Popocatépetl Ash — the 31 Aug–1 Sep Pulse Closed Inside Twelve Hours, No Airport Inside a Reported Ash Area, September 2026
- The pulse is over. Three advisories in 11 hours 26 minutes — 31 Aug 22:37 UTC, 1 Sep 04:08 UTC, then a 1 Sep 10:03 UTC advisory reporting no ash detected. Popocatépetl keeps emitting in pulses, so check your airline on the day; wind moves ash.
- No airport on our exposure list fell inside any reported or +6-hour forecast ash area during the pulse — Puebla (PBC) 30 km, Mexico City (MEX) 66 km, Felipe Ángeles (NLU) 90 km, Toluca (TLC) 106 km. No ash-related closure was in effect at any of them.
- Reported ash tops were FL210 then FL200 (~21,000 and ~20,000 ft). Mexico City's field sits at 7,316 ft, so arrivals and departures work through that band in the terminal area — which is why the reported flight level, not just the 66 km, is the number dispatchers plan around.
- En-route traffic never moved. On 31 August, the only complete day containing the pulse, whole-airspace flight counts ran 1.04× expectation over MMEX and 1.07× over MMFR. There was no drawdown at MEX to recover from — which is why this close-out carries no recovery curve.
Advisory timeline — how active is this episode
Popocatépetl is a persistent emitter, not an eruption-headline event: VAAC Washington has issued 109 ash advisories on 37 separate days over the last 90, in irregular pulses rather than a steady stream. The monthly base rate is remarkably flat — our record covers three complete months so far and each one landed on exactly 39 advisories (June across 14 days, July across 11, August across 13). What was unusual in late August was the gap structure, not the volume: after the 15–17 August pulse the emitter went quiet for seven days — its longest run without an advisory since early July — produced three advisories on 25–26 August, then fell quiet for another four before restarting at 22:37 UTC on 31 August. That restart is the pulse this page tracked, and it is now closed: three advisories in total (6 across 4 days in the last fortnight), reported ash tops of FL210 then FL200, and a 10:03 UTC advisory on 1 September reporting no ash detected. Daily advisory count, 15 August – 1 September (zero = no advisory issued that day):
Fig. 1 — Popocatépetl daily VAAC advisory count, VAAC Washington (public) · * partial day · Data as of 1 Sep 2026, 16:18 UTC · Source: VAAC Washington advisories (public).
Airports in the exposure zone
| Airport | Size | Distance from vent | FIR |
|---|---|---|---|
| Hermanos Serdán (Puebla) (PBC) | medium | 30 km · outside every reported ash area | MMFR |
| Mexico City Benito Juárez (MEX) | large | 66 km · outside every reported ash area | MMFR |
| Felipe Ángeles Intl (NLU) | large | 90 km · outside every reported ash area | MMFR |
| Adolfo López Mateos (Toluca) (TLC) | medium | 105 km · outside every reported ash area | MMFR |
Distance alone does not determine impact — wind direction and the flight levels the ash reaches decide which airports and routes are affected on a given day. FIR = Flight Information Region.
What our monitoring showed: the pulse opened and closed between two complete traffic days
FlySafe tracks observed departures and arrivals at 493 airports from crowdsourced ADS-B receiver data. Observed counts are a floor, not a total — receiver coverage varies — but an airport measured against its own baseline is a reliable shape. Mexico City Benito Juárez's last 30 days:
Fig. 2 — Observed daily movements at MEX, 1 Aug – 30 Aug (latest complete UTC day). Data as of 1 Sep 2026, 16:18 UTC · Source: FlySafe monitoring (ADS-B receiver network).
- 30-day observed window at Mexico City (MEX), 1–30 August. Across the 28 days that are not receiver-coverage days, the airport averaged ~862 movements/day. The last complete day, 30 August, reads 737 — −15% against that mean, and −8% against the same-weekday median (~800/day) our airport monitor uses, which classifies the day as normal.
- Two days in the window are coverage, not traffic. 29 August (374) is flagged automatically by our global feed-health rule; 26 August (660) is not, but shows the same shape. Both read low at nearly every airport on this feed on exactly those dates — Felipe Ángeles (NLU) 67 and 42 against a ~112 norm, Toluca (TLC) 5 and 4 against ~27, Tijuana (TIJ) 202 and 139 against ~298, and Guatemala City (GUA, another country) 327 and 171 against ~384 — while FIR-level en-route counts over central Mexico ran 1.02× and 1.01× expectation on the same two days. Aircraft do not stop flying in Guatemala because of a volcano in Mexico; a receiver feed does thin out everywhere at once.
- The remaining drift is country-wide and modest. Comparing 2–15 August with 24–30 August and excluding those two coverage days, observed movements fell 8% at MEX — against 14% at Cancún (1,247 km from the vent), 10% at Tijuana (2,367 km), 5% at Monterrey and 4% at Toluca, while Felipe Ángeles was flat and Guatemala City rose 5%. MEX is not the outlier in that set. We have not measured a cause for the wider decline — the end of the Mexican summer school holidays is the obvious candidate, and receiver coverage varies too — but the geography rules the volcano out: a hazard 66 km from Mexico City does not reach 2,367 km to Tijuana.
- Honesty note — why this curve stops on 30 August. The pulse began at 22:37 UTC on 31 August and the closing advisory landed at 10:03 UTC on 1 September, so it opened after the last complete movement day in this feed and closed before the next one settled. No during-pulse airport movement day exists to publish, and we will not draw a partial one. The measurement that does cover the pulse is the FIR-level en-route count for 31 August, in the section below. These are observed movements from a crowdsourced ADS-B receiver network — a floor, not the airport's official total.
That is the whole point of holding an airport against its own baseline and then checking it against neighbours: the number that looked like impact here is a country-wide pattern with two coverage-thin days inside it. Live view: MEX airport brief.
What it means for flights
A restart is not an escalation — read the gap structure, not the headline. Persistent emitters do not disrupt in bursts the way an eruption spike does; they redraw the ash map a little, in pulses, for months. Popocatépetl's volume has been almost mechanically stable: exactly 39 advisories in each of June, July and August, spread over 14, 11 and 13 days respectively, and 109 on 37 days across the last 90. What moved in late August was the spacing — a seven-day gap (18–24 Aug), the emitter's longest run without an advisory since early July, then two days of emission and a four-day gap (27–30 Aug). The 31 August restart resumed that pattern and ran for three advisories before the ash stopped being detectable; on this record it was a normal pulse boundary, not a step change. The number that would signal something different is the cadence itself: a sustained 10-plus advisories a day is a different volcano than the one this chart shows.
Why FL200–FL210 carries more information than the 66 km. Ash reported to FL210 and then FL200 (~21,000 and ~20,000 ft) sits above what most persistent emitters produce, and it lands in an awkward place for central Mexico specifically. High-altitude traffic crossing the country at cruise passes above the reported layer and is largely unaffected. But Mexico City's field elevation is 7,316 ft, and Puebla and Toluca are higher still — so departures and arrivals spend their climb and descent inside the band the advisories describe, in the terminal area rather than en route. That is why regional dispatchers watch the reported flight level and the drift direction in each advisory, not the distance from the summit alone. The 31 August advisory put the ash moving north at 5 kt; the 04:08 UTC advisory on 1 September had it drifting west of the summit with the forecast winds veering from the south-east over the following six hours — the kind of detail that decides which arrival streams get re-planned on a given morning.
What the FIR-level data says — and what it does not. Across the Mexican FIRs MMEX and MMFR, our observed-versus-expected en-route flight ratio ran between 0.92 and 1.07 (MMEX) and 0.94 and 1.26 (MMFR) on every complete day from 25 to 31 August. That is a FIR-level measurement — whole-airspace, not per-route — and it says there was no broad en-route avoidance signature over central Mexico across the pulse. Automated multi-signal analysis has likewise registered no volcano-category airspace-disruption event in MMEX since 7 August. Neither of those is a statement about any individual flight: a single arrival can still be re-sequenced or a runway direction changed on a day when the whole-FIR count reads normal. 1 September is excluded from that range because it is still in progress — a partial day is not a daily ratio.
The traffic dip you would misread as impact. Observed movements at MEX are down about 15% on the last complete day against the 30-day mean, and a volcano 66 km away makes that look causal. It is not, and the check is a control group. Comparing 2–15 August with 24–30 August and excluding two coverage-thin days, MEX fell 8% — while Cancún fell 14% (1,247 km from the vent) and Tijuana 10% (2,367 km), Monterrey 5%, Toluca 4%, Felipe Ángeles was flat and Guatemala City rose 5%. Whatever drives that country-wide softening — the end of the Mexican summer school holidays is the obvious candidate, and we have not measured it — it is not a plume 66 km from one of those airports. The two thin days themselves (26 and 29 August) read low at nearly every airport on this feed, in Guatemala as well as Mexico, while en-route counts held near expectation on both — the shape of receiver coverage, not of aircraft staying on the ground. We publish the raw counts and the correction together, because a number that would be wrong to cite is worse than no number.
What we watch for escalation. Three signals would reopen this page: an advisory-cadence inflection — the current 1–4 a day becoming a sustained 10-plus; reported ash tops climbing well above FL210; or advisory ash areas reaching the MEX, PBC or NLU terminal areas. Our detection layer checks all three every three hours alongside the observed-movement counts and the FIR-level en-route ratios, and this page, the MEX brief and the live ash map all update from the same cycle. Until one of those thresholds moves, the operating decision on any given day stays where it belongs — with the airport, the airline and the Mexican authorities.
What the data showed — pulse closed 1 Sep. We close this pulse on the two signals that opened it: how high the ash was reported, and whether it reached an airport. The pulse ran 11 hours 26 minutes, from the 22:37 UTC advisory on 31 August to the 10:03 UTC advisory on 1 September that reported ash not detectable on satellite products. It produced three advisories and two ash-bearing ones, with reported tops of SFC/FL210 then SFC/FL200 — and all three carried “no volcanic ash expected” in every forecast layer they published. Testing each reported and +6-hour forecast polygon point-by-point, Puebla (PBC, 30 km), Mexico City (MEX, 66 km), Felipe Ángeles (NLU, 90 km) and Toluca (TLC, 106 km) fell inside none of them. The operational number is the recovery that was not needed: on 31 August — the one complete day containing the pulse — FIR-level en-route flight counts ran 1.04× expectation over MMEX and 1.07× over MMFR, and MEX's last complete movement day before it (30 August, 737 movements) classifies as a normal day against its own same-weekday median. Two honesty notes, both load-bearing. First, there is deliberately no airport recovery curve here: the pulse opened after the last complete movement day and closed before the next one settled, so no during-pulse movement day exists in our feed, and 26 and 29 August are receiver-coverage days we exclude rather than read as traffic. Second, Popocatépetl has not stopped — it is a persistent emitter, September has already added two advisories, and new emission is possible at any time; this closes the pulse and the near-field exposure question, not the volcano. Automated multi-signal analysis keeps it under continuous watch and this page reopens the moment the cadence, the reported flight levels, or the reach of the ash areas cross the thresholds above. Advisory cadence, reported ash tops and ash-area polygons: VAAC Washington (public advisories); movement and en-route counts: FlySafe ADS-B receiver network, a floor rather than an official total.
Passenger questions, answered from the data
As of 1 September 2026, the 31 August–1 September ash pulse is closed: VAAC Washington's 10:03 UTC advisory reported no ash detected on satellite products. During the pulse, the reported and forecast ash areas did not reach Mexico City's Benito Juárez (MEX), 66 km west of the vent, and no ash-related closure was in effect there. Observed traffic ran inside its normal band on the last complete day. Popocatépetl emits in pulses, so conditions can shift with the next one and with wind — check your airline's status on the day of travel.
Because it is a persistent emitter rather than a single eruption: emission comes in pulses with quiet gaps between them, and VAAC Washington issues advisories when a plume is detected — and another advisory when it is no longer detectable. After the 15–17 August pulse there were seven days without an advisory, two days of emission on 25–26 August, four more quiet days, then a restart at 22:37 UTC on 31 August, a second advisory at 04:08 UTC on 1 September, and a third at 10:03 UTC reporting no ash detected. Across the last 90 days the pattern totals 109 advisories on 37 separate days.
FL200–FL210 is about 20,000–21,000 ft — below the cruise levels of traffic crossing Mexico, so high overflights are largely unaffected. It matters most in the terminal area: Mexico City's field sits at 7,316 ft, so arrivals and departures climb and descend through that band. Our FIR-level measurement across the pulse showed no broad en-route change: on 31 August, whole-airspace flight counts ran 1.04× expectation over MMEX and 1.07× over MMFR. Dispatchers file around advisory areas and regulators close airspace if ash reaches it.
Aircraft do not fly through reported ash — dispatchers file around the advisory areas, and regulators close airspace if ash reaches it. We publish the measured picture: advisory cadence, reported flight levels (FL210 then FL200 in this pulse), which airports fell inside the reported ash polygons (none of ours did), observed airport movements and FIR-level en-route flight counts. Routing and go/no-go decisions belong to operators and the relevant authorities, not to us.
For media — the full kit
Everything on this page — charts, CSV, quotes — is published under CC BY 4.0: free to reuse with attribution to FlySafe.zone. Charts come in dark (social/web) and light (print/light layouts) versions; attribution and license are embedded in every file.
“Popocatépetl's 31 August–1 September ash pulse closed 11 hours 26 minutes after it began, with VAAC Washington issuing three advisories — reported ash tops of FL210 and then FL200, followed by a 10:03 UTC advisory on 1 September reporting no ash detectable on satellite — and none of the reported or forecast ash areas contained Mexico City's Benito Juárez airport 66 km away, Puebla, Felipe Ángeles or Toluca, while FIR-level en-route flight counts over central Mexico ran 1.04 and 1.07 times expectation on 31 August, according to flight-monitoring service FlySafe.zone.”
Boilerplate — about the source (for your article footer)
FlySafe.zone monitors airspace risk and observed airport activity worldwide: 334 flight-information regions scored daily, GPS-interference signals, volcanic-ash exposure and daily observed movements at 493 airports from the crowdsourced ADS-B receiver network. Methodology: flysafe.zone/methodology.
Need a different cut of the data, a comment, or numbers for another airport? Leave an email — we answer fast during active events.

