Issue 018 - Spaceflight - Kinetic energy
How much energy did 20 reentering Starlinks release?
AP reported that SpaceX launched Starship Flight 13 on July 24, 2026, carrying 20 advanced Starlink satellites. SpaceX said the satellites would follow the same suborbital trajectory as Starship and demise on reentry roughly 20 minutes after deployment. Estimate the kinetic energy they carried into the atmosphere.
The problem
Estimate the combined kinetic energy of the 20 Starlink V3 test satellites as they begin atmospheric reentry.
Convert your answer into one or two familiar equivalents, such as gasoline, tons of TNT, or annual household electricity consumption.
Is this a genuinely large atmospheric energy release, or does it mainly look dramatic because the energy is released high in the atmosphere over a long path?
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 assumed each satellite was about 100 kg, probably less, and that they would be traveling at nearly 20,000 m/s, which was my recollection for orbital velocity. The appropriate equation is kinetic energy:
KE = 1/2 x mass x velocity^2For 20 satellites:
total mass ~= 20 satellites x 100 kg/satellite
~= 2 x 10^3 kg
speed ~= 2 x 10^4 m/s
speed^2 ~= (2 x 10^4 m/s)^2
~= 4 x 10^8 m2/s2
KE ~= 1/2 x (2 x 10^3 kg) x (4 x 10^8 m2/s2)
~= 4 x 10^11 JIf a liter of gasoline contains about 4.5 x 10^7 J, that represents a bit less than 10,000 liters of gasoline.
If monthly household energy use is about 1,000 kWh, then:
1,000 kWh ~= 1,000 x 3.6 x 10^6 J
~= 3.6 x 10^9 J/month
annual household electricity ~= 4.3 x 10^10 J/year
satellite KE equivalent ~= (4 x 10^11 J) / (4.3 x 10^10 J/year)
~= 9 household-yearsIn the grand scheme of energy being released into the atmosphere, this would not be a major contributor. Solar energy alone is about 1,400 W/m2. If only two-thirds of Earth's surface were getting direct exposure, that would be roughly:
solar power estimate ~= (3 x 10^13 m2) x (1.4 x 10^3 J/m2/s)
~= 4.2 x 10^16 J/sThat is about 100,000 times as much energy as the satellite reentry every second.
Calibration Score
Matt's Calibration Score: 70 / 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. Satellite mass was low and velocity was high, but the errors canceled.
Model: 30/30. Kinetic energy was exactly the right model.
Math: 10/10. The arithmetic was clean.
Result: 20/30. The final energy estimate landed in the corrected order of magnitude.
Grounding facts
A ton of TNT equivalent sounds alarming, but the delivery shape matters. A few hundred tons of TNT released as a compact surface explosion is a very different event from the same order of energy smeared out through thin upper atmosphere over many kilometers of reentry path.
Velocity dominates kinetic-energy problems because it is squared. Being off by 2x in speed changes energy by 4x. Being off by 3x changes energy by 9x. That is why the wrong velocity memory peg can do real damage, even when the equation is right.
After checking sources
Check and recalibrate
The structure was right: kinetic energy is the whole game. The two biggest corrections point in opposite directions. Full-size Starlink V3 satellites are far larger than Matt's 100 kg starting point; public reporting and V3 spec summaries put them around the 1 to 2 metric ton range. But orbital speed is about 7 to 8 km/s, not 20 km/s, and Flight 13's payloads were on a suborbital trajectory.
Use a central estimate of 2,000 kg per satellite and 7 km/s:
satellite count ~= 20
mass per satellite ~= 2 x 10^3 kg
total mass ~= 20 x 2 x 10^3 kg
~= 4 x 10^4 kg
reentry speed ~= 7 x 10^3 m/s
speed^2 ~= (7 x 10^3 m/s)^2
~= 4.9 x 10^7 m2/s2
~= 5 x 10^7 m2/s2
KE ~= 1/2 x (4 x 10^4 kg) x (5 x 10^7 m2/s2)
~= 1 x 10^12 JA reasonable Fermi bracket is about 4 x 10^11 to 1.5 x 10^12 J, depending mostly on the assumed satellite mass and entry speed. Matt's answer of 4 x 10^11 J is low, but still in the right order of magnitude because the mass underestimate and velocity overestimate partly cancel.
Now convert. EIA's motor-gasoline heat-content peg is about 120,000 Btu per gallon, or about 1.3 x 10^8 J/gallon. That makes the corrected energy:
gasoline equivalent ~= (1 x 10^12 J) / (1.3 x 10^8 J/gallon)
~= 8 x 10^3 gallons
~= 3 x 10^4 litersFor TNT equivalent, use 1 ton TNT as 4.184 x 10^9 J:
TNT equivalent ~= (1 x 10^12 J) / (4.2 x 10^9 J/ton)
~= 2.4 x 10^2 tons TNTFor household electricity, EIA's current household peg is about 10,500 kWh per year:
annual household electricity ~= 1.05 x 10^4 kWh/year
x 3.6 x 10^6 J/kWh
~= 3.8 x 10^10 J/year
household-year equivalent ~= (1 x 10^12 J) / (3.8 x 10^10 J/year)
~= 2.6 x 10^1 household-yearsThe important context: 10^12 J is huge by everyday standards and tiny by planetary-atmosphere standards. It is hundreds of tons of TNT equivalent, but it is not released at one point like a bomb. It is spread across 20 separate objects, high altitude, minutes of reentry, long paths through thin air, fragmentation, heating, radiation, and ablation.
Compared with solar input, it is negligible. Earth intercepts sunlight on the order of 10^17 joules per second, so 10^12 J is only a few microseconds of incoming solar energy. The reentry can look dramatic because it concentrates visible heating into bright streaks, not because it meaningfully heats the global atmosphere.
Post-check reflection
Matt's reflection
I underestimated the Starlink satellite mass by at least an order of magnitude, which would have been a bigger issue if my recollection of orbital velocity had not been so bad. I am not sure why I was off by so much. The two errors mostly canceled out.
The corrected number was near enough to what I estimated that it did not have a big impact on my impression of the news item. In terms of how much energy it introduces to the atmosphere, this really is not a big deal.
Recommended memory peg
Remember low-Earth-orbit speed is about 8 km/s, 1 kWh = 3.6 x 10^6 J, 1 gallon of gasoline is about 1.3 x 10^8 J, and 1 ton TNT is about 4.2 x 10^9 J. For reentry problems, start with KE = 1/2mv^2 and treat speed as the fragile assumption.
Reader results
Bars show how submitted estimates sort into the answer choices from the gut-check prompt.