An offshore wind incident does not begin with a clean population list. It begins with people at different stages of work. One technician may be waiting at the marine base, another may be aboard a crew transfer vessel, a contractor team may be inside a turbine, and a supervisor may be returning to a service operation vessel. The same person can cross several operational boundaries in one shift, and every crossing changes the answer to a simple question: who is responsible for confirming that person’s status now?1, 2
The latest G+ incident data shows why those joins deserve attention. Its member dataset covered 69.2 million hours of offshore wind work in 2025 across countries including Denmark, Germany, and the United Kingdom. All 19 reports coded to the communication work process were hazards or near misses. For transfers to or from vessels, 46 of 64 reports were hazards or near misses. G+ also cautions that access, movement, and transition exposure can appear across several work-process categories, so the transfer category alone does not describe the whole risk.3
The regional context is developing as well. In the United Kingdom, the Maritime and Coastguard Agency updated its Hub Emergency Response Co-operation Plan template to version 3 in July 2026 and says these plans should cover construction, operation, and decommissioning.4 Denmark’s maritime authority describes the International Code of Safety for Ships Carrying Industrial Personnel as a mandatory framework for people transported to and from offshore work.5 In Germany, the Federal Maritime and Hydrographic Agency and coastal occupational-safety authorities published a draft joint offshore-rescue standard in May 2026, focused on emergency medical care and immediate rescue in the German exclusive economic zone.6 These are different instruments with different legal effects. Together, they reinforce the operational need to connect vessel safety, site response, personnel information, and external rescue arrangements. This article is an operating-design discussion, not legal advice, and each organisation must establish the requirements that apply to its assets, vessels, people, and jurisdictions.
Why does accountability break at operational boundaries?
A conventional roster answers an administrative question: who was expected to work? Incident accountability asks a more exacting question: what is the latest verified state of each person who may be affected? Those two answers separate quickly when a sailing is delayed, a technician boards a different vessel, a transfer is postponed, a team changes turbine, or worsening conditions leave people temporarily offshore.
The vulnerable point is usually not the record itself. It is the handover between records. A marine-base list may show that a person checked in. The vessel record may show who embarked. Marine coordination may hold the transfer plan. A work-management system may show the assigned turbine. A contractor supervisor may know that the work party changed. Each source can be accurate for its own purpose while the combined operating picture is already wrong.
G+ emergency-response guidance says the scope of a plan should be agreed in terms of geography, assets, and participants, including employees, contractors, and subcontractors. It also recommends one accountable organisation and a responsible individual, while recognising that each employer, contractor, aircraft operator, and vessel owner retains its own responsibilities.1 That is a useful model for accountability data. One coordinated picture does not erase separate duties. It makes the interfaces between them visible.
An organisation should therefore treat each movement as a change of accountable state. “Expected at base”, “checked in”, “embarked”, “on board”, “cleared to transfer”, “confirmed on structure”, “awaiting recovery”, “returned to vessel”, and “ashore” are not interchangeable. The exact terminology should match the operator’s procedures, but each state should have a source, time, and responsible role. If two sources disagree, the discrepancy should remain visible and receive an owner rather than being silently overwritten.
This approach also prevents location from carrying more meaning than it can support. A recent device position may help establish where to look, but it does not prove that its user is safe, that the device remains with them, or that they completed a transfer. Accountability is strongest when direct confirmation, operational status, and location evidence are distinguishable.
What should the operating picture prove at each stage?
The shore-base stage should establish the planned population and whether the person has actually presented for the movement. The vessel stage should establish who is aboard, under whose authority, and for which work plan. The transfer stage should establish that the handover was authorised and completed. The offshore-work stage should establish which asset or defined area the person is associated with, who supervises the work, and what recovery plan applies. The return stage should reconcile everyone back to the vessel and, eventually, ashore.
This is not a demand for one technology to generate every fact. Some facts may come from authorised vessel data, some from a marine coordinator, some from a supervisor, and some from a direct response. The design task is to decide which source can establish each state, when that evidence becomes stale, and who resolves an exception. Suitable authorised data should only be brought together through agreed permissions, formats, security controls, and approved integration or import methods.
Evidence quality matters because the word “accounted for” can conceal several meanings. A person may be scheduled, detected, reported present by a supervisor, directly confirmed safe, or physically verified at a muster point. The operating picture should preserve those distinctions. During an incident, leaders need to see not only the total confirmed safe, but also the people whose status is inferred, out of date, disputed, or still unknown.
The same discipline applies to communications. A message prepared by an operator, one sent into a channel, one delivered to a device, and one acknowledged by a person are different events. A Check-In response can provide useful direct evidence. A non-response provides a clear exception for an operator to handle under the organisation’s standard operating procedure. Neither a delivery indicator nor silence should be converted automatically into a welfare conclusion.
A useful operating picture is therefore a chain of claims with evidence, not a collection of coloured markers. It should let an authorised user answer where a person was expected to be, what has since been verified, how recently it was verified, which organisation or role supplied the evidence, and who owns the next action if confidence is insufficient.
Who owns the truth when contractors and vessel operators are involved?
Offshore wind relies on interfaces between asset owners, operators, original equipment manufacturers, vessel companies, specialist contractors, and emergency services. G+ guidance does not treat that complexity as a reason to blur accountability. It recommends that one organisation lead the project-level emergency plan and that one responsible individual be recognised by participants, while the other organisations continue to discharge their own responsibilities.1
ISO 22320 applies to incidents involving one organisation or several organisations that continue to use their own structures. It places roles, responsibilities, tasks, resource management, joint direction, and cooperation among the core components of incident management.7 The practical implication is that a shared picture needs named decision rights. It is not enough for everyone to have access to the same screen if nobody knows whose confirmation is authoritative at a particular stage.
Responsibility should follow the operation. At the shore base, an authorised departure role may confirm the people released to the vessel. Once aboard, the master retains authority over the vessel and those on it. At the transfer point, the vessel and offshore structure need an agreed readiness decision. After transfer, an offshore supervisor or other defined role confirms receipt and worksite responsibility. If the person returns by a different route or vessel, the new chain must be recorded rather than assumed.
The Maritime and Coastguard Agency states that the master has ultimate responsibility for people on board and may halt a transfer if safety could be compromised. Its current guidance also calls for clear communication between the responsible officer on the offshore structure and the vessel bridge, properly trained personnel, a safe route, and a job safety analysis.8 These controls should not be collapsed into a generic “transfer complete” button. The accountability record should reflect the decision made by the people who hold the authority.
Contractors need equal clarity. Their employer may maintain the personnel record, the vessel may control embarkation, and the wind-farm operator may coordinate the site response. Before work begins, the parties should agree who provides the current population, who may change a status, who follows up a missing confirmation, and how an urgent contractor update enters the shared incident record. A contractual boundary should not become an information gap.
How should weather change the accountability plan?
Weather is not merely a go or no-go input at the start of a shift. It can change the meaning of a person’s status after they have travelled offshore. A technician who is safely on a structure but cannot be recovered within the planned window is accounted for, yet their exposure, supplies, communications, supervision, and recovery options have changed. Treating the original work plan as current would produce a reassuring count and a weak response.
The G+ offshore wind transfer guideline recommends reviewing metocean data before a vessel leaves port, defining allowable weather windows, and monitoring actual conditions. Its example controls also include an allowance for forecast uncertainty, emergency supplies on structures, and planning for the possibility that people cannot be recovered because conditions deteriorate.2 The decision trail should therefore extend beyond the departure forecast.
Before sailing, the plan should identify the latest safe recovery point, the assumptions behind it, and the person authorised to review it. After transfer, the operating picture should retain the people exposed on each structure, the planned recovery method, the current weather decision, and the next review time. If retrieval is brought forward, delayed, or changed, that update should reach the work party, the vessel, marine coordination, and the accountable response role without creating four different versions of the truth.
A weather decision should also distinguish status from consequence. “Transfer suspended” describes the immediate control. It does not say whether technicians remain safely sheltered, require additional supplies, need another communications provision, or should stop work in preparation for recovery. Those are separate actions with separate owners.
Dynamic risk assessment is relevant here. G+ reviewers examined narratives from nearly 2,000 incidents and hazard observations in 2025 and found a recurring pattern in which routine work continued without a pause to reassess changed local conditions. The pattern crossed vessels, turbines, ports, and other work areas.3 An accountability system cannot perform that judgement for a competent person, but it can preserve the changed conditions, the decision, and the people affected by it.
What must happen at the vessel-to-turbine transfer?
The transfer is both a physical operation and a change in the operational responsibility for confirming status. It should proceed only through the approved procedure and the responsible people’s judgement. From an information perspective, the minimum outcome is a two-sided confirmation: the vessel records that the person left its accountable population, and the receiving side records that the person arrived under its control.
G+ guidance calls for communication checks before and immediately after transfer. For walk-to-work arrangements, it also calls for confirmation of the numbers of people on both sides of the gangway before disconnection. More generally, it recommends effective personnel tracking, marine coordination, and emergency-response arrangements, together with regular practice by vessel and structure personnel.2
The current UK industrial-personnel guidance adds an important authority boundary. It requires clear communication between the vessel bridge and the supervising responsible officer on the offshore installation during transfer, and it confirms the master’s authority to stop the operation.8 A digital workflow should support that exchange without suggesting that software has authorised the transfer.
Recent safety experience also argues against treating the gangway as a routine doorway. The UK Health and Safety Executive reported serious injuries involving motion-compensated gangways and advised wind-farm clients, operators, principal contractors, duty holders, and vessel operators to address safe deployment, use, retraction, training, and emergency actions as part of a wider safe system of work.9
If a transfer is interrupted, the system of record should not force a neat outcome. It should be possible to show that the operation started, was stopped, and left the person on a defined side of the boundary. If reports conflict, the status should remain unresolved until the authorised roles reconcile it. This is a small design detail with considerable incident value. It prevents a partially completed movement from disappearing between the vessel count and the turbine count.
How should communications work when the primary channel weakens?
Offshore communications should be designed around essential information, not around the assumption that one channel will always remain available. The UK Cabinet Office’s resilience guidance says technical solutions, communication processes, and organisational arrangements need equal attention. It also states that no technical solution is available all the time and recommends diverse, layered fallback arrangements.10
For personnel accountability, the critical communication task is more precise than “make contact”. The receiving role may need the identity of the person or group, the relevant asset or coordinates, the time, the current status, the assistance required, and the next intended action. G+ emergency-response guidance uses the same operational core for incident notification, including who, what, where, when, weather, the number of people at risk, assistance required, initial actions, and on-scene resources.1
The primary channel may differ between the port, vessel, and turbine. The fallback should therefore be designed for the task and the location. It should define which smaller set of information must still pass when bandwidth or availability is reduced, which person monitors the alternate route, how receipt is confirmed, and how information is reconciled later. Two devices do not create resilience if they share the same dependency, nobody monitors the second path, or the message cannot be interpreted.
The AtlasNXT app can support alerts, Check-Ins, Panic, Overwatch, and location-enabled functions where appropriate. Compatible satellite devices can support selected users beyond cellular coverage.
ReachScore™ can expose communication-readiness gaps before they become incident surprises. The useful question is whether the chosen communication arrangements are appropriate for the people, locations, and response tasks in scope, including the people who may remain on an offshore structure when the normal plan changes.
What should change when an incident is declared?
Routine operations can tolerate separate systems if their handovers are disciplined. An incident reduces that tolerance. A marine coordinator may need to work with a vessel master, an offshore supervisor, an internal incident lead, external responders, contractor management, and people ashore. Each participant needs information suited to their role, but material updates cannot be allowed to form disconnected histories.
G+ guidance recommends an event log, clear command roles, a responsible individual with authority to take emergency-response decisions, and contingency arrangements when primary role holders are unavailable. It also describes formal search-and-rescue coordination, including recorded handovers and regular situation reports when multiple resources are involved.1 The organisation’s internal system should support that structure without replacing statutory distress, maritime, aviation, or rescue communications.
The first incident picture should connect the population baseline to current evidence. It should show who could be affected, the last verified stage for each person, current communications, unresolved discrepancies, and the source and time of each material update. It should also show which actions have been assigned, who accepted them, what is blocking progress, and when the next review is due.
AtlasNXT’s Incident Room can keep significant updates, tasks, communications, decisions, and status changes with one event. The platform does not decide when an incident should be escalated, direct a rescue, or replace the organisation’s command structure. Those decisions remain with authorised people applying the emergency plan and the relevant standard operating procedures.
Decision records should explain material changes. If a turbine team remains sheltered rather than being recovered, the record should preserve the available weather information, the advice received, the decision-maker, the action taken, and the review condition. If a contractor’s status is accepted through a supervisor rather than direct confirmation, the evidence type should remain visible. A cleaner dashboard is not more valuable than an honest account of uncertainty.
How can location support safety without becoming workforce surveillance?
Location can help answer a defined incident question, but it should not become a substitute for purpose, authority, or direct welfare confirmation. The relevant question may be whether a person reached a turbine, which work party could be affected by an exclusion area, or where a vessel was when the last status was confirmed. Each purpose may justify a different collection window, precision, and audience.
For Denmark and Germany, Article 5 of the General Data Protection Regulation requires personal data to be processed lawfully, fairly, and transparently, collected for specified purposes, and limited to what is necessary. It also requires accuracy, storage limitation, security, and accountability.11 The UK Information Commissioner’s Office similarly advises employers to define the purpose of worker monitoring, use the least intrusive means, be transparent, and assess the legal basis and proportionality.12
This means continuous location should not be the unexamined default. Depending on the purpose, workforce arrangement, organisational policy, and applicable law, location may be live, event-led, or consent-based where appropriate. Consent is not automatically the correct legal basis for employment-related processing, and the choice of product setting does not establish legal compliance. The organisation remains responsible for determining its lawful basis, notices, access, retention, and worker-engagement arrangements.
AtlasNXT Remits can define geographic responsibility and authorised views. This can help an operator give a marine team visibility of the relevant site while avoiding unnecessary access to unrelated people or locations. The same principle should govern imported information. Only suitable, authorised data should enter through approved integrations or imports, with permissions, formats, and security controls agreed in advance.
Transparency can also strengthen trust. If technicians and contractors understand when location is active, why it is being used, who can see it, and when it stops, they are better placed to identify a wrong assignment or stale status.12 The purpose is not to make people permanently observable. It is to make safety-critical uncertainty visible at the point where someone is responsible for resolving it.
What should drills prove before the next difficult day?
A drill that confirms the alarm can be heard and the usual names can be counted has limited diagnostic value. A stronger exercise tests the boundaries where real operations fragment. It might delay a sailing after the shore list has been prepared, move a contractor to another vessel, stop a transfer midway, strand a work party after the weather window changes, remove the primary communications channel, and require a shift handover while several statuses remain unresolved.
G+ emergency-response guidance says performance standards should be measurable and auditable, and that exercises should test whether the standards in an emergency plan can be achieved. It also recommends post-exercise reporting with participants, objectives, findings, recommendations, status, and follow-up action.1 The UK Maritime and Coastguard Agency requires the current Hub Emergency Response Co-operation Plan approach to be used in site safety-management planning across the construction, operational, and decommissioning phases.4
The accountability measures should examine decision quality as well as speed. How long did it take to establish a trustworthy baseline? How many statuses depended on inference? Could the team identify who remained on each side of an interrupted transfer? Did every exception have one accepted owner? Could the incoming shift distinguish current evidence from old information? Did the alternate communication path carry the minimum required message? Could leaders see uncertainty without reading every operational update?
The exercise should continue through closure. Every person should reach a final verified state or move into an explicitly owned follow-up case. Every material action should be completed, cancelled with a reason, or transferred. Every significant decision should remain linked to the evidence available when it was made. Corrective actions should receive owners and review dates, applying the same accountability discipline to improvement that the organisation expects during an incident.
What does a joined-up model look like in AtlasNXT?
A joined-up model does not mean forcing the port, vessel, turbine, contractors, and emergency services into one undifferentiated process. It means preserving the joins between them. Each source and role keeps its defined authority, while the people coordinating safety can see the current population, the strength of the evidence, the open exceptions, and the next owned action.
AtlasNXT can support this model as a coordination layer. Remits can define geographic responsibility and authorised views. Check-Ins can show responses and non-responses for operators to handle under the agreed procedure. The app can support alerts, Panic, Overwatch, and location-enabled functions where appropriate. Compatible satellite devices can extend provision to selected users beyond cellular coverage. The Incident Room can keep material updates, tasks, communications, decisions, and status changes together as the response develops.
None of those capabilities removes the transfer authority of the master, the responsibilities of the offshore operator and contractors, or the role of emergency services. The value is a more coherent account of what each authorised person has confirmed and what still needs to happen.
The practical test is straightforward. If conditions change while technicians are spread between a shore base, vessels, turbines, and an offshore substation, can the incident lead identify who may be affected, which status is verified, who supplied the evidence, what communication route remains available, who owns each unresolved case, and when the position will be reviewed? If that answer depends on reconciling several private messages and static lists under pressure, the accountability design needs work before the next sailing.
Book a free AtlasNXT demo to explore how a joined-up personnel-accountability and incident-response model could support your offshore wind operations.
References
1. G+ Global Offshore Wind Health and Safety Organisation, “Integrated Offshore Emergency Response: good practice guidelines for offshore renewable energy developments”, second edition, November 2023. https://www.gplusoffshorewind.com/__data/assets/pdf_file/0008/671399/G-integrated-offshore-emergency-response-G-IOER.pdf
2. G+ Global Offshore Wind Health and Safety Organisation, “Offshore wind farm transfer: good practice guidelines”, second edition, October 2024. https://www.gplusoffshorewind.com/__data/assets/pdf_file/0008/763523/Good-practice-guideline-Offshore-wind-farm-transfer.pdf
3. G+ Global Offshore Wind Health and Safety Organisation, “2025 incident data report”, June 2026. https://www.gplusoffshorewind.com/?a=1821764
4. UK Maritime and Coastguard Agency, “Offshore renewable energy installations: emergency response planning”, updated 22 July 2026. https://www.gov.uk/government/publications/offshore-renewable-energy-installations-orei
5. Danish Maritime Authority, “New international safety standard for ships transporting offshore personnel”, 21 March 2024. https://www.soefartsstyrelsen.dk/nyheder/2024/mar/ny-international-sikkerhedsstandard-for-skibe-der-transporterer-offshore-personale
6. German Federal Maritime and Hydrographic Agency, “Offshore Rescue: publication and start of consultation on the draft joint standard by the BSH and the occupational safety and health authorities of the coastal federal states”, 7 May 2026. https://wp.bsh.de/en/blog/2026/05/07/offshore-rescue-publication-and-start-of-consultation-on-the-draft-joint-standard-by-the-bsh-and-the-occupational-safety-and-health-authorities-of-the-coastal-federal-states/
7. International Organization for Standardization, “ISO 22320:2018 Security and resilience: emergency management, guidelines for incident management”, confirmed current in 2024. https://www.iso.org/standard/67851.html
8. UK Maritime and Coastguard Agency, “MGN 701 (M): application of the industrial personnel code and the special purpose ships code”, 11 December 2025. https://www.gov.uk/government/publications/mgn-701-m-application-of-the-industrial-personnel-code-and-the-special-purpose-ships-code/mgn-701-m-application-of-the-industrial-personnel-code-and-the-special-purpose-ships-code
9. UK Health and Safety Executive, “Risk of serious injury from motion compensated gangways”, Safety Notice ED02-2024, May 2024. https://www.hse.gov.uk/safetybulletins/serious-injury-motion-compensated-gangways.htm
10. UK Cabinet Office, “Telecoms resilience”, updated 5 December 2019. https://www.gov.uk/guidance/telecoms-resilience
11. European Union, “Regulation (EU) 2016/679, Article 5: principles relating to processing of personal data”. https://eur-lex.europa.eu/eli/reg/2016/679/oj
12. UK Information Commissioner’s Office, “Data protection and monitoring workers”, under review following the Data (Use and Access) Act. https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/employment/monitoring-workers/data-protection-and-monitoring-workers/
