified + local filters

iSpring RCC7AK

~US$300 + import · ~US$95–110/yr
  • NSF/ANSI 58 certified, 6-stage with alkaline remineralization
  • Universal filters restocked on Mercado Libre for US$7–15; membranes US$25–41
  • Auto-ranging voltage, no transformer needed
Best all-in: real certification, longevity-grade, and you never depend on a US supply chain for filters.
Best · peak performance

Waterdrop G3P800 / AquaTru

Waterdrop ~US$850 · AquaTru ~US$450 · ~US$105–160/yr
  • Tankless, confirmed 100–240V (imports cleanly), ~99% PFOA/PFOS reduction, smart TDS readout
  • AquaTru is the certification leader (83 contaminants, PFAS-certified) if you prefer countertop, no plumbing
  • Proprietary filters, stockpile a 1 to 2 year supply on import
Peak performance, with one honest trade-off: a proprietary filter supply you have to manage.

The honest baseline

A premium NSF/ANSI 53 carbon block (~US$130–200) fixes the one thing your water demonstrably has wrong (trihalomethanes) plus chlorine, lead and long-chain PFAS. If simplicity mattered more than completeness, that is genuinely enough. You are stepping up to RO to also cover short-chain PFAS, nanoplastics and a future salinity event, none of which OSE currently measures in your supply. That is a defensible insurance purchase, not a fix the water is crying out for.

01

What's actually in your OSE water

Montevideo runs almost entirely on surface water from the Santa Lucía river system (Paso Severino to Aguas Corrientes), not groundwater. That single fact decides most of this: it makes arsenic a non-issue and creates the chlorination-byproduct problem that is the real story here.

1.1–2.2×
THM levels vs the legal limit at Montevideo network points, Jan to Feb 2026
<1 mg/L
Fluoride, water is not fluoridated; Uruguay fluoridates salt instead
30–40
Typical sodium mg/L in a normal year, soft, low-mineral water
0
Public PFAS measurements for Montevideo tap, a genuine data gap

The threat model, contaminant by contaminant

ContaminantIn your OSE water?Why it mattersWhat removes it
Trihalomethanes (chlorination byproducts)Real & documentedThe only recurring exceedance. In the 2026 spike, drought forced OSE to draw more downstream, higher-bromide water, and chlorinating it produced brominated THMs. Dose-dependent bladder-cancer association on chronic exposure.Activated carbon (plus RO on the brominated ones)
Chlorine + taste/odorRoutine residualAn intended disinfectant residual, not an exceedance; the main day-to-day taste complaint, not a health risk at these levels.Activated carbon
LeadMaybe, at the tapNot in the network; leaches from legacy lead fixture joints in older Carrasco homes. Highest toxicity if present.NSF 53 carbon or RO · test at tap
PFAS "forever chemicals"Unknown / unmonitoredNo Uruguay tap data exists. Precautionary target for a longevity build; surface supply is not the classic high-PFAS scenario.RO (incl. short-chain); carbon erratic
Microplastics / nanoplasticsPresumed, globalEmerging cardiovascular signal (NEJM 2024), association not proven cause; cheap to filter regardless.RO (nano); sub-micron carbon (micro)
FluorideNot a concern hereLow and not added. Overexposure in Uruguay is an interior-groundwater story, not Montevideo.RO, if ever present
ArsenicNon-issueA groundwater/well problem in the interior; your surface supply sidesteps it.RO, if ever present
NitrateCompliantFine in Montevideo; exceedances are in specific interior towns.RO / ion exchange
Salinity (sodium/chloride)Drought risk onlyNormally low; the 2023 crisis forced emergency limits (Na 200 to 440). Only RO helps in a repeat.Reverse osmosis

Uruguay's standard is UNIT 833:2008 (Decreto 375/011). Its arsenic limit (20 µg/L) is twice WHO's 10 µg/L, and its lead limit (30 µg/L) is three times WHO's. Note the 30 µg/L lead figure was a transitional 2011 value scheduled to converge to WHO's 10 µg/L by 2021, so treat it as a ceiling, not today's aim.

The one real event to know: the 2023 salinity crisis

A multi-year drought dropped Paso Severino to roughly 1.6% of capacity (about 1.1 of 67 million m³), forcing OSE to blend brackish estuary water. The Ministry of Health temporarily raised sodium limits 200 to 440 mg/L and chloride 250 to 720 mg/L, and advised pregnant women and people with kidney, heart or liver disease to avoid the tap. It normalized after rains. RO is the only filtration that would help in a repeat, which is a genuine, Montevideo-specific reason it earns its place.

Out of scope, but worth one line

THMs are volatile, so a hot shower is a separate inhalation and skin-contact exposure that a drinking filter never touches. You scoped this to the kitchen, which is right for the health payoff, but if you ever extend, a shower carbon filter is where the next marginal THM reduction lives.

02

What actually matters for health

We weighted the evidence rather than the marketing. Ranked for your specific situation, a chlorinated surface supply with documented THM spikes, the priority order is clear, and it is not the order the filter industry sells.

Evidence-weighted priority (for OSE water)

  1. Disinfection byproducts (THMs): the only documented exceedance; a dose-dependent bladder-cancer association. Carbon's job.
  2. Lead from plumbing: highest per-unit toxicity, no safe level; depends on your home, so test.
  3. Arsenic / nitrate: potent but effectively absent in Montevideo's surface supply.
  4. PFAS: serious toxicology, but unmeasured locally; precautionary only.
  5. Fluoride: a real IQ signal above 1.5 mg/L, but your water is low and unfluoridated.
  6. Microplastics: alarming headlines, immature causal evidence, but cheap to filter.

What the experts actually do

Bryan Johnson (Blueprint): multi-stage RO plus deionization plus coral-calcium remineralization to pH ~8; drops TDS ~300 to 25 then re-mineralizes; tests his water monthly via SimpleLab. Targets PFAS, microplastics, metals, VOCs. He does not campaign on fluoride.

HoWon Noh: RO is the gold standard; personally uses AquaTru plus a shower filter. His sharp counter-point, which the evidence backs: don't fear remineralization or over-obsess over water minerals, your diet supplies far more calcium and magnesium than water ever will.

Where they and the water-science literature (WHO, EPA, EWG, the Duke PFAS study, the NEJM microplastics paper) agree: RO for full-spectrum removal, always remineralize, and verify by testing.

Two "expert endorsements" we caught as fabricated

During verification, our agents flagged two claims that circulate as fact and dissolve on inspection: "Cloud RO, rated best under-sink RO by Wirecutter" (Wirecutter has never published an RO pick; its actual pick is a carbon filter, and this line lives only on the vendor's own affiliate pages), and the "Bryan Johnson Blueprint Water Purification System" (a third-party paid press release, not a real Blueprint product). If an AI or a blog repeats either, it did not check the source. This is why the whole guide was adversarially verified.

03

The technology, and why RO for your spec

No single technology removes every target class, so the answer is a multi-stage stack. But the honest nuance below matters: the stage that fixes your documented problem is mostly the carbon, not the famous membrane.

TechnologyTHMsFluoridePFASLead / metalsMicroplastics
Reverse osmosisCarbon + membraneYes ~85–95%+Yes, all chain lengthsYesYes, incl. nano
Activated / catalytic carbon blockYes (limited capacity)NoLong-chain only, erraticLead yes (NSF 53)Micro if sub-micron
DistillationPartial (volatiles carry over)Yes >99%Yes ~99%YesYes
UVNoNoNoNoNo
Ion exchange / softenerNoOnly special resinOnly PFAS-selective resinSoftener: noNo
KDF mediaPartialNoNoYes (metals)No
UltrafiltrationNoNoNoNoMicro yes, nano weaker
Activated alumina / bone charNoYes 70–98%NoSomeNo

Fluoride blind spot: plain carbon, UV, KDF, ultrafiltration and softeners do not remove fluoride. Only RO, distillation or a dedicated fluoride medium does. This is the most common gap in "5-stage" filters sold as complete, though for low-fluoride Montevideo it happens not to bite.

The nuance that reframes everything

A single RO membrane only partly rejects the smallest THM, chloroform (it is a small, uncharged, volatile molecule), so the carbon stage is the reliable THM barrier in any home RO system. The membrane does reject the larger brominated THMs well, and those were the species that spiked in 2026, but either way you are paying for a carbon stage to be safe.

What RO uniquely adds for Montevideo: short-chain PFAS and nanoplastics (both unmeasured here), salinity insurance (a real local phenomenon), and the ability to strip the water to a blank slate and remineralize it to a chosen profile. Given your explicit "remove all four, no budget cap" brief, that is worth buying. Just know it as insurance and future-proofing, not remediation, and note the two real costs: it wastes 2 to 3× the water it makes, and it strips already-soft water that you must then remineralize.

The required stack

  • 1 · Sediment pre-filter (5 µm): protects everything downstream, catches particulates and larger microplastics.
  • 2 · Catalytic carbon block: removes chlorine and THMs and shields the membrane (chlorine destroys RO membranes). This is the workhorse for your actual water.
  • 3 · Reverse osmosis membrane: the full-spectrum barrier for PFAS (all chain lengths), lead, nitrate, salinity and microplastics.
  • 4 · Post-carbon polish: final taste and residual VOCs after the tank.
  • 5 · Remineralization: non-negotiable; see the next section.
  • 6 · UV (optional): redundant for municipal supply with an intact membrane; add only if you want microbial belt-and-braces.
04

Remineralization, the longevity piece

Do not drink bare RO water long-term. The strongest reason is not nutrition, it is chemistry: demineralized water is aggressive and leaches lead and copper out of plumbing. WHO's own assessment (Kozisek) says so plainly.

Target profile

Magnesium20–30 mg/L
Calcium40–80 mg/L
TDS≥100, toward 200–300
Ca : Mg ratio≤ ~2 : 1
pH7.2–7.8

Mg/Ca targets are WHO-published expert assessment (Kozisek), not a binding standard. The pH and Ca:Mg targets come from corrosion-control and nutrition literature. Treat them as sensible aims, not law.

How to actually hit it

  • Inline calcite plus a touch of Corosex (MgO) after the membrane, for automatic calcium, pH lift and anti-corrosion buffering.
  • A separate magnesium source, because calcite adds mostly calcium. ConcenTrace ionic drops at ~8 drops/L land magnesium near the ~28 mg/L target; they add no calcium, so pair them with the calcite stage. They are low-sodium but add chloride and have shown a lithium flag in independent testing, so dose conservatively and buy a third-party-tested batch.
  • Verify with a cheap TDS/hardness check, since cartridge output drifts and a precise profile is not realistic without measuring.

The balanced verdict (both camps are half-right)

Remineralize primarily for corrosion control, that part is non-negotiable plumbing chemistry. The nutritional benefit is real but modest, and HoWon Noh is correct that your diet, not your water, supplies most of your calcium and magnesium (water is under ~10% of daily intake). So add minerals and a magnesium hedge, but don't pretend a home cartridge delivers a lab-precise health profile, and don't neglect your plate.

One longevity-specific footnote: silica

RO also strips silica, the one water constituent with a genuine longevity rationale: silicon-rich water (>30 mg/L Si) modestly raises urinary aluminum excretion in small human studies. The effect is real but small and diet-replaceable, so don't buy a system for it, but an occasional bottle of high-silica mineral water is a cheap hedge if aluminum is on your radar.

05

The systems, ranked

Independent lab-testers (Water Filter Guru, Quality Water Lab) converge on a clear order. We weighted your two stated priorities heavily: filter availability in Uruguay, and genuine third-party certification.

SystemTypeCertificationRemineralizeFilters~Annual filter cost
iSpring RCC7AKTank, 6-stageNSF/ANSI 58Yes, alkalineUniversalUS$95–110
APEC RO-PH90Tank, 6-stageWQA Gold Seal (NSF 58)Yes, calciteUniversalUS$75–100
Waterdrop G3P800TanklessNSF 42/53/58/372Add-onProprietaryUS$160
AquaTruCountertopNSF 42/53/58/401 + PFASAlkaline variantProprietaryUS$100–110
Home Master TMAFCTank, 7-stageNone (system)Yes, full-contactProprietaryUS$75–99
Local UY tank ROTank + UVNoneYesLocal shopsUS$35–55/kit

Certification note: these RO brands are absent from NSF's own database only because they certify via IAPMO R&T or WQA to the same ANSI standards, that is not a red flag. AquaTru is the one countertop RO also certified for PFAS reduction. The genuinely uncertified system is Home Master (materials tested, system not), which is why it drops despite its excellent remineralization.

Why iSpring wins for you

Certified, remineralized, non-electric or auto-ranging, and its universal 10-inch cartridges plus a standard 1812 membrane are on Mercado Libre for a few dollars. You never depend on an overseas filter supply. The trade-off is a storage tank (mild TDS creep) and 75 GPD, minor for a two-person home.

When to go AquaTru / Waterdrop

AquaTru is the certification and lab-test leader (83 contaminants, ~100% of detected health-effect contaminants removed, PFAS-certified) and needs no plumbing. Waterdrop G3P800 is the best tankless and imports cleanly on 100–240V. Both use proprietary filters, so buy a 1 to 2 year supply on import and reorder ahead.

What we ruled out

Home Master (no system certification), Frizzlife (no NSF), and any "5-stage" carbon unit sold as complete. Also: Wirecutter's real pick is a carbon filter (Aquasana AQ-5200), and its water expert is openly skeptical of RO for municipal water, useful context, not a reason to skip RO given your brief.

06

Buying in Uruguay, local vs import

The core dependency of the whole plan, that you can rebuy filters here, is verified solid. Standard cartridges and membranes are abundant and cheap on Mercado Libre Uruguay.

Buy local (fastest, guaranteed resupply)

  • Osmoaqua, Montevideo, +598 2902 4030; under-sink RO from ~US$200, filter and membrane resupply section.
  • Carrasco Import, 7-stage US$269, 8-stage + UV US$335; maintenance service $499 UYU (3-filter, semi-annual) or $3,490 UYU (8-filter, annual).
  • Green Pure, RO equipment, filters and spares.

Premium local systems run about US$200–330 from these specialists (cheaper generic units also exist on Mercado Libre). No NSF certification, but the same carbon-plus-RO-plus-remineralization architecture.

Filter resupply, verified on Mercado Libre

10" carbon/sediment cartridgeUS$7–15
6-month cartridge kitUS$34–55
1812 membrane (75–100 GPD)US$25–41
Full 5-stage kit + membraneUS$49–84

1812-format 75 GPD membranes sell under Membrane Solutions, Greenfilter and compatible brands. That standard size is why the universal-cartridge iSpring/APEC route is safe.

Check your water pressure before you choose a tier

Many Montevideo buildings and upper-floor apartments run on a rooftop gravity tank, which can deliver well under the ~40 psi an unpumped tank RO needs to work properly. Measure static pressure first with a cheap gauge on a tap. If it is low, either add a booster/permeate pump, or choose a pumped tankless unit (Waterdrop, AquaTru run their own pump and are pressure-independent). This is often the real deciding factor between the local/iSpring tank tiers and a pumped unit.

Importing a premium unit

The two customs routes (post 1 May 2026)

Franquicia: up to 3 shipments/year, US$800/year total cap. A shipment whose goods originate in the US (proven by the invoice) pays no IVA up to US$200; above that, IVA applies.

Simplified 60%: a flat 60% of invoice value per shipment (US$20 minimum), up to US$800 and 20 kg, no customs broker. Best for a whole ~US$300 unit in one box.

Above US$800 you need a DUA and a despachante. Since 1 May 2026 the courier, not you, pays the tax to Aduana.

Voltage & plugs (220V / 50Hz)

  • Waterdrop G3P800 and iSpring pumped units auto-range 100–240V, only a plug adapter is needed.
  • Non-pumped tank RO (iSpring RCC7-series, APEC ROES) needs no power at all, just ~40+ psi feed pressure.
  • Only APEC's specific RO-PUMP-120V forces a step-down transformer, avoid that SKU.
  • Every US unit needs a physical Type A/B to Type C/F/L plug adapter regardless.

Bottom line on buying

Import the iSpring RCC7AK under the simplified 60% regime, then forget the US supply chain, its filters cost a few dollars here. Or buy a local tank RO today if you want it handled this week, when you call, ask specifically for a "6 or 7-stage under-sink RO with standard 10-inch cartridges, a 50 to 100 GPD membrane, a pressurized tank, and an added calcite remineralization stage," and confirm the cartridges are the universal size so you are not locked into proprietary refills locally either. Reserve the proprietary-filter systems (Waterdrop, AquaTru) for when you specifically want tankless or top-tier certification and will stockpile filters.

07

Install, upkeep & the cost calculator

Under-sink RO is a half-day plumber job (cold-water tee, drain saddle, dedicated faucet). Filters are a tool-free swap. Here is what it actually costs to run, and how it compares to bottled water.

Maintenance rhythm

  • Sediment + carbon pre-filters, every 6 to 12 months (sooner drives THM protection, replace on schedule, not on taste).
  • RO membrane, every 2 to 5 years.
  • Post-carbon + remineralization, every 6 to 12 months.
  • Turn the TDS pen into a test: rejection = (1 minus product/feed TDS). A new membrane rejects ~95 to 98%; when it drops below ~85% or product TDS climbs steadily, the membrane is failing, replace it.
  • Tank hygiene, post-carbon water has no chlorine residual, so a storage tank can slowly grow biofilm; flush the first glass after long idle periods and sanitize the tank yearly. One more quiet point for tankless.

Two honest running costs

Water waste. Legacy RO wastes 3 to 4 L per filtered litre; good modern units are ~2:1 and the best tankless ~1:1. In a supply with a recent drought history, pick a low-waste unit (Waterdrop, or add a permeate pump to a tank system).

Store and drink from glass, not plastic. RO strips nanoplastics upstream, but a plastic pitcher or PET bottle re-introduces them at the last step; keep a glass carafe at the tap. Tank systems also store water against a butyl bladder, another small reason a tankless unit or a first-draw flush suits the plastics-conscious.

Cost calculator

Estimate a 5-year cost of ownership and compare it to bottled water.

1 L12 L
Filtered water, 5-yr total
Cost per litre
Bottled would cost

Upfront figures are the unit's US retail price. For an imported unit, add the ~60% simplified-regime import charge plus a plumber's install, see section 06.

08

Test your water first

Before you spend, spend a little to know your actual numbers, especially at-tap lead (a home-specific risk) and, if you care, PFAS (the local unknown). This is a phone-and-visit errand in Montevideo, not a click.

Now · ~US$15–30

A TDS/EC pen plus chlorine/hardness strips from Mercado Libre for a same-day baseline and to verify a filter later. Honest limit: this tells you nothing about lead, PFAS, fluoride or THMs, it only reads bulk dissolved solids.

This week · lab panel

Call an accredited Montevideo lab for a targeted UNIT-833 subset (metals incl. lead/arsenic, nitrate, fluoride, hardness, THMs, microbiological):

  • Ecotech · +598 2403 2130
  • Cristar Zerbi · +598 2900 7505
  • ALTIX · +598 2208 6700 (backup)

No lab publishes prices, ask for a quote. Order a subset, not the full panel.

PFAS · ~US$85

Cyclopure mail-in kit, it ships a sorbent disc rather than a water sample, so it dodges liquid-customs problems and has documented Latin-America use. Email them to place an international order. Beats Tap Score (US$335, ships liquid, customs-risky). No local lab offers PFAS, and UNIT-833 doesn't cover it.

Skip microplastics testing, there is no dependable consumer test anywhere; just filter for them.

REF

Sources & method

How this was built

11 parallel research agents covered OSE water quality, toxicology, filtration technology, the top systems, remineralization, PFAS/microplastics/fluoride removal, testing, Uruguay availability and import, running costs, and the Bryan Johnson and HoWon Noh protocols.

10 adversarial verification agents, in two passes, then attacked the load-bearing claims: is RO overkill for this water, are the certifications real, are universal filters truly buyable in Uruguay, are the remineralization targets sound, and a full re-audit of every number, price and phone on the page.

That verification is what caught the two fabricated endorsements, corrected the customs, THM-chemistry and lead-limit details, and confirmed the filter-supply linchpin, rather than accepting the first plausible answer.

Primary sources (selected)

  • OSE, Trihalometanos + Calidad de agua; Observatorio OSE
  • URSEA muestreos, Montevideo/área metropolitana
  • UNIT 833:2008 · Decreto 375/011
  • El Observador, THM exceedances 2026 (OSE causes)
  • WHO / Kozisek, Health risks from demineralised water (2005); Calcium & Magnesium in drinking water (2009)
  • US EPA, PFAS treatment tech + POU/POE effectiveness; WaterSense RO
  • Herkert et al. 2020, ES&T Letters (Duke/NC State PFAS filter study)
  • Marfella et al. 2024, NEJM (microplastics & cardiovascular risk)
  • NTP 2024 fluoride monograph; JAMA Pediatrics 2025 meta-analysis
  • NSF/ANSI 53 & 58 standards; IAPMO R&T & WQA listings
  • Blueprint (Bryan Johnson) protocol; howonnoh.com
  • Water Filter Guru & Quality Water Lab (independent lab tests)
  • Aduana Uruguay (DNA), MEF franquicias; Decreto 50/026
  • Mercado Libre Uruguay listings (filters, membranes, meters)
  • Osmoaqua, Carrasco Import, Green Pure; Ecotech, Cristar Zerbi, ALTIX; Cyclopure
How to read this: figures are current as of mid-2026 and were adversarially verified, but water quality varies by season and by home. Confirm your own numbers with a test, and confirm live prices and the exact NSF listing of any specific model before buying. This is a research synthesis, not professional water-safety or medical advice.