Most water-treatment guides say "use all three methods together" — fine for a stocked retreat but useless when you are next to a creek with one filter, half a bottle of bleach, and a nearly empty camp stove. The real question is which method to use first given your threat, source, and actual resources. For detailed technique, see how to boil water safely — but read this first to understand when boiling is actually the right call.
This article breaks down each method by what it kills, what it misses, speed, and cost. For deeper context on the technical differences between filtration and purification ratings or the exact bleach dosing protocol for emergency bleach water purification , those articles go deep on each method. Here we map every method to the scenario where it wins — and where it fails.
The Core Tradeoff: Pathogen Kill Rate vs Operational Cost
Every method trades three things: kill completeness, cost, and storage safety. Boiling kills bacteria, viruses, and protozoa (including Cryptosporidium) but burns fuel and produces water too hot to drink immediately. Chemical disinfection is fast and cheap, leaving residual protection in sealed containers, but fails against Cryptosporidium and weakens in cold or turbid water. Mechanical filtration removes protozoa and bacteria in seconds with no heat, but lets most viruses pass through.
Understanding failure modes matters more than a "best" answer. The right method in a grid-down suburban setting may be wrong with glacial meltwater and no fuel. The CDC and EPA publish scenario-specific guidance for this reason — a one-size answer can be fatal [1][2].
Boiling — When It Wins, When It Doesn't
Boiling is the only method with universal pathogen kill. Bacteria die at 158°F (70°C); viruses at similar temps; Cryptosporidium (chlorine-resistant) at 140°F (60°C); Giardia likewise. By a rolling boil at 212°F (100°C), all biological pathogens are dead. CDC guidance: bring water to rolling boil for one minute, three minutes above 6,500 feet where lower boiling point reduces thermal kill [1].
Where boiling fails: chemical contamination. Heavy metals, pesticides, solvents — boiling concentrates these as water evaporates. Near agriculture or industry, boiling alone is insufficient. Also, fuel is finite: a 16 oz propane canister boils 40–50 liters. At two liters per person daily, that is one week for one person; a six-person month-long scenario burns significant fuel on water. Boiling works best when fuel is available, the source is biologically suspect but chemically clean, and you treat for immediate consumption.
Filtration — What It Removes, What It Misses
Mechanical filtration traps particles through ceramic, hollow-fiber, or activated-carbon membranes. Pore size in microns is key. Giardia and Cryptosporidium average 4–15 and 4–6 microns; bacteria 0.2–10 microns. A 0.2-micron absolute filter removes both. Viruses average 0.02–0.3 microns — small enough to pass most portable filters rated 0.1–0.2 microns because ratings are nominal, not absolute. EPA purifier standards require 4-log (99.99%) virus reduction plus bacteria and protozoa removal [2]. A device removing only protozoa and bacteria is a filter, not a purifier.
Filtration excels at speed and convenience: gravity filters produce liters per hour hands-free; squeeze filters work in seconds; no heat or chemicals. Activated carbon removes chlorine taste and some pesticides. Filtration is the first step for turbid sources because sediment exhausts chlorine and blocks UV. Where it fails: clogging in silt (pre-filter with bandana to extend cartridge life), freezing (ice cracks media undetectably), and viral threats in high-risk areas.
Chemical Disinfection — Chlorine and Iodine
Chemical disinfection oxidizes pathogen cell membranes. Chlorine — as bleach, sodium hypochlorite, or calcium hypochlorite granules — kills bacteria and viruses in 30 minutes at 60°F+ (15°C+). EPA guidance: 8 drops of 6% bleach per gallon of clear water, doubled if cloudy [2]. Critically, standard chlorine doses don't reliably kill Cryptosporidium. EPA guidance is explicit: if Crypto is a concern, boil or filter instead.
Iodine works in 30–60 minutes, pockets easily, shares chlorine's Crypto limitation, and raises thyroid concerns (limit to weeks; avoid if thyroid-compromised). Chemical disinfection's big advantage: residual protection. Sealed chlorine-treated water suppresses regrowth for months; FEMA says six months is safe [3]. Boiled water recontaminates on air exposure; filtered water has no residual.
Decision Framework — Which Method First
- Identify your water source type
Municipal tap water that is flowing normally: start with chlorine to extend shelf life for storage. Tap water during a flood or pipe-damage event: treat as unknown biological source — filter or boil, do not trust the tap. Surface water (creek, lake, pond): assume protozoa and bacteria; assume viral risk if near any human habitation or agricultural land. Rainwater collected in a clean container: low pathogen risk but add chlorine for storage residual.
- Assess your Cryptosporidium risk
If your source is downstream of agricultural operations, sewage overflow, or known Crypto outbreaks in the area, Cryptosporidium is a real threat. In that case, eliminate chlorine and iodine as your primary method — they do not kill Crypto at standard doses. Your options are boiling, a mechanical filter that physically removes oocysts (0.1–0.2 micron rated), or UV at NSF/ANSI 55 Class A doses. If Crypto risk is low (high-elevation pristine source, treated municipal water), chemical disinfection remains on the table.
- Assess your viral risk
In North America, backcountry surface water carries low viral risk. In any post-flooding scenario, urban disaster, or international environment with uncertain sewage infrastructure, assume elevated viral risk. If viral risk is elevated, a standard hollow-fiber filter is not enough on its own — pair it with chemical disinfection or UV, or boil. If viral risk is low, a hollow-fiber filter alone covers the protozoa and bacteria threat.
- Check your fuel and battery inventory
Boiling is only viable if you have fuel to burn. UV is only viable if you have batteries or power. If both are scarce or rationed, chemical disinfection becomes your primary workhorse — it is the only method that requires no energy input beyond stirring. Plan your fuel budget: at approximately 40–50 liters per 16 oz propane canister for boiling, a 6-person group needs roughly 3 canisters per week on water alone. If you cannot sustain that supply, build chemical disinfection into your primary protocol.
- Check your water turbidity
If the source water is visibly murky, silty, or carries significant sediment, pre-treatment is not optional — it is required before chemical or UV treatment. Run the water through a mechanical filter, or improvise a pre-filter with layered cloth, sand, and charcoal. Turbidity shields pathogens from chlorine contact and UV exposure. A clear-looking source after pre-filtering improves the effectiveness of every downstream treatment method.
- Factor in storage requirements
If you are treating water for immediate consumption, any method works. If you are treating water for storage in sealed containers, only chemical disinfection (chlorine specifically) provides residual protection that suppresses regrowth over days and weeks. Boiled or filtered water with no chemical residual should be stored sealed and consumed within 24 hours if possible, or re-treated before longer-term storage. For rotation-stock water stored 3–6 months, FEMA recommends treating with chlorine and sealing in food-grade containers [3].
- Layer when the threat exceeds any single method
If your source has both a Cryptosporidium risk and a viral risk (flood scenario, sewage overflow area), no single method covers both gaps — chlorine misses Crypto, and most filters miss viruses. Layer: filter first to remove protozoa, sediment, and bacteria, then add chlorine to kill viruses and provide storage residual. If your UV device is available and the source water is clear post-filtration, UV instead of or in addition to chlorine provides the broadest kill spectrum without chemical taste.
When to Layer Methods
Layering addresses gaps, not insurance. Best combinations: filter-then-disinfect and pre-filter-then-UV. Filter-then-disinfect handles floods and urban disasters: run source through a 0.1-micron hollow-fiber filter (removes turbidity, Giardia, Crypto, bacteria), then add chlorine. The filter removes particles that shield viruses and waste chlorine; chlorine kills viruses that passed. Result: safe, sealed, with residual protection. FEMA assumes this approach for month-long water stockpiles [3].
Pre-filter-then-UV handles turbid water before UV: run through any coarse filter (bandana over coffee filter) first. UV needs clear water — turbidity above rated NTU lets pathogens hide in shadows. Pre-filtering takes 60 seconds and prevents false confidence. Filter-then-boil is rarely needed; if boiling, the filter adds nothing for safety. Use pre-filtration before boiling only to cut sediment and taste, not for safety.
Frequently Asked Questions
Does boiling water remove chemicals and heavy metals?
No. Boiling kills biological pathogens — bacteria, viruses, protozoa — but it cannot remove dissolved chemicals, heavy metals, nitrates, or pesticides. Boiling actually concentrates these contaminants slightly as water evaporates. If your source water may contain chemical contamination from agricultural runoff or industrial proximity, use an activated-carbon filter rated for chemical reduction, or source water from a different location entirely.
How much bleach do I add to make water safe to drink?
The EPA recommends 8 drops of unscented liquid chlorine bleach at 6–8.25% sodium hypochlorite per gallon of clear water, doubled to 16 drops for cloudy water. Stir and let stand 30 minutes. The water should have a faint chlorine smell — if it does not, repeat the dose and wait 15 more minutes. Bleach stored over a year loses potency; double the dose regardless of water clarity.
Can I use a Brita-style pitcher filter to purify emergency water?
No. Standard pour-through pitcher filters using activated carbon are designed for taste and odor improvement of already-safe municipal water, not pathogen removal. In an emergency with biologically unsafe water, a Brita filter will not make it safe. You need a filter rated for bacteria and protozoa removal to NSF Standard 244 or the EPA microbiological purifier standard, or one of the other treatment methods in this article.
Does boiling or chlorine kill Cryptosporidium?
Boiling does — Cryptosporidium oocysts are inactivated at 140°F (60°C), well below the boiling point, so a rolling boil for one minute guarantees kill. Standard chlorine doses used for emergency disinfection do not reliably kill Cryptosporidium; the EPA emergency guidance states this explicitly. UV light at NSF/ANSI 55 Class A doses also kills Crypto. A 0.1-micron filter physically removes the oocysts. Iodine shares the same limitation as chlorine.
How long does chemically treated water stay safe in storage?
Properly treated water stored in clean, sealed food-grade containers in a cool dark location remains safe for up to six months per FEMA guidance. After six months the chlorine residual diminishes enough that you should re-treat or rotate the supply. Use opaque containers to slow chlorine degradation, and store away from gasoline, pesticides, or other volatile chemicals that can permeate plastic over time.