Issue 067 - Water infrastructure - Mass-flow logistics
Can water tankers replace a city's water network?
UNICEF says Gaza emergency water service has relied on a mix of trucked water, desalination plants, wells, and limited network supply. Suppose 2 million people depended entirely on tanker-truck delivery.
Energy and infrastructureAbout 1 minute
Sources checked October 2, 2026. Figures and circumstances may have changed.
The problem
If 2 million people depended entirely on tanker trucks for drinking, cooking, and basic hygiene water, about how many tanker deliveries per day would be needed?
Then consider whether trucking can substitute for a piped water network for a full year.
Because Fermi problems target an order of magnitude, I normally use no more than two significant digits and write most calculations in scientific notation; the Fermi reference explains both conventions.
Grounding facts
The Sphere humanitarian water standard uses 15 litres per person per day as a survival-level allocation for drinking, cooking, and personal hygiene, and says higher quantities up to about 20 litres should be targeted where needs are greater.
UNICEF reported that one Gaza filling point had previously supplied 2 million litres per day before operations there were suspended, and that UNICEF and partners were replacing that volume from desalination plants at substantial daily cost.
For truck capacity, a practical round value is 10,000 litres for a medium-to-large water tanker. Some larger tankers can carry closer to 20,000 litres, but 10,000 litres is a useful conservative Fermi peg.
After checking sources
Checked answer and calculation
Start with the Sphere minimum:
daily emergency water
~= 2 x 10^6 people x 15 L/person/day
~= 3 x 10^7 L/day
~= 30 million L/dayWith 10,000-litre tanker deliveries:
deliveries per day
~= 3 x 10^7 L/day / 1 x 10^4 L/delivery
~= 3 x 10^3 deliveries/dayIf larger 20,000-litre tankers dominated, the answer would drop to about 1,500 deliveries per day. So a reasonable checked range is 1,500 to 3,000 tanker deliveries per day for emergency-level water.
The truck count depends on cycle time. If one truck manages three full delivery cycles per day, that is still on the order of:
trucks needed
~= 3,000 deliveries/day / 3 deliveries/truck/day
~= 1,000 tanker trucksOrdinary household water use is much larger. EPA says the average American uses about 82 gallons per day at home, or roughly 300 litres per person per day. At that level:
ordinary household water
~= 2 x 10^6 people x 3 x 10^2 L/person/day
~= 6 x 10^8 L/dayThat is about 60,000 ten-thousand-litre deliveries per day. So trucking can plausibly meet emergency survival volumes if logistics, fuel, security, filling points, and distribution sites function. It cannot provide normal city-scale household water use for 2 million people in anything like an ordinary way.
Before checking sources
Matt's original estimate
This is the unverified estimate Matt wrote before checking sources, not the checked answer.
I assumed the absolute basic amount of water necessary to cover drinking, cooking and hygiene should be about 12 L daily per person. If there are 2 million people that need to be supported, that's 12 L x 2 x 10^6 people = about 2.4 x 10^7 L daily.
Next, I assumed a truck might be able to carry on average about 10 tons of water, maybe as much as 20 tons, which works about to between 10^4 and 2 x 10^4 L of water. Dividing that into the daily need, that means between 1.2 and 2.4 x 10^3, or 1,200 to 2,400, truckloads of water would be necessary daily. That's an enormous undertaking, and I don't think it would be possible to accomplish once, much less daily all year.
I expect the water needs will have to be supplemented by reliable piping in order to satisfy the needs of that many people.
Reasoning score
Matt's reasoning score: 90 / 100
Higher is better: earn points for useful facts, a sound reasoning approach, correct math, and a final estimate close to the sourced answer. The owl meter shows percent full of it: 100 minus the reasoning score.
Useful facts: 30/30. The 12 L/person/day assumption was very close to the 15 L Sphere minimum, and the 10,000 to 20,000 L tanker-capacity range was right for this scale of estimate.
Reasoning approach: 30/30. The model was exactly the right one: people times litres per person per day, divided by litres per tanker trip.
Math: 10/10. The arithmetic followed correctly from the assumptions.
Final estimate: 20/30. The delivery-count estimate landed in the checked range, but the original interpretation was too pessimistic about whether emergency-level trucking could be sustained. It is possible at massive effort, though far below ordinary household water service.
Post-check reflection
Matt's reflection
Looks like my estimates and math were pretty close to the real world answers, so I was properly calibrated in terms of the volume.
Where I went wrong was my assumptions about how viable this is: sounds like this is very close to the volume and pace of water aid Gaza has been receiving, so it is possible. It does fall very short of standard household water usage, so we shouldn't aim to keep people at this level, but at least basic needs will be met.
Recommended memory peg
For emergency water logistics, remember 15 L/person/day as the Sphere survival-level target and 10,000 L as a useful water-tanker delivery size.
Reader results
Bars show how submitted estimates sort into the answer choices from the gut-check prompt.