Summary

  • 33 CFR 101.305 sends a suspicious activity, a breach of security and a transportation security incident to the National Response Center, or to the local Captain of the Port, "without delay" — and no reviewed provision prints an interval in hours or days.
  • The centre publishes workbooks it calls initial, unvalidated pollution and railroad data, while its five security-related report categories are distributed only to government offices holding a maritime security coordinator's written approval.

Read 33 CFR 101.305 the way an operator reads it, looking for the clock. Paragraph (a) covers activities that may result in a transportation security incident and ends at the National Response Center. Paragraph (b) covers breaches of security and ends at the same desk. Both use the same two words: "without delay". Paragraph (c) sends a transportation security incident to the local Captain of the Port, or to the cognizant District Commander where part 106 applies, also "without delay". Paragraph (d) tells the caller what to have ready: their own name and contact details, the suspicious or responsible party's name and contact details, the location as specifically as possible, and a description of what happened.

No reviewed provision of section 101.305 states an interval in hours, minutes or days. Not twenty-four hours, not seventy-two, not one. The only temporal standards in the material reviewed for this piece are "without delay" in that section and "immediately" in a neighbouring one, and the Coast Guard's own 2024 guidance restates the duty without supplying a unit. Most reporting regimes worth reading keep a number somewhere, and the professional reflex is to find it and plan around it. Here the number is not there, and the log an operator keeps is the only clock this instrument will ever have.

The definitions decide who holds the first judgement. Section 101.105 defines a breach of security as an incident that has not resulted in a transportation security incident but in which security measures have been circumvented, eluded or violated; a transportation security incident is one resulting in significant loss of life, environmental damage, transportation system disruption or economic disruption in a particular area. The first definition is broad and carries no severity floor. Everything from a defeated door reader upward can qualify, and the person who decides whether the reader was defeated runs the gate.

Section 101.305 sits in a three-section subpart called Communication (Port—Facility—Vessel), as the subpart shows. One of its neighbours, 101.310, allows alert systems such as the ship security alert system required under SOLAS chapter XI-2, and AIS, to augment communication and to be listed in a security plan. The subpart therefore contemplates a silent alarm and still routes the account of the event through a telephone call to a watch desk.

The Coast Guard's NVIC 02-24, dated 21 February 2024, does the operational work the section leaves undone. It states that regulated entities must report a breach of security or a suspicious activity to the centre "without delay", that they may also report directly to the local Captain of the Port, and that doing so "does not relieve an owner or operator from the requirement to notify the NRC". It treats a breach of telecommunications, computer or network security measures as a breach of security where those systems support security-plan functions or where exploitation could contribute to a transportation security incident, and it separates routine spam and phishing that do not breach a system's defences from the reportable case, while noting that such events may still be suspicious activity. Its checklist — source, location, facility, vessel or port type, summary, impact — is not identical to the four fields in 101.305(d). The fifth field lives in guidance, not in the regulation.

Then there is the second corridor. 33 CFR 6.16-1 requires evidence of sabotage, subversive activity, or an actual or threatened cyber incident involving any vessel, harbor, port or waterfront facility — including the data, networks, programs and systems on it — to be reported "immediately" to the Federal Bureau of Investigation, to the Cybersecurity and Infrastructure Security Agency for any cyber incident, and to the Captain of the Port. The January 2025 cybersecurity rule added a third route to the same watch desk. Subpart F defines a reportable cyber incident by consequence: substantial loss of confidentiality, integrity or availability; disruption or significant adverse impact on the ability to run the business or deliver goods and services, including potential impact on public health or safety; disclosure of nonpublic personal information of a significant number of individuals; other potential operational disruption to critical infrastructure; or incidents that may lead to a transportation security incident. Where an entity has not reported under 6.16-1, or is not subject to it, section 101.650(g)(1) sends that reportable cyber incident to the National Response Center, again "without delay". Subpart F covers owners and operators of U.S.-flagged vessels, facilities and offshore facilities with a security plan; foreign-flagged vessels subject to part 104 are outside it.

Count the addressees for one maritime security event: the National Response Center under 101.305(a), 101.305(b) and 101.650(g)(1); the local Captain of the Port or District Commander under 101.305(c); the FBI and CISA under 6.16-1. Four offices. The 2025 rule's own account of the interaction is that breaches of security and suspicious activity are already addressed under 101.305 while the new subpart addresses reportable cyber incidents (Federal Register, 17 January 2025). So one event can sit in two of those four offices under two different verbs, and neither verb carries an hour.

Now turn the instrument around and look at the desk. The centre's own front page describes itself without flinching: "The National Response Center (NRC) is not a response agency. It serves as an emergency call center that fields INITIAL reports for pollution and railroad incidents and forwards that information to appropriate federal/state agencies for response" (nrc.uscg.mil). The same page invites anyone witnessing an oil spill, a chemical release or a maritime security incident to call, and states that the spreadsheets it posts contain "INITIAL incident data that has not been validated or investigated by a federal/state response agency." The Environmental Protection Agency is more explicit about the scope: the centre, staffed around the clock by the Coast Guard, "also takes maritime reports of suspicious activity and security breaches" in United States waters and territories (EPA). There is no contradiction between those two pages, and the gap between them is the point: a call centre built to take releases also receives the security reports of the regulated maritime world, and calls its own published data unvalidated.

A briefing deck for Regional Response Team 5, hosted on rrt5.org, fills in the machinery. It lists a separate port security reporting line alongside the pollution number, describes watchstanders completing reports in the IRIS database, and describes that database queueing notifications to EPA and Coast Guard on-scene coordinators and to state, local, tribal and federal agencies. The port security line does not appear in 101.305. The number an operator may be told to call is not the number the rule prints, and only one of them is in the regulation.

What comes back is where this stops being a routing curiosity and becomes a finding. The centre's own application form for receiving its reports divides its notification criteria into non-security and security-related groups. The security group lists suspicious activity, security breach, cyber incident, terrorism and trespasser. To receive maritime security reports, a non-federal agency must obtain written approval from the designated local Federal Maritime Security Coordinator defined in 33 CFR 103.205; the coordinator may require a Department of Homeland Security non-disclosure agreement; profiles are created for government entities only, one per office. The form's sentences are not reproduced here as quotations, but its structure is plain: the security stream is a permissioned channel, not a published one. What obligation the centre carries after it forwards a security report is not stated in any provision reviewed here. The duty described in the regulation ends at the call.

The published workbooks do not close that gap. In the material reviewed for this piece — the centre's homepage, the RRT5 briefing deck and one third-party description of the annual workbooks — no security category appears in the workbook's incident-type vocabulary, which runs fixed, vessel, unknown sheen, railroad non-release, mobile, pipeline, storage tank, railroad, platform, aircraft and continuous, and the report-received timestamp sits in a separate sheet from the incident date and time (Apify's field documentation). That is third-party documentation of an official dataset, not the centre's data dictionary, and the centre's own dictionary was not opened in this review. Whether security reports also appear, redacted, in the public workbooks cannot be established from what was read. What can be said is that the five security categories travel by a different, gated route.

The regulation's own text is stable: the same section, the same three contact means and the same "without delay" appear in the 2024 annual edition of 33 CFR part 101 and in the Cornell mirror of 101.305. Nothing here alleges that a report was late, mishandled or improperly circulated; the claim is narrower and harder to argue with, because an operator's entire compliance story on this instrument rests on a call and the instrument prints no unit for it. Nor is this the first time this publication has written about the United States Coast Guard. Our only earlier article on the service, published 23 July 2026, read the Integrated Deepwater acquisition programme as an accountability test across mission needs, contractor integration, asset cost, schedule and field evidence. It asked what the Coast Guard buys. It did not open a reporting instrument and never mentions the National Response Center. This piece asks the other question: what a maritime operator must tell the service, and where the call lands.

Which leaves a small control recommendation. A return receipt: a per-report reference the reporter can record and, if it comes to it, point to, and a published field that says what became of the report at the level of the case rather than the year. The instrument already takes four fields from the caller and prints nothing it gives back. Until one exists, the best account of this channel is the one the reviewed text supports exactly: four offices, zero hours, and a public record that stops at the approved office door.