Backpacking Water Filter vs Purifier: What Each Removes

More than a million people a year.

That is how many Americans get sick from Giardia, according to the CDC, which adds that more people in this country get sick from it than from any other parasite living in the gut. It is also the number that sells every backpacking water filter on the shelf, and for once the number and the product line up. A Giardia cyst measures 8 by 19 microns. The hollow-fiber membrane in an ordinary filter has pores around 0.1 micron. The cyst is two hundred times too wide to pass, and it does not.

Here is the figure the packaging leaves out. An enteric virus — norovirus, for instance — averages 0.03 micron. Three times smaller than the pore it is supposed to be caught by. It goes straight through, on every pass, in every brand.

Those two numbers are the whole subject. Everything below is a consequence of them.

Filter or purifier: the whole difference is a virus

Two words get used as if they were interchangeable in catalogs, and they are not. The distinction is not a matter of quality, or price, or how serious the manufacturer is. It is one row in a table.

A filter is mechanical. Water goes through a barrier with holes of a stated size, and anything wider than the hole stays behind. The CDC’s page for hikers and campers gives two thresholds worth writing on the back of your hand: a filter with an absolute pore size of 1 micron or smaller removes parasites but not viruses or all bacteria, and a filter at 0.3 micron or smaller removes bacteria and parasites, but not viruses. Both sentences end the same way.

A purifier is whatever also closes the virus row. That can be a far finer membrane — the same CDC page notes that reverse osmosis filters remove parasites, bacteria and viruses — or it can be chemistry, ultraviolet light, or a pot on a stove. The mechanism is not the point. The coverage is.

Which turns a shopping problem into a reading problem. A product calling itself a purifier without naming a pore size or a test standard has told you nothing; one calling itself a filter has told you the truth about its single limit, which is more useful. If your trip genuinely needs the virus row closed — a river below a town, water abroad, a flooded municipal system — then a purifier-grade filter that states virus removal^aff. is a different purchase from the one most hikers need, and it should say so on the box.

The rest of the time, the honest answer is that American backcountry water is a parasite problem. The CDC’s Cryptosporidium page calls cryptosporidiosis the leading cause of waterborne disease in the United States, with symptoms starting two to ten days after infection and lasting one to two weeks. That is what a membrane is for.

What a backpacking water filter removes, threat by threat

Six things can be in the water you are about to drink, and six families of treatment exist to deal with them. No single pairing is complete, which is why a ranked list of “best filters” is the wrong shape for this decision.

The sizes come from the CDC’s travel-medicine chapter on water disinfection, and they are the only specifications here that will still be accurate in ten years: viruses 0.03 micron, E. coli 0.5 by 2 to 8, Cryptosporidium oocysts 4 to 6, Giardia cysts 8 by 19. Organisms do not get redesigned every season. Products do.

The grid below is built on those sizes. Pick the trip you are packing for and the rows that do not apply go quiet, so each technology is scored against the threats you will meet.

Threat coverage grid — six threats, six technologies

Pick the trip. The grid decides which rows count.

Nothing here is a product comparison. Every cell says what a category of treatment does to one thing that can be in the water, with the page it comes from. Choose a trip and the rows that are not in play go quiet, so each column is scored on the threats you will actually meet.

Day hike, national forest creek Trek below a grazing allotment Desert canyon, silty potholes Alpine snowmelt above treeline River below a town, or travel abroad Home, under a boil-water notice

No trip picked yet — every row is lit, which is the honest default and the least useful view.

In the water	Hollow-fiber filter 0.1 µm — squeeze, straw, gravity, pump	Purifier-grade filter reverse osmosis or virus-rated	Chlorine dioxide tablets, contact time per the label	Bleach or iodine the cheap chemical	UV pen clear water only	Boiling 1 minute, 3 above 6,500 ft
Bacteria E. coli, 0.5 × 2–8 µm	✓ Bacteria — Hollow-fiber filter — Covered	✓ Bacteria — Purifier-grade filter — Covered	✓ Bacteria — Chlorine dioxide — Covered	✓ Bacteria — Bleach or iodine — Covered	✓ Bacteria — UV pen — Covered	✓ Bacteria — Boiling — Covered
Giardia cysts, 8 × 19 µm	✓ Giardia — Hollow-fiber filter — Covered	✓ Giardia — Purifier-grade filter — Covered	✓ Giardia — Chlorine dioxide — Covered	~ Giardia — Bleach or iodine — Partly — read the cell	✓ Giardia — UV pen — Covered	✓ Giardia — Boiling — Covered
Cryptosporidium oocysts, 4–6 µm	✓ Cryptosporidium — Hollow-fiber filter — Covered	✓ Cryptosporidium — Purifier-grade filter — Covered	~ Cryptosporidium — Chlorine dioxide — Partly — read the cell	✗ Cryptosporidium — Bleach or iodine — Not covered	✓ Cryptosporidium — UV pen — Covered	✓ Cryptosporidium — Boiling — Covered
Enteric viruses norovirus and kin, 0.03 µm	✗ Enteric viruses — Hollow-fiber filter — Not covered	✓ Enteric viruses — Purifier-grade filter — Covered	✓ Enteric viruses — Chlorine dioxide — Covered	✓ Enteric viruses — Bleach or iodine — Covered	✓ Enteric viruses — UV pen — Covered	✓ Enteric viruses — Boiling — Covered
Cloudy or silty water not an illness — the thing that stops the rest working	✓ Cloudy or silty water — Hollow-fiber filter — Covered	✓ Cloudy or silty water — Purifier-grade filter — Covered	✗ Cloudy or silty water — Chlorine dioxide — Not covered	✗ Cloudy or silty water — Bleach or iodine — Not covered	✗ Cloudy or silty water — UV pen — Not covered	~ Cloudy or silty water — Boiling — Partly — read the cell
Chemicals, cyanotoxins fuel, metals, salts, the toxin from a green mat	✗ Chemicals, cyanotoxins — Hollow-fiber filter — Not covered	— Chemicals, cyanotoxins — Purifier-grade filter — No claim made here	✗ Chemicals, cyanotoxins — Chlorine dioxide — Not covered	✗ Chemicals, cyanotoxins — Bleach or iodine — Not covered	— Chemicals, cyanotoxins — UV pen — No claim made here	✗ Chemicals, cyanotoxins — Boiling — Not covered
Covers this trip	—	—	—	—	—	—

✓ covered · ~ partly, read the cell · ✗ not covered · — no claim made here. Click any cell for the sentence it comes from and the page that published it.

The row nothing on the grid closes

One line stays open no matter which column you choose, and it is worth stating plainly because the industry is quiet about it.

Dissolved chemistry is untouchable. The EPA puts it in one sentence: boiling or disinfection will not destroy other contaminants, such as heavy metals, salts, and most other chemicals. Heat does not remove a chemical from water — it removes water from around the chemical, which is the opposite of what you want.

Cyanotoxins belong in the same row. At Grand Canyon, the National Park Service writes that standard backpacking water filters do not remove dissolved cyanotoxins, and tells hikers to avoid collecting water directly from or immediately adjacent to visible cyanobacterial mats. There is no gear answer to that. You walk upstream, or you walk to a different drainage, and you keep the dog out as well, because a dog drinks first and asks nothing.

Squeeze, straw, gravity, pump: one membrane, four chores

Here is the thing catalogs blur: a squeeze filter, a straw, a gravity bag and most pumps are the same technology in four housings. Hollow fiber, around 0.1 micron, identical removal profile. What differs is the labor, and labor is what decides whether a filter gets used on the fourth evening or quietly abandoned.

The straw removes the container from the equation. You lie down at the creek and drink through it. No waiting, almost no weight — and it fills nothing, not a pot, not a bottle, not tomorrow morning. A day-hike object, and a backup for everyone else.

The squeeze filter is the straw with a pouch. Scoop dirty water, screw the cartridge on, push. It fills a bottle in a couple of minutes and the work scales with the number of people drinking. For one or two, it is the right answer almost every time.

The gravity bag removes the human. Fill the reservoir, hang it from a branch, go set up the tent, come back to clean water. Nobody is squeezing anything. The reservoir and the extra hose cost you a few ounces and a bit of pack volume, and from about three people or eight liters a day the arithmetic flips hard in its favor. If you are outfitting a group, buy the gravity filter with a hanging reservoir^aff. and skip the debate.

The pump is the one that still earns its weight in awkward water. Shallow seeps, a puddle between rocks, a trickle you cannot submerge a bag in — a pump draws through a hose from water nothing else can collect. It also gives you a rate you control, which matters when the group is waiting. The cost is moving parts, seals, and an arm. A pump filter with an intake hose and prefilter^aff. is a desert and canyon tool more than a forest one.

Which one for which trip

Two on a weekend: squeeze, plus a tube of tablets. Four on a five-day trek: gravity at camp, squeeze for the daytime refill. Canyon country, where the water is a pothole and not a stream: pump, with a prefilter and a settling step first. Solo day hikes: a straw and a full bottle from home.

That last one is not a cop-out. The filter is plan B and the water on your back is plan A, exactly as the go bag reasons about its collapsible jug — carry a little, carry the means to get more.

The two ways a filter stops working, and only one gives a warning

A membrane fails in two directions, and the two are easy to confuse because they produce opposite symptoms.

Clogging is the honest failure. Silt builds on the dirty side, the flow drops, you backflush, the flow comes back. Eventually it does not, and that is the end of the cartridge. The whole process announces itself and never once lets contaminated water through. Annoying, and safe.

Cracking is the silent one. After its first use a hollow-fiber cartridge stays full of water. Water expands when it freezes. The fibers are finer than a hair, they split from the inside, and nothing about that is visible when you look at the housing. What you notice — if you notice anything — is that the filter runs faster than it used to. Better flow, no cleaning in between. That is the tell, and it reads like good news.

There is no field test. No dye, no taste, no pressure check that a hiker can run at a trailhead. Which is why the rule is blunt: a cartridge that may have frozen is a cartridge you replace, and an unknown winter in a car trunk counts as may have frozen.

The sheet below sorts the two apart, and it deliberately does not add up points. One disqualifying line beats four reassuring ones.

Filter retirement check — the veto sheet

Four good answers do not outvote one bad one.

Go and get the filter out of the drawer, then tick what is true of it. This sheet does not add up points, because a membrane does not fail gradually — it fails invisibly, and the reassuring lines do not cancel the disqualifying ones.

Trust it Service it first Retire it

Any one of these retires the cartridge

A veto, not a point. Ticking one ends the sheet whatever else is ticked.

It has been below freezing at some point since its last use, still wet inside. Nobody can say where it spent last winter — the trunk, the garage, the shed. Clean water still runs slowly through it after a full backflush. The housing is cracked, or a gasket or O-ring is missing. It suddenly runs faster than it used to, with no cleaning in between.

These are fixable before Saturday

Nothing here condemns the filter. All of it is an evening at the kitchen sink.

Flow has dropped, and it has not been backflushed since the last trip. It went into the pack damp at the end of the season and smells musty. It was used on visibly silty water with no prefilter and no settling step. The intake hose or the dirty-water bag has not been rinsed this year.

These are the reasons to trust it

Worth ticking honestly. They are also the habits that keep a cartridge alive.

Stored dry, indoors, at room temperature since the last trip. Used only on clear, moving water. Well under the capacity printed on its own care card. There is a tube of chlorine dioxide tablets in the same pocket.

Two habits make most of this moot. On a cold night the filter sleeps in the sleeping bag with you; between trips it lives indoors, dry, not in the garage and not in the trunk. The rest is rinsing: a pair of collapsible dirty-water bags^aff. keeps the intake side away from the side you drink from, which is the mistake that turns a working filter into a pointless one.

Chemicals and UV: the other half of the kit

A chemical treatment weighs almost nothing and covers precisely the row a membrane cannot. That is why it belongs in the lid pocket of every pack, including packs that already hold a good filter.

Chlorine dioxide is the family to carry. The CDC’s hiking page says chemical disinfectants kill viruses and bacteria, then adds the distinction that matters: chlorine dioxide tablets can kill the parasite Giardia and are somewhat effective at killing Cryptosporidium, whereas chlorine bleach and iodine kill bacteria and viruses but do not work well against parasites. A tube of chlorine dioxide purification tablets^aff. is about the size of a lip balm and does not freeze, break, or clog. Follow the contact time on the label, because the contact time is the product.

UV, and the water that defeats it

Ultraviolet light is the other purifier that fits in a pocket. The CDC’s travel chapter is generous about it: ultraviolet radiation kills bacteria, viruses, and both Giardia and Cryptosporidium oocysts in water. All four rows, no waiting beyond a minute or so. Then comes the condition that decides whether you should own one — UV light does not work well in cloudy water, because small particles block the germs from the light. A silty pothole defeats it, and it defeats it silently. A rechargeable UV purifier pen^aff. makes sense for clear water and for travel where viruses are the real threat; in canyon country it is a battery you are carrying for nothing.

Where the official pages stop, and where this one does

Two limits are worth naming before the trip planning. The federal guidance is not a single voice, and it does not always agree with itself. And nothing on this page is medical advice, which matters more here than on most gear subjects.

American backcountry has rules the catalog does not mention

The gear sold here is global; the water is not. Three American specifics change how the grid above gets used.

Grazing allotments. Enormous areas of national forest are open to cattle and sheep, and the drainage below one is the textbook case for Giardia and Cryptosporidium: ruminant parasites, in water that looks perfect. Clear, cold and moving is not evidence of anything.

Sediment and the desert. In canyon country the problem is not what is alive in the water, it is what is suspended in it. Grand Canyon’s guidance includes a clearing step before treatment — settle the water, strain it, then filter or boil — precisely because clay defeats both membranes and light. The EPA’s version for households is the same idea with kitchen materials: if water is cloudy, let it settle and filter it through a clean cloth, paper towel, or coffee filter.

Altitude, and a federal disagreement. The CDC says boil one minute, three minutes above 6,500 feet. The EPA says three minutes above 5,000 feet. That is a fifteen-hundred-foot gap between two federal agencies, and it lands squarely on the Rockies and the Sierra. Between those two lines, boil three minutes. The extra two minutes cost a little fuel and save you looking anything up at 9,000 feet with cold hands. A small canister stove and pot^aff. is what makes boiling an option at all above treeline, and it is the only method with no gap in any row of the grid.

The same filter at home, and where that reasoning stops

A hiking filter in a kitchen drawer is a genuinely useful object, and it is also the object most likely to be over-trusted. Both halves matter.

What it does at home: a boil-water notice usually follows a main break, a pressure loss or a flood pushing surface water into a distribution system, and the organisms that come with that are the same ones the membrane was built for. On the third day of a notice, being able to make tap water drinkable without burning fuel is worth a lot. That reasoning sits alongside the stored supply rather than replacing it, and the sizing for the supply itself is in emergency water storage.

Where it stops, in three places.

Volume, first. A squeeze filter sized for one thirsty hiker is the wrong shape for a family of four, who need gallons rather than bottles. This is the only context on this page where weight does not matter and volume decides everything, so the gravity bag that looked like a luxury on the trail becomes the obvious tool — hung from a door handle, working while you do something else. A household-scale gravity filter^aff. is that same object built for a kitchen instead of a branch. The household’s other water posts are counted in the 72-hour kit checklist and assembled in the 72-hour emergency kit.

Viruses, second. A municipal system contaminated by sewage is exactly the case a 0.1-micron membrane does not cover, and it is far more plausible at home than on a ridge. Boiling or chemical treatment has to follow the filter, in that order.

Chemicals, third, and this one is absolute. After a flood the question is not whether the water is boiled but whether the advisory says do not drink at all. A do-not-drink advisory is not a stronger boil notice; it is a different instruction, and no pot and no cartridge answers it. Which notice is in force reaches you over local stations and NOAA Weather Radio when the cell network is congested — the reason the emergency weather radio is in the kit, and the reason an extended power outage is a water problem before it is a light problem.

Why the household horizon is three days rather than one is worked through in why 72 hours; which hazards your address is actually exposed to starts in natural disaster preparedness. The shopping view sits on the emergency kit page.

What I would carry, and what I would leave at the store

The carbon-only bottle is the object I would not buy. Activated carbon binds molecules — chlorine, some pesticides, bad taste — and does nothing to organisms, which is why a listing that mentions carbon and improved taste but never states a pore size is describing an office bottle. It is not a backpacking water filter. It is a filter for water that was already safe.

The six-nines claim on the box deserves the same shrug. Removing 99.9999 percent of bacteria is true of any competent membrane and says nothing about viruses. Look for the pore size, and for the word absolute in front of it.

So: a hollow-fiber squeeze filter and a tube of chlorine dioxide tablets, for almost everybody, almost always. A gravity bag when the group grows. A pump when the water is a puddle. A stove, because it has no gaps. And the discipline that costs nothing — the filter sleeps warm, dries fully, and gets retired the first time nobody can vouch for its winter.

The one place that same calculation reappears off the trail is the bag by the front door, where a filter and an empty jug do the work of eight pounds of bottles at a fraction of the weight — counted line by line in the go bag checklist.

What I still find strange is how rarely any of this reaches the packaging. Two numbers settle the question, and you have to go to the CDC to find them. The household side of the same problem is in the blog, post by post.