Issue 017 - Agricultural biosecurity - Biological supply chain
How hard is it to distribute 500 million sterile flies?
Reuters reported that the U.S. is testing new fly-production methods in the rush to fight New World screwworm, a flesh-eating livestock pest. USDA currently disperses about 100 million sterile flies per week, while researchers estimate roughly 500 million per week may be needed to suppress the expanding infestation.
The problem
Estimate the total mass and packed transport volume of 500 million adult sterile New World screwworm flies.
Then estimate how many aircraft release flights would be required to distribute them each week.
At this scale, is producing the flies likely to be harder than physically transporting and releasing them?
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.
Before checking sources
Matt's first pass
I had to make a lot of assumptions for this one. First, for flies, I assumed they were about 10^-4 kg each and could be safely stored in about 1 cubic centimeter for each fly.
total flies ~= 5 x 10^8
mass per fly ~= 1 x 10^-4 kg/fly
total fly mass ~= (5 x 10^8 flies) x (1 x 10^-4 kg/fly)
~= 5 x 10^4 kg
~= 50 metric tonsFor volume, there are 1 million cubic centimeters in a cubic meter. If one fly needs 1 cm3, then 1 m3 can safely contain 1 million flies.
packed volume ~= (5 x 10^8 flies) / (1 x 10^6 flies/m3)
~= 5 x 10^2 m3Next, I assumed I did not have to pay much attention to the geographical distribution of those flies, which is probably not true, and that all that matters is efficient aircraft distribution. I assumed a large cargo aircraft might hold about 50 m3 of containers on average.
cargo flights ~= (5 x 10^2 m3) / (50 m3/flight)
~= 10 flightsIf all we are asking is the cost of loading and releasing them from flights, and not careful distribution across some area, then I think producing 50 metric tons of flies is the more challenging task.
Calibration Score
Matt's Calibration Score: 65 / 100
Higher is better: earn points for accurate pegs, sound models, correct math, and a result close to the sourced answer. The image shows percent full of it: 100 minus the Calibration Score.
Pegs: 10/30. Fly mass was close enough, but packing density and aircraft-release assumptions were weak.
Model: 30/30. Count times mass, volume per organism, and release flights is the right model.
Math: 5/10. The geographic-distribution constraint was knowingly skipped.
Result: 20/30. The final production-versus-transport conclusion stayed broadly right.
Grounding facts
Five hundred million is an enormous count, but for fly-sized insects the mass is closer to truckload scale than freight-train scale. At 20 mg each, 1 million flies is about 20 kg, so 500 million flies is about 10 metric tons.
The operational surprise is that volume and mass are not the only constraints. If releases must be spread over a broad barrier zone or infestation front, flights become about coverage and timing. A single huge aircraft loaded to the ceiling is not automatically useful if the flies need to be released evenly across many flight lines.
After checking sources
Check and recalibrate
Matt's fly-mass estimate was high but still Fermi-useful. Screwworm flies are in the same broad size class as large housefly-like insects. A good rough mass range is about 20 to 50 mg per fly, or 2 x 10^-5 to 5 x 10^-5 kg.
total flies ~= 5 x 10^8
mass per fly ~= 2 x 10^-5 to 5 x 10^-5 kg
total mass ~= (5 x 10^8) x (2 x 10^-5 to 5 x 10^-5 kg)
~= 1 x 10^4 to 2.5 x 10^4 kg
~= 10 to 25 metric tonsThat is a lot of biomass, but it is not a cargo-airlift problem by itself. The volume correction is larger. USDA/DVIDS release-operation material describes dispersal boxes holding about 2 million chilled flies. That implies:
release boxes ~= (5 x 10^8 flies) / (2 x 10^6 flies/box)
~= 2.5 x 10^2 boxesIf each box is roughly a large cooler or tray-stack container, the packed transport volume is likely tens of cubic meters, not 500 m3. Matt's 1 cm3 per active fly assumption might be defensible for loose active insects, but actual release logistics use chilled, immobilized flies packed into containers.
The flight estimate changes for a different reason: a release aircraft is not just a cargo plane. Sterile flies have to be distributed across a target zone, often along flight lines, at a useful density. Payload is only one constraint; route coverage, release rate, aircraft speed, weather, staging airports, crews, and daily operations all matter.
If one release sortie carries and disperses about 2 to 10 million flies, then:
release sorties ~= (5 x 10^8 flies/week) / (2 x 10^6 to 1 x 10^7 flies/sortie)
~= 50 to 250 sorties/weekA central Fermi answer is therefore about 100 release flights per week, with a broad plausible range from a few dozen to a few hundred depending on aircraft capacity and release geography. Current operations around tens of millions of flies per week scale naturally into a small-fleet operation if the target rises to 500 million per week.
So: physical transport and release are real logistical work, but production still looks like the harder bottleneck. USDA's public materials describe major investments to increase sterile-fly production toward the 500-million-per-week level, and the Reuters hook is specifically about production methods being tested to close that gap. The planes matter, but the biology factory is probably the harder system to scale.
Post-check reflection
Matt's reflection
I overestimated fly mass but was close enough for Fermi work. My space required to safely keep them was a massive overestimate, but could be fine depending on the state of the flies: pupae, sedated or chilled adults, and so on.
I also underestimated the number of flights necessary, but I knew I was doing so. I anticipated that the geographic distribution would make maximal packing of a large plane impossible, and that the job would need to be accomplished using a larger number of smaller planes.
Otherwise, my mental models were pretty close. This does not have a big impact on my impression of the news item.
Recommended memory peg
Remember a fly-sized insect is roughly 10 to 50 mg, and 1 million 20-mg insects weigh about 20 kg. For sterile-insect releases, use sorties ~= total insects / insects released per sortie; for screwworm, a useful public logistics peg is about 2 million chilled flies per dispersal box.
Reader results
Bars show how submitted estimates sort into the answer choices from the gut-check prompt.