Zeravshan irrigation district: salinity, sodicity and drainage
The Zeravshan district has 4,600 ha of irrigated cotton and wheat in a semi-arid lowland with 210 mm annual rainfall and 1,850 mm pan evaporation. Water arrives by unlined canal; canal seepage and flood irrigation have raised the water table from 3.5 m in 2005 to 1.9 m in 2024. Root-zone ECe has risen from 2.0 to 11.2 dS/m and sodium adsorption ratio from 4 to 14. Infiltration has slowed and the surface crusts after irrigation. Wheat yields are down 38% and cotton down 22% against 2005. There is no drainage system. Canal water has an ECw of 0.9 dS/m. Options costed are: (A) apply 25% more irrigation water to push salts down; (B) subsurface tile drains on 800 ha at US$2,400/ha plus gypsum at 4 t/ha; (C) convert 1,200 ha of the worst land to salt-tolerant forage and rotate with fallow-leaching; (D) drip irrigation on all 4,600 ha. Wheat yield begins to fall above ECe 6 dS/m at about 7% per dS/m; cotton above ECe 7.7 dS/m at about 5% per dS/m.
1. Explain the mechanism linking the unlined canal and flood irrigation to ECe 11.2 dS/m, and why option A is likely to make matters worse.
Read the worked answer
The mechanism is secondary salinisation driven by a rising water table:
- Unlined canal plus flood irrigation deliver more water than the crops transpire → deep percolation.
- The water table rises 3.5 m → 1.9 m; above roughly 2 m, capillary rise reaches the root zone in most loams.
- Evaporation is 1,850 mm against 210 mm rainfall, so water leaves the surface by evaporation and leaves its salt behind. Salts accumulate as a stock, not a flow - this is why the problem persists after irrigation stops.
- ECe rises to 11.2 dS/m. Wheat loss = (11.2 − 6) × 7% ≈ 36%, close to the observed 38%; cotton loss = (11.2 − 7.7) × 5% ≈ 18%, close to 22%. The crop data corroborate the soil data.
- SAR 14 adds a second problem: sodium disperses clay, the surface seals and crusts, and infiltration falls. Less water enters, so salts cannot be leached - a positive feedback that accelerates degradation.
Option A fails because the district has no drainage outlet. Adding 25% more water raises the water table further, increases waterlogging, and drives more capillary rise; with SAR 14 much of the extra water ponds or runs off. More water without somewhere for it to go is the cause of the problem, not the cure.
2. Calculate the leaching requirement needed to bring the root zone to the wheat threshold, and use it to judge between options B and C.
Read the worked answer
Leaching requirement: LR = ECw / (5 × ECe − ECw) = 0.9 / (5 × 6 − 0.9) = 0.9 / 29.1 = 3.1%.
So about 3% of applied water must pass below the root zone and be carried away. Without drains there is nowhere for it to go except back into the water table, which is exactly how the district got here. Drainage is a precondition for leaching, not an alternative to it.
Option B: 800 ha × US$2,400 = US$1.92 million capital, plus 3,200 t of gypsum. Gypsum is not optional: at SAR 14, calcium must displace sodium on the clay before water can move, and the displaced sodium then has to leave through the drains. B fixes the mechanism on the land where irrigation can still pay.
Option C: converting 1,200 ha to salt-tolerant forage accepts lower value per hectare but costs little and produces fodder; however it does not lower the water table, so adjacent land keeps degrading.
Judgement: a staged combination, not a choice. B on the highest-value 800 ha, where the capital can be recovered from cotton and wheat; C on the 1,200 ha already at or beyond the economic margin, because draining land that cannot repay US$2,400/ha is poor value. Critical unresolved question: where the saline effluent goes. Disposal to a river or evaporation ponds transfers the problem downstream, and that must be resolved before construction.
3. State what you would monitor, over what period, and what result would show the reclamation is failing.
Read the worked answer
Variables: root-zone ECe sampled in 0.2 m increments to 1.0 m at fixed georeferenced points, four times a year (spring, mid-season, post-harvest, winter); SAR and infiltration rate; water table depth in piezometers across the district; drained water volume and its salinity; crop yield per hectare; and gypsum application records.
Period: reclamation of sodic, saline soil is slow - expect measurable ECe decline in two to three years and a return to the wheat threshold within five to eight. A single season tells you nothing.
Failure indicators:
- Water table does not fall below about 2.5 m despite drainage → the drains are under-designed or the canal is still leaking.
- ECe falls near the drains but rises between them → drain spacing is too wide.
- SAR stays high while ECe falls → insufficient gypsum, and infiltration may collapse further (this is the classic reclamation failure, and it is dangerous because ECe looks improved while the soil is still dispersive).
- Drained water volume is far below the calculated 3.1% leaching fraction → leaching is not actually happening.
Decision rule: if the water table has not dropped by year three, stop extending drainage and line the canal first. Fixing the seepage is cheaper per hectare than draining water that should never have arrived.