---
date: 2026-09-26
issue: 64
title: Alcatel Submarine Networks Develops Early Warning Sensing for Subsea Cables
standfirst: New Distributed Acoustic Sensing technology aims to detect anchor and
  trawl gear vibrations up to 15 minutes before impact. The system seeks to provide
  real-time monitoring for transoceanic fiber lines.
jurisdictions:
- AU
- INT
- RU
- DE
- FI
- SE
tags: []
vertical: cablewatch
---

## Situation

- **INDIGO West**: open since 8 Aug, 49 days, per AusInternet. Operator status: Partial Service Disruption.  
  Latest update, 21 Sep 05:49 UTC (5 days ago): “Scheduled to repair the fault on Indigo West off the coast of Perth has completed. Unfortunately the existing shunt fault on Indigo West in Singapore has deteriorated into a full cable break and services remain down. Repairs off Singapore is expected to be on 16 Oct 2026 until 26 Oct 2026.”
- **Australia Singapore Cable**: open since 4 Sep, 22 days, per Global Secure Layer. Operator status: Identified.  
  Latest update, 4 Sep 06:49 UTC (22 days ago): “We observed short flaps occurring on the ASC submarine cable system at 04/09/2026 03:50:33 UTC. However a hard down event followed at 04/09/2026 06:19:12 UTC Due to the Indigo West cable system also being offline for shunt fault repairs, traffic will take a longer path which will result in higher latency between Australia and Singapore. We are waiting for further information from Vocus regarding this outage. Information received will be shared via our status page.”
- **INDIGO West**: notice posted by Superloop, already on its page when monitoring began 14 Sep; the notice carries no date of its own.  
  Notice: “We are currently experiencing a service disruption affecting the Indigo West link due to instability. Our team is actively investigating and working to resolve the issue. Updates will be provided as more information becomes available. Support remains available via phone and email during this time.”
- **INDIGO West**: works scheduled 2 Oct 00:00 UTC to 8 Oct 23:00 UTC, per AusInternet: “Indigo West (Perth-Singapore) Shunt Fault Repairs (Singapore)”
- **Measured latency** (RIPE Atlas, best measured path over 24 hours against each route's best sustained floor over 36 days): Perth to Singapore 191.6ms, up 145.1ms on its 46.6ms floor; Sydney to Japan 100.4ms, up 43.0ms on its 100.5ms floor; Sydney to Singapore 144.6ms, up 48.1ms on its 96.5ms floor.
- **Elsewhere**: Cloudflare Radar lists an outage in GY, cause cable cut, 23 Sep 18:45 UTC to 24 Sep 06:00 UTC: “A cut fibre-optic cable caused a significant drop in Internet traffic from One Communications (AS19863) in Guyana.” **New since the last brief.**

<sub>No model writes this section. Operator updates are quoted from each operator's own status page and dated as the operator dated them. No operator has posted an update or a new notice on these since the last brief (25 Sep 18:42 UTC). Lines marked Elsewhere are Cloudflare Radar's own outage annotations, from anywhere in the world, with the cause as Radar labels it; this archive attributes nothing to anyone. Latency is measured between cities, not on any named cable: several systems serve each route and traffic moves to a survivor within seconds of a fault. Detail is in Standing status at the end of the brief.</sub>

## Today

Alcatel Submarine Networks (ASN) is developing an upgrade to Distributed Acoustic Sensing (DAS) technology that uses wavelength-division multiplexing to detect vibrations across the entire length of transoceanic fiber lines [1]. While current DAS is limited to the first and last 200 km of a cable, this new system is designed to identify threats up to 2 km and 15 minutes before impact [1]. The technology is intended to provide real-time diagnosis for systems such as the Indigo West and Indigo Central lines off the coast of Australia, which recently suffered outages of uncertain causes [1].

## Subsea Sabotage and Insurance Precedents

NATO reported foiling a Russian attempt to sabotage undersea internet and energy cables near Svalbard in September 2026 [2]. The operation allegedly involved Russian GUGI submarines testing a secret weapon designed to sever deep-sea infrastructure, which reports suggest would have taken repair ships two to three weeks to reach the breakage points [2]. 

In a separate legal development, a London Commercial Court ruled that Nord Stream AG's €579 million insurance claim was voided because the 2022 explosions were 'occasioned by war' [3]. The court's judgment establishes that uncertainty regarding the specific perpetrator does not prevent a 'war exclusion' from triggering, a precedent that affects other Baltic infrastructure including the Balticconnector, Estlink 2, and cables near the Yi Peng 3 and Eagle S [3].

## Petabit-class Infrastructure and Manufacturing Expansion

Meta, NEC, and Sumitomo Electric have announced a collaboration to develop "Petal," which is described as the world's first commercial petabit-class transoceanic submarine cable system [4]. The project intends to utilize Sumitomo's 2-core fiber with 48 pairs of spatial channels to double the capacity of previous state-of-the-art systems [4]. To support the commercialization of multi-core fiber, Lightera has secured investment to nearly triple its submarine optical fiber manufacturing capacity in Denmark and the US by 2029 [4].

Analysts suggest these developments indicate a shift toward hyperscaler-dominated infrastructure designed to meet AI-driven data demands [4]. However, the move toward ultra-high capacity cables creates a significant risk of single points of failure in geopolitically sensitive waters [4].

## References

1. [Undersea Cables Learn to Feel Anchors Before Impact - IEEE Spectrum](https://spectrum.ieee.org/undersea-cables-acoustic-sensing) (spectrum.ieee.org)
   Published: 26 Sep 2026 · spectrum.ieee.org · Claim type: analyst_assessment · Subject: fault, analysis, repair · Systems named: Indigo West, Indigo Central · Jurisdictions: AU
2. [NATO foils Russian undersea cable sabotage plot near Svalbard as GUGI submarines are chased away - Karmactive](https://karmactive.com/nato-foils-russian-undersea-cable-sabotage-plot-svalbard) (karmactive.com)
   Published: 26 Sep 2026 · karmactive.com · Claim type: analyst_assessment · Subject: fault, vessel_activity, analysis · Systems named: undersea internet and energy cables · Vessels named: GUGI submarines · Jurisdictions: INT, RU
3. [How a London Court Quietly Decided Who Pays for Baltic Sea Sabotage - Modern Diplomacy](https://moderndiplomacy.eu/2026/09/25/baltic-sea-cable-sabotage-insurance-gap-nord-stream-war-exclusion-ruling) (moderndiplomacy.eu)
   Published: 25 Sep 2026 · moderndiplomacy.eu · Claim type: analyst_assessment · Subject: analysis, precedent · Systems named: Nord Stream, Balticconnector, Estlink 2 · Vessels named: NewNew Polar Bear, Yi Peng 3, Eagle S · Jurisdictions: INT, DE, FI, SE, LV
4. [Subsea Daily News#5](https://note.com/fine_alpaca6065/n/na9be5823672d?hl=en) (note.com)
   Published: 24 Sep 2026 · note.com · Claim type: official_statement · Subject: policy, analysis · Systems named: Petal · Jurisdictions: INT

---

## Standing status

**Operator incident timelines** (quoted from each operator's own status page, not summarised by any model):

- **INDIGO West** · AusInternet — network status · Partial Service Disruption, open since 8 Aug
  - `21 Sep 05:49 UTC` Identified: Scheduled to repair the fault on Indigo West off the coast of Perth has completed. Unfortunately the existing shunt fault on Indigo West in Singapore has deteriorated into a full cable break and services remain down. Repairs off Singapore is expected to be on 16 Oct 2026 until 26 Oct 2026.
  <sub>page last checked 2026-09-26 18:12 UTC · https://ausinternet.com.au/status/</sub>
- **Australia Singapore Cable** · Global Secure Layer — network status · Identified, open since 4 Sep
  - `4 Sep 06:49 UTC` Identified: We observed short flaps occurring on the ASC submarine cable system at 04/09/2026 03:50:33 UTC. However a hard down event followed at 04/09/2026 06:19:12 UTC Due to the Indigo West cable system also being offline for shunt fault repairs, traffic will take a longer path which will result in higher latency between Australia and Singapore. We are waiting for further information from Vocus regarding this outage. Information received will be shared via our status page.
  <sub>page last checked 2026-09-26 18:13 UTC · https://status.gslnetworks.com.au/incidents/wqmz5wt7l8qj</sub>
- **INDIGO West** · Superloop — outages · active notice, on the page when monitoring began 14 Sep 21:44 UTC (the page gives no date)
  - We are currently experiencing a service disruption affecting the Indigo West link due to instability. Our team is actively investigating and working to resolve the issue. Updates will be provided as more information becomes available. Support remains available via phone and email during this time.
  <sub>page last checked 2026-09-26 18:12 UTC · https://www.superloop.com/outages/</sub>
- **INDIGO West** · AusInternet — network status · works scheduled 2 Oct 00:00 UTC to 8 Oct 23:00 UTC: “Indigo West (Perth-Singapore) Shunt Fault Repairs (Singapore)”
  <sub>page last checked 2026-09-26 18:12 UTC · https://ausinternet.com.au/status/</sub>

<sub>Also polled for changes, but carrying no incident listing, so nothing is said about their state: ACMA protection zone — Perth, ACMA protection zone — Sydney, AJC — cable system maintenance, AJC — news, SUBCO, SUBCO — INDIGO West, SUBCO — OAC, Southern Cross Cable Network, Telstra — service status, Vocus — network status.</sub>

**Incident threads:**

- `perth_indigo_2026_08` (INDIGO West, INDIGO Central) — 18 documents, last 2026-09-26 (quiet 1d)

**Measured path state** (RIPE Atlas traceroute over 24h, no model involved):

- **perth to singapore** — best path **191.6ms, +145.1ms on its best sustained 46.6ms**; typical path 276.9ms (+81.5ms per probe against its own history, 22 probes); ocean crossing 149.9ms; queuing 0.6ms; 36 days of history; clean crossing (22 probe(s), 46 reading(s) excluded as below the corridor's physical floor)
- **sydney to japan** — best path 100.4ms; typical path 180.8ms (+43.0ms per probe against its own history, 16 probes); ocean crossing 134.9ms; queuing 0.5ms; 36 days of history; clean crossing (16 probe(s))
- **sydney to nz** — best path 26.7ms; typical path 34.5ms (-0.1ms per probe against its own history, 8 probes); ocean crossing 27.7ms; queuing 0.2ms; steady against 36 days of history; clean crossing (8 probe(s))
- **sydney to singapore** — best path **144.6ms, +48.1ms on its best sustained 96.5ms**; typical path 229.8ms (+37.4ms per probe against its own history, 22 probes); ocean crossing 152.1ms; queuing 0.7ms; 36 days of history; clean crossing (22 probe(s))

<sub>A corridor is a city pair, not a cable. Several systems serve each of these routes and IP routing does not reveal which one carries a packet, so a healthy corridor is not evidence that any particular cable is intact: traffic reroutes onto a survivor within seconds of a fault. Per-probe detail is in `cablewatch segments`.</sub>

**Every cable route leaving the mainland** (measured hourly, no model involved):

- **Australia to Guam** — best 70.5ms against a 52.0ms floor; 12/24 paths direct (normally 12) <br><sub>Australia-Japan Cable (AJC), Japan-Guam-Australia South (JGA-S), PIPE Pacific Cable-1 (PPC-1)</sub>
- **Dili to Australia** — best 49.4ms against a 7.1ms floor; 3 paths measured <br><sub>Timor-Leste South Submarine Cable (TLSSC), North-West Cable System</sub>
- **Melbourne to Tasmania** — best 10.1ms against a 5.9ms floor; 8/17 paths direct (normally 8) <br><sub>Basslink, Bass Strait-1, Bass Strait-2</sub>
- **Perth to Jakarta** — best 63.9ms against a 29.5ms floor; 42 paths measured <br><sub>INDIGO-West, Australia-Singapore Cable (ASC)</sub>
- **Perth to Oman** — best 111.2ms against a 85.0ms floor; 2/4 paths direct (normally 2) <br><sub>Oman Australia Cable (OAC) · only system on this route</sub>
- **Perth to Singapore** — best 51.5ms against a 38.3ms floor; 8/67 paths direct (normally 23) <br><sub>INDIGO-West, Australia-Singapore Cable (ASC), Darwin-Jakarta-Singapore Cable (DJSC)</sub>
- **Sydney to Fiji** — best 37.9ms against a 31.5ms floor; 6/6 paths direct (normally 6) <br><sub>Southern Cross Cable Network (SCCN), Southern Cross NEXT</sub>
- **Sydney to Hawaii** — best 189.1ms against a 80.0ms floor; 2 paths measured <br><sub>Telstra Endeavour, Southern Cross Cable Network (SCCN), Hawaiki</sub>
- **Sydney to Honiara** — best 38.3ms against a 28.0ms floor; 2/2 paths direct (normally 2) <br><sub>Coral Sea Cable System (CS²) · only system on this route</sub>
- **Sydney to Japan** — best 99.0ms against a 76.7ms floor; 18/94 paths direct (normally 25) <br><sub>Australia-Japan Cable (AJC), Japan-Guam-Australia South (JGA-S)</sub>
- **Sydney to New Zealand** — best 23.4ms against a 21.1ms floor; 48/65 paths direct (normally 50) <br><sub>Southern Cross Cable Network (SCCN), Southern Cross NEXT, Hawaiki, Tasman Global Access (TGA) Cable</sub>
- **Sydney to Noumea** — best 22.3ms against a 19.3ms floor; 6/9 paths direct (normally 6) <br><sub>Gondwana-1 · only system on this route</sub>
- **Sydney to Port Moresby** — best 37.7ms against a 26.9ms floor; 3/6 paths direct (normally 3) <br><sub>Coral Sea Cable System (CS²), PIPE Pacific Cable-1 (PPC-1)</sub>
- **Sydney to US West Coast** — best 133.7ms against a 118.3ms floor; 282/302 paths direct (normally 283) <br><sub>Southern Cross Cable Network (SCCN), Southern Cross NEXT, Hawaiki, Telstra Endeavour</sub>

<sub>A route is a pair of places, not a cable: several systems serve most of these and IP routing does not reveal which one carried a packet, so a healthy route is not evidence that every cable on it is intact. A path counts as direct when it is within reach of the physical floor between the two places. Routes with no vantage at the far end are listed in latency_watch.yaml under `uncovered`.</sub>

**Cloudflare Radar outages attributed to a cable cut in the last 7 days:**

- `2026-09-23` GY — NETWORK — A cut fibre-optic cable caused a significant drop in Internet traffic from One Communications (AS19863) in Guyana.

<sub>Cause is Cloudflare Radar's own classification of its own observation, reproduced as they state it. It is not this archive's finding and it attributes nothing to anyone.</sub>

<sub>Collection: 580 successful fetches in 24h, 39 errors across 11 source(s). Generated 2026-09-26 18:24 UTC.</sub>
