Phosphate industry wastewater treatment in Morocco is changing because of one decision: OCP is preparing to move phosphogypsum from sea outfalls to a land storage facility at Jorf Lasfar. Until now, the acidic water that carries phosphogypsum there has gone to sea with it. Once gypsum is stored on land, that water stays on site, and fluoride, phosphate, acidity and rainfall all become a water-treatment problem.
This page explains the main water streams in a phosphate and fertilizer chain, what is known about each in Morocco, and the treatment logic that applies wherever phosphoric acid is made. Brine Consulting, an Amsterdam-based independent industrial water advisory firm, sells no treatment equipment, so it can size a gypsum-stack water surplus, or show that there is none, without a plant to sell; it advises phosphate and fertilizer sites remotely or through a defined engagement.
Where the water goes in a phosphate chain
| Stage | Moroccan sites | What the water picks up | Where it usually ends up |
|---|---|---|---|
| Mining and beneficiation (washing, flotation) | Khouribga, Benguerir, Youssoufia | Fine clays and phosphate particles, flotation reagents | Thickened and recycled; make-up water increasingly from desalination or treated municipal wastewater |
| Phosphoric acid (wet process) | Jorf Lasfar, Safi | Acidity, phosphate, fluoride, sulphate, metals released from the rock | Recycled in the acid plant; the excess leaves with the phosphogypsum |
| Phosphogypsum handling | Jorf Lasfar, Safi | The same acidic process water, carried as slurry | The sea, or a land stack with a process-water pond |
| Gas scrubbing in acid and fertilizer plants | Jorf Lasfar, Safi | Fluorine compounds, ammonia | Recovery as fluorosilicic acid where economic; otherwise neutralisation |
| Utilities | All sites | Cooling and boiler blowdown, desalination brine | Sea outfall at the coast; reuse or evaporation inland |
The table is a simplification. In our view, two streams dominate the treatment problem: the phosphogypsum water and, where membranes are used, the salt that ends up in their reject.
Phosphogypsum: the stream that sets the scale
The wet process makes phosphoric acid by attacking phosphate rock with sulphuric acid, and calcium sulphate (phosphogypsum) is the by-product. The US EPA cites an industry estimate of 5.2 tonnes of phosphogypsum for every tonne of phosphoric acid, and notes that about 80% of the radium-226 in the rock ends up in it.
For the phosphoric acid unit on the El Jadida coast, a 2023 study reported about 20 million tonnes of phosphogypsum a year, discharged untreated into coastal waters as a liquid slurry (Arhouni et al., Journal of Radioanalytical and Nuclear Chemistry, 2023). The same study measured fluoride in the liquid effluents at 3.28 to 927 mg/L, found that cadmium largely transfers to the phosphoric acid rather than the gypsum, and recorded that OCP wants to stop discharging phosphogypsum to the sea. Worldwide, it cites roughly 58% of phosphogypsum as stacked and 28% as discharged to coastal waters.
That plan now has financing. In May 2026 the IFC approved and signed a loan of up to EUR 95 million towards an inland phosphogypsum storage facility at Jorf Lasfar, with a total cost of about EUR 190 million, a capacity of up to 22 million tonnes and a height of 60 metres, intended to store phosphogypsum “instead of discharging it into the ocean via outfalls” (IFC project 51794).
Does moving phosphogypsum to land create a wastewater problem?
Only if the water balance has a surplus, and on a large stack the slurry water and rain make a surplus a real possibility. Public documents give the storage capacity but not how much slurry water the facility will handle, so the hypothetical balance below shows why that water is the real design question.
Assume, for illustration, that phosphogypsum is pumped as a slurry of 25% solids by weight. Each tonne of gypsum then arrives with about 3 tonnes of water. (If the gypsum is instead filtered and trucked to a dry stack, far less water arrives with it and rainfall becomes the main term; the public documents do not say which method is planned.)
At the 20 million tonnes a year reported for the El Jadida unit, that is about 60 million m³ of water a year, or roughly 164,000 m³ a day, passing through the stack system. Most of it must be decanted and returned to the acid plant.
Filling the full 22 million tonne capacity would bring about 66 million m³ of slurry water through the facility. On these two figures, the facility holds about 13 months of the reported output of one unit; the public documents do not say what share of Jorf Lasfar’s output it will take.
Rain adds to the inventory: every 100 mm of rain on every 100 hectares of stack and pond is 100,000 m³ of extra acidic water, before evaporation is subtracted.
For scale, OCP’s J2K pipeline is built to carry up to 80 million m³ a year of desalinated water from Jorf Lasfar inland (IFC project 49203). A slurry loop of similar size, mostly recycled but with any surplus needing treatment before it leaves the site, is not a side issue.
The practical consequences for a land stack, at Jorf Lasfar or anywhere else:
The process water balance decides whether the site ever needs to discharge. If rain and slurry water exceed what the acid plant and evaporation can take back, the surplus must be treated. Size that surplus from local rainfall and the stack’s footprint before anything else.
The pond water is not ordinary wastewater. It is acidic and carries phosphate and fluoride. Neutralising it with lime precipitates calcium fluoride and calcium phosphates and makes a sludge that also needs a home.
Seepage and lining become permanent obligations. The US EPA counts more than 70 phosphogypsum stacks in the US, from 2 to 324 hectares and 3 to 60 metres high. A stack is a long-lived structure, so its water management has to be planned for closure as well as operation. Florida’s Piney Point shows the cost of getting that wrong: after a leak in the stack in April 2021, operators released more than 200 million gallons (about 760,000 m³) of nutrient-rich process water into Tampa Bay to stop it bursting, and the closure now relies on a 3,300-foot injection well taking about 1 million gallons a day (WUSF, April 2023).
Fluoride and cadmium
Fluoride comes from the rock and leaves partly in the gas scrubbers and partly in the process water. Lime or calcium dosing removes most of it as calcium fluoride, but the solubility of calcium fluoride limits how low a single precipitation step can go, so strict limits usually need a second stage or polishing. Our guide to fluoride removal from wastewater compares the options. Where scrubber liquor is concentrated enough, recovering fluorosilicic acid as a product usually beats neutralising it, in our view.
Cadmium follows the phosphoric acid more than the gypsum, according to the 2023 El Jadida study, so it matters more for product quality than for the gypsum water. Where cadmium does reach a discharge, Morocco’s general limit is 0.2 mg/L. Precipitation and sulphide treatment are covered in our guide to heavy metal removal.
Which Moroccan discharge rules apply?
Discharges to rivers, wadis and groundwater at the inland mining sites need an authorisation from the basin agency under Law 36-15 and must meet the general limits of joint order 2942-13 (in force since 1 January 2018) unless a sector order applies. Those general limits include conductivity of 2,700 µS/cm and cadmium of 0.2 mg/L, and they also list fluoride (joint order 2942-13, BO 6202).
Discharges to the sea from Jorf Lasfar and Safi fall under Law 81-12 on the coastline, which requires an authorisation of up to five years and a discharge register. The decree that will set numeric marine limits was still under review in February 2026 (Le360, 27 February 2026), and we found no published version by 4 November 2026.
Our page on industrial wastewater treatment in Morocco sets out these rules route by route, with dates, and our North Africa overview covers Egypt and the wider region.
OCP’s water programme: desalination and reuse
OCP’s water subsidiary, OCP Green Water, reported 320 million m³ of installed non-conventional water capacity in 2025, a target of 610 million m³ by 2027, and that the group’s goal of relying on non-conventional water had been met two years ahead of the 2027 deadline (OCP Green Water, via IDRA, 14 July 2025). The figures below are as OCP and its lender report them:
Jorf Lasfar to Khouribga: desalinated water travels inland through the J2K pipeline, 219 km long according to the IFC, with a capacity of up to 80 million m³ a year (IFC project 49203).
Safi: a new desalination plant was still under development in July 2025, and the Safi to Gantour pipeline it is meant to feed was slated for commissioning in 2026.
Gantour (Benguerir and Youssoufia): supplied with treated wastewater from the Marrakech treatment plant since 15 June 2025, over more than 80 km.
Cities: OCP Green Water also supplies drinking water to Safi, El Jadida and southern Casablanca. The ministry reported in July 2025 that the El Jadida and Safi plants had run partly since 2023 and should reach full capacity by the end of 2026 (Hespress, 15 August 2025).
Two consequences for wastewater work. Washing plants fed with treated municipal effluent inherit its nutrients, salts and variability, so the reuse water needs a specification, not just a volume. And every desalination plant returns brine to the sea; that side of the programme is covered on desalination brine in Morocco.
Choosing a treatment route, stream by stream
| Stream | What makes it hard | Usual first step | When membranes or evaporation come in |
|---|---|---|---|
| Beneficiation water | Fine clays that settle slowly | Thickening and flocculation, then recycle | Rarely; salt builds up only if make-up water is saline |
| Phosphogypsum pond water | Acidity, fluoride, phosphate, very large volumes | Return to the acid plant; lime neutralisation of the surplus | When the surplus must meet a salinity limit or be reused; scaling from calcium sulphate and calcium fluoride makes this a pilot-first decision |
| Scrubber liquors | Fluorine compounds, ammonia | Recovery as fluorosilicic acid or ammonium products where concentrations allow | Rarely needed if recovery works |
| Cooling and boiler blowdown | Salts concentrated by evaporation | Reuse within the site | When an inland site has no discharge route |
| Treated municipal effluent used as make-up | Nutrients, salts, variability | Agreed quality specification with the supplier | When the process needs low-salinity water |
Pond water rarely goes straight to reverse osmosis: it is acidic and saturated with calcium sulphate and calcium fluoride, which scale membranes quickly, so neutralisation and testing come first. In our view, the size of the surplus, the lime demand and the sludge it makes drive the treatment cost more than the choice of membrane or evaporator.
Phosphorus in these streams can also be a resource: our post on phosphorus recovery from wastewater covers where recovery is realistic. For any stream where salt is the binding constraint, high-TDS wastewater treatment sets out the options.
What an independent review covers on a phosphate site
We work from the data a site shares, remotely or through a defined engagement:
a whole-site water balance including the gypsum circuit, rainfall and evaporation;
screening of neutralisation, fluoride recovery, membrane and evaporation options, with treatability and pilot testing where scaling risk is high;
review of supplier proposals and lender documents for water and brine scope.
Useful first data: acid plant and gypsum circuit flows, a year of pond water analyses with fluoride, phosphate and sulphate, rainfall records, and the current discharge authorisation.
Before the stack design is frozen
The surplus question is easiest to settle before the stack lining, pond volume and return lines to the acid plant are fixed. If you are planning a gypsum stack, a new acid line or a switch to reuse water, book a call about your water balance, send us your flows and pond analyses or email christos@brineconsulting.com.
Written by the Brine Consulting advisory team. The Brine Consulting advisory practice includes Edwin Muller, External Advisor, a veteran industrial water expert with more than 25 years of experience across water, energy, and heavy industry, including major roles in refineries, gas, petrochemicals, power generation, hydrogen, and ammonia projects, and Osis G. Kalache, P.Eng., Senior Consultant, Wastewater Process Engineering & Optimization, a senior wastewater process engineering and optimization expert with more than 17 years of experience in industrial and conventional wastewater treatment. Meet our experts.