Last week, I attended CalDesal’s Sacramento conference where I gave a paper on the Top Ten Desalination Disasters of all time. Although the title is somewhat incendiary, it raises a serious question: what is the best way of managing risk for big-ticket infrastructure projects?
For the record, my Top Ten Disasters are:
- Palm Jumeirah, UAE (2007) – A demand miscalculation; they built it, no one came
- Jeddah 1, Saudi Arabia (1966) – Acid attacked MSF plant
- Santa Barbara, California (1992) – Rained out
- Point Lisas, Trinidad (1999) – EPC costs spiral and parties dispute
- Ad Dur RO, Bahrain (1990) – Pretreatment failure for large SWRO
- Carlsbad, California (1998-?) – 14 years in development and startup date still uncertain
- Tampa Bay, Florida (2002) – Technical, commercial foul-ups
- Carboneras, Spain (2002) – Farmers failed to pay for desalted water
- Lok On Pai Plant, Hong Kong (1972) – A big, stand-alone MSF that never ran
and, the number one desal disaster is:
- Victorian Desalination Plant, Wonthaggi, Australia (2012) – $1 billion EPC cost overrun
I also have a long list of older desalination plants — both thermal and membrane plants — which also failed spectacularly, mostly because the technology did not perform according to plan, but the amounts of money involved were smaller, so they don’t make the top ten.
Before I get deluged with complaints from those involved in building these plants, I should point out that Wonthaggi, Carboneras, Point Lisas and Palm Jumeirah are all great plants and deserve all of the accolades that they have received from GWI, WDR and others. Carlsbad has every opportunity of becoming a great plant as well (but the cost over-run on permitting should not be ignored). Ad Dur and Tampa Bay are now great plants, but they did have their difficulties.
The point that I’m trying to make is that the biggest risk in the desal business is not on the technology or operations side of a project—it is on the demand side. Hong Kong, Carboneras, Santa Barbara and Palm Jumeirah all entailed a significant waste of money because the demand for water never came as expected, and as a result, tens of millions of dollars were wasted.
Contractual arrangements, such as the BOT model, can effectively manage risk on the supply side. You can still have EPC cost blow-outs and you can have operating problems, but these can be managed to protect the client.
For example, the cost over-runs at Point Lisas did not have a significant impact on the Water and Sewerage Authority of Trinidad and Tobago’s profit and loss account because the developers picked up the tab. However, nothing could have protected WASA if there had been no water demand in the Point Lisas industrial estate.
The fundamental dilemma that must be faced when building a desal plant is that while water demand will vary, the infrastructure necessary to build a desal plant is fixed. If it rains and the reservoirs are full, the desal plant is switched off. But the interest payments on the hundreds of millions of dollars spent to build a tunneled intake will continue.
If a water agency contracts a desalination plant that it does not use, it ends up wasting a whole lot of money, no matter what happens. The situation could be even more dire if there is no desal plant available to pick up the slack if it stops raining. Whoever finds a way of effectively managing that risk is sure to make a lot of money.Christopher Gasson