Pricing Search Start free trial
// EVENT · GNSS INTERFERENCE CLOSED 14 NOV · STEADY STATE DOCUMENTED, NO RECOVERY IN THE CELL — 61°N 26°E 93.1% ACROSS 1–13 NOV · EFIN 6.7% (FIR-LEVEL, QUIETER THAN SEP–OCT) · HEL TRAFFIC NORMAL Data as of 14 Nov 2026, 09:58 UTC

GNSS Interference Inland over Southeastern Finland — 86–100% of Aircraft Flagging Degraded Navigation, September 2026

← All airspace events

Closed 14 November 2026: the inland pocket held — 93.1% (229 of 246 aircraft) across 1–13 November in the 61°N 26°E cell, 88–94% on each of 10–13 November — while the FIR-level Finland ratio drifted quieter (6.7% over 1–13 November) and Helsinki traffic stayed rated normal; the close-out is the last section of this page. Written 11 November: Over the 61°N 26°E cell — inland southeastern Finland, north-east of Helsinki — 77–100% of aircraft broadcast degraded navigation integrity on every observed day from 1 September to 9 November. After the softest stretch of the run — 86% (19 of 22 aircraft) on 27 October, 82% (23 of 28) on 28 October and 77% (10 of 13) on 29 October, the lowest reading since 13 August — the cell read 94%, 91%, 100%, 89% and 100% over 30 October–3 November, then 88% (22 of 25) on 4 November, 83% (15 of 18) on 5 November, 94% (16 of 17) on 7 November, 100% (19 of 19) on 8 November and 100% (25 of 25) on 9 November (FlySafe monitoring, crowdsourced ADS-B network; a floor, not a total; 14 September and 6 November are gaps in our per-cell feed). Pooled, 27–29 October read 82.5% (52 of 63), the lowest three-day figure since 21 August, 1–3 November 96.2% (51 of 53) 3–5 November 90.0% (54 of 60) 5, 7 and 8 November 92.6% (50 of 54) and 7–9 November 98.4% (60 of 61); three soft days were not a level, and one soft day on eighteen aircraft was not one either. The zone's western side stayed soft: 61°N 25°E read 60% (26 of 43) on 4 November and 62% (26 of 42) on 5 November, under its 69.7% September floor four days running from 2 November — the longest such run since the series opened on 10 August — and 71% (25 of 35), 76% (32 of 42) and 84% (31 of 37) on 7–9 November, back inside; 61°N 24°E read 51% (23 of 45) on 4 November, under its 56.1% floor a third day running — a second three-day run after 26–29 October, which we report as a softening of the edge, not a step west — 60% (26 of 43) on 5 November, back inside, 48% (19 of 40) on 7 November, under it again, and 70% (28 of 40) and 68% (25 of 37) on 8–9 November, back inside; Tampere-Pirkkala's cell (61°N 23°E), under its 18–48% band for one day on 3 November (12%, 3 of 26), read 24% (9 of 38) and 24% (7 of 29) on 4–5 November, back inside — softer for a day, not further west. 61°N 27°E read 94% (15 of 16) and then 80% (12 of 15), under its range on a thin sample. On the Gulf row: Tallinn's cell read 57% (86 of 150) and 53% (83 of 156), twelve days running under its 72–83% September range, its series low still the 49% of 3 November; Helsinki-Vantaa's own cell (60°N 24°E) read 53% (81 of 152), inside its 49–72% 1–29 September range, then 41% (59 of 145) on 5 November, under it a third time and a new low for a series that opened on 10 August; 60°N 25°E read 51% and then 42% (50 of 119), under its 48% floor a fourth time; 60°N 23°E 44% and then 36% (34 of 94), under its 39% floor a fifth time; and 60°N 26°E, the August pocket over the Gulf, read 97% (37 of 38) and then 87% (40 of 46) — under its 92–100% September range for the first time since 1 September and the lowest reading of its series. Across the whole 59–62°N, 21–30°E box the flagged share read 57.8% on 4 November, 58.0% on 5 November and 62.2% on 7 November, when Tallinn's cell read 54% (79 of 146), a thirteenth observed day running under its floor, Helsinki-Vantaa's 45% (60 of 132), under its range a fourth time, and 60°N 26°E 98% (40 of 41), back inside; the week of 26 October–1 November closed at 57.3%, the lowest week since the series opened, in a week when the share across our whole feed also had its lowest (4.8% against 5.4–5.9% in every other full week since August), so it is not cleanly regional. On 8 November the box read 64.6%, back inside its 64–74% daily September range for the first time since 25 October, with Tallinn's cell at 63% (104 of 166), a fourteenth observed day under its floor, and Helsinki-Vantaa's at 59% (86 of 147), back inside. The week of 2–8 November closed at 59.3% on six observed days, the box's second-lowest week, under every September week (68.7–70.2%); the feed-wide share pooled 4.8% — inside its 4.7–6.4% September range read day by day, but its second-lowest week since August, under every September week (5.4–5.7%), and on 8 November itself 4.5%, under that daily range. The test this page set — the box under its September range with the feed-wide share inside its own — is met on the daily reading and not on the weekly one, so we do not call the week regional. On 9 November, the first day of the next test week, the box read 58.5%, under its September range again, with Tallinn's cell at 55% (84 of 153), a fifteenth observed day under its floor, and Helsinki-Vantaa's at 42% (62 of 147), under its range a fifth time — the same count as on 2 November — while the feed-wide share read 4.8%, inside its daily September range. These Gulf-row cells carry 87–166 aircraft a day, so they are measurements; we attribute nothing. At FIR scale: EFIN read 6.6% on 4 November (3,391 aircraft), inside the 6.0–9.6% post-spike band, 5.5% on 5 November (3,125), under it for a day, as on 26 October and 2 November, 6.3% on 6 November (3,368), back inside, 6.4% on 7 November (2,780), 7.7% on 8 November (3,084), 6.7% on 9 November (3,326) and 8.3% on 10 November (3,449) — the FIR feed's newest day, a day ahead of the per-cell feed; its 7-day average is 6.8%, up from the 6.3% of 1–7 November, the lowest seven-day figure since the start of August, and under its 8.2% 90-day average. The spike of 8–9 October (16.4% then 10.4%) did not become a run, and a day under the band is not a break. Observed traffic at Helsinki-Vantaa (HEL) is still rated normal for the weekday: 424 movements on Monday 9 November (221 departures, 203 arrivals; settled on 13 November at exactly that count), the lowest of thirteen Mondays; 428 on Sunday 8 November, settled, the fourth-lowest of thirteen Sundays; Saturday 7 November settled at 365, the lowest full-count Saturday, level with 31 October; Friday 6 November settled at 436, the lowest of thirteen Fridays; Thursday 5 November at 419, the lowest of thirteen Thursdays; Wednesday 4 November settled at 424, the lowest of eleven full-count Wednesdays; Tuesday 3 November settled at 410, the lowest full-count Tuesday of the window; Monday 2 November at 429, then the lowest of twelve Mondays; Sunday 1 November at 410, the lowest of twelve Sundays; Saturday 31 October at 365, a floor from a day our network-wide flight feed ran short, not ranked; Friday 30 October at 438, then the lowest of twelve Fridays; Thursday 29 October at 437, then the lowest of twelve Thursdays; Wednesday 28 October at 430; Tuesday 27 October at 428, flagged as a partial count — a one-hour, network-wide gap in our flight feed around 13:00 UTC never filled; Monday 26 October at 438 and Sunday 25 October at 416. The first sixteen days of the northern winter timetable sit at or near the bottom of their weekday ranges, while our network-wide flight feed ran 9–14% under the week before on each of 31 October–4 November, 2–7% under on 5–6 November and above it on 7–9 November, against a short week, and Helsinki's own counts 0–4% under on 1–6 November and within 4% either way on 7–9 November; we report that as a pattern in the count and attribute nothing to it. For one cycle the bars for 9–11 October were drawn from a different receiver network's count; they were restored to the chart's own feed on 14 October and the slip stays disclosed. Tuesday 29 September stays plotted at 350 but flagged as partial: a two-hour, network-wide hole in our flight feed that afternoon never filled. GNSS interference is a navigation-layer signal, not a runway closure; aircraft continue to operate on inertial and ground-based navigation, and procedures and go/no-go decisions belong to operators and authorities.
// Flying to Helsinki or across southern Finland? The short version
  • This brief closed on 14 November as a documented steady state: the inland 61°N 26°E pocket was still flagging 88–94% of aircraft on 10–13 November, the FIR-level Finland ratio ran lower than in September or October, and Helsinki traffic stayed rated normal. The live regional GPS-interference map keeps reading the area daily.
  • Helsinki (HEL) is operating normally — observed movements stayed around their ~442/day reference through the escalation and read 424 on Monday 9 November, settled, the lowest Monday of the window, after 428 on Sunday 8 November and a settled 365 on Saturday 7 November, the lowest full-count Saturday: the first sixteen days of the winter timetable sit at or near the bottom of their weekday ranges (rated normal on every day whose verdict we have read), reported, not attributed. Helsinki's own airspace cell jumped to 88% on 8 October — its highest since the 30–31 August spike — then fell back to 70%, 57%, 65%, 63%, 60%, 69%, 66%, 72%, 68%, 71%, 74%, 72%, 71%, 67%, 64%, 73%, 62%, 50%, 55%, 51%, 52%, 47%, 56%, 61%, 42%, 53%, 53%, 41%, 45%, 59% and 42% over 9 October–9 November (6 November is a gap in our per-cell feed), within or at the top of its September range on most days, a shade above it on 19 and 24 October, and under it on 30 October and 2, 5, 7 and 9 November — the first two on days when the share across our whole feed was also low, the last three with the feed-wide share back inside its September range, 5 November's the lowest reading of its series; it was back inside on 8 November and under again on 9 November.
  • The strong interference has moved inland: it now sits over the southeastern Finnish lake district, north-east of the capital, rather than only over the Gulf of Finland.
  • The zone has a sharp western edge, and it moves — on 9 November 100% of aircraft were flagged at 26°E (25 of 25), 84% at 25°E (inside its September range, after four days under its floor on 2–5 November), 68% at 24°E (inside) and 37% in Tampere's cell (10 of 27, inside its September band — softer than the core, not further west); the one-day spread of 8 October into Helsinki's and Tallinn's cells, south-west of the pocket, receded, and over 26 October–9 November (6 November missing) the same cells read 50%, 55%, 51%, 52%, 47%, 56%, 61%, 42%, 53%, 53%, 41%, 45%, 59% and 42% and 64%, 62%, 63%, 61%, 57%, 67%, 58%, 55%, 49%, 57%, 53%, 54%, 63% and 55% — Helsinki's a new low on 5 November, back inside on 8 November and under again on 9 November, Tallinn's under its floor fifteen observed days running — the softest stretch on that side of the zone since August, on days when the whole region, and over 29 October–2 November our whole feed, read softer.
  • GPS interference is a navigation-integrity issue, not a closure event — aircraft fly on back-up systems; check your airline on the day.

Advisory timeline — how active is this episode

Our detection layer reads the navigation-integrity fields aircraft broadcast — the consistent fingerprint of GNSS interference. The cell that has escalated is 61°N 26°E, inland southeastern Finland — the lake district between Lahti and the eastern border, roughly 130 km north-east of Helsinki, not the Gulf of Finland pocket we tracked in August. When the per-cell series opened on 10 August this cell read 60–80% on thin samples; from 1 September to 9 November it has run 77–100% on every observed day, on samples that grew through September (16 aircraft on 1 September, 21–31 a day from 4 to 28 September, 16 on 29 September, then 23, 19, 21, 11, 19, 19, 14, 25, 18, 20, 23, 27, 27, 22, 29, 28, 27, 17, 25, 29, 19, 30, 20, 25, 24, 25, 21, 22, 28, 13, 18, 11, 17, 19, 17, 25, 18, 17, 19 and 25 on the last forty-one days observed). Seventy-nine of the ninety days observed here have come in at 85% or above. 9 November read 100% (25 of 25 aircraft), 8 November 100% (19 of 19) and 7 November 94% (16 of 17), back above 85% — 6 November is missing from our per-cell feed — after 83% (15 of 18) on 5 November, the third-lowest reading since 1 September, on a thin sample, 88% (22 of 25) on 4 November and 94% (17 of 18), 91% (10 of 11), 100% (17 of 17), 89% (17 of 19) and 100% (17 of 17) over 30 October–3 November — five days back up after the softest stretch of the run — 86% (19 of 22) on 27 October, 82% (23 of 28) on 28 October and 77% (10 of 13) on 29 October, the last the lowest reading since 13 August, on the thinnest sample since 3 October; the 84% (21 of 25) of 18 October, which had replaced the 86% (18 of 21) of 2 October as the floor, held for ten days, and the eight days between read 90% (26 of 29), 89% (17 of 19), 93% (28 of 30), 100% (20 of 20), 92% (23 of 25), 100% (24 of 24), 100% (25 of 25) and 95% (20 of 21): the thirty-seven observed days between 2 October and 9 November read 91%, 100%, 100%, 93%, 96% (24 of 25), 94% (17 of 18), 90% (18 of 20), 96% (22 of 23), 96% (26 of 27), 100% (27 of 27), 95% (21 of 22), 100% (29 of 29), 100% (28 of 28), 93% (25 of 27), 100% (17 of 17), 84% (21 of 25), 90% (26 of 29), 89% (17 of 19), 93% (28 of 30), 100% (20 of 20), 92% (23 of 25), 100% (24 of 24), 100% (25 of 25), 95% (20 of 21), 86% (19 of 22), 82% (23 of 28), 77% (10 of 13), 94% (17 of 18), 91% (10 of 11), 100% (17 of 17), 89% (17 of 19), 100% (17 of 17), 88% (22 of 25), 83% (15 of 18), 94% (16 of 17), 100% (19 of 19) and 100% (25 of 25). The 3 October and 31 October bars rest on eleven aircraft each — the thinnest samples since 30 August — and the 29 October bar on thirteen, the next-thinnest since 1 September, so read all three as directions, not measurements; the 5 November bar rests on eighteen, the 7 November bar on seventeen, the 8 November bar on nineteen and the 9 November bar on twenty-five. 14 September and 6 November are not plotted: our per-cell feed holds no data for either day in any cell — collection gaps, not readings. 9 November is the newest bar: the per-cell feed is written once a day, for the day before. The 84% of 18 October was the first day below 86% since the run opened on 1 September, and the eight days after it (89–100%) showed it did not hold; 27 October's 86% (on twenty-two aircraft), 28 October's 82% (on twenty-eight, a full-sized sample for this cell) and 29 October's 77% (on thirteen, too thin to weigh alone) were three soft days in a row, on days when the whole region read softer — see below; pooled, they read 82.5% (52 of 63 aircraft), the lowest three-day figure since 21 August. On 30 October the cell read 94% (17 of 18), on 31 October 91% (10 of 11), on 1 November 100% (17 of 17), on 2 November 89% (17 of 19) and on 3 November 100% (17 of 17), climbing back the way it did after 18 October, then 88% (22 of 25) on 4 November, 83% (15 of 18) on 5 November, 94% (16 of 17) on 7 November, 100% (19 of 19) on 8 November and 100% (25 of 25) on 9 November; pooled, 28–30 October read 84.7% (50 of 59), still the second-lowest three-day figure since 1 September, 29–31 October 88.1% (37 of 42), 31 October–2 November 93.6% (44 of 47) 1–3 November 96.2% (51 of 53), back where the cell ran before the soft days, 3–5 November 90.0% (54 of 60) the three observed days 4, 5 and 7 November 88.3% (53 of 60), 5, 7 and 8 November 92.6% (50 of 54) and 7–9 November 98.4% (60 of 61). Three soft days were not a level, and six days back at 85% or above said the cell had settled where it was; we read that as no change, not as a recovery from something, and 5 November's 83%, on eighteen aircraft, as one soft day after them — logged, not called, until a second day says otherwise; 7–9 November's 94% (16 of 17), 100% (19 of 19) and 100% (25 of 25) were not that second day. October closed at 93.9% pooled (all aircraft over the month), against 95.2% for September; November so far reads 94.3% (148 of 157). Across 8–26 October this cell barely moved except for the 18 October dip (94%, 90%, 96%, 96%, 100%, 95%, 100%, 100%, 93%, 100%, 84%, 90%, 89%, 93%, 100%, 92%, 100%, 100%, 95%), then read 86%, 82% and 77% on 27–29 October and 94%, 91%, 100%, 89%, 100%, 88%, 83%, 94%, 100% and 100% on 30 October–9 November (6 November missing), while the cells to its south rose sharply on 8 October and have since fallen back — see below. Honesty note: this per-cell series begins 10 August, so it is 90 observed days deep, and daily percentages on a cell that sees 6–37 aircraft are noisy by construction — the thinnest day in the series, 30 August, rests on six aircraft; the 100% on 18 September is 31 of 31, the 100% on 25 September is 23 of 23, the 92% on 27 September is 24 of 26, the 91% on 28 September is 21 of 23, the 94% on 29 September is 15 of 16, the 91% on 30 September is 21 of 23, the 100% on 1 October is 19 of 19, the 86% on 2 October is 18 of 21, the 91% on 3 October is 10 of 11, the 100% on 4 October is 19 of 19, the 100% on 5 October is 19 of 19, the 93% on 6 October is 13 of 14, the 96% on 7 October is 24 of 25, the 94% on 8 October is 17 of 18, the 90% on 9 October is 18 of 20, the 96% on 10 October is 22 of 23, the 96% on 11 October is 26 of 27, the 100% on 12 October is 27 of 27, the 95% on 13 October is 21 of 22, the 100% on 14 October is 29 of 29, the 100% on 15 October is 28 of 28, the 93% on 16 October is 25 of 27, the 100% on 17 October is 17 of 17, the 84% on 18 October is 21 of 25, the 90% on 19 October is 26 of 29, the 89% on 20 October is 17 of 19, the 93% on 21 October is 28 of 30, the 100% on 22 October is 20 of 20, the 92% on 23 October is 23 of 25, the 100% on 24 October is 24 of 24, the 100% on 25 October is 25 of 25, the 95% on 26 October is 20 of 21, the 86% on 27 October is 19 of 22, the 82% on 28 October is 23 of 28, the 77% on 29 October is 10 of 13, the 94% on 30 October is 17 of 18, the 91% on 31 October is 10 of 11, the 100% on 1 November is 17 of 17, the 89% on 2 November is 17 of 19, the 100% on 3 November is 17 of 17, the 88% on 4 November is 22 of 25, the 83% on 5 November is 15 of 18, the 94% on 7 November is 16 of 17, the 100% on 8 November is 19 of 19 and the 100% on 9 November is 25 of 25. Daily share of aircraft flagging degraded navigation in the cell:

94
1 Sep
89
2 Sep
91
3 Sep
93
4 Sep
96
5 Sep
90
6 Sep
96
7 Sep
96
8 Sep
100
9 Sep
96
10 Sep
93
11 Sep
94
12 Sep
100
13 Sep
92
15 Sep
96
16 Sep
93
17 Sep
100
18 Sep
96
19 Sep
100
20 Sep
96
21 Sep
95
22 Sep
100
23 Sep
95
24 Sep
100
25 Sep
100
26 Sep
92
27 Sep
91
28 Sep
94
29 Sep
91
30 Sep
100
1 Oct
86
2 Oct
91
3 Oct
100
4 Oct
100
5 Oct
93
6 Oct
96
7 Oct
94
8 Oct
90
9 Oct
96
10 Oct
96
11 Oct
100
12 Oct
95
13 Oct
100
14 Oct
100
15 Oct
93
16 Oct
100
17 Oct
84
18 Oct
90
19 Oct
89
20 Oct
93
21 Oct
100
22 Oct
92
23 Oct
100
24 Oct
100
25 Oct
95
26 Oct
86
27 Oct
82
28 Oct
77
29 Oct
94
30 Oct
91
31 Oct
100
1 Nov
89
2 Nov
100
3 Nov
88
4 Nov
83
5 Nov
94
7 Nov
100
8 Nov
100
9 Nov
88
10 Nov
92
11 Nov
94
12 Nov
90
13 Nov

Fig. 1 — share of aircraft broadcasting degraded navigation integrity in the 61°N 26°E cell (%) — crowdsourced ADS-B network · Data as of 14 Nov 2026, 09:58 UTC · Source: FlySafe monitoring — crowdsourced ADS-B navigation-integrity data (public).

Airports in the exposure zone

AirportSizeOwn-cell flag rate (13 Nov)FIR
Helsinki Vantaa (HEL) large — · ≈58% of aircraft flagged (own cell) EFIN
Lappeenranta (LPP) medium — · 100% flagged — 5 of 5 aircraft, too thin to read EFIN
Tampere-Pirkkala (TMP) medium — · ≈26% of aircraft flagged (own cell) EFIN
Lennart Meri Tallinn (TLL) medium — · ≈64% of aircraft flagged (own cell) EETT

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 shows: Helsinki traffic normal while the navigation layer stays lit

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. Helsinki Vantaa's last 30 days:

Fig. 2 — Observed daily movements at HEL, 13 Aug – 9 Nov (latest complete UTC day). Data as of 14 Nov 2026, 09:58 UTC · Source: FlySafe monitoring (ADS-B receiver network).

  • Across the 13 Aug–9 Nov window plotted above: ~442 observed movements/day, averaged over the 88 days other than 29 September — that one day is a partial count, plotted and flagged below rather than dropped; leaving out 27 October as well, a partial count with a smaller hole, gives the same ~442, and leaving out 31 October too, a floor from a short day in our feed, about 443. The 442 reference line this page has used since it opened matches it.
  • Through the September escalation in the cell (1–30 Sep): 456 / 445 / 458 / 446 / 385 / 452 / 476 / 378 / 440 / 469 / 452 / 402 / 434 / 483 / 453 / 446 / 510 / 465 / 410 / 443 / 475 / 446 / 451 / 474 / 448 / 417 / 447 / 469 / 350 / 441 observed movements, then 480 / 455 / 395 / 449 / 478 / 457 / 457 / 465 / 453 / 398 / 445 / 480 / 447 / 437 / 480 / 461 / 427 / 459 / 457 / 448 / 458 / 467 / 459 / 412 / 416 / 438 / 428 / 430 / 437 / 438 / 365 over 1–31 October, then 410 / 429 / 410 / 424 / 419 / 436 / 365 / 428 / 424 over 1–9 November — around baseline on every full day, with Saturdays and Sundays running lower, as they have all window, and the first days of the winter timetable lower still (see below). Monday 9 November, at 424 (221 departures, 203 arrivals), is the newest bar. It settled at its final pass on 13 November at exactly the count this page published on 11 November, its stored daily row matching the recount: 424 is the lowest of thirteen Mondays, 5 under the settled 429 of 2 November (the others: 442 / 471 / 472 / 476 / 483 / 475 / 469 / 478 / 480 / 457 / 438); the HEL airport brief rates it −3% on its 438 Monday reference (verdict: normal). Sunday 8 November, at 428 (229 departures, 199 arrivals), ten departures more than the 418 this page published on 10 November with the note that it might still gain, settled at its final pass on 12 November at exactly 428, its stored daily row back level with the recount. Our network's total for the day now sits about 12% over 1 November's — itself a short day — and 4% under 25 October's, and its last two hours run above 1 November's. Settled, 428 is the fourth-lowest of thirteen Sundays, above the 410 of 1 November, the 416 of 25 October and the 422 of 16 August (the others: 444 / 453 / 452 / 434 / 443 / 447 / 449 / 445 / 459), −4% on the 445 Sunday reference the HEL airport brief quoted for it (verdict: normal). Saturday 7 November has settled at 365 (192 departures, 173 arrivals), exactly the recount this page published on 10 November: the lowest of twelve full-count Saturdays, 20 under the 385 of 5 September (the others: 399 / 410 / 397 / 385 / 402 / 410 / 417 / 395 / 398 / 427 / 412), and exactly level with 31 October's floor — a sign that floor may have sat close to a full count after all; we keep 31 October flagged and unranked all the same. Our network's total for the day sits about 3% over 31 October's and 8% under 24 October's, but its ratio to the Friday before is the same as 24 October's (0.90), so we see no short pass in it. It reads −11% on the 412 Saturday reference the HEL airport brief quoted for 31 October; the brief has not scored 7 November. Friday 6 November has settled at 436 (233 departures, 203 arrivals), four departures more than the 432 this page published on 8 November with the note that it might still gain: the lowest of thirteen Fridays, 2 under the 438 of 30 October (the others: 468 / 459 / 451 / 446 / 452 / 465 / 448 / 455 / 453 / 461 / 459 / 438), −5% on the 459 Friday reference the HEL airport brief quoted for it (verdict: normal). Thursday 5 November has settled at 419 (218 departures, 201 arrivals), exactly the recount this page published on 8 November: the lowest of thirteen Thursdays, 18 under the 437 of 29 October (the others: 442 / 462 / 479 / 458 / 469 / 510 / 474 / 480 / 465 / 480 / 467 / 437), −10% on the 467 Thursday reference the HEL airport brief quoted for it (verdict: normal). Wednesday 4 November has settled at 424 (220 departures, 204 arrivals), exactly the recount this page published on 7 November: the lowest of eleven full-count Wednesdays, 6 under the 430 of 28 October (the others: 442 / 445 / 440 / 446 / 451 / 441 / 457 / 437 / 458 / 430; 26 August, 391 with a one-hour hole in it, is not counted), −7% on the 457 Wednesday reference the HEL airport brief quoted for it (verdict: normal). Tuesday 3 November has settled at 410 (216 departures, 194 arrivals), exactly the count this page published on 5 November: the lowest full-count Tuesday of the window, 24 under the 434 of 25 August (the others: 435 / 456 / 453 / 446 / 457 / 447 / 448), −8% on the 448 Tuesday reference the HEL airport brief quoted for it (verdict: normal). Monday 2 November has settled at 429 (223 departures, 206 arrivals), exactly the recount this page published on 5 November: then the lowest of twelve Mondays, 9 under the 438 of 26 October (the others: 442 / 471 / 472 / 476 / 483 / 475 / 469 / 478 / 480 / 457 / 438), −10% on the 478 Monday reference the HEL airport brief quoted for 26 October. Sunday 1 November has settled at 410 (223 departures, 187 arrivals), three departures more than the 407 this page published on 4 November with the note that it might still gain: the lowest of twelve Sundays, 6 under the 416 of 25 October (the others: 422 / 444 / 453 / 452 / 434 / 443 / 447 / 449 / 445 / 459 / 416), −8% on the 445 Sunday reference the HEL airport brief quoted for it (verdict: normal). Saturday 31 October has settled at 365 (191 departures, 174 arrivals), exactly the count this page published on 4 November, and stays a floor: the short pass across our whole flight feed that day never filled — the network's total still sits about 11% under the Saturday before, its hours from 15:00 UTC 14–22% short — so 365, 20 under the 385 of 5 September, the window's lowest Saturday, is plotted and not ranked. On the 412 Saturday reference the HEL airport brief quoted for it on 2 November, it reads −11%. Friday 30 October settled at 438 (231 departures, 207 arrivals), then the lowest of twelve Fridays, 8 under the 446 of 4 September (the others: 468 / 459 / 451 / 446 / 452 / 465 / 448 / 455 / 453 / 461 / 459), −4% on the 455 Friday reference the HEL airport brief quoted for 23 October. Thursday 29 October settled at 437 (234 departures, 203 arrivals), then the lowest of twelve Thursdays, 5 under the 442 of 13 August (Thursday reference 480, −9%, normal). Wednesday 28 October settled at 430 (227 departures, 203 arrivals), then the lowest of ten full-count Wednesdays, 7 under the 437 of 14 October (Wednesday reference 457, −6%, normal). Tuesday 27 October has settled at 428 (225 departures, 203 arrivals), exactly the count this page published on 30 October, and stays flagged as a partial count — its hole is permanent: at its final pass our flight feed still holds about 5,500 flights worldwide in the 13:00 UTC hour against about 7,300 and 7,700 in the same hour on the days either side, with the same two near-empty ten-minute stretches (56 and 52 flights). Helsinki's own count in that hour is 13 movements against 25 on 26 October and 20 on 28 October. It sits under every full-count Tuesday before it (435 / 434 / 456 / 453 / 446 / 457 / 447 / 448) and above the settled 410 of 3 November, but with a known hole in it we do not rank it; it reads −4% on the 448 Tuesday reference the HEL airport brief quoted for it. Monday 26 October settled at 438 (235 departures, 203 arrivals), then the lowest of eleven Mondays, 4 under the 442 of 17 August (Monday reference 478, −8%, normal). Sunday 25 October settled at 416 (228 departures, 188 arrivals), then the lowest of eleven Sundays, 6 under the 422 of 16 August (Sunday reference 449, −7%, normal). The winter-timetable pattern has run sixteen days. The first sixteen days of the northern winter timetable season, which began on 25 October, include the window's lowest Sunday (1 November, settled, under 25 October's, also settled); its lowest settled Monday (2 November, under 26 October's, also settled); two Tuesdays at the bottom — 3 November, settled, the lowest full-count Tuesday, and 27 October, settled but with a hole in it, under every full-count Tuesday before it; two Wednesdays at the bottom of the full-count range — 28 October and 4 November, both settled, the second lower again; two Thursdays at the bottom — 29 October, settled, then the lowest of twelve, and 5 November, settled, lower again; two Fridays at the bottom — 30 October, settled, then the lowest of twelve, and 6 November, settled, lower again; two Saturdays at 365, 20 under any earlier Saturday — 31 October, settled on a day our whole flight feed ran short, and 7 November, settled; a third Sunday, 8 November, fully collected, the fourth-lowest of its weekday; and a second Monday, 9 November, settled, lower again than 2 November. The calendar week of 26 October–1 November totals 2,946 movements against 3,117 the week before (−5%), with a partial day and a short-feed floor in it; the week of 2–8 November totals 2,911, against 2,946 the week before (−1%, the earlier week carrying the partial 27 October and the 31 October floor) and 3,117 the week before that (−7%). (The previous version of this sentence compared 1–3 November with 25–27 October, 1,249 against 1,282, and called those the first three days of the week; they straddle two weeks.) That is a pattern in the count, sixteen days long, and we report it as one. We still attribute nothing to it: the HEL airport brief rated normal for its weekday every one of these days whose verdict we have read; a schedule change is the ordinary candidate for a step at the start of a timetable season, and we have not checked timetables; our network-wide flight feed ran 9–14% under the same weekday a week earlier on each of 31 October–4 November, then about 7% on 5 November and 2% on 6 November, and ran 3%, 12% and 6% over it on 7–9 November, against a week that had itself run short, while Helsinki's own counts ran 0–4% under on 1–6 November, level on 7 November, 4% over on 8 November and 1% under on 9 November, so the feed's shortfall does not set Helsinki's numbers, though it may shave a few movements off them; and the soft traffic began on 25 October, while the tracked cell still read 100%, and ran on through 9 November; the cell read 83% on 5 November, 94% on 7 November and 100% on 8 and 9 November (6 November is missing from our per-cell feed). Saturday 24 October has settled at 412 (223 departures, 189 arrivals), exactly the recount this page published on 27 October, third of eleven Saturdays inside a 385–427 range. Friday 23 October has settled at 459 (243 departures, 216 arrivals), one departure more than the 458 this page published on 26 October with the note that it might still gain; against the window's other Fridays (468 / 459 / 451 / 446 / 452 / 465 / 448 / 455 / 453 / 461 / 438 / 436) it ties 21 August for fourth of thirteen (Friday reference 455, +1%, normal). Thursday 22 October has settled at 467 (244 departures, 223 arrivals), exactly the count this page published on 26 October, seventh of thirteen Thursdays (Thursday reference 480, −3%, normal). Wednesday 21 October settled at 458 (243 departures, 215 arrivals), the highest of eleven full-count Wednesdays (the others: 442 / 445 / 440 / 446 / 451 / 441 / 457 / 437 / 430 / 424), 1 above the 457 of 7 October (Wednesday reference 451, +2%, normal) — one Wednesday a movement above the previous top is not a signal. Tuesday 20 October has settled at 448 (237 departures, 211 arrivals), exactly the count this page published on 23 October; against the window's other full-count Tuesdays (435 / 434 / 456 / 453 / 446 / 457 / 447 / 410) it is fourth of nine, in a 410–457 range (Tuesday reference 447, 0%, normal). Monday 19 October settled at 457 (249 departures, 208 arrivals), the fifth-lowest of thirteen Mondays (the others: 442 / 471 / 472 / 476 / 483 / 475 / 469 / 478 / 480 / 438 / 429 / 424), 12 under the next one up (469); Mondays have run high all window, so on the page's flat 442 line it still reads +3% — one Monday at the soft end of a 41-wide weekday range is not a signal. Sunday 18 October settled at 459 (245 departures, 214 arrivals), the highest of thirteen Sundays (the others: 422 / 444 / 453 / 452 / 434 / 443 / 447 / 449 / 445 / 416 / 410 / 428), +3% on its 447 same-weekday reference (verdict: normal). Saturday 17 October settled at 427 (232 departures, 195 arrivals), the highest of eleven Saturdays (the others span 385–417), +4% on its 410 same-weekday reference (verdict: normal). Friday 16 October settled at 461 (241 departures, 220 arrivals), third of thirteen Fridays, level with its 455 same-weekday reference; Thursday 15 October at 480, second of thirteen Thursdays tied with 1 October; Wednesday 14 October at 437, now the third-lowest of eleven full-count Wednesdays, 3% under its 451 reference (verdict: normal); Tuesday 13 October at 447; Monday 12 October at 480, second of thirteen Mondays; Sunday 11 October at 445, Saturday 10 October at 398, Friday 9 October at 453 and Thursday 8 October at 465.
  • A correction carried forward: for one cycle (12 October) the three bars for 9 to 11 October were drawn from a different receiver network's count (477 / 406 / 444) rather than the feed every other bar on this chart comes from, the 8 October bar was described in the text at that network's 475 while plotted at 465, and the weekday comparisons in this section were made against that other feed. Those bars were restored on 14 October to the chart's own feed (453 / 398 / 445) and have since settled at exactly those values; the picture they show — an ordinary week at Helsinki — is unchanged, and the slip stays disclosed rather than silently overwritten. A smaller one from the 21 October cycle: the list of Sundays this section compared 18 October against carried a Monday's count (469) in place of two Sundays (449, 445); the list above is the corrected one, and 18 October's rank moved from third to first as a result.
  • Coverage notes: 29 September stays plotted at 350 and stays a partial count — its hole is permanent: its rollup row settled at exactly the 350 the recount gives (185 departures, 165 arrivals); between 13:00–15:00 UTC our flight feed holds 8 Helsinki movements against 72 in the same two hours the prior Tuesday, and the day's network total has not moved from 111,000. The bar sits 21% below the line because of our feed, not because of Helsinki: read it as a floor with a known hole in it. Two older bars carry smaller holes: 8 September (378) is missing most of the 13:00 UTC hour, and 26 August (391) has no records at all between 06:00 and 07:00 UTC. Both are network-wide gaps, not at Helsinki. The HEL airport brief flags no gap days in its current 30-day window (11 October–9 November); the stored rows for 28 October–7 November have settled at exactly the counts this page plots, 8 November's was cut back to 373 on its latest refresh, 55 under its recount (the bar carries the recount), and 9 November's matches its recount. 27 October stays flagged as a partial count, and its hole is now permanent: the network-wide gap around 13:00 UTC described above did not fill at its final pass, and its stored row settled at exactly the 428 the recount gives; it is a one-hour hole, smaller than 29 September's two-hour one and of the same kind as 8 September's, and the bar is plotted and flagged rather than dropped. 31 October has settled at 365 and stays a floor from a short day across our whole feed (see above); 10–13 November are not plotted: the nightly ingest of our flight feed failed network-wide on 12–14 November, leaving about 13,000–16,000 flights worldwide on each of those days against 107,000–119,000 on the days before, so they are unreadable rather than low. Observed counts are a floor, not a total.
  • A note on the baseline: HEL's airport brief compares each day with the same weekday — its latest day is Monday 9 November, 424 against a Monday reference of 438 (−3%, normal), and its daily avg 7d is 415, over 3–9 November (2,906 movements), now that 8 November's stored row has settled level with its recount. The ~442 on this page is the window-level reference, weekday-independent, kept stable so the series stays comparable across updates; on it, 9 November reads −4% (424 vs 442) — a Monday, settled — 8 November −3% (428 vs 442) — a Sunday — 7 November −17% (365 vs 442) — a Saturday — 6 November −1% (436 vs 442) — a Friday — 5 November −5% (419 vs 442) — a Thursday — 4 November −4% (424 vs 442) — a Wednesday — 3 November −7% (410 vs 442) — a Tuesday — 2 November −3% (429 vs 442) — a Monday — 1 November −7% (410 vs 442) — a Sunday — 31 October −17% (365 vs 442) — a Saturday, on a short day in our feed — 30 October −1% (438 vs 442) — a Friday — 29 October −1% (437 vs 442) — a Thursday — 28 October −3% (430 vs 442) — a Wednesday — 27 October −3% (428 vs 442) — a Tuesday with a hole in it — 26 October −1% (438 vs 442) — a Monday, the weekday that runs highest all window — 25 October −6% (416 vs 442) and 24 October −7% (412 vs 442) — a Sunday and a Saturday — 23 October +4% (459 vs 442), 22 October +6% (467 vs 442), 21 October +4% (458 vs 442), 20 October +1% (448 vs 442), 19 October +3% (457 vs 442), 18 October +4% (459 vs 442) — a Sunday, the window's strongest — 17 October −3% (427 vs 442), a Saturday effect, as the window's eleven Saturdays span 385 to 427 — and 16 October +4% (461 vs 442). The window's twelve other Fridays span 438 to 468, with 6 November's settled 436 under all of them; its twelve other Thursdays 437 to 510, with 5 November's settled 419 under all of them; its twelve other Mondays 429 to 483, with 9 November's 424, settled, under all of them; its thirteen Sundays 410 to 459, with 8 November's 428 fourth from the bottom; its eleven other full-count Saturdays 385 to 427, with 7 November's settled 365 under all of them and 31 October's 365 a floor from a short day, not ranked; its nine full-count Tuesdays 410 to 457, the low the settled 410 of 3 November (27 October, settled at 428 with a one-hour hole in it, is flagged partial and not counted); its eleven full-count Wednesdays 424 to 458, the low the settled 424 of 4 November. The 510 of Thursday 17 September remains the top of the window.
  • The point, and it survives the churn: GNSS interference does not surface in the movement count. It surfaces in the navigation-integrity layer — aircraft broadcasting degraded position quality — not in whether flights depart. Every large swing in this section has been a collection artefact on our side, corrected in the open, and the sixteen soft days that opened the winter timetable (416, 438, 428, 430, 437, 438, 365, 410, 429, 410, 424, 419, 436, 365, 428 and 424) began while the tracked cell read 100%, and every one whose verdict we have read the HEL airport brief rated normal for its weekday; the bars show no interference signature while the cell above ran at 77% or more on every one of the 68 days observed since 1 September. 8 October — the day Helsinki's own cell read 88% and the FIR-wide figure jumped to 16.4% — came in at 465 movements, inside the window's Thursday range; 12 October, the day the tracked cell read 100% (27 of 27 aircraft), at 480, second of thirteen Mondays; 14 and 15 October, two more days at 100% in the cell (29 of 29, then 28 of 28), at 437 and at 480; 18 October, then the cell's softest day at 84%, at 459, the window's highest Sunday; 19 October, the cell back at 90% and Helsinki's own cell at 74%, at 457; 20 October, 89% in the cell, at 448, level with its weekday reference; 21 October, 93% in the cell, at 458, the highest of ten full-count Wednesdays; 22 October, 100% in the cell, at 467, seventh of thirteen Thursdays; 23 October, 92% in the cell, at 459, fourth of twelve Fridays; 24 October, 100% in the cell, at 412, third of eleven Saturdays; 25 October, 100% in the cell again, at 416 — then the lowest of eleven Sundays, on a day the cell read the same 100% it read on 12 and 14 October at 480 and 437; 26 October, 95% in the cell, at 438, then the lowest of eleven Mondays; 27 October, 86% in the cell, at a partial 428; 28 October, 82% in the cell, at 430, then the lowest of ten full-count Wednesdays; 29 October, 77% in the cell on a thin sample, at 437, then the lowest of twelve Thursdays; 30 October, 94% in the cell, at 438, then the lowest of twelve Fridays; 31 October, 91% in the cell on a thin sample, at a short 365; 1 November, 100% in the cell (17 of 17), at 410, the lowest Sunday; 2 November, 89% in the cell, at 429, then the lowest Monday; 3 November, 100% in the cell (17 of 17) again, at 410, the lowest full-count Tuesday; 4 November, 88% in the cell (22 of 25), at 424, the lowest full-count Wednesday; 5 November, 83% in the cell (15 of 18), at 419, the lowest Thursday; 6 November, a day missing from our per-cell feed, at 436, the lowest Friday; 7 November, 94% in the cell (16 of 17), at 365, the lowest full-count Saturday; 8 November, 100% in the cell (19 of 19), at 428; 9 November, 100% in the cell (25 of 25), at 424, the lowest Monday — the soft traffic days began two days before the cell's own soft days and outlasted them, so the two do not line up; still no interference signature in the count.

Live view: HEL airport brief.

What it means for flights

The news is the geography, not the number. August's brief documented a pocket over the eastern Gulf of Finland at 60°N 26°E. That cell is still running high — 100% (80 of 80) on 18 October, then 98%, 94%, 96%, 96%, 100% (45 of 45), 98% (43 of 44), 96% (44 of 46), 97% (38 of 39), 100% (39 of 39), 100% (39 of 39), 100% (39 of 39), 97% (31 of 32), 100% (35 of 35), 100% (42 of 42), 95% (36 of 38), 97% (35 of 36) and 97% (37 of 38) over 19 October–4 November, and 87% (40 of 46) on 5 November — under its 92–100% September range for the first time since 1 September and the lowest reading of its series, on a full-sized sample: one day, logged, 98% (40 of 41) on 7 November, back inside, 96% (45 of 47) on 8 November and 97% (29 of 30) on 9 November (6 November is a gap in our per-cell feed) — the same 39 of 39 on three straight days, 27–29 October; exact three-day repeats occur in a handful of cells across our feed (four of about 4,200 with 30 or more aircraft), so we do not read it as a stuck row, and the run ended on 30 October — but the band of near-total flagging now extends a full degree north over land: 61°N 26°E at 100% (25 of 25) on 9 November, 100% (19 of 19) on 8 November, 94% (16 of 17) on 7 November, 83% (15 of 18) on 5 November, 88% (22 of 25) on 4 November, 100% (17 of 17) on 3 November, 89% (17 of 19) on 2 November, 100% (17 of 17) on 1 November, 91% (10 of 11) on 31 October and 94% (17 of 18) on 30 October, back up after 77% (10 of 13) on 29 October — its lowest reading since 13 August, on a sample too thin to weigh alone — 82% (23 of 28) on 28 October, its lowest since 21 August, 86% (19 of 22) on 27 October, 95% (20 of 21) on 26 October, 100% (25 of 25) on 25 October, 100% (24 of 24) on 24 October, 92% (23 of 25) on 23 October, 100% (20 of 20) on 22 October, 93% (28 of 30) on 21 October, 89% (17 of 19) on 20 October, 90% (26 of 29) on 19 October and the 84% (21 of 25) of 18 October that had been its softest day, with 61°N 25°E at 79%, 70%, 74%, 79%, 78%, 74%, 84%, 90%, 83%, 59%, 71%, 67%, 76%, 87%, 73%, 69%, 61%, 60%, 62%, 71%, 76% and 84% over 18 October–9 November (6 November missing). 61°N 27°E, which dipped to 80% on ten aircraft on 6 October, read 94%, 92% and then 100% on ten straight days (9–18 October); 19–21 October broke that run at 94% (17 of 18), 92% (11 of 12) and 96% (23 of 24), inside the 90–100% range it held through September, 22 October read 85% (17 of 20), below that range for the first time since 6 October, 23 October 95% (18 of 19), back inside it, 24 October 94% (17 of 18), 25 October 100% (18 of 18), 26 October 100% (16 of 16), 27 October 95% (20 of 21), 28 October 85% (17 of 20), below that range again, the same 17 of 20 as on 22 October, 29 October 86% on seven aircraft, 30 and 31 October 100% on six each and 1 November 100% on ten, all too thin to read, 2 November 73% (11 of 15), under its range on a sample nearly as thin, 3 November 92% (11 of 12), back inside on another, 4 November 94% (15 of 16) and 5 November 80% (12 of 15), under it again on a thin sample, and 7–9 November 100% (10 of 10, 11 of 11, then 16 of 16) — see the edge discussion below. That is the southeastern Finnish lake district, well inland, on the arrival and departure tracks between Helsinki and the northeast. A pocket over water affects transiting traffic; a pocket over the mainland sits under domestic and regional routings as well. This page tracks the extent, not just the peak — and on 8 October the extent is what moved, for a day.

A hard western edge is a real finding. Interference has a footprint, and this one stops sharply. On 9 November — the per-cell feed's newest day — the flag rate ran 100% at 26°E (25 of 25 aircraft), 84% at 25°E (31 of 37), 68% at 24°E (25 of 37) and 37% at 23°E (10 of 27) — Tampere-Pirkkala's cell, roughly 160 km west of the high zone; on 8 November it ran 100% (19 of 19), 76% (32 of 42), 70% (28 of 40) and 32% (7 of 22); on 7 November (6 November is missing from the feed in every cell, a collection gap) it ran 94% (16 of 17), 71% (25 of 35), 48% (19 of 40) and 23% (5 of 22); on 5 November it ran 83% (15 of 18), 62% (26 of 42), 60% (26 of 43) and 24% (7 of 29); on 4 November it ran 88% (22 of 25), 60% (26 of 43), 51% (23 of 45) and 24% (9 of 38); on 3 November 100% (17 of 17), 61% (23 of 38), 52% (16 of 31) and 12% (3 of 26); on 2 November it ran 89% (17 of 19), 69% (25 of 36), 55% (18 of 33) and 33% (8 of 24); on 1 November it ran 100% (17 of 17), 73% (24 of 33), 68% (26 of 38) and 35% (7 of 20); on 31 October 91% (10 of 11, a thin sample), 87% (26 of 30), 56% (18 of 32) and 26% (6 of 23); on 30 October it ran 94% (17 of 18), 76% (29 of 38), 60% (21 of 35) and 21% (6 of 28); on 29 October — the row's second-thinnest day since 1 September — it ran 77% at 26°E (10 of 13), 67% at 25°E (18 of 27), 39% at 24°E (14 of 36) and 21% at 23°E (4 of 19). The edge is real but it does not hold a fixed line, and the last forty-two observed days at 24°E show why we do not read it off a day or two: 60% → 82% → 74% → 70% → 73% → 79% → 78% → 80% → 63% → 75% → 87% → 65% → 89% → 87% → 76% → 79% → 76% → 64% → 70% → 72% → 82% → 54% → 59% → 60% → 76% → 60% → 74% → 70% → 47% → 45% → 55.6% → 39% → 60% → 56% → 68% → 55% → 52% → 51% → 60% → 48% → 70% → 68% (28 September–9 November; 6 November is a gap). The two days above the cell's 56–83% September range, 10 and 11 October (31 of 35 and 39 of 45 aircraft — its highest readings since 31 August), were followed by seven days inside it, 76% to 82%, on 12–18 October, and then by 54% (25 of 46) on 19 October — the first reading below the September floor since 1 September — 59% (27 of 46) and 60% (31 of 52) on 20–21 October, back inside at the bottom of the range, 76% (31 of 41) on 22 October, back in the middle of it, 60% (26 of 43) on 23 October, inside again near the bottom, 74% (31 of 42) on 24 October and 70% (33 of 47) on 25 October. Same shape as the rise, in the other direction: one day out, then back in. On 26 October it went out again: 47% (24 of 51 aircraft), below the floor for the second time and its lowest since 18 August — and it stayed out: 45% (17 of 38) on 27 October, 55.6% (25 of 45) on 28 October, a shade under the 56.1% floor, and 39% (14 of 36) on 29 October, its lowest since 17 August. Three days running below the floor is the run this page said it would report, and we reported it as that — a softening of the edge, not a step west; the fourth day deepened it, and on 30 October the cell came back inside its range at 60% (21 of 35 aircraft), ending the run at four days, stayed inside on 31 October at 56% (18 of 32), a fraction above the floor, and on 1 November at 68% (26 of 38), then went under it again on 2 November at 55% (18 of 33) on 3 November at 52% (16 of 31) and on 4 November at 51% (23 of 45) — a second run of three days under the floor, seven days under it since 26 October — came back inside on 5 November at 60% (26 of 43), and on 7 November read 48% (19 of 40), under the floor again — eight days under it since 26 October — and on 8 and 9 November 70% (28 of 40) and 68% (25 of 37), back inside — the step test is about readings above the range, and Tampere's cell stayed inside its band throughout but for one day, 3 November, when it went under it (see below). (Earlier versions of this page called 19 October the first reading below that floor since the series opened; it was the first since 1 September — the cell read 22–51% on eight August days, before the escalation.) The sharp drop-off on this row stays where it was, between 24°E and 23°E — 68% against 37% (10 of 27 aircraft) on 9 November, 70% against 32% (7 of 22) on 8 November, 48% against 23% (5 of 22) on 7 November, 60% against 24% (7 of 29) on 5 November, 51% against 24% (9 of 38) on 4 November, 52% against 12% (3 of 26) on 3 November, 55% against 33% (8 of 24) on 2 November, 68% against 35% (7 of 20) on 1 November, 56% against 26% (6 of 23) on 31 October, 60% against 21% (6 of 28) on 30 October, 39% against 21% (4 of 19) on 29 October, 55.6% against 31% (9 of 29) on 28 October, 45% against 29% (8 of 28) on 27 October and 47% against 26% (8 of 31) on 26 October, smaller steps because the 24°E side softened, after 70% against 30% on 25 October, 74% against 33% on 24 October, 60% against 24% on 23 October, 76% against 36% on 22 October and three earlier days when the step was smaller for the same reason; a softer edge is not a westward step, and we did not call it one. Tampere's cell, which read 50% on 10 October (matching its 31 August reading, just above the 18–48% band it held through September), has read 35%, 35%, 40%, 35%, 39%, 31%, 29%, 36%, 38%, 27%, 38%, 36%, 24%, 33%, 30%, 26%, 29%, 31%, 21%, 21%, 26%, 35%, 33%, 12%, 24%, 24%, 23%, 32% and 37% since — inside the band on all but one of those days: under it on 3 November, 12% (3 of 26 aircraft), its first reading below the band since 1 September and its lowest since 25 August, and back inside at 24% (9 of 38, then 7 of 29) on 4 and 5 November 23% (5 of 22) on 7 November, 32% (7 of 22) on 8 November and 37% (10 of 27) on 9 November: one day, on a thin sample, on the soft side; a step west would show as readings above the band, not below it. At 25°E the last forty-five observed days read 97% → 98% → 86% → 81% → 83% → 85% → 88% → 90% → 100% → 95% → 95% → 92% → 90% → 97% → 93% → 95% → 98% → 93% → 88% → 91% → 96% → 92% → 83% → 79% → 70% → 74% → 79% → 78% → 74% → 84% → 90% → 83% → 59% → 71% → 67% → 76% → 87% → 73% → 69% → 61% → 60% → 62% → 71% → 76% → 84% (25 September–9 November; 6 November is a gap): a dip, a climb back, two softer days, a climb back, six soft days and then 84% (36 of 43) on 24 October, 90% (35 of 39) on 25 October, its highest since 16 October, and 83% (33 of 40) on 26 October; then 59% (26 of 44) on 27 October — below the cell's 70–98% September range (its floor is 69.7%) for the first time since 1 September, and its lowest since 13 August — 71% (29 of 41) on 28 October, a fraction back inside, 67% (18 of 27) on 29 October, under the floor a second time in three days, on a thinner sample, 76% (29 of 38) on 30 October, back inside, 87% (26 of 30) on 31 October, its highest since 25 October, 73% (24 of 33) on 1 November, 69.4% (25 of 36) on 2 November, a fraction under the floor a third time, 61% (23 of 38) on 3 November, under it a fourth time and two days running for the first time since 1 September, then 60% (26 of 43) on 4 November and 62% (26 of 42) on 5 November — four days running under the floor, the longest such run since the series opened on 10 August — 71% (25 of 35) on 7 November, back inside, ending the run at four observed days, 76% (32 of 42) on 8 November and 84% (31 of 37) on 9 November, its highest since 31 October. The 70% (33 of 47) of 19 October had been a fraction inside that floor; 74%, 79%, 78% and 74% followed it, then 84%, 90% and 83%. Logged, not called a retreat: until now this cell had climbed back from every dip since late September; this time it climbed back a fraction for one day, fell under again, came back inside on 30 October, stayed inside on 31 October and 1 November, slipped a fraction under again on 2 November stayed clearly under for three days, on 38–43 aircraft a day, and came back inside on 7–9 November. Until 26 October, every easing on this edge since late September had reversed: a drop at 24°E from 76% to 62% over 21–23 September was back at 80% the next day; the 24°E fall to 60% on 28 September, logged as a direction to watch rather than a trend, was back at 82% a day later; Tampere's one-day drop to 22% on 30 September read 30% again on 1 October; the tracked cell's own 86% on 2 October was followed by 91% and two days at 100%, and its 84% on 18 October by 90%, 89%, 93%, 100%, 92%, 100%, 100% and 95% — its 86%, 82% and 77% of 27–29 October were the next test of that pattern, and over 30 October–3 November it read 94% (17 of 18), 91% (10 of 11), 100% (17 of 17), 89% (17 of 19) and 100% (17 of 17) — the pattern held again, and 4–5 November's 88% and 83% are the next test of it — 7–9 November's 94% (16 of 17), 100% (19 of 19) and 100% (25 of 25) the answer so far; and 6 October's softer 61°N row — every cell from 22°E to 27°E below its 5 October figure, 27°E down to 80% on ten aircraft — was gone within a day, 27°E reading 94% on 7 October, 92% on 8 October and 100% on every day from 9 to 18 October. The rises reverse just as readily: the 24°E readings of 10–11 October were back inside the September range a day later, and Helsinki's own cell climbed four days running to 77% on 1 October and fell to 63% the next — so a run of rises is not a trend either. That is the honest lesson about reading a trend off a day or two of a thin cell. The row to the south rose for a day and has fallen back. On 8 October the 60°N cells west of the Gulf pocket all read their highest since the 30–31 August spike — 84% at 25°E, 88% at 24°E (Helsinki), 74% at 23°E, 65% at 22°E, 50% at 21°E — as did Tallinn's cell across the Gulf at 89%. By 5 November they read 42%, 41%, 36%, 28%, 14% and Tallinn 53% — Tallinn's cell under its September floor for a twelfth day running (see below), Helsinki's cell at a new low for its series and 60°N 25°E, 60°N 23°E and 60°N 21°E under their floors too, a day after all but Tallinn's had read inside their ranges — twelve days after Helsinki's cell and Tallinn's sat a shade above their September tops; three of these cells have poked out above their tops on single days since: 60°N 21°E, which sat above its 32% September top for four days (50%, 40%, 37%, 43% over 8–11 October), fell to 22–26% over 12–17 October, jumped to 46% (26 of 57 aircraft) on 18 October, read 22% on 19 October, 34% (23 of 68) on 20 October — above the top a second time in three days — and 25%, 28%, 21%, 20%, 18%, 17%, 26% and 17% on 21–28 October, 14.5% (9 of 62) on 29 October, a fraction under its September floor, 10% (6 of 58) on 30 October, under it again, 30% (18 of 60) on 31 October, back inside near its top, 19% on 1 and 2 November, 22% on 3 and 4 November and 14% (9 of 64) on 5 November, a fraction under its floor a third time; two excursions above the top, neither followed by a second day; 60°N 25°E read 78% (86 of 111 aircraft) on 14 October, above its 75% September top, 66% the next day, 76% (109 of 144) on 16 October — a fraction above that top again — 56% on 17 October, 77% (161 of 209) on 18 October — above the top for a third separate day — then 75% (96 of 128, a fraction inside), 67%, 73%, 60%, 52%, 73%, 73%, 57% and 52% on 19–27 October, 49.5% (46 of 93) on 28 October, a shade above its 48% September floor and then its lowest since 8 September, 55% (56 of 101) on 29 October, 43% (55 of 128) on 30 October — under that floor for the first time, its lowest since 13 August — 40.5% (45 of 111) on 31 October, under it a second day and then the lowest reading in a series that opened on 10 August, 71% (63 of 89) on 1 November, back inside, 46.5% (53 of 114) on 2 November, under the floor a third time, 54% (61 of 113) on 3 November and 51% (64 of 126) on 4 November, back inside, and 42% (50 of 119) on 5 November, under the floor a fourth time; 60°N 23°E read 64% (65 of 102) on 20 October, above its 62% September top for the first time since 8 October, then 59%, 60%, 44% and 49% on 21–24 October and 36% (39 of 107) on 25 October — below its 39% September floor, its lowest since 23 August — and 28% (30 of 107) on 26 October, below it a second day running and the lowest reading in a series that opened on 10 August, then 39.2% (40 of 102) on 27 October — a fraction above that floor — 41% (43 of 105) on 28 October, back inside at the bottom, 36% (36 of 101) on 29 October, under the floor a third time, 43% (43 of 101) on 30 October, back inside, 42% (37 of 88) on 31 October and 46% (45 of 98) on 1 November, inside again, 32% (32 of 100) on 2 November, under the floor a fourth time and the third-lowest reading of its series, 42% (38 of 91) on 3 November and 44% (41 of 94) on 4 November, back inside, and 36% (34 of 94) on 5 November, under the floor a fifth time; and Tallinn's cell read 78% (174 of 223) on 18 October, 68% (145 of 212) on 19 October — below its 72% September floor, as it was on 1, 5, 7, 13 and 15 October — 74% on 20 October, 72% (137 of 191) on 21 October, a hair inside, 80% (176 of 220) on 22 October, 78% (156 of 200) on 23 October and 85% (164 of 193) on 24 October — above its 83% September top for the first time since 8 October, on the same day Helsinki's own cell read a shade above its top (73%, 107 of 147), 71% (128 of 181) on 25 October, a shade under its floor again, with Helsinki's cell back at 62%, and 64% (114 of 179) on 26 October — a second day under the floor, the first back-to-back pair since 17–18 August and its lowest since 12 August, with Helsinki's cell at 50% (74 of 149), inside its September range near the floor, its lowest since 6 September; then 62% (104 of 168) on 27 October — at the time the lowest reading in a series that opened on 10 August — 63% (102 of 161) on 28 October, 61% (94 of 153) on 29 October, then a new series low, 57% (88 of 154) on 30 October, lower again and six days running under the floor, 67% (104 of 156) on 31 October, up but under the floor a seventh day, 58% (94 of 163) on 1 November, 55% (85 of 155) on 2 November, a new series low, 49% (68 of 140) on 3 November, lower again, 57% (86 of 150) on 4 November and 53% (83 of 156) on 5 November — twelve days running under the floor, with Helsinki's cell at 55% (79 of 144), 51% (77 of 151), 52% (75 of 145), 47% (72 of 154) — under its 49% September floor for the first time and then the lowest reading of its own series, which also opened on 10 August — 56% (72 of 129) and 61% (90 of 148), back inside its range, 42% (62 of 147) on 2 November, under the floor a second time and a new low for its series, 53% (72 of 137) and 53% (81 of 152) on 3 and 4 November, back inside, and 41% (59 of 145) on 5 November, under the floor a third time and lower again. On 7 November, after a day missing from our per-cell feed, Tallinn's cell read 54% (79 of 146), under its floor on a thirteenth observed day in a row, Helsinki's 45% (60 of 132), under its floor a fourth time, 60°N 25°E 51% (60 of 117), back inside, and 60°N 23°E 36% (27 of 74), under its floor a sixth time. On 8 November Tallinn's cell read 63% (104 of 166), under its floor a fourteenth observed day running, while Helsinki's read 59% (86 of 147), 60°N 25°E 59% (78 of 132) and 60°N 23°E 45% (39 of 87), all three inside their ranges. On 9 November Tallinn's cell read 55% (84 of 153), under its floor a fifteenth observed day running, Helsinki's 42% (62 of 147) — under its floor a fifth time, the same count as on 2 November, level with it as the second-lowest reading of its series — 60°N 25°E 52% (50 of 97), inside, and 60°N 23°E 35% (37 of 106), under its floor a seventh time. These are well-sampled cells (46–220 aircraft a day), so these are measurements, not noise; what they show is a one-day surge that has unwound into a row of cells wobbling around the edges of their September ranges — 24 October was the latest wobble on the high side, and the second day above at both Helsinki and Tallinn that would have been the first sign of a repeat of 8 October did not come. 26 October was the softest day on this row since August: every cell from 25°E west, and Tallinn's, read lower than the day before, while the August pocket's own Gulf cell (60°N 26°E) held at 97% (38 of 39). 27 October–5 November did not undo it on the Gulf side: Tallinn's cell set five series lows and stayed under its floor, Helsinki's cell and 60°N 25°E went under their own floors for the first time on 30 October — Helsinki's came back inside on 31 October and 1 November, went under again on 2 November to a new series low, came back inside on 3 and 4 November and went under a third time on 5 November, to a lower one, while 60°N 25°E fell to a series low on 31 October and was under again on 2 and 5 November, inside between — 60°N 23°E moved either side of its floor, and the Gulf pocket read 100% (39 of 39) on 27–29 October, 97% (31 of 32) on 30 October, 100% on 31 October and 1 November (35 of 35, 42 of 42), 95% (36 of 38) on 2 November, 97% on 3 and 4 November (35 of 36, 37 of 38) and 87% (40 of 46) on 5 November, under its September range for the first time since 1 September — and 98% (40 of 41), 96% (45 of 47) and 97% (29 of 30) on 7–9 November, back inside. The easing also reached the pocket's own row — the tracked cell read 86%, 82% and 77% on 27–29 October — but that row came back on 30 October–1 November (94%, 91% and 100% at 26°E, 76%, 87% and 73% at 25°E, 60%, 56% and 68% at 24°E), on 2 November the tracked cell held at 89% while 25°E and 24°E slipped a fraction under their floors again, and on 3 November it read 100% while both went further under (61% and 52%) and Tampere's cell fell under its own band (12%); on 4 November it read 88%, with 25°E and 24°E still under (60% and 51%) and Tampere back inside (24%), and on 5 November 83%, with 25°E under a fourth day (62%), 24°E back inside (60%) and Tampere inside (24%). The whole 59–62°N, 21–30°E box averaged 57%, 60%, 59%, 56%, 51%, 60%, 58%, 55%, 58%, 58%, 58%, 62%, 65% and 59% flagged on 26 October–9 November (6 November missing), all but 8 November under its 64–74% September range — 30 October the lowest day since the series opened, 2 November the third-lowest, and 8 November, at 64.6%, back inside it for the first time since 25 October, before 58.5% on 9 November. The calendar week of 26 October–1 November closed at 57.3%, the lowest week since the series opened (the previous low, 10–16 August, 62.8%; every week since then pooled 67.6–72.5%) — the call this page said it would make once 1 November was in. It closed against a feed-wide low: the flagged share across our feed worldwide pooled 4.8% that week, against 5.4–5.9% in every other full week since August, and the box and the feed-wide share both fell by about a sixth on the week before (69.3% to 57.3%, 5.75% to 4.77%); the feed-wide share read 4.1% on 30 October and 1 November, its two lowest days since 1 September, and 4.4% on 2 November. So the week's low is measured, and it is not cleanly regional — the Gulf-row lows of 30 October and 2 November came on days of low readings across our whole feed. On 3–5 November the feed-wide share read 5.0%, 5.2% and 5.0%, inside its 4.7–6.4% September range, while the box read 58.4%, 57.8% and 58.0%, under its own — three days of a soft region against a feed reading normally, and 7 November a fourth (62.2% against 5.0%); on 8 November it went the other way, the box back inside its range at 64.6% while the feed-wide share fell to 4.5%, under its own. The week of 2–8 November closed at 59.3% in the box on six observed days — its second-lowest week, under every September week (68.7–70.2%) — and at 4.8% feed-wide, inside the feed-wide September range read day by day but the second-lowest week since August, under every September week (5.4–5.7%). So the second soft week is measured too, and again not cleanly regional: the box and the feed-wide share both sat at their second-softest weeks. On 9 November, the first day of the week of 9–15 November, the box read 58.5%, under its range again, with the feed-wide share at 4.8%, inside its own — the fifth such day since 3 November. Fifteen observed days running under the floor at Tallinn, eight at 24°E since 26 October and a four-day run at 25°E that ended on 7 November are a measured easing on the edge of the zone, read against that feed-wide dip and, since 3 November, against a feed-wide share back inside its range on most days. It began on the day after the clocks went back and the winter timetable began, which we note and do not treat as an explanation. We report both directions; we call neither a westward advance nor a retreat. The gradient itself is the useful operational fact: the degradation is not blanket Finnish airspace, it is a bounded eastern-and-inland zone with measurable edges, and where those edges move is what our detection layer is watching.

Helsinki's own cell: a September range, a rise that reversed — and one day at 88% that did not hold. On 30–31 August, Helsinki-Vantaa's airspace cell ran 90–95% of aircraft flagged. Over 1–29 September it ran 49–72%, dipping to 49% on 6 September; it then read 75% (128 of 171 aircraft) on 30 September and 77% (142 of 185) on 1 October — four rises in a row — before falling back inside its September range: 63%, 73%, 67%, 62%, 64% and 60% (107 of 178) over 2–7 October. We logged that rise as a direction to watch rather than a trend, and it did not become one. On 8 October the cell read 88% (146 of 166 aircraft) — its highest since 31 August's 90%, on a full-sized sample, and on the same day as the jump in the FIR-wide figure. On 9 October it read 70% (109 of 155), on 10 October 57% (82 of 143), then 65% (105 of 162), 63% (102 of 162), 60% (93 of 156), 69% (111 of 162), 66% (118 of 178), 72% (129 of 178), 68% (102 of 150) and 71% (140 of 198) on 11–18 October, back inside its September range — 8 October was one day. On 19 October it read 74% (129 of 175), above the 72% top of its 1–29 September range for the first time since 8 October; 20 October read 72% (119 of 165), level with that top, 21 October 71% (128 of 181), 22 October 67% (113 of 168) and 23 October 64% (99 of 154), inside it, and 24 October 73% (107 of 147), a shade over the top again; 25 October read 62% (94 of 151), back inside, 26 October 50% (74 of 149) — still inside, about a point above the 49% floor and its lowest since 6 September — and 27–29 October 55% (79 of 144), 51% (77 of 151) and 52% (75 of 145), near the floor; on 30 October it read 47% (72 of 154) — under its September floor for the first time and the lowest reading in a series that opened on 10 August, on a day the flagged share across our whole feed was also at its lowest since 1 September; on 31 October it read 56% (72 of 129) and on 1 November 61% (90 of 148), back inside its range; on 2 November it read 42% (62 of 147), under the floor a second time and a new low for its series, on a day the feed-wide share was also low (4.4%); on 3 and 4 November it read 53% (72 of 137, then 81 of 152), back inside; on 5 November it read 41% (59 of 145), under the floor a third time and lower again, on a day the feed-wide share sat inside its September range (5.0%); on 7 November, after a day missing from our per-cell feed, it read 45% (60 of 132), under the floor a fourth time and its third-lowest reading, the feed-wide share again inside its range (5.0%); on 8 November it read 59% (86 of 147), back inside, with the feed-wide share low again (4.5%); on 9 November it read 42% (62 of 147), under the floor a fifth time — the same count as on 2 November, level with it as the second-lowest reading of its series — with the feed-wide share back inside its range (4.8%). Two separate days a shade over the top, neither followed by a second, and five days under the floor — 30 October and 2, 5, 7 and 9 November, the last three either side of a day missing from our feed and a day back inside — on a cell that has shown before how readily it swings: logged, not called. So two things remain true at once: the strong zone shifted inland and away from the capital's immediate airspace while remaining as intense as before where it sits, and the capital's own cell spends most days well below that — 8 October the exception so far. Averaged across the whole Finland FIR (EFIN) the last seven observed days come out at 6.8%, under an 8.2% 90-day average, with 29 September at 9.3%, then 8.5% on 30 September, 9.4% on 1 October, 8.6% on 2 October, 10.4% on 3 October — on 2,652 aircraft, the FIR's thinnest count since 5 September — 9.1% on 4 October, 7.0% on 5 October, 7.3% on 6 October, 6.2% on 7 October — its lowest since 9 September — 16.4% on 8 October, on 3,309 aircraft — its highest since the 20.0% of 31 August — 10.4% on 9 October, on 3,377 aircraft, 6.7% on 10 October, on 2,720 aircraft, then 8.3% on 11 October, on 3,020 aircraft, 7.1% on 12 October, on 3,353 aircraft, 7.6% on 13 October, on 3,200 aircraft, 8.0% on 14 October, on 3,553 aircraft, 8.7% on 15 October, on 3,631 aircraft, 9.1% on 16 October, on 3,544 aircraft, 8.0% on 17 October, on 2,955 aircraft, 9.6% on 18 October, on 3,244 aircraft — at the top of the band but not above it — then 7.2% on 19 October, on 3,543 aircraft, 8.7% on 20 October, on 3,462 aircraft, 7.8% on 21 October, on 3,658 aircraft, 8.6% on 22 October, on 3,463 aircraft, 6.2% on 23 October, on 3,389 aircraft — its lowest since 7 October — 7.7% on 24 October, on 2,916 aircraft — 25 October is missing from the FIR feed, a gap in every FIR rather than a reading — 5.4% on 26 October, on 3,533 aircraft, under the band's floor and its lowest since 10 August, 6.2% on 27 October, on 3,417 aircraft, 8.7% on 28 October, on 3,447 aircraft, 6.8% on 29 October, on 3,451 aircraft, 7.9% on 30 October, on 3,461 aircraft, 6.8% on 31 October, on 2,889 aircraft, 7.3% on 1 November, on 3,129 aircraft, and 5.9% on 2 November, on 3,378 aircraft — under the band's floor for a day, its lowest since 26 October — 6.1% on 3 November, on 3,051 aircraft, back inside the band, 6.6% on 4 November, on 3,391 aircraft, 5.5% on 5 November, on 3,125 aircraft — under the floor again for a day, its lowest since 26 October — 6.3% on 6 November, on 3,368 aircraft, back inside the band, 6.4% on 7 November, on 2,780 aircraft, 7.7% on 8 November, on 3,084 aircraft, 6.7% on 9 November, on 3,326 aircraft, and 8.3% on 10 November, on 3,449 aircraft, the FIR feed's newest day. The seven-day average eased from 9.2% to 8.1% when 8 October's 16.4% dropped out of the window, to 7.9% once 9 October's 10.4% followed it, read 8.4% over 16–22 October, 8.0% over 17–23 and 18–24 October, 7.2% over 21–28 October, 7.1% over 22–29 October, 7.0% over 23–30 October, 7.1% over 24–31 October, 27 October–2 November and 28 October–3 November, 6.8% over 29 October–4 November, 6.6% over 30 October–5 November, 6.4% over 31 October–6 November, 6.3% over 1–7 November — the lowest seven-day figure since the start of August — 6.4% over 2–8 November and 6.5% over 3–9 November, and reads 6.8% now (4–10 November), with one day of that window, 5 November, under the band's floor. One FIR-wide figure hides the pocket; the pocket alone overstates the FIR. We label which scale each number belongs to for exactly that reason.

Why 90% flagged is a detection signal, not a grounding. The figure counts aircraft broadcasting downgraded navigation integrity in their ADS-B position-quality fields — the consistent fingerprint of GNSS interference. It does not mean those aircraft cannot navigate: airliners cross-check satellite positioning against inertial reference systems and ground-based aids (DME/DME, VOR) and fall back to them when GNSS degrades. What the signal tells us is where interference is concentrated and how persistent it is, measured from the aircraft themselves. It is also why Helsinki's departure board shows nothing: wind closes runways and ash threatens engines, but interference degrades a position source that crews and dispatchers have documented back-ups for.

Small cells, honest arithmetic. The 61°N 26°E cell is quiet airspace by traffic volume — 6 to 37 aircraft a day in our feed. A 100% reading on six aircraft (30 August) carries far less weight than the 100% on twenty-eight aircraft recorded on 9 September, and we mark the thin days rather than smoothing them away — and a day with no data at all, like 14 September, is left out rather than filled in. That cuts both ways on the newest bars: 29 September's 94% rests on sixteen aircraft, 30 September's 91% on twenty-three, 1 October's 100% on nineteen, 2 October's 86% on twenty-one, 3 October's 91% on just eleven, the thinnest sample since 30 August, so it is a direction, not a measurement, the 100% of 4 and 5 October on nineteen each, 6 October's 93% on fourteen, 7 October's 96% on twenty-five, 8 October's 94% on eighteen, 9 October's 90% on twenty, 10 October's 96% on twenty-three, 11 October's 96% on twenty-seven, 12 October's 100% on twenty-seven, 13 October's 95% on twenty-two, 14 October's 100% on twenty-nine, 15 October's 100% on twenty-eight, 16 October's 93% on twenty-seven, 17 October's 100% on seventeen, 18 October's 84% — the softest day of the run until 28 October — on twenty-five, 19 October's 90% on twenty-nine, 20 October's 89% on nineteen, 21 October's 93% on thirty, 22 October's 100% on twenty, 23 October's 92% on twenty-five, 24 October's 100% on twenty-four, 25 October's 100% on twenty-five, 26 October's 95% on twenty-one, 27 October's 86% on twenty-two, 28 October's 82% on twenty-eight, a full-sized sample for this cell, so a measurement, not noise, 29 October's 77% — the lowest reading of the run — on thirteen, the thinnest sample since 3 October, so a direction, not a measurement, 30 October's 94% on eighteen, 31 October's 91% on eleven, level with 3 October as the thinnest sample since 30 August, so a direction, not a measurement, 1 November's 100% on seventeen, 2 November's 89% on nineteen, 3 November's 100% on seventeen, 4 November's 88% on twenty-five and 5 November's 83% — the third-lowest reading since 1 September — on eighteen, a thin sample, so a direction, not a measurement, 7 November's 94% on seventeen, 8 November's 100% on nineteen and 9 November's 100% on twenty-five (6 November, like 14 September, has no data and is left out). Seventy-nine of the ninety days observed in this cell have come in at 85% or above. The same discipline governs the traffic section below, where 26 October, published at 432 on 28 October as a count still filling, gained six late-arriving departures and settled at 438 — then the window's lowest Monday — where 28 October, published at 424 as a count still filling, gained six more departures and settled at 430, where 29 October, published at 434 as a count still filling, gained three more departures and settled at 437 — then the window's lowest Thursday — where 27 October, a day with a one-hour network-wide gap in our feed that never filled, is flagged as a partial count, as this page said it would be, rather than ranked through, where 30 October, published at 438 as a count that might still gain, settled at exactly 438, where 31 October, held back for a cycle as a count still filling, settled at 365 on a day our whole flight feed ran short and is plotted as a floor rather than ranked, where 1 November, published at 407 as a count that might still gain, gained three departures and settled at 410 — the window's lowest Sunday — where 2 November, published at 429 as a count that might still gain, settled at exactly 429 — then the window's lowest Monday — where 3 November, published at 410 as a count that might still gain, settled at exactly 410 — the window's lowest full-count Tuesday — where 4 November, published at 422 as a count still filling, gained two departures and settled at 424 — the window's lowest full-count Wednesday — where 5 November, published at 416 as a count still filling, gained three departures and settled at 419 — the window's lowest Thursday — where 6 November, published at 432 as a count that might still gain, gained four departures and settled at 436 — the window's lowest Friday — where 7 November, published at 365 as a count that might still gain, settled at exactly 365 — the window's lowest full-count Saturday — where 8 November, published at 418 as a count still filling, gained ten departures and settled at 428, where 9 November, published at 424 as a count that might still gain, settled at exactly 424 — the window's lowest Monday — where three bars drawn for one cycle from a different receiver network's count were restored to the chart's own feed and the slip stays disclosed, where a wrong Sunday list from the 21 October cycle was corrected in the open, where 29 September — a day with a known two-hour hole in our flight feed — is plotted as the partial count it is rather than passed off as a quiet Tuesday, and where two older bars with smaller holes are disclosed. The same caution applies to the airport table below: Lappeenranta's cell reads 100% on six aircraft (6 of 6), too few to be a reading at all. The reason the September readings are the ones we are publishing on is that the sample roughly doubled while the share stayed put: 16 aircraft on 1 September, 21 to 31 a day from 4 to 28 September, then 16, 23, 19, 21, 11, 19, 19, 14, 25, 18, 20, 23, 27, 27, 22, 29, 28, 27, 17, 25, 29, 19, 30, 20, 25, 24, 25, 21, 22, 28, 13, 18, 11, 17, 19, 17, 25, 18, 17, 19 and 25 on the last forty-one days observed, the rate never dropping below 77%, and that on the thinnest of them. Growing denominator, near-unchanged ratio — that is what turns a noisy cell into a signal. Every figure here is an aggregate of crowdsourced receiver telemetry: a floor, not a total.

What would change the picture. Two measured things are watched continuously: the high-flag zone widening further north or west — the Tampere cell and Helsinki's own cell are the tell-tales; and the FIR-level EFIN ratio breaking decisively out of the 6.0–9.6% band it has held since the 30–31 August spike rather than oscillating inside it, as it did over 23 September–7 October (8.3%, 8.2%, 7.2%, 9.6%, 8.4%, 8.3%, 9.3%, 8.5%, 9.4%, 8.6%, 10.4%, 9.1%, 7.0%, 7.3%, 6.2%). On 8 October one of each moved, for one day: EFIN read 16.4%, its highest since 31 August, and Helsinki's own cell 88%, its highest since the same spike — while the Tampere cell held at 47%, inside its September band. We said a run of days above 9.6%, or the seven-day average crossing it, would be the break, and set the test for 10 October: a reading above about 10.8% would lift the average over 9.6%, and a third straight day above the band would be a run. Neither happened: EFIN read 6.7% on 10 October, back inside the band, and the seven-day average eased to 9.0%. Our call: no break — a two-day spike, 16.4% then 10.4%, the same shape as the 30–31 August spike, which also fell back after two days; 11 October to 10 November read 8.3%, 7.1%, 7.6%, 8.0%, 8.7%, 9.1%, 8.0%, 9.6%, 7.2%, 8.7%, 7.8%, 8.6%, 6.2%, 7.7%, 5.4%, 6.2%, 8.7%, 6.8%, 7.9%, 6.8%, 7.3%, 5.9%, 6.1%, 6.6%, 5.5%, 6.3%, 6.4%, 7.7%, 6.7% and 8.3% (25 October is missing from the FIR feed) — 18 October touched the ceiling of the band and the next day fell away from it, and 26 October, 2 November and 5 November each dipped under its floor for a day, none followed by a second — 10 November, the newest day, read 8.3%, inside it. The seven-day average, 9.2% while 8 October was still inside the window, eased to 8.1% once it dropped out and to 7.9% once 9 October followed — 8.4% over 16–22 October, 8.0% over 17–23 and 18–24 October, 7.2% over 21–28 October, 7.1% over 22–29 October, 7.0% over 23–30 October, 7.1% over 24–31 October, 27 October–2 November and 28 October–3 November, 6.8% over 29 October–4 November, 6.6% over 30 October–5 November, 6.4% over 31 October–6 November, 6.3% over 1–7 November, the lowest since the start of August, 6.4% over 2–8 November, 6.5% over 3–9 November and 6.8% now (4–10 November), under the 90-day figure — and it never crossed 9.6%. Helsinki's own cell read 57% on 10 October, then 65%, 63%, 60%, 69%, 66%, 72%, 68%, 71%, 74%, 72%, 71%, 67%, 64%, 73%, 62%, 50%, 55%, 51%, 52%, 47%, 56%, 61%, 42%, 53%, 53%, 41%, after a day missing from our per-cell feed, 45%, 59% and 42% — two days, 19 and 24 October, a shade above its September top, neither followed by a second, 26–29 October near its floor, 30 October under it, 31 October and 1 November back inside, 2 November under it again, 3 and 4 November back inside, 5 November under it a third time, the lowest reading of its series, 7 November under it a fourth time, 8 November back inside and 9 November under it a fifth time. The watch then moved to the other tell-tale: on 10 October 61°N 24°E read 89%, above its 56–83% September range for the second time in three days, and Tampere's cell 50%, just above its 18–48% band for the first time since the spike, on twenty-two aircraft — and we said a second day above those ranges at both 24°E and 23°E would be the first sign of the edge stepping west. It was not met: on 11 October 24°E read 87%, a second day above its range, but Tampere's cell fell to 35%; on 12 October 24°E itself fell back to 76% (31 of 41 aircraft), inside its range, with Tampere at 35% (11 of 31); on 13 October they read 79% (38 of 48) and 40% (15 of 38), on 14 October 76% (42 of 55) and 35% (12 of 34), on 15 October 64% (36 of 56) and 39% (13 of 33), on 16 October 70% (35 of 50) and 31% (8 of 26), on 17 October 72% (18 of 25) and 29% (6 of 21), on 18 October 82% (27 of 33) and 36% (9 of 25) — seven days inside at both — and then 24°E went out the other side: 54% (25 of 46) on 19 October, below its September floor, with Tampere at 38% (10 of 26); 59% (27 of 46) and 27% (8 of 30) on 20 October; 60% (31 of 52) and 38% (15 of 39) on 21 October; 76% (31 of 41) and 36% (10 of 28) on 22 October; 60% (26 of 43) and 24% (8 of 33) on 23 October; 74% (31 of 42) and 33% (8 of 24) on 24 October; 70% (33 of 47) and 30% (7 of 23) on 25 October; 47% (24 of 51) and 26% (8 of 31) on 26 October, 24°E below its floor a second time; 45% (17 of 38) and 29% (8 of 28) on 27 October; 55.6% (25 of 45) and 31% (9 of 29) on 28 October; 39% (14 of 36) and 21% (4 of 19) on 29 October; 60% (21 of 35) and 21% (6 of 28) on 30 October; 56% (18 of 32) and 26% (6 of 23) on 31 October; 68% (26 of 38) and 35% (7 of 20) on 1 November; 55% (18 of 33) and 33% (8 of 24) on 2 November; 52% (16 of 31) and 12% (3 of 26) on 3 November; 51% (23 of 45) and 24% (9 of 38) on 4 November; 60% (26 of 43) and 24% (7 of 29) on 5 November; 48% (19 of 40) and 23% (5 of 22) on 7 November; 70% (28 of 40) and 32% (7 of 22) on 8 November; 68% (25 of 37) and 37% (10 of 27) on 9 November. One cell above its range for two days, then below it on 19 October, on four days running from 26 October and on three more running over 2–4 November, and again on 7 November; the other inside throughout but for 3 November, when it went under its band for a day, on twenty-six aircraft — the soft side again. The step test stays open as set: a second day above both ranges, together, remains the sign we would report, and it has not come. The other half of the test was met on 28 October: three days running below the 24°E floor is the run of days this page said it would report, and we reported it — as a softening of the edge, not a step; 29 October made it four, and on 30 October the cell came back inside its range; 2, 3 and 4 November put it under again — a second run of three days, which we report as that: the edge softening again, not a step west — and on 5 November it came back inside at 60%, before reading 48% on 7 November, under the floor again, and 70% and 68% on 8–9 November, back inside. Softening on the Gulf side is reported the same way: Tallinn's cell sat under its September floor on fifteen observed days running, 25 October–9 November (6 November missing), with new series lows of 57% on 30 October, 55% on 2 November and 49% on 3 November; Helsinki's own cell and 60°N 25°E went under theirs for the first time on 30 October, a day the share across our whole feed was also at its lowest since 1 September — Helsinki's came back inside on 31 October and 1 November, went under again on 2 November, to 42%, came back inside on 3 and 4 November and went under a third time on 5 November, to 41%, the lowest of its series, while 60°N 25°E fell to 40.5% on 31 October and was under again on 2 and 5 November; 60°N 23°E was under on seven days, 25–26 and 29 October and 2, 5, 7 and 9 November; the August pocket's own Gulf cell, 60°N 26°E, read 87% (40 of 46) on 5 November, under its September range for the first time since 1 September; and on 27–29 October the softening reached the pocket's own row, which came back at 26°E on 30 October–4 November while 25°E, 24°E and Tampere's cell all read under their September floors on 3 November, and read 83% at 26°E on 5 November, on eighteen aircraft. The box's calendar week of 26 October–1 November closed at 57.3%, the lowest week of its series, in the week the feed-wide share also had its lowest — so we report the easing as measured and do not call it regional. On 3–5 and 7 November the box read 58.4%, 57.8%, 58.0% and 62.2% while the feed-wide share sat inside its September range (5.0%, 5.2%, 5.0% and 5.0%); on 8 November the box read 64.6%, back inside its own range, and the feed-wide share 4.5%, under its. The week of 2–8 November closed at 59.3% in the box and 4.8% feed-wide (six observed days; 6 November is a gap in our per-cell feed). We said a week closing under the box's September range with the feed-wide share inside its own would be the first we could call regional, noting how close the feed-wide week sat to its own low: the feed-wide week sits inside the September range only when that range is read day by day — as a week it is the second-lowest since August, under every September week (5.4–5.7%), just as the box's is its second-lowest, under every September week (68.7–70.2%). Our call: not cleanly regional — a second soft week in the box, measured, alongside a second soft week across our whole feed. The next test is the week of 9–15 November: a box week under its September range with the feed-wide week back inside the 5.4–5.7% of September's weeks would be the first regional signal, and we will say so. On its first day, 9 November, the box read 58.5%, under its September range, and the feed-wide share 4.8% — under the 5.4–5.7% of September's weeks, inside their daily range; one day of seven, so no call yet. The tracked cell itself barely moved over 8–26 October — 94%, 90%, 96%, 96%, 100%, 95%, 100%, 100%, 93%, 100%, 84%, 90%, 89%, 93%, 100%, 92%, 100%, 100% and 95% — and then read 86% (19 of 22), 82% (23 of 28) and 77% (10 of 13) on 27–29 October — the last its lowest since 13 August, but on the thinnest sample since 3 October; pooled, 82.5% (52 of 63), the lowest three-day figure since 21 August; three days, logged, not called a level — and over 30 October–3 November it read 94% (17 of 18), 91% (10 of 11), 100% (17 of 17), 89% (17 of 19) and 100% (17 of 17), back up, as it came back after 18 October, then 88% (22 of 25) on 4 November and 83% (15 of 18) on 5 November — one soft day on a thin sample, logged — 94% (16 of 17) on 7 November, 100% (19 of 19) on 8 November and 100% (25 of 25) on 9 November, back above 85%. 61°N 25°E read below the tracked cell on 4–7 October, above it on 8 and 9 October, level with it on 10 October (95% against 96%), above it on 11 October (98% against 96%) and below it on every day from 12 October to 5 November (93%, 88%, 91%, 96%, 92%, 83%, 79%, 70%, 74%, 79%, 78%, 74%, 84%, 90%, 83%, 59%, 71%, 67%, 76%, 87%, 73%, 69%, 61%, 60% and 62% against 100%, 95%, 100%, 100%, 93%, 100%, 84%, 90%, 89%, 93%, 100%, 92%, 100%, 100%, 95%, 86%, 82%, 77%, 94%, 91%, 100%, 89%, 100%, 88% and 83%) — still no sign of the pocket's peak drifting a degree west; if anything the cell to the west of the peak has softened, falling under its own September floor on 27 and 29 October, by a fraction on 2 November and clearly on 3, 4 and 5 November — four days running, the longest run under it since the series opened — came back inside at 71% (25 of 35) on 7 November and read 76% (32 of 42) on 8 November and 84% (31 of 37) on 9 November. 61°N 27°E, at 80% on 6 October, read 94%, 92% and then 100% on ten straight days (9–18 October), then 94%, 92% and 96% on 19–21 October — inside the 90–100% range it held through September — 85% (17 of 20) on 22 October, below it for the first time since 6 October, 95% (18 of 19) on 23 October, back inside, 94% (17 of 18) on 24 October, 100% (18 of 18) on 25 October, 100% (16 of 16) on 26 October, 95% (20 of 21) on 27 October, 85% (17 of 20) on 28 October, below the range a second time — two separate days on twenty aircraft, logged, not called — 86% (6 of 7) on 29 October, 100% (6 of 6) on 30 and 31 October and 100% (10 of 10) on 1 November, all too thin to read, 73% (11 of 15) on 2 November, under its range on a sample nearly as thin, 92% (11 of 12) on 3 November, back inside on another, 94% (15 of 16) on 4 November and 80% (12 of 15) on 5 November, under it again on a thin sample, and 100% on 7–9 November (10 of 10, 11 of 11, then 16 of 16). A third signal — regulatory action, meaning NOTAMs suspending satellite-based approach procedures at Finnish airports — would be the clearest operational marker, but we flag its absence honestly: our NOTAM feed carries no EFIN items at all — checked against the set as refreshed on 11 November, after it did not refresh between 4 and 27 October; it holds only notices currently in force, so its silence here is a gap in our coverage, not evidence that nothing has been issued. This page updates from the same monitoring cycle as the HEL brief and the live regional GPS-interference map.

What the data showed — close-out, 14 November. This brief opened on 7 September to put on record that GNSS interference over Finland had moved inland, north-east of Helsinki, and to carry the 61°N 26°E series forward. It closes on 14 November with no recovery to report in the cell: the inland pocket did not end, it held — and we close it as a documented steady state, the way the Riga brief was closed in October. What has gone quiet is the FIR-level reading. The trigger cell. Over 10–13 November the 61°N 26°E cell read 88% (15 of 17), 92% (22 of 24), 94% (17 of 18) and 90% (27 of 30) — 91.0% (81 of 89) over the four days. Across 1–13 November it aggregates to 93.1% (229 of 246) on twelve observed days (6 November is a gap in our per-cell feed), against 93.9% (635 of 676) across October, 95.2% (680 of 714) across September and 87.2% (307 of 352) across 10–31 August. Since 1 September there are seventy-two observed days, every one between 77% and 100%, sixty-eight of them at 85% or above. Its softest stretch, 86%, 82% and 77% on 27–29 October, was back above 85% the next day, as every soft day of the series has been. There is no inflection in it, in either direction. The FIR-level figure. The Finland FIR (EFIN) ratio — the coarser reading, labelled as FIR-level — has not been above the 9.6% top of the band it has held since the 30–31 August spike on any day since 18 October. It read 6.0% on 12 November and 5.5% on 13 November, its lowest since 26 October; it averaged 6.7% over 1–13 November, against 8.1% in September and 8.2% in October, and 7.0% over the last seven days (7–13 November) against a 90-day average of 8.2%. That is the signal that has wound down: a quieter FIR-wide picture while the inland pocket inside it stays lit. Helsinki's own cell. 60°N 24°E read 74% (112 of 152), 72% (102 of 142), 62% (81 of 130) and 58% (88 of 153) over 10–13 November — all four inside its 49–75% September range, after five days under it between 2 and 9 November. Across 1–13 November it aggregates to 55.1% (955 of 1,732), against 65.1% in both September and October: the softest month of its series, measured over thirteen days, not called a trend. The test this page set. The week of 9–15 November was to give the first regional signal if the box over southern Finland and the Gulf ran under its September weeks while the share across our whole feed came back inside September's weekly 5.4–5.7%. At close, five of its seven days are in: the box at 63.2% (5,366 of 8,489 aircraft), under every September week (68.7–70.2%), and the feed-wide share at 4.9%, still under September's weeks. The test is not met; the week ends after this page closes, and we leave it unanswered rather than call it. Helsinki traffic. Monday 9 November settled on 13 November at exactly the 424 this page published (221 departures, 203 arrivals) — the lowest of thirteen Mondays, 5 under 2 November's 429; the HEL airport brief rates it −3% against its 438 Monday reference (verdict: normal). Sunday 8 November settled at 428. The series closes at 89 days averaging ~442 observed movements a day, excluding the partial 29 September; 1–24 October averaged about 451 and the sixteen days from 25 October, when the winter timetable began, about 419 — reported, not attributed. 10–13 November are not plotted: the nightly ingest of our flight feed failed network-wide on 12–14 November (about 13,000–16,000 flights worldwide a day against 107,000–119,000 on the days before), so those days are unreadable rather than low. The crowdsourced ADS-B network behind our navigation-integrity figures — a separate feed with its own counting — logged 446, 412, 424 and 438 movements at HEL on 10–13 November, level with its own 393–444 on the non-Saturday days of the fortnight before; a cross-check, not plotted. NOTAMs. Our NOTAM set, refreshed on 14 November, still carries no item for the Finland or Tallinn FIRs. It holds only notices currently in force, so the silence is a gap in our coverage, not evidence that nothing was issued. Why close now. The FIR-level reading has stayed inside its band for nearly four weeks and drifted lower, Helsinki traffic is rated normal, and the cell series has said the same thing since the start of September; a page that repeats a steady state adds nothing new. The live regional GPS-interference map keeps reading the cell daily. If the pocket widens west into the Tampere or Helsinki cells, or the EFIN ratio breaks above 9.6% on a run of days, the detection layer will raise it and a new brief will open.

Passenger questions, answered from the data

Is GPS jamming affecting flights to Helsinki right now?

This brief closed on 14 November 2026. Through 13 November the inland cell was still flagging 88–94% of aircraft a day while the FIR-level Finland ratio ran lower than in September or October; the rest of this answer was written on 11 November. As of 9 November 2026, the strongest interference sits inland over southeastern Finland, north-east of Helsinki, where 77–100% of aircraft flagged degraded navigation on every observed day from 1 September to 9 November; 86% (19 of 22 aircraft) on 27 October, 82% (23 of 28) on 28 October and 77% (10 of 13) on 29 October — the last on a sample too thin to weigh alone — were together the softest stretch since August, on days when the whole region read softer, and over 30 October–3 November the cell read 94%, 91%, 100%, 89% and 100% (17 of 18, 10 of 11, 17 of 17, 17 of 19 and 17 of 17), back where it ran before, then 88% (22 of 25) and 83% (15 of 18) on 4–5 November 94% (16 of 17) on 7 November, 100% (19 of 19) on 8 November and 100% (25 of 25) on 9 November (6 November is a gap in our per-cell feed). Helsinki-Vantaa's own airspace cell usually runs lower — 49–72% over 1–29 September and 60–73% over 2–7 October — and on 8 October it read 88%, its highest since the 90–95% of 30–31 August, on the same day the Finland-wide figure jumped; it was back at 70%, 57%, 65%, 63%, 60%, 69%, 66%, 72%, 68% and 71% over 9–18 October, read 74% on 19 October — a shade above its September top — then 72%, 71%, 67% and 64% on 20–23 October, 73% on 24 October, a shade above that top again, 62% on 25 October, back inside, 50%, 55%, 51% and 52% on 26–29 October, near the bottom of its range, 47% on 30 October, a little below it, 56% and 61% on 31 October and 1 November, back inside, 42% on 2 November — under its range again, on a day of low readings across our whole feed — 53% on 3 and 4 November, back inside, 41% on 5 November, under it a third time and the lowest reading of its series, 45% on 7 November, under it a fourth time, 59% on 8 November, back inside, and 42% on 9 November, under it a fifth time. Observed traffic at HEL is still rated normal for the weekday, reading 424 movements on Monday 9 November — settled, the lowest Monday of the window — after 428 on Sunday 8 November, the fourth-lowest Sunday, a settled 365 on Saturday 7 November, the lowest full-count Saturday, a settled 436 on Friday 6 November, the lowest Friday, a settled 419 on Thursday 5 November, the lowest Thursday, a settled 424 on Wednesday 4 November, the lowest full-count Wednesday, a settled 410 on Tuesday 3 November, the lowest full-count Tuesday, 429 on Monday 2 November, then the lowest Monday, 410 on Sunday 1 November, the lowest Sunday, 365 on Saturday 31 October, a floor from a day our whole flight feed ran short, 438 on Friday 30 October, then the window's lowest Friday, 437 on Thursday 29 October, then its lowest Thursday, 430 on Wednesday 28 October, then its lowest full-count Wednesday, a partial 428 on Tuesday 27 October, carrying a one-hour gap in our feed, 438 on Monday 26 October and 416 on Sunday 25 October. Aircraft in the affected area broadcast degraded navigation but continue operating on back-up systems. See the live HEL brief, and check your airline's status on the day of travel.

Has the GPS interference over Finland spread?

It has moved rather than grown uniformly. In August our strongest readings were over the eastern Gulf of Finland; since September the near-total flag rates extend a degree further north over the Finnish mainland — 94%, 91%, 100%, 89%, 100%, 88%, 83%, 94%, 100% and 100% at 61°N 26°E over 30 October–9 November (17 of 18, 10 of 11, 17 of 17, 17 of 19, 17 of 17, 22 of 25, 15 of 18, 16 of 17, 19 of 19 and 25 of 25 aircraft; 6 November is a gap in our per-cell feed), after 86%, 82% and 77% on 27–29 October, the softest stretch of the run, the last on a thin sample. On 8 October, for one day, elevated readings also spread west along the Gulf: Helsinki's own cell read 88%, Tallinn's 89% and 61°N 24°E 87%; by 23 October Helsinki's cell was at 64% and every 60°N cell west of the pocket, Tallinn's included, was inside its September range; on 24 October Helsinki's cell read 73% and Tallinn's 85%, both a shade above their September tops, and on 25 October 62% and 71% — one day, not repeated; over 26 October–5 November they read 50%, 55%, 51%, 52%, 47%, 56%, 61%, 42%, 53%, 53% and 41% and 64%, 62%, 63%, 61%, 57%, 67%, 58%, 55%, 49%, 57% and 53%, Tallinn's under its September floor twelve days running, its low the 49% of 3 November, Helsinki's at a new low for its series on 5 November — the softest stretch on that side of the zone since August, over 29 October–2 November on low readings across our whole feed, and on 5 November, when 60°N 26°E itself read 87%, its first reading under its September range since 1 September, with the feed-wide share back inside its own. 61°N 24°E read 89% and 87% on 10–11 October — above its 56–83% September range for two consecutive days — then 76%, 79%, 76%, 64%, 70%, 72% and 82% on 12–18 October, back inside it, then 54% on 19 October, its first reading below that range since 1 September, 59% and 60% on 20–21 October, back inside at the bottom, 76% on 22 October, 60% on 23 October, 74% on 24 October, 70% on 25 October, 47%, 45%, 55.6% and 39% on 26–29 October — below the range four days running, a softening of the edge — 60% and 56% on 30–31 October and 68% on 1 November, back inside, 55%, 52% and 51% on 2–4 November, under it again three days running, 60% on 5 November, back inside, 48% on 7 November, under it again, and 70% and 68% on 8–9 November, back inside. Tampere-Pirkkala's cell (61°N 23°E) read 35%, 35%, 40%, 35%, 39%, 31%, 29%, 36%, 38%, 27%, 38%, 36%, 24%, 33%, 30%, 26%, 29%, 31%, 21%, 21%, 26%, 35% and 33% over 11 October–2 November, inside its 18–48% September band, then 12% (3 of 26 aircraft) on 3 November, under it for the first time since 1 September, 24% on 4 and 5 November, 23% on 7 November, 32% on 8 November and 37% on 9 November, back inside — softer for a day, not further west: the drop-off between 24°E and 23°E holds and the edge has not stepped west. At FIR level, EFIN read 5.4% on 26 October — under its 6.0–9.6% post-spike band for a day — then 6.2%, 8.7%, 6.8%, 7.9%, 6.8% and 7.3%, 5.9% on 2 November, under the band for a day again, 6.1% and 6.6% on 3–4 November, back inside, 5.5% on 5 November, under it for a day a third time, 6.3% on 6 November, back inside again, 6.4% on 7 November, 7.7% on 8 November, 6.7% on 9 November and 8.3% on 10 November; the 7-day average is 6.8%, up from the 6.3% of 1–7 November, the lowest since the start of August, and under the 8.2% 90-day average, with both spike days (8–9 October) long out of the window.

Is it dangerous to fly when there's GPS interference?

Aircraft carry independent navigation systems — inertial reference and ground-based aids — that crews and air-traffic control use when satellite positioning degrades, and regulators publish procedures for interference-prone airspace. We publish the measured picture: where interference is concentrated, how persistent it is and how it is moving. Operational and go/no-go decisions belong to operators and authorities.

Why does Helsinki's traffic look normal if 90% of aircraft are flagging?

Because the two measure different layers. The figure counts aircraft broadcasting degraded position quality in a specific airspace cell; the traffic figure counts departures and arrivals at the airport. Interference degrades a navigation source with documented back-ups — it does not close a runway — so it leaves no signature in the movement count. Helsinki averaged about 442 observed movements a day over the 88 days to 9 November (excluding 29 September, a partial count). The first sixteen days of the winter timetable sit at or near the bottom of their weekday ranges — 416 on Sunday 25 October, 438 on Monday 26 October, a partial 428 on Tuesday 27 October, 430 on Wednesday 28 October, 437 on Thursday 29 October and 438 on Friday 30 October, all settled, the last then the window's lowest Friday; 365 on Saturday 31 October, a floor from a day our whole flight feed ran short; 410 on Sunday 1 November, 429 on Monday 2 November and 410 on Tuesday 3 November, settled, the window's lowest Sunday, then its lowest Monday, and its lowest full-count Tuesday; 424 on Wednesday 4 November, settled, the window's lowest full-count Wednesday; 419 on Thursday 5 November and 436 on Friday 6 November, settled, each lower than any earlier count for its weekday; 365 on Saturday 7 November, settled, level with 31 October and under every other Saturday; 428 on Sunday 8 November, the fourth-lowest Sunday; and 424 on Monday 9 November, settled, under every earlier Monday — reported, not attributed. 29 September is plotted at 350 (185 departures, 165 arrivals) but flagged as partial — a two-hour, network-wide hole in our flight feed that afternoon never filled. For one cycle the bars for 9–11 October were drawn from a different receiver network's count; they were restored to the chart's own feed and the slip stays disclosed on the chart.

Why is this brief closed if aircraft are still flagging over southeastern Finland?

Because the cell has said the same thing for ten weeks and the wider reading has gone quiet. Over 1–13 November the 61°N 26°E cell read 93.1% (229 of 246 aircraft), against 93.9% in October and 95.2% in September — no inflection either way. The FIR-level Finland ratio has stayed under the 9.6% top of its band on every day since 18 October and averaged 6.7% over 1–13 November, against 8.1% in September and 8.2% in October. Helsinki traffic is rated normal. We close it as a documented steady state, not as a recovery: the interference over that cell did not end. The live regional GPS-interference map keeps reading it daily, and a new brief opens if the pocket widens west into the Tampere or Helsinki cells or the FIR-level ratio breaks above its band on a run of days.

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.

// Copy-paste quote

“Over inland southeastern Finland, north-east of Helsinki, 77–100% of aircraft broadcast degraded navigation integrity on every observed day from 1 September to 9 November 2026 — a zone that has shifted north from the Gulf of Finland and stops sharply to the west, with the core cell at 83–100% over 30 October–9 November after its three softest days since August, and Tampere-Pirkkala's cell at 37% — while the wider Finland flight-information region read 8.3% on 10 November, inside its 6.0–9.6% post-spike band, and its 7-day average is 6.8%; Helsinki-Vantaa's observed traffic is still rated normal for the weekday, reading 424 movements on Monday 9 November, the lowest Monday of the window if the count holds, according to flight-monitoring service FlySafe.zone.”

Chart: observed traffic at HEL
PNG · 1600px · attribution baked in
Observed daily movements at HEL — light version preview
Chart: advisory cadence
PNG · 1600px · attribution baked in
share of aircraft broadcasting degraded navigation integrity in the 61°N 26°E cell (%) — crowdsourced ADS-B network — light version preview
↓ Data behind the charts — CSV CC BY 4.0 · dates + daily values, ready for your own chart
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.

No newsletters — we only reply to this request. Press page
// Cite this
"FlySafe.zone — GNSS Interference Inland over Southeastern Finland — 86–100% of Aircraft Flagging Degraded Navigation, September 2026", https://flysafe.zone/events/gnss-hel-2026-09/, data as of 14 Nov 2026, 09:58 UTC — free to cite with a link. Observed-traffic figures are FlySafe’s own ADS-B-derived counts (floor, not total). Navigation-integrity figures are FlySafe-derived from the crowdsourced ADS-B receiver network (a floor, not a total); FIR-level interference ratios are from the same source and are labelled as FIR-level. Method: /methodology/ · Press: /press/.

Related coverage