Off-Grid Water Treatment: Test First, Then Buy the Filter

Nobody should buy a water filter before they have a lab report. That is the whole argument of this article and the reason the old version of it was wrong.

Here is why it matters more off grid than anywhere else. If you are on a city supply, somebody is testing your water to a federal standard and telling you when it fails. If you are on a private well, nobody is. EPA says it plainly: the quality and safety of drinking water from private domestic wells is not regulated by the federal government under the Safe Drinking Water Act, nor by most state governments. Around 15 percent of the country, more than 43 million people, drink from one anyway.

Which means the first purchase is not a filter. It is a test.

Test first, and test for the right things

EPA's annual recommendation is total coliform bacteria, nitrate, total dissolved solids and pH. CDC gives the same four and adds that you should contact your health department to find out what else to test for based on where you live. Both say to test immediately after flooding, construction nearby, any repair to the well system, or any change in taste, colour or odour.

Take the second half of CDC's advice seriously, because the four-analyte panel contains no arsenic, no uranium, no fluoride and no radon. A buyer in the arid West who runs only the standard annual panel will miss the contaminants most likely to be above a health limit in their own water.

Two of our ten states already do some of this for you. North Carolina requires the local health department to test a newly constructed private well within 30 days of issuing the certificate of completion, and the list is long: arsenic, barium, cadmium, chromium, copper, fluoride, lead, iron, magnesium, manganese, mercury, nitrates, nitrites, selenium, silver, sodium, zinc, pH and bacterial indicators. Oregon requires the seller of real estate with a well to test for arsenic, nitrates and total coliform on accepting an offer, report the results to the buyer, and report them to the state health authority within 90 days.

That Oregon rule is worth knowing if you are buying there. The data may already exist for the parcel you are looking at, and the state holds it. Minnesota has tested every newly drilled well for arsenic since 2008. Arizona publishes a recommended schedule rather than a requirement: arsenic, bacteria, fluoride, lead, nitrates, total dissolved solids and uranium on a new well, then bacteria, nitrates, arsenic, fluoride and uranium every five years.

Use a state-certified drinking water laboratory. EPA maintains the index of state certification programs and their certified lab lists, and most state health departments publish a local list or run the testing themselves.

What is actually in rural groundwater

USGS sampled domestic wells across the principal aquifers and found that about 23 percent had at least one contaminant above a maximum contaminant level or health-based screening level, with no single contaminant exceeding a benchmark in more than 8 percent of wells. Another 48 percent had something above a secondary, aesthetic standard.

The important part is where it comes from. Man-made organic compounds, meaning pesticides and solvents, were detected in 60 percent of wells and exceeded a health benchmark in less than 1 percent. Nitrate was the only contaminant from man-made sources above a benchmark in more than 1 percent of wells. Everything else driving that 23 percent is rock.

So articles that frame rural well risk as runoff and the neighbour's septic tank are describing the minority case. The usual problem is the aquifer.

Arsenic deserves its own paragraph because it is the one that bites hardest in our footprint. USGS estimates about 2.1 million Americans draw from private wells with arsenic above the 10 microgram limit, and names much of the West plus Atlantic southeast coastal states among the high-risk areas. A Desert Research Institute study of more than 170 northern Nevada households on private wells found 22 percent above the EPA limit, some reaching levels 80 times the threshold, and only 41 percent of those wells had any treatment at all. Minnesota has detected arsenic in about 40 percent of wells drilled since 2008, with about 10 percent of private wells above the limit. Oregon's transaction-testing dataset puts 9.9 percent of 11,596 tests above the limit. In Texas it runs above 20 parts per billion in the High Plains and West Texas, and drought makes it worse, because exposing the aquifer to oxygen increases arsenic concentrations.

Nitrate is the one with an immediate and specific danger. EPA's own language: infants below six months who drink water with nitrate above the limit could become seriously ill and, if untreated, may die. Minnesota reports more than 10 percent of private wells above the limit in some townships.

Radon is a confusing one worth getting right, and most articles get it wrong. There is no federal limit for radon in drinking water. EPA proposed 300 picocuries per litre with an alternative level of 4,000 back in 1999 and the rule was never finalised, so any vendor presenting those as the EPA limit is quoting a proposal. The risk is also not what people assume: of estimated radon deaths attributable to drinking water, EPA puts 89 percent at lung cancer from breathing radon released into indoor air during showering and washing, and 11 percent at stomach cancer from drinking it. North Carolina's rule of thumb is that 10,000 picocuries per litre in water contributes about 1 picocurie per litre to indoor air.

And manganese is not purely cosmetic, whatever the secondary standard implies. EPA publishes a lifetime health advisory of 0.3 mg/L, and Minnesota advises 100 micrograms per litre or less where an infant drinks the water or formula made with it, because long exposure to high manganese is associated with problems in memory, attention and motor skills.

What each kind of treatment actually does

Minnesota's health department says the governing sentence better than we could: no single treatment unit can remove all contaminants in water. Which means the useful way to read any product page is to ask what it does not do, and here are the four answers that catch people out most often.

Reverse osmosis is the broadest tool and it has an off-grid problem nobody mentions. It removes most dissolved and suspended inorganics including arsenic, lead, uranium and much of the nitrate, though Penn State puts nitrate rejection at 83 to 92 percent rather than all of it. It needs a sediment prefilter because hard water fouls the membrane. And the waste ratio is the killer: between 80 and 90 percent of the water entering a standard system leaves as wastewater, with EPA putting it at 5 to 10 gallons wasted per gallon treated for standard units.

On a municipal supply that is a bill. On a cistern or hauled water it is a budget. Run the arithmetic against your storage before you install one.

Distillation removes nitrate, bacteria, viruses, heavy metals and radionuclides and is ineffective against volatile organics. It is also finished as an off-grid option the moment you look at the power: about 800 watts to distil a quart, with countertop units yielding under three or four gallons a day.

Arsenic is a special case worth understanding before you buy anything. The treatment depends on whether the arsenic is the trivalent or pentavalent form, and the trivalent form is uncharged below pH 9.2, so adsorptive media and anion exchange barely see it. It has to be oxidised first. Chlorine, permanganate, ozone and solid-phase oxidants do that. Aeration alone, chloramine and ultraviolet light by itself do not. Minnesota's four approved approaches are all phrased with pre-oxidation attached: reverse osmosis with preoxidation, distillation, adsorptive media with preoxidation, and anion exchange with preoxidation. A standard lab report giving total arsenic does not tell you which form you have.

And iron, manganese and hydrogen sulfide all get solved the same way, which is a good illustration of why chemistry rather than product category drives the choice: an oxidising filter converts all three from dissolved to solid and then filters the particles out. One mechanism, three problems. If the rotten egg smell is only in the hot water, the cause is the water heater's magnesium anode rod rather than the well, and you swap the rod.

Where to treat, and what the certification means

Point of use treats one tap. Point of entry treats everything coming into the building. The split is not only about cost, it is about exposure. EPA states that point-of-use devices should not be used for radon or most volatile organics, because they give no protection against inhalation or skin contact at untreated taps such as the shower. Anything that is only an ingestion risk, including arsenic, lead, nitrate, fluoride and uranium, can be handled at one tap. Anything that volatilises, or that damages plumbing and appliances, has to be handled where the water comes in.

Certification is how you tell marketing from performance, and the numbers are specific. NSF/ANSI 42 is aesthetic effects: chlorine, taste, odour, particulate, iron, manganese and total dissolved solids. NSF/ANSI 53 is health effects, with more than 50 contaminant reduction claims. 58 is reverse osmosis, 62 is distillation, 44 is cation exchange softeners, 401 is emerging contaminants, and 55 is ultraviolet.

Notice that iron and manganese sit under 42, the aesthetic standard. A filter can carry a completely legitimate NSF/ANSI 42 certification and have zero certified health-contaminant performance. Check which standard and which specific claim, not just that there is a logo.

The sharpest trap in the category is NSF/ANSI 55. Class A delivers at least 40 millijoules per square centimetre and is the class for disinfecting water of unknown microbiological quality. Class B is 16 and is intended only to reduce non-pathogenic nuisance organisms in already-disinfected water, which Ohio's health department says outright is unsuitable for primary disinfection of a private system. A unit advertised as NSF 55 certified with no class stated tells you nothing.

And one more. EPA does not certify home treatment devices. EPA says in its own consumer fact sheet that EPA registration, unlike that of NSF, the Water Quality Association or UL, is not intended to guide consumers in selecting a water treatment device and is not an endorsement of the product. "EPA registered" on a box usually refers to pesticide registration for something like silver-impregnated media. Penn State Extension is blunter: vendors touting EPA certification should be avoided, and be cautious of anyone running a colourful home water test to pressure you into buying on the spot.

Rainwater and hauled water, which are their own problem

If your supply is a cistern rather than a well, the treatment train is different and the Texas Water Development Board publishes a specific one: a 5-micron fibre cartridge filter, then a 3-micron activated carbon cartridge, then ultraviolet, with a UV unit rated at 12 gallons a minute sufficient for a standard setup and lamps replaced after a maximum of 10,000 hours. The chlorination alternative is a free chlorine residual of 2 parts per million with a few minutes of contact time.

Texas A&M's extension service makes the same recommendation in general terms and adds the principle worth carrying: use a multi-barrier approach where more than one method of treatment is used. For pathogen removal by filtration alone, CDC's pore-size thresholds are an absolute rating of 1 micron or smaller for parasites and 0.3 micron or smaller for bacteria, with viruses requiring nanofiltration or reverse osmosis. Note absolute rather than nominal: EPA tells consumers to look for the absolute rating, which is the largest hole, not the nominal rating, which is the average one.

Test the stored water, not just the source. The Texas manual recommends quarterly testing at minimum if the water is used for drinking, and testing both before and after treatment so you know the treatment is doing anything. The whole rainwater legal picture, which varies more by state than almost anything else we write about, is in what each state actually allows on water catchment.

The order to do this in

Where we fit

We sell the land, not the equipment, and we are not going to tell you what is in the water under a parcel because we have not tested it and neither has anyone else. What we will do is give you the parcel number and the county so you can call the health department, ask what the known regional problems are, and find a certified lab before you commit.

If a seller tells you the water is fine, ask for the lab report and the date. If there isn't one, that is not a red flag by itself, because most rural parcels have never been tested. It just means the testing is now your job and it should happen before the well goes in, not after. The rest of the water question, from well permits to what you are even allowed to pump, is in the guide to living off the grid.

If you are shopping with a well or a cistern in the plan, our off-grid land listings are the place to start and the health department call is free.

See All Available Land

The cheapest thing in this entire article is the lab test, and it is the only step that tells you which of the expensive things you actually need.

Everything else is a response to a result you do not have yet. Get the result first and most of the catalogue stops being relevant.