The salinity budget: what your water leaves for the fertilizer
Irrigation water in this region arrives with salt already in it, and fertigation adds more. Both come out of the same allowance, because the crop responds only to the total. This review works the FAO salinity equations through the water actually delivered in the Jordan Valley, shows which crops are losing yield before a bag is opened, and sets out what is left to spend on fertilizer.
What this review covers.
- Kind
- Review of published work, not an Infert trial
- Framework
- FAO Irrigation & Drainage Paper 29 (Ayers & Westcot)
- Crop response
- Maas–Hoffman threshold and slope, as tabulated by FAO
- Water data
- Published Jordan Valley salinity figures
- Arithmetic
- Ours, from the published equations — shown in full
- Compiled
- August 2026
One number the crop responds to
A root does not distinguish between the salt that came down the canal, the salt left behind by the last three irrigations, and the salt injected this morning as fertilizer. It responds to the osmotic potential of the solution around it, and that is what electrical conductivity measures. Salinity is therefore a budget: the water spends part of it, the fertilizer spends part of it, and the crop has a fixed allowance that depends on what it is.
From the canal to the root zone
Soil salinity is higher than the water that made it, because the crop takes the water and leaves the salt. FAO takes ECe — the conductivity of the soil saturation extract, which is what the tolerance tables are written against — as about 1.5 times ECw, the conductivity of the applied water, assuming a 15–20% leaching fraction and a normal root-zone extraction pattern. That factor of 1.5 is the bridge between the number you can measure at the emitter and the number the crop actually experiences.
What each crop tolerates, and what it loses after that
Maas and Hoffman described crop response as two straight lines: no loss up to a threshold, then a fixed percentage of yield lost per unit of salinity above it. FAO tabulates both numbers for 81 crops, and the spread is the useful part. Date palm tolerates 4.0 dS/m before losing anything and then loses 3.6% per unit; onion starts losing at 1.2 and loses 16% per unit. On the same water, in the same season, one of those is comfortable and the other is in trouble.
The water that is actually delivered here
Published figures for the Jordan Valley put the northern King Abdullah Canal at roughly 500–750 mg/L total dissolved solids — about 0.8–1.2 dS/m — while blended water from the King Talal Dam, freshwater mixed with treated wastewater and the supply for much of the central and southern valley where vegetables and date palms are grown, runs nearer 1200–1500 mg/L, about 1.9–2.4 dS/m. The consequence shows up in the soil: one transect reported topsoil ECe rising from 4.5 to 14.1 dS/m between the north of the valley and the south.
Worked through: what the water alone costs
Take those three water qualities, convert each to ECe by multiplying by 1.5, and apply each crop's threshold and slope. The second table below is that arithmetic and nothing else — it can be reproduced with a calculator in a minute. It says something worth knowing before any fertilizer decision is made: on mid-valley water at 1.4 dS/m, tomato has lost nothing and onion has already lost about 14% of its yield. On southern blended water, pepper is down a quarter and date palm is still untouched.
And then the fertilizer
Fertigation adds dissolved salts to the water and therefore to the budget. As a first approximation FAO puts 1 dS/m at roughly 700 mg/L of dissolved salts, so injecting a fully ionic soluble fertilizer at 1 g/L is on the order of an extra 1.4 dS/m at the emitter — enough on its own to move mid-valley water into ground where sensitive crops lose real yield. Two qualifications matter. Urea does not dissociate in water, so it adds nitrogen without adding conductivity, which is exactly why it behaves differently from an equivalent nitrate in the tank. And the conversion is approximate for everything else, because conductivity per gram differs by salt. The practical instruction is therefore not a table but a measurement: run the injector, read EC at the emitter, and subtract the reading with the injector off. The difference is what the programme is spending.
Buying the allowance back
Where the water is saltier than the crop tolerates, the way out is leaching — applying more water than the crop uses, so salts move below the root zone. FAO gives the requirement as LR = ECw ÷ (5 × ECe − ECw), with ECe the salinity you intend to hold the root zone at, and the total application as ET ÷ (1 − LR). At ECw 1.4, holding a pepper crop at its 1.5 dS/m threshold: LR = 1.4 ÷ (7.5 − 1.4) = 0.23. A leaching fraction of about 23% requires about 30% more applied water than crop ET, because total application is ET ÷ (1 − LR). For tomato at its 2.5 threshold, the same water needs 1.4 ÷ (12.5 − 1.4) = 0.13. In a country irrigating with treated wastewater because the fresh water has run out, that is a real constraint — and a reason to fit the crop and the fertilizer form to the water rather than assume the water can be fixed.
What we do with this
We ask for a water analysis before writing a programme, and we would rather quote a lower-EC route — nitrate and urea nitrogen where it suits, potassium sulfate rather than chloride where chloride is already high — than sell a higher-analysis product into water that cannot carry it. Where the arithmetic says the crop is the problem rather than the fertilizer, we will say that too, and it is not a sales conversation.
Salt tolerance: the threshold, and the loss per unit above it
| Crop | Threshold ECe (dS/m) | Yield lost per dS/m above it |
|---|---|---|
| Barley | 8.0 | 5.0% |
| Wheat | 6.0 | 7.1% |
| Squash, zucchini | 4.9 | 10.5% |
| Date palm | 4.0 | 3.6% |
| Tomato | 2.5 | 9.9% |
| Cucumber | 2.5 | 13% |
| Alfalfa | 2.0 | 7.3% |
| Potato | 1.7 | 12% |
| Maize | 1.7 | 12% |
| Pepper | 1.5 | 14% |
| Grape | 1.5 | 9.6% |
| Lettuce | 1.3 | 13% |
| Orange | 1.3 | 13.1% |
| Onion (bulb) | 1.2 | 16% |
| Melon | 1.0 | 8.4% |
FAO Irrigation & Drainage Paper 29 (rev. 1), Annex 1, Table A1.1 — Maas & Hoffman / Maas & Grattan data, on an ECe basis.
Yield lost to the water alone, before any fertilizer
| Crop | ECw 0.8 — north KAC | ECw 1.4 — mid range | ECw 2.2 — southern blend |
|---|---|---|---|
| Date palm | none | none | none |
| Tomato | none | none | 7.9% |
| Cucumber | none | none | 10.4% |
| Potato | none | 4.8% | 19.2% |
| Pepper | none | 8.4% | 25.2% |
| Onion (bulb) | none | 14.4% | 33.6% |
Our arithmetic, from the FAO relations: ECe = 1.5 × ECw, then loss = slope × (ECe − threshold), using the thresholds in the table above.
The three water qualities are points inside the published ranges, not measurements of a particular farm. Reproducible with a calculator: at ECw 1.4, ECe = 2.1, so pepper = 14 × (2.1 − 1.5) = 8.4%.
Key takeaways.
- Ask for the water analysis before the programme, not after the season.
- Multiply water EC by 1.5 to get the number the crop responds to.
- On 1.4 dS/m water, tomato has spent nothing and onion has already lost about 14%.
- Measure EC at the emitter with the injector running — that difference is your fertilizer.
- Where leaching is the answer, these leaching fractions require about 14–30% more applied water than crop ET. Budget it.
References & further reading.
Ayers, R.S. & Westcot, D.W. (1985). Water quality for agriculture. FAO Irrigation and Drainage Paper 29 (rev. 1). Rome: FAO — the ECe/ECw relation, the leaching requirement, and the crop tolerance annex.
FAO 29 ↗FAO. Annex 1: Crop salt tolerance data (Table A1.1), after Maas & Hoffman (1977) and Maas & Grattan (1999).
FAO crop tolerance table ↗Maas, E.V. & Hoffman, G.J. (1977). Crop salt tolerance — current assessment. Journal of the Irrigation and Drainage Division, ASCE 103(2), 115–134.
Maas & Hoffman 1977 ↗Talozi, S. et al. (2026). Mitigating water salinity in the Jordan Valley. Advances in Agriculture — reports King Abdullah Canal and King Talal blended water salinity.
Jordan Valley water ↗
