Library › Part B · Article 8 in the series · Level 1 · 18 min read

Water quality: what is in your water and why it matters

Standfirst: After reading this article you will be able to read a water report line by line, sort each constituent into useful, harmless up to a limit, or a cause of trouble, compare tap, rain, reverse-osmosis and well water, and decide with published limits whether a given water suits lettuce and how much calcium and magnesium it already brings to your recipe.

Before you read: How roots take up nutrients.

Water is never just water

A nutrient solution is water plus dissolved ions, and the water you start with already contains some, so it is already a partial nutrient solution. Building a recipe starts with finding out what is in it.

The reason is geology. Rain forms with almost nothing dissolved in it: across Texas, total dissolved solids in rainwater range from 2 mg/L to 20 mg/L, against 100 ppm to more than 800 ppm in municipal supplies⁹. The difference is what the water picked up from soil and rock. Water from limestone aquifers can carry as much as 100 mg/L of calcium and 30 mg/L of magnesium, which Jones calls not unusual, often with sodium, bicarbonate, sulfate, chloride and sometimes boron².

Some of those ions are plant nutrients and some are not. Resh asks for an analysis of at least calcium, magnesium, iron, boron, molybdenum, carbonate, sulfate and chloride before any water is used, so that correspondingly less of each goes into the recipe¹. In his injector example the raw water held 30 ppm of calcium and 20 ppm of magnesium, both counted toward the final 200 ppm of calcium and 40 ppm of magnesium¹.

How to read a water report

Municipal utilities publish an annual water quality report, and any agricultural laboratory will analyse a sample. Jones advises a complete analysis before use, repeated whenever the source changes, plus an organic analysis for water from a river, a shallow well or another surface source, because pesticides affect plant growth even at low concentrations².

A note on units. One milligram per litre is one part per million by weight, because a litre of dilute solution weighs about one kilogram^1,9. Laboratories also report the major ions in milliequivalents per litre (meq/L), which counts charge rather than weight; Resh calls it the most meaningful way to express the major constituents of water¹. To convert, divide mg/L by the atomic weight to get mmol/L, then multiply by the ion’s charge; the atomic weights used here are calcium 40.08, magnesium 24.31, sodium 22.99 and chlorine 35.45¹. Conductivity in dS/m, mS/cm or mmho/cm is one and the same unit, and Resh’s rule of thumb is about 650 ppm of salt per 1 mS/cm; Jones quotes 650 to 700 ppm^1,2.

Two lines are easy to confuse. Hardness is caused by calcium and magnesium³. Alkalinity is the capacity to neutralise acid; laboratories report it either as bicarbonate alkalinity or as total alkalinity expressed as calcium carbonate². Resh describes hardness as a measure of the bicarbonate content¹, which reflects how often the two travel together: ground water from limestone and dolomite dissolves calcium and magnesium carbonates, so it arrives both hard and alkaline¹. The grower needs them separately: the calcium is welcome, the bicarbonate is the problem.

Sorting the constituents

Useful: calcium, magnesium and sulfate

Resh states that hard waters are normally just as suitable as soft water, because calcium and magnesium are essential and the amount in hard water is ordinarily much less than a nutrient solution contains¹. Most hard waters carry them as carbonates or sulfates; sulfate is itself an essential nutrient, carbonate is not¹. Jones adds that hard water may supply part or all of the crop’s calcium and magnesium, but only an elemental analysis can decide that². A guideline for plug growers treats 50 to 125 mg/L of calcium and 5 to 25 mg/L of magnesium in the water as nutritionally beneficial².

Harmless up to a limit: sodium and chloride

Chloride is a micronutrient and sodium at most a beneficial element, so a plant needs little of either (What plants are made of). Harm at higher concentration comes by two routes. The first is osmotic: the total solute concentration lowers water uptake, which Resh names as the cause of the inhibitory effect of saline solutions, with symptoms of stunting, smaller and darker green leaves, marginal leaf burn, and bluing or bleaching¹. The second is specific toxicity: ions taken up with the water accumulate in the leaves as the water transpires away, over many days or weeks⁵. Sodium injury appears as leaf burn along the outer edges, first on older leaves; chloride injury starts at the leaf tips⁵.

Published limits for source water differ, and the differences are informative.

Three meq/L is 69 mg/L of sodium or 106 mg/L of chloride^1,5. The spread reflects the system: 30 to 50 mg/L of sodium for rockwool culture against 180 mg/L for general use², for a reason explained below.

Crops differ. Resh notes that watercress and mint tolerate more sodium and chloride than tomatoes and cucumbers¹. The FAO salt-tolerance table gives lettuce full yield up to an irrigation-water EC of 0.9 dS/m, 90% at 1.4 dS/m and 75% at 2.1 dS/m, against 1.3 dS/m for spinach and 1.7 dS/m for tomato and cucumber at full yield⁵. Those values are for soil, where the root zone ends up about 1.5 times saltier than the water⁵. In recirculating soilless culture, Sonneveld and van der Burg found thresholds of 2.3 to 3.5 dS/m for tomato, cucumber and sweet pepper, with yields falling 2.3% to 7.6% per dS/m above them, measured on the whole solution, nutrients included⁶.

Trouble: alkalinity

Carbonate and bicarbonate make the pH rise or stay high, and once present they make it resist dropping¹. This is buffering, and a little is welcome: Resh cites Smith (1987), who advised growers to hold 30 to 50 ppm of carbonate and bicarbonate against sudden pH swings, adding potassium carbonate or bicarbonate to very pure water that has none¹. More than that costs acid. Whipker and co-workers set 40 to 65 ppm of alkalinity as calcium carbonate as the desired level and 150 ppm as the upper limit²; Farnham’s irrigation guideline puts bicarbonate below 40 mg/L in the no-problem class and 40 to 520 mg/L in the increasing-problem class². The remedy can be as simple as acidifying the water²; how much acid, and why alkalinity rather than pH is the number to watch, is the subject of Alkalinity and buffering.

Trouble: boron

Boron has the narrowest margin of any element. Resh gives the normal requirement in solution as 0.30 mg/L and warns that some waters carry two to three times that or more, in which case it must be removed by reverse osmosis; boron toxicity is relatively common with saline waters¹. Hydroponic and greenhouse guidelines cluster between 0.3 and 1 mg/L: below 0.3 mg/L desired and 0.5 mg/L upper limit in Whipker’s table, 0.7 mg/L maximum for rockwool culture, below 1 mg/L for general hydroponic use, and 1.0 mg/L as the edge of the no-problem class for irrigation². The FAO table of boron tolerance, built on field crops in soil, lists lettuce as moderately tolerant at 2.0 to 4.0 mg/L⁵. The gap is not a contradiction: in a nutrient solution the boron in the water adds to the boron in the fertiliser, and in a recirculating system it accumulates, so the source-water limit must sit well below the plant’s tolerance.

Trouble: iron and manganese

Groundwater without oxygen can hold several milligrams per litre of ferrous iron and still be clear when pumped; in air the iron oxidises to the ferric form and turns the water reddish-brown³. Above 0.3 mg/L it stains laundry and fixtures³. Manganese stains above 0.1 mg/L, and at 0.2 mg/L it often coats pipes and sloughs off as a black precipitate³. The same chemistry means that iron and manganese in the water are not reliably available to the plant: they precipitate, and the precipitate ends up on filters and roots. Greenhouse guidelines set 1 mg/L as the upper limit for each, with Whipker’s table allowing 5 mg/L of iron and 2 mg/L of manganese as ceilings². The iron a recipe relies on is supplied as a chelate for this reason; see Micronutrients and chelates.

Trouble: chlorine and chloramine

Utilities disinfect with chlorine, which most people taste well below the health-based guideline of 5 mg/L, some at 0.3 mg/L³. Many now use chloramine instead, formed by adding ammonia to chlorinated water; monochloramine is the only useful chloramine disinfectant, less effective than free chlorine but persistent, which is why it is chosen to hold a residual through long distribution systems³. A residual designed to survive the pipe network will also survive the trip to your tank. Date and co-workers grew lettuce in a nutrient solution carrying 0.5 mg/L of chlorine as hypochlorous acid together with 0.67 meq/L of ammonium, which converts it to chloramine, and saw root injury and growth inhibition unless a reducing agent was added to remove the chlorine¹⁰.

Why sodium matters more in a recirculating system

Plants take up water much faster than sodium. In a drain-to-waste system the sodium leaves with the drainage; in a recirculating system it stays, and every litre of make-up water adds more. Resh describes the process: closed systems add raw water to the returning solution to replace what the plants transpired, so more ions arrive with every addition, and if the raw water carries significant levels this eventually leads to toxicity; the salts accumulate until the solution has to be changed¹.

Raviv and Lieth draw the practical conclusion: 100% recirculation is not yet possible where the supply water contains ions the plant cannot use, and growers must keep the freedom to discard solution once an ion such as sodium has accumulated to a toxic level⁴. Sonneveld and van der Burg found that absorption of sodium and chloride differed with crop and concentration, so some crops reach the limit sooner⁶. This is why Resh ties long-lived recirculating solutions to very pure raw water or reverse-osmosis treatment¹. Managing a recirculating solution over time shows how to track sodium with a simple mass balance.

Tap, rain, reverse osmosis and well water

Tap water is the easiest to characterise, because the utility has measured it; the disinfectant residual is the one constituent no other source carries³.

Rainwater is the nearest thing to pure water a grower can collect for free: soft, and in Texas 2 to 20 mg/L of dissolved solids before collection⁹. What it contains afterwards depends on the roof. A survey of 53 rainwater tanks in South Australia found that roof material was related to the levels of lead, zinc, copper and cadmium⁷; 47 tanks, 88.6%, exceeded the drinking-water guideline of 0.01 ppm for lead at least once, the highest lead value was 3.24 ppm, and household filters did not bring the metals below guideline concentrations, apart from two cadmium samples⁷. Tank pH ranged from 4.5 to 8.4, and acidic rainwater dissolves metals from roof and tank materials⁷, lead solder in older gutters and copper flashings among them⁹. Rain also has almost no buffering, which is why Resh passes on the advice to add 30 to 50 ppm of bicarbonate to very pure water¹.

Reverse osmosis forces water through a membrane that rejects most dissolved ions. The Nebraska extension guide lists calcium, magnesium, sodium, chloride, iron and nitrate among what it removes, gives a rejection example of 85%, states that most household units recover only 20% to 30% of the feed water as product, and notes that RO water can be more corrosive than untreated water⁸. Resh recommends RO where the raw water carries too much boron or too much sodium chloride¹. For the recipe this is the mirror image of hard water: calcium and magnesium are near zero, so the salts supply the full amounts, and some bicarbonate may need adding back¹.

Well water is the least predictable. Jones gives 100 mg/L of calcium and 30 mg/L of magnesium as not unusual upper values for limestone aquifers², and Resh notes that some well waters carry so much magnesium that none needs adding¹. Wells can also deliver sulfide, with its rotten-egg smell², ferrous iron that oxidises in the tank³, and pesticides where a shallow well sits in farmed land². A well needs a full laboratory analysis first.

Worked example: does this water suit lettuce?

Take the following analysis. The calcium and magnesium values are the raw-water values from Resh’s injector example¹. The other values are chosen for this example, inside the desired column of Whipker’s guideline table except alkalinity, which we set between the desired level and the upper limit on purpose². This is an example, not a real sample.

Step 1, salinity. An EC of 0.45 dS/m is below Farnham’s 0.75 dS/m no-problem limit² and below the 0.9 dS/m at which the FAO table still gives lettuce full yield⁵. By Resh’s rule of thumb, 0.45 mS/cm × 650 ≈ 290 ppm of dissolved salts, below Farnham’s 480 mg/L^1,2.

Step 2, sodium and chloride. 25 mg/L of sodium is below the 30 to 50 mg/L rockwool maximum, the 50 ppm desired level and the 69 ppm upper limit; 30 mg/L of chloride is below the 50 to 100 mg/L rockwool maximum and the 70 to 71 mg/L limits². In milliequivalents, 25 ÷ 22.99 = 1.09 meq/L of sodium and 30 ÷ 35.45 = 0.85 meq/L of chloride, both under the FAO 3 meq/L line^1,5. The water passes, with the recirculation caveat above.

Step 3, boron. 0.20 mg/L is below the 0.3 mg/L desired level, the 0.5 mg/L upper limit and the 0.7 mg/L rockwool maximum². It is also two-thirds of Resh’s 0.30 mg/L requirement, so the recipe should add only 0.30 − 0.20 = 0.10 mg/L of boron from boric acid¹.

Step 4, iron and manganese. 0.10 mg/L of iron is below the 0.3 mg/L staining threshold and 0.02 mg/L of manganese below the 0.1 mg/L threshold³; both are far under the 1 mg/L greenhouse limits². Neither value is counted toward the recipe, because the iron will not stay in solution³.

Step 5, alkalinity. 120 mg/L as CaCO₃ is above the desired 40 to 65 ppm but below the 150 ppm upper limit². Whipker’s table equates 150 ppm with 3 meq/L, so 120 ppm is 2.4 meq/L of acid-neutralising capacity². This water will push the pH up and hold it there; Alkalinity and buffering shows how much acid that takes.

Step 6, what the water contributes. Calcium: 30 mg/L ÷ 40.08 = 0.75 mmol/L¹. Magnesium: 20 mg/L ÷ 24.31 = 0.82 mmol/L¹. Against Resh’s targets of 200 ppm calcium and 40 ppm magnesium, the water supplies 30 ÷ 200 = 15% of the calcium and 20 ÷ 40 = 50% of the magnesium, so the salts need to supply 170 mg/L of calcium and 20 mg/L of magnesium¹. The 20 mg/L of sulfate is counted toward the recipe in the same way¹.

Step 7, disinfectant. 0.5 mg/L of free chlorine is a tenth of the drinking-water guideline³. Lettuce roots were injured at 0.5 mg/L of chlorine once ammonium had turned it into chloramine¹⁰, so check the utility report for chloramine, which persists³, and let the water stand or treat it before use.

Verdict. The water suits lettuce. Subtract 30 mg/L of calcium, 20 mg/L of magnesium, 20 mg/L of sulfate and 0.20 mg/L of boron from the recipe¹, budget acid for 2.4 meq/L of alkalinity², and in a recirculating system watch sodium^1,4.

Key numbers

Common mistakes

1. Building a recipe without a water analysis. The water already contains calcium, magnesium, sulfate and sometimes boron, and the full recipe on top over-supplies them^1,2.
2. Treating pH as the number that matters. The pH of water with few ions is hard even to measure; alkalinity decides how far the pH drifts and how much acid holds it^1,2.
3. Reading a sodium value that passes every limit as safe for a recirculating system. The plant leaves sodium behind while it takes water, and every top-up adds more^1,4.
4. Counting the iron in well water as part of the recipe. Ferrous iron oxidises in air and precipitates, so it stains and clogs instead of feeding the plant³.
5. Switching to reverse-osmosis or rain water without changing the recipe. Those waters bring almost no calcium, magnesium or bicarbonate, so the salts must supply all of it^1,8,9.
6. Assuming roof-collected rain is as pure as rain. Roof and gutter materials add lead, zinc and copper⁷.

Summary

The water you start with is a partial nutrient solution, so building a recipe begins with its analysis. Calcium, magnesium and sulfate in the water are nutrients and are subtracted from the recipe. Sodium and chloride are harmless up to limits that vary by system, from about 30 to 50 mg/L of sodium for rockwool culture to 180 mg/L for general use, and the stricter figures apply wherever the solution recirculates, because sodium accumulates with every top-up². Alkalinity pushes the pH up and resists correction, boron has the narrowest safe margin, iron and manganese precipitate and stain, and a chloramine residual persists. Rain and reverse-osmosis water are nearly empty and need calcium, magnesium and buffering added; tap water comes with a report; well water needs a full analysis. In the worked example, 30 mg/L of calcium and 20 mg/L of magnesium in the water supplied 15% of the calcium and half the magnesium of a standard recipe¹.

Next in the series: pH: what it is and why 5.5 to 6.5.

References

1. Resh, H. M. (2013). Hydroponic Food Production: A Definitive Guidebook for the Advanced Home Gardener and the Commercial Hydroponic Grower, 7th ed. CRC Press, Boca Raton. ISBN 978-1-4398-7869-9.
2. Jones, J. B. Jr. (2005). Hydroponics: A Practical Guide for the Soilless Grower, 2nd ed. CRC Press, Boca Raton. Chapter 7, Tables 7.1 to 7.6, which reproduce Verwer & Wellman (1980), Farnham, Hasek & Paul (1985, UC Davis Leaflet 2995), Waters, Geraldson & Woltz (1972), Smith (1999), Faulkner (1998) and Whipker, Dole, Cavins & Gibson (2003, Ball Redbook vol. 2).
3. World Health Organization (2011). Guidelines for Drinking-water Quality, 4th ed. WHO, Geneva. Chapter 10, Acceptability aspects, and Annex 5.
4. Raviv, M. and Lieth, J. H. (eds.) (2008). Soilless Culture: Theory and Practice. Elsevier, Amsterdam. Chapter 13, Growing plants in soilless culture: operational conclusions. https://doi.org/10.1016/B978-044452975-6.50015-0
5. Ayers, R. S. and Westcot, D. W. (1985, rev. 1994). Water Quality for Agriculture. FAO Irrigation and Drainage Paper 29. FAO, Rome. Table 1, Table 4, Table 16 and section 4. https://www.fao.org/4/t0234e/t0234e00.htm
6. Sonneveld, C. and van der Burg, A. M. M. (1991). Sodium chloride salinity in fruit vegetable crops in soilless culture. Netherlands Journal of Agricultural Science, 39(2), 115–122. https://doi.org/10.18174/njas.v39i2.16546
7. Chubaka, C. E., Whiley, H., Edwards, J. W. and Ross, K. E. (2018). Lead, zinc, copper, and cadmium content of water from South Australian rainwater tanks. International Journal of Environmental Research and Public Health, 15(7), 1551. https://doi.org/10.3390/ijerph15071551
8. Dvorak, B. I. and Skipton, S. O. Drinking Water Treatment: Reverse Osmosis. University of Nebraska–Lincoln Extension, NebGuide G1490. https://extensionpubs.unl.edu/publication/g1490/
9. Texas Water Development Board (2005). The Texas Manual on Rainwater Harvesting, 3rd ed. Austin, TX. pp. 1, 6, 21. https://www.twdb.texas.gov/publications/brochures/conservation/doc/RainwaterHarvestingManual_3rdedition.pdf
10. Date, S., Hano, F., Naba, K., Terabayashi, S. and Fujime, Y. (2005). Effects of adding reducing agent to the nutrient solution containing chloramine on the occurrence of root browning and growth in lettuce. Horticultural Research (Japan), 4(3), 297–301. https://doi.org/10.2503/hrj.4.297 (English abstract)

This article was drafted with the help of AI tools and reviewed and edited by PonicLabs. Sources are listed above; tell us if you find an error.