Deployment Assessment: Zaporozhye Nuclear Power Plant

Site Overview

Zaporozhye Nuclear Power Plant (ZNPP) is Europe's largest nuclear facility, comprising six VVER-1000 reactors with a combined installed capacity of 5,700 MW. Located in Enerhodar, Zaporizhzhia Oblast, Ukraine, the plant is operated by Energoatom and has been under Russian military occupation since March 2022. All six reactors have been in cold shutdown since September 2022. The site sits at the intersection of active armed conflict, nuclear safety obligations under IAEA oversight, and a near-total absence of verified autonomous or robotic system deployments — a combination that defines one of the highest-risk infrastructure profiles currently assessable.

CARVER composite score: 45/60 — the highest theoretically achievable for a site of this type, driven by perfect scores in Criticality, Effect, and Recognizability. DRES composite: 7.2 (HIGH). These scores are not projections; they reflect current, verifiable conditions.

Autonomous early-warning and perimeter monitoring systems are not enhancements at this site — they are the minimum viable detection layer.

---

Why This Site Matters Now

The operational logic is straightforward: a 5.7 GW nuclear plant in a conflict zone, with reactors in cold shutdown dependent on external power for cooling, represents a radiological release scenario with consequences measurable in decades and across international borders. The 290,047 people within 25 km are the immediate exposure population; the actual consequence radius for a Fukushima-class event extends to hundreds of kilometers across southeastern Europe.

Three factors make the 12–24 month window analytically distinct from prior periods:

1. Sustained conflict proximity. The front line has remained within operational strike range of the plant since 2022. Shelling incidents have been documented by the IAEA at and near the site on multiple occasions, though no attack events are recorded against the specific facility in the dataset underlying this assessment.

2. Zero verified autonomous system deployments. Despite a robotics relevance score of 10/10 (a standalone robotics-applicability measure, not a CARVER dimension) and explicit identification of inspection robotics as a gap, no deployed autonomous systems — C-UAS, inspection robots, perimeter sensors, or ground-based autonomous platforms — are publicly confirmed at ZNPP. This is the primary operational finding of this assessment.

3. Regulatory coverage exists; enforcement does not. Ukraine's nuclear regulatory framework nominally applies, but the occupying force's operational control over the site creates a de facto enforcement vacuum. IAEA monitoring missions have accessed the site intermittently, but their instrumentation and personnel presence are not equivalent to a deployed autonomous monitoring capability.

---

CARVER Analysis

Component	Score	Implication
Criticality	10/10	Loss of cooling function risks core damage across six units simultaneously
Accessibility	5/10	Military occupation reduces adversarial accessibility from outside; insider threat elevated
Recuperability	1/10	Near-zero — warzone conditions preclude meaningful recovery timeline
Vulnerability	9/10	External power dependency, degraded maintenance staffing, conflict-induced stress on systems
Effect	10/10	Radiological release consequence is transnational and multi-generational
Recognizability	10/10	Global media saturation since March 2022

Recuperability at 1/10 is the score that most directly governs procurement urgency. A site that cannot recover from a disruption event within any operationally meaningful timeframe has no margin for detection-and-response latency. Autonomous early-warning and perimeter monitoring systems are not enhancements at this site — they are the minimum viable detection layer.

---

DRES Assessment

Composite: 7.2 (HIGH)

Sub-domain	Score	Operational Reading
Air	4.7	Moderate aerial threat exposure; FPV and loitering munition use is documented across the broader conflict theater
Ground	15.1	Elevated ground-domain threat; military occupation creates both insider and perimeter breach vectors
Subsurface	17.1	Highest sub-score; reflects infrastructure interdependency and underground utility vulnerability
Criticality	4.69	Normalized criticality within DRES framework; consistent with CARVER reading
Hardening	17.1	Physical hardening of reactor containment structures is substantial, but balance-of-plant and external power infrastructure is not equivalently hardened
Target Profile	15.1	Globally elevated; the site has been explicitly referenced in conflict-related communications by multiple state actors

The Air sub-score of 4.7 warrants specific attention. Across the Ukrainian conflict theater, FPV drones and loitering munitions have been employed against infrastructure targets at ranges and with accuracy profiles that were not operationally credible at conflict onset in 2022. The ZNPP cooling tower and external transformer yard represent soft targets within the plant boundary. A C-UAS gap at a site with Air DRES of 4.7 and no verified deployments is a quantifiable exposure, not a theoretical risk.

---

Verified Deployments: Primary Finding

No autonomous or robotic systems are publicly confirmed as deployed at Zaporozhye NPP.

This finding is assessed with MODERATE CONFIDENCE — absence of public evidence at a site under military occupation and active information control is not equivalent to confirmed absence of deployment. However, the following conditions make the gap operationally significant regardless of classification status:

• IAEA mission reports through early 2025 reference manual inspection protocols and do not describe autonomous monitoring platforms.
• No procurement announcements, contract awards, or system integration disclosures referencing ZNPP have been identified in open sources.
• Energoatom's public communications reference IAEA coordination and manual safety checks; no robotic inspection program is cited.

For a site scoring 10/10 on robotics applicability — where human access to reactor internals and spent fuel pools is constrained by both radiological dose rates and active conflict — the absence of deployed inspection robotics is an operational gap with direct safety implications, independent of the threat environment.

Specific capability gaps, by domain:

Domain	Gap	Risk Driver
C-UAS	No verified deployment	FPV/loitering munition threat to cooling and power infrastructure
Inspection robotics	No verified deployment	Human access to reactor buildings constrained by dose and conflict
Perimeter autonomous sensing	No verified deployment	Ground DRES 15.1; occupation creates non-standard perimeter threat model
Subsurface/utility monitoring	No verified deployment	Subsurface DRES 17.1; external power supply lines are critical single points

---

Procurement and Threat Exposure: 12–24 Month Outlook

C-UAS. The conflict theater has demonstrated sustained drone employment against infrastructure targets. Any return of Ukrainian operational control — or any negotiated access arrangement — would trigger immediate C-UAS procurement requirements. Systems capable of operating in GPS-degraded environments (consistent with documented electronic warfare conditions in Zaporizhzhia Oblast) and compliant with nuclear facility RF emission constraints are the relevant procurement category. MODERATE CONFIDENCE that C-UAS procurement will be a first-order requirement within 12 months of any change in site control status.

Inspection robotics. The IAEA has publicly identified the need for remote monitoring capability at ZNPP on multiple occasions. Radiation-hardened inspection platforms with manipulator arms — capable of operating in VVER reactor geometry — represent the most technically constrained procurement category. Lead times for radiation-hardened robotic systems from qualified vendors are typically 18–36 months from contract award. This means procurement decisions made in the current window determine operational capability in 2027–2028. HIGH CONFIDENCE that this gap will drive procurement regardless of conflict resolution timeline, given accumulated deferred maintenance and inspection backlog.

Perimeter and subsurface sensing. Ground DRES of 15.1 and Subsurface DRES of 17.1 indicate that the highest-probability near-term threat vectors are not aerial. Fiber-optic distributed acoustic sensing (DAS) for subsurface utility monitoring and autonomous ground sensor networks for perimeter coverage are procurement categories with shorter lead times (6–18 months) and lower regulatory friction than radiation-hardened robotics. MODERATE CONFIDENCE these will be early procurement priorities under any stabilization scenario.

Regulatory and funding pathway. FEMA C-UAS grant mechanisms do not apply to non-U.S. sites. Applicable funding pathways include IAEA Technical Cooperation, EU Nuclear Safety Cooperation instruments, and bilateral defense assistance programs. Energoatom's procurement authority is currently constrained by occupation status; shadow procurement through Ukrainian government channels or international intermediaries is the operative model for any near-term capability introduction.

---

Analyst Notes

The CARVER score of 45 places ZNPP at the upper bound of assessable sites in this framework. The combination of maximum Effect and Criticality scores with a Recuperability score of 1 means that the standard risk calculus — probability times consequence, mitigated by recovery capacity — effectively collapses to consequence alone. No autonomous system deployment changes the underlying nuclear physics or the conflict environment. What autonomous systems change is detection latency and human exposure in the inspection and monitoring functions that are currently being performed manually, under degraded conditions, by a reduced workforce.

The 15,420 people within 5 km and 290,047 within 25 km are the population at direct acute risk in a release scenario. The transnational consequence population is not bounded by these figures.

---

Confidence: MODERATE | Assessment Valid Until: 2027-04-22

Confidence limited by: military occupation information control, absence of IAEA instrumentation disclosure, and unverifiable current site access conditions. Structural findings — CARVER/DRES scores, deployment gap, population exposure — are HIGH CONFIDENCE. Procurement timeline projections are MODERATE CONFIDENCE.

---