The TLE and SPG4 propagator alone should not be used for Lifetime predictions. Lifetime predictions should be done using higher accuracy propagators like LOP for Long Term propagation or HPOP for shorter term propagation. HPOP includes high levels of fidelity and includes Drag models, SRP models and high accuracy gravity models. This does requires knowledge of the vehicle, here's why. The lifetime of any resident space object is difficult to predict. The difficulty stems from the dependence on the atmospheric drag analysis, which includes calculation of density profiles, solar flux, atmosphere, the mass, surface Area relative to the atmosphere and to the Sun, its unknown/uncontrolled attitude, etc. To get a better prediction, tracking or supplemental data is necessary. Tracking of the object will provide a higher accuracy solution and it will have a better understanding of drag and atmospheric effects. With a higher quality solution, you can get a better lifetime prediction. The best solution achieved with repeated tracking of an object is about 1 rev, so +/- 90 min re-entry window. This best solution is only achieved days before the re-entry event. Here is how to use the Lifetime Tool in STK:
https://agiweb.secure.force.com/faqs/articles/Keyword/Best-practices-for-using-Satellite-Lifetime-Tool-and-HPOP-to-predict-satellite-decayHere is a code Example how to take a TLE then use Astrogator’s HPOP to predict the decay estimation:https://agiweb.secure.force.com/code/articles/Code_Sample/TLE-Satellite-Deorbit-Estimation-using-HPOPI ran some analysis on a re-entry. Iridium 96 satellite re-entered May 30th. View the image below. The black and green lines are Apogee and Perigee of the Orbit solution using only the TLE from Jan 1st 2020. The Blue and Purple are the Apogee and Perigee data for the solution that used every TLE update from January – June. During this 6 months, you can see deviations of the two solutions. The most striking part is the large deviation occurring in April/May this is when the Drag is the highest and the re-entry occured. The satellite Re-enters in May as seen in the sharp dropoff in late May, but the other solution from Jan1st's TLE predicts that it will remain in orbit for longer. IRIDIUM 96 (ID 27376)
Launched: Feb 11, 2002
ReEntered: MAY 30, 202006:17 UTC ± 1 HOURThe SGP4 propagator cannot handle re-entry prediction. Look at the below image, it shows the Jan 1st TLE propagated till it reaches 0km. The motion of the SGP4 propagator at lower altitudes is not correct. The drag is too high at lower altitudes to remain in orbit. The threshold is around 80 km where the force of drag is so high that the object is slowed and it falls down to Earth. Higher fidelity models like HPOP can show that high drag force effecting the motion.
How to get a better Re-Entry prediction: SSA centers normally use Long-term Orbit Propagator like AGI’s LOP to predict approximately when the vehicle will re-enter. Once the time of re-entry gets below 60 to 45 days, then they will switch to a High Precision propagator like AGI’s HPOP. They track this satellite often to get a better solution and update their prediction with every pass. They wait for the re-entry to occur. They know the re-entry occurred, when they cannot track it on the next expected pass, a “no show” has occurred so the vehicle re-entered. AGI's ODTK and AGI's SSA Software Suite includes these propagators and additional capability for Higher accuracy re-entry analysis.