Systematic Methodology for Mitigating Radon Risk and Enhancing Fruition of Underground Sites: The Case-Study of Herculaneum’s Theatre

Abstract

Radon exposure poses a significant health risk in underground cultural heritage sites, where limited ventilation and prolonged visitor presence can lead to high radon exposures. While previous studies have concentrated on monitoring and mitigation strategies, few have developed a comprehensive approach that ensures both safe and sustainable site use. This research introduces an innovative methodology that integrates periodic/seasonal radon risk assessment with risk-informed access management based on periodic monitoring and time tracking. This approach is based on: (i) periodic monitoring to obtain representative concentrations; (ii) the calculation of permissible stay durations using a dose-based framework; (iii) implementation via access registration (badges) and procedural measures; and (iv) the application of mitigation measures when concentrations exceed limits (otherwise, the dose is evaluated in accordance with the applicable reference levels). This strategy was implemented and validated at the Roman Theatre in Herculaneum, a unique case study characterised by complex architectural constraints (as the theatre is completely underground) and high cultural significance. Results from years of monitoring, along with ongoing campaigns, demonstrate that this methodology not only reduces radon-related health risks but also enhances visitor experience. This integrated framework provides a replicable model for balancing conservation, safety, and accessibility in underground heritage sites.

1. Introduction

1.1. Historical Background

The Roman theatre of Herculaneum, as mentioned in an inscription that was frequently displayed on the entrance arches of the monument, was constructed during the Augustan age. Placed at the NW of Herculaneum, it was designed by the architect Numisius and funded by the duumvir Annius Mammianus Rufus [1,2,3]. The structure was located near the Forum and featured an external façade composed of two rows of arches, supported by pilasters topped with Doric capitals [4,5]. The semicircular seating area, which could accommodate 2500 people, was built entirely above ground and divided into three sections according to specific rules1, reflecting different social classes. The arrangement of seats was determined by factors such as merit, social rank, age, and gender, which helped reinforce the social hierarchy through the visual representation of one’s status [6,7]. The bourbon tunnels illustrate this distinction:

• The ima cavea, which is closest to the stage, featured marble steps for movable chairs (subsellia) and was designated for city authorities.
• The media cavea, equipped with tuff seats (Figure 1a), was accessible to the middle class.
• The summa cavea was set apart from the media cavea by a podium decorated with marble pilasters and rectangular panels, and it was reserved for less prominent individuals and women.

Eminent figures of the city and distinguished guests occupied special boxes of honour (tribunalia) located on either side of the stage, above the two vaulted entrance corridors (parodoi).

Here, two marble bases recall the original presence of honorary statues dedicated to the city’s benefactors: Marcus Nonius Balbus, tribune of the plebs in 32 BC, praetor and proconsul of Crete and Cyrene [8] (Figure 1b) and Appius Claudius Pulcher [9,10,11], consul in 38 BC and member of one of Rome’s most important senatorial families. One of the two accessible tribunalia features a white marble slab floor and remnants of elegant wall decorations. At the top of the cavea, where a velarium protected spectators from sunlight and improved acoustics [12] were a sacellum and two aedicules that likely displayed statues of emperors and imperial family members [13]. The stage (pulpitum) is situated in front of the cavea and orchestra, both of which retain traces of marble flooring. The supporting wall (proscaenium) remains largely intact, though it lacks its original marble decoration, featuring rectangular and semicircular niches (Figure 1c). The monumental stage backdrop, known as the scaenae frons, is a brick structure featuring a central apse for the porta regia and two valvae hospitales on the sides. It has two levels marked by niches, pilaster strips, and columns, and was originally decorated with polychrome marble and statues. After the stripping activity of the 18th century, only a few architectural fragments and imprints on the mortar surfaces for the marble slab bedding remain from this sumptuous decoration. Two spacious rooms (versurae) with well-preserved 4th style frescoes flanked the stage (Figure 1d). Spectators gathered there to socialize and enjoy refreshments during intermissions or after performances.

This building was buried under over 20 m of volcanic material from the Vesuvius eruption in 79 AD until 1710, when farmer Ambrogio Nocerino, known as Enzechetta, discovered marbles while digging a well. The artifacts were later bought by Emanuele Maurizio di Lorena, Duke of Elboeuf, to decorate his new home in Portici. To uncover the rich marble cladding and beautiful statues from the ancient theatre, d’Elboeuf had tunnels dug into the volcanic tuff. Thirty years later, in 1738, King Charles of Bourbon commissioned Joaquín Roque de Alcubierre to continue the excavations. This project was groundbreaking in its scale and discoveries, marking the beginning of modern archaeological research of the cities buried by Vesuvius [3,14]. Starting from the Theatre, Alcubierre expanded the underground tunnel network to the south and created an initial plan. This risky endeavour allowed explorations of the volcanic mud covering the city and gradually covered the entire ancient centre. Assisting him, Swiss engineer Karl Weber recognized the city’s urban layout’s importance and developed accurate surveys of the bourbon excavations while organizing the tunnel network with updated plans. The innovative graphic documentation approach, as seen in the Alcubierre and Weber theatre plans from 1738 and 1763 [15], highlights the modern aspects of Bourbon investigations. Excavation teams understood the importance of creating accurate plans for the main structures found in the tunnels2.

The Theatre of Herculaneum became a key stop on the Grand Tour [3] for young aristocrats and artists from the European elite between the eighteenth and nineteenth centuries3. To improve access, rooms and a ramp to the underground tunnels were arranged in 1750. These spaces were remodelled in 1849 in the Pompeian style and restored in 1865 on the initiative of the archaeologist Giuseppe Fiorelli.

The Theatre, well-preserved yet lacking in valuable materials and furnishings, is located around 25 m underground between the buildings of the new Herculaneum, linking the archaeological park to the historic city Centre. After nearly 20 years of closure, it was reopened to the public in 2018 by the new administration established following the 2016 ministerial reform4. Restoration efforts ensured safety and allowed the community to reclaim this important symbol of their history and cultural identity.

The new visitor route allows the public to explore the “four lives” of the Theatre: as a refuge during World War II, a stop on the Grand Tour, the site of early tunnel explorations, and a Roman show building. It also features notable geological formations, including impressive stalactites and stalagmites created by dripping calcareous water in the Roman-era rooms.

1.2. Radon as a Limiting Factor for Fruition in Underground Sites

The fruition of underground sites is usually limited by the need to maintain proper conservation conditions for the artifacts [16,17]. The presence of visitors can increase relative humidity values and carbon dioxide concentration, which contribute to the degradation of the artifacts. This degradation can manifest in various forms, such as salt crystallization [18,19,20,21], the growth of microorganisms, damage to wall paintings, and deterioration of the rock [18,19,20,21]. In the case of the Herculaneum Theatre, issues such as percolation of rainwater from the surrounding soil, detachment of stone materials, and inadequate ventilation have compounded the challenges. Furthermore, the volcanic nature of the surrounding rocks has resulted in elevated radon concentration, posing potential health risks for the staff. This was the reason the site was shut down in the late 1990s.

Radon is classified as a Group I human carcinogen by the International Agency for Research on Cancer (IARC) in Monograph 43 [22]. WHO classifies radon as the second leading cause of lung cancer after cigarette smoking. Inhaling radon particles and their decay products can damage lung cells, increasing the risk of cancer. Radon exposure does not cause immediate or visible symptoms, and health effects may appear years later. Epidemiological studies conducted in residential settings have provided direct evidence linking radon exposure to an increased risk of lung cancer. The relationship between radon exposure and risk is linear, meaning there is no threshold level below which the risk is eliminated. The risk remains significant even at low concentrations of radon, especially with constant and prolonged exposure. The Excess Relative Risk (ERR) associated with long-term average exposure (around 30 years) is approximately 16% for every 100 Bq/m³ increase in radon concentration, according to the World Health Organization (WHO) [23].

According to the European Directive 59/2013/EURATOM [24], the limit values for average annual radon concentrations in homes and workplaces are set at 300 Bq m⁻³ for existing buildings and 200 Bq m⁻³ for buildings built after 31 December 2024.

To reduce the risk of cancer due to radon, gas concentrations in enclosed spaces should be as low as possible. Mitigation can be achieved by preventing the entry of gas from the subsoil and promoting the escape of that which has penetrated living spaces. The methods for preventing radon from entering buildings can be categorized as passive or active [25,26,27]. For new buildings, passive techniques include placing an impermeable membrane between the soil and the structure. For existing buildings, it involves sealing entry routes such as cracks, floor-wall joints, and service passages, including electrical, thermal, and hydraulic. Another active technique involves depressurizing the soil beneath a building. This is accomplished by creating a small well in the soil, which is enclosed by permeable walls. The well is connected to an extraction system that expels gas outside, thereby preventing it from entering the building. If it is not possible to avoid the entry of radon, it is possible to reduce the concentration of the gas by introducing external air through periodic ventilation. This can be achieved by using natural or mechanical ventilation systems. Mechanical ventilation, which allows the incoming air flow to be controlled, is particularly effective in ensuring efficient dilution of radon within the environment. Obviously, in underground sites, especially those dug into volcanic materials, the radon risk is even greater and increases if the environments are poorly ventilated [28,29,30,31]. In Italy, the actions to be implemented to mitigate the risks associated with radon are regulated by the National Action Plan for Radon 2023–2032 [32].

1.3. Aim of the Paper

The Roman Theatre of Herculaneum, which was buried by the eruption of Vesuvius in 79 AD and is only partially excavated beneath the modern city, poses unique challenges for public access and safety. The accessible areas, including a network of tunnels and parts of the cavea and scaena, significantly increase the risk of radon exposure. This risk led to the site being closed to public access in the 1990s. Unlike other underground archaeological sites, conventional mitigation techniques, such as large-scale ventilation or structural modifications, are impractical in this case due to strict conservation constraints and the complex underground layout. This paper introduces a novel integrated approach that transcends traditional monitoring by combining dynamic radon risk management with adaptive visitor access strategies, specifically designed for underground cultural heritage environments. The methodology has been developed and refined over years of experimental campaigns to ensure safe visits to the site without compromising its integrity.

The following sections detail: (i) the innovative framework used to characterize and quantify radon risk under real operating conditions; (ii) the results of multiple experimental campaigns (some of them still in progress) that validate this approach; (iii) the proposed interventions and implemented strategies that allow for controlled, sustainable access to the theatre; and (iv) the broader implications of this methodology for the enjoyment of the theatre and similar heritage sites. By addressing occupational health safety and cultural accessibility within a single integrated model, this research fills a crucial gap in current radon mitigation practices and assists archaeologists and engineers in finding replicable solutions for sites where conventional techniques cannot be applied.

2. Materials and Methods

2.1. Legislation-Driven Transferable Protocol (Procedural Steps)

The operational workflow adopted in this study is guided by the assessment phases mandated by Italian legislation, which requires radon measurements representative of the annual average concentration, accounting for seasonal and diurnal variability. To ensure transferability to other subterranean heritage sites, the protocol is structured as follows:

• Site screening and zoning (Observation stage): Identify access areas and openings, document constraints for mitigation, and define the zones relevant to workers and visitors.
• Monitoring design (Analysis stage): Perform a technical site inspection and define the monitoring plan; where workers are present, assign passive dosimeters and record individual occupancy/working hours.
• Long-term monitoring for annual average (Expertise stage): Deploy passive dosimeters at representative points and process data to obtain six-monthly and annual mean concentrations.
• Regulatory decision logic: Compare the annual average concentration with the action level and, where required, verify compliance through dose assessment.
• Access management (Operational control): Translate the dose-based evaluation into permissible stay durations and implement controlled access through time tracking (e.g., entry badges) and periodic updates of the limits.

The following sections discuss the application of this methodology to the underground archaeological context of the Roman Theatre in Herculaneum, starting with a description of the site and moving to the timeline of interventions and the investigation methodologies employed.

2.2. The Theatre Geometry

The complex geometry of the theatre is shown in Figure 2. The route includes a tunnel that connects the entrance to the large Enzechetta well, along with a staircase carved into the pyroclastic mud, extending over an area of 76 m². The Roman walkable surfaces are represented by the following measurements: 167 m² for the ambulatory of the media cavea, 79 m² for the radial stairs, 34 m² for the ima cavea, 108 m² for the media cavea, 28 m² for the summa cavea, 46 m² for the two tribunals, 60 m² between the proscenium and orchestra, and 165 m² for the pulpitum.

In total, these areas add up to 763 m² (800 m² if the surface area of the tunnels connecting the various spaces are considered). The visitor route only includes 155 m² of walkway, 20 m² of radial stairs, 27 m² of media cavea, 6 m² of ima cavea, 60 m² between the proscenium and the orchestra, and 104 m² of pulpitum, totalling 372 m² of effective visiting area (from the initial 2000 m²).

2.3. The Investigation Timeline

The investigation timeline is crucial not only for reconstructing the sequence of actions undertaken to ensure the safe reopening of the Theatre, but also for demonstrating how these steps have progressively contributed to the development of a systematic protocol for radon risk assessment and sustainable site fruition, reflecting substantial advancements in mitigation procedures over the past decades.

In 2018, the management of the Herculaneum Archaeological Park made the decision to open the ancient theatre to the public. They aimed to implement a solution that would minimize the risk of radon exposure to visitors and staff. This approach focuses on reducing both the dose of radon exposure and the time spent in affected areas, ensuring a safe, systematic, and easily manageable experience for everyone at the site.

In this context, it is necessary to act both on exposure time and on the dose absorbed by individuals. The absorbed dose is expressed in Gray (Gy), and is defined as:

$$D={\displaystyle \frac{d\epsilon }{dm}}$$

(1)

The challenge was significant because this objective had never been achieved for several reasons. First, the high relative humidity negatively impacted the lifespan of the lighting systems. Second, the very poor ventilation increased the risk associated with elevated radon concentrations. Below are the main investigation phases, one of which is still ongoing.

• In 1998, the Archaeological Superintendency of Pompeii tasked the Interdepartmental Centre for Engineering for Cultural Heritage (CIBeC) at the University of Naples Federico II with a study. The study’s purpose was to determine the necessary safety measures—specifically regarding ventilation issues—to open the theatre to the public, ensuring the well-being of both visitors and staff.
• In 1999, the Radiation Protection Service of the University Federico II of Naples conducted a radon concentration monitoring that involved four distinct points of the monument (on behalf of the CIBeC).
• In 2008, the Herculaneum Conservation Project5 (HCP) commissioned further monitoring from the Ionizing Radiation Laboratory—Natural Radioactivity Sector of the Higher Institute for Prevention and Safety at Work (ISPESL). This later study aimed to provide a general indication for technical consultancy regarding radon risk diagnostics.
• Scheduled in 2017, a 2018–2019 monitoring campaign used personal dosimeters on theatre personnel to determine absorbed radon doses. The goal was to inform safety measures for public reopening, pending full annual radon monitoring, with checks occurring monthly in 2018 and annually throughout 2018–2019.
• Environmental monitoring campaign started in 2024 (still ongoing).

The methodologies adopted in each phase will be described in the following sections.

2.3.1. The Ventilation Study Carried out in 1998

The studies conducted in 1998 were mainly focused on the ventilation of the theatre.

Air changes in the theatre primarily rely on the open well, two ventilation pipes at street level, and two access doors (Figure 2). Since both doors are typically closed and sit above the scaena, the well is essentially the primary source of outdoor air entering the space. To determine the extent of the radon-related risk, it was essential to calculate the air change rate n, as defined by Equations (2) and (3).

$$n={\displaystyle \frac{q_{\sup }}{V}}$$

(2)

$${\tau }_n={\displaystyle \frac{1}{n}}={\displaystyle \frac{V}{q_{\sup }}}$$

(3)

Due to the Theatre’s architectural configuration and the influence of climatic conditions of supply air, the only viable method for estimating the air change rate was the tracer gas technique. This method involves introducing a gas with suitable properties (the tracer) into the indoor environment and monitoring its concentration over time [33,34]. The experimental layout is shown in Figure 3.

Table 1. Gas tracer technique: methods, hypotheses, and equations for the concentration of the tracer over time C(t) and the evaluation of the air change rate n [34]. q is the emission rate in m³ h⁻¹.

Method	Hypotheses	Initial Condition	C(t)	n	Note
Step down	q = 0 n = cost.	C(0) = C0	$C_0\cdot e^{-nt}$	${\displaystyle \frac{\ln \left(C_0\right)-\ln \left(C\left(t\right)\right)}{t}}$	Gas is premixed with air
Step up	q = cost. n = cost.	C(0) = 0	${\displaystyle \frac{q}{nV}}+\left(C_0-{\displaystyle \frac{q}{nV}}\right)e^{-nt}$	${\displaystyle \frac{q}{VC\left(\infty \right)}}$	Gas is injected with a known and constant flow rate, up to a constant concentration C(∞)
Steady state concentration	$C=\overline{C}$	C(0) = 0	${\displaystyle \frac{q\left(t\right)}{Vn\left(t\right)}}$	${\displaystyle \frac{q\left(t\right)}{V\overline{C}}}$	Gas is injected with a known and modulable flow rate to maintain a constant concentration

outlines the three tracer release techniques and the expected concentration trends, assuming perfect mixing. The corresponding equation describing the concentration decay and the formula for calculating n(t) under perfect mixing conditions are also specified. Any gas with precisely measurable concentration can serve as a tracer, provided it is non-reactive, non-flammable, non-toxic, not naturally abundant in the air, has a density close to that of air, and is cost-effective.

Table 2. Most common gas tracer used over time.

Tracer	Molecular Weight (kg kgmol−1)	Boiling Point (°C)	Range of Detection (ppm)	Base Concentration (ppm)	GWP
CO2	44	−56.6	0.05–2000	Variable	1
N2O	44	−88.5	0.05–2000	0.03	265
CCl2F2 (R12)	121	−29.8	0.05–2000 0.001–0.05	-	10,900
SF6	146	−50.8	0.05–2000 0.00002–0.5	-	22,800
C6F14	338	57.0	10−8	-	9300

lists commonly used tracer gases. Fluorine-based gases are no longer used due to their high GWP (Global Warming Potential).

Given the spatial complexity and extent of the Theatre, the study was confined to the ima cavea, the area of highest visitor interest and directly connected to the central well (Figure 2). To estimate the air change rate n, the constant concentration method was adopted, employing sulphur hexafluoride (SF₆) as the tracer gas the use of which was permitted at the time.

As indicated in Table 1, calculating the air change rate requires an accurate determination of the volume of the environment. In this case, the volume estimation was particularly challenging due to the irregular geometry of the walls and the presence of numerous blind tunnels. To address this, a one-meter square grid was superimposed onto the floor plan of the ima cavea (Figure 4a) covering a longitudinal span of 55 m along the proscenium and extending 20 m orthogonally. By measuring the vertical height at the centre of each grid cell, the total volume was approximated at 930 m³, with an estimated uncertainty of ±10%.

To ensure adequate mixing of the tracer gas with ambient air, two portable, adjustable fans (each with a mass < 60 kg, 1 kW of total electrical power consumption, single-phase) were employed. These devices provided an air velocity of 0.50 m s⁻¹ at 15 m, promoting uniform dispersion of the tracer.

The tracer gas injection, sampling, and monitoring system comprised the following instrumentation:

• Two Brüel & Kjær (Nærum, Denmark) Type 1302 Multi-Gas Monitors;
• One Brüel & Kjær (Nærum, Denmark) Type 1303 Multipoint Sampler and Doser;
• One Innova AirTech Instruments (Ballerup, Denmark) Type 1303 Multipoint Sampler;
• Two personal computers equipped with Brüel & Kjær (Ballerup, Denmark) 7620 software (B&K 7620 Version 3.5) for system control and data acquisition.

The measurement campaign required the transport and installation of over 100 m of PTFE tubing for sampling, along with the sulphur hexafluoride (SF₆) cylinder, portable fans, support tables, and mounting structures for the various sampling instruments.

As shown in Figure 4b, SF₆ concentrations were sampled at thirteen distinct locations, primarily distributed across the proscenium and stage areas. To ensure effective dispersion, the tracer gas was introduced at two injection points positioned at opposite ends of this zone.

All sampling points, except P1 and P12, were positioned at a height of 1.50 m above ground level, corresponding to the typical breathing zone. To verify adequate mixing conditions of the tracer gas and exclude the onset of stratification phenomena, P1 and P12 were placed at 3.90 m and 0.20 m above the ground, respectively. Sampling point P4, located at the geometric centre of the study area, was used by the system to regulate the tracer gas injection rate.

The Superintendency required that all operations be conducted within a single day and during regular public opening hours of the archaeological site. Consequently, the available time for measurements was limited, especially when accounting for the setup and transportation of equipment.

The air change rate measurement was conducted on 9 January 1998, between 12:30 and 18:00. The measurements were conducted in the absence of visitors, as the theatre was closed to the public at the time due to safety requirements. The procedure followed these main phases:

• The tracer gas was introduced using the constant concentration method.
• A set-point concentration of 3 ppm for sulphur hexafluoride (SF₆) was established.
• At approximately 16:00, the sampling probe at point P12 was relocated to the summa cavea, near the ventilation shaft, to assess vertical distribution.
• The measurement campaign concluded at 18:00 to allow sufficient time—approximately one hour—for the dismantling and removal of equipment from the theatre.

2.3.2. Radon Concentration Monitoring Campaign Developed in 1999

The monitoring activity of the radon concentration commissioned by CIBeC in 1999 in spring over the period from 18 to 21 May (in the absence of visitors) was carried out in the four points (A, B, C D) shown in Figure 5.

To this purpose four E-PERM^® systems (Rad Elec Ink, Frederik, MD, USA) were used. The system utilizes an electret, a permanently polarized dielectric material, within an ion chamber. Radon gas entering the chamber decays, ionizing the air, and the resulting ions are attracted to the electret, causing a voltage decrease that is proportional to the radon concentration and exposure time.

2.3.3. Radon Concentration Spot Measurements Carried out in 2008

The HCP-commissioned monitoring activity involved a 5 h spot measurement of radon concentration, temperature, and humidity using an AlphaGUARD (Laurus Systems, Ellicott City, MD, USA) ionization chamber (sampling interval: 10 min). The measurement point was situated in the tunnels behind the scaenae frons (see Figure 5, point I).

2.3.4. The Monitoring Campaign Carried out over the Period 2018–2019

The Systematic Assessment Strategy Formulated and Adopted

The strategy implemented since 2017 aligns with ISO 15265 [35] for assessing thermal stress risk [36] according to the SOBANE (Screening, OBservation, ANalysis, Expertise) method [37]. This approach offers significant advantages by promoting a proactive and participatory process that integrates prevention, observation, and analysis before transitioning to expert-level evaluation. By progressively involving stakeholders and focusing on practical solutions at each stage, the SOBANE framework ensures effective risk management, minimizes unnecessary interventions, and supports continuous improvement.

Table 3. Comparison of the characteristics of the 3 stages of the strategy SOBANE.

Modalities	Stage 1 Observation	Stage 2 Analysis	Stage 3 Expertise
When?	when a “problem” is detected	more complicated cases	very complex cases
How?	qualitative observations	ordinary measurements	specialised measurements
Cost?	Low	average	high
Duration (order of magnitude)	2 h	1 day	a few days
By whom?	workers + management from the company	same + specialists	same + specialists + experts
Competency
work situation:	high	average	low
ergonomics:	average	high	specialised

summarizes the three phases of the adopted methodology.

During the observation phase, the various activities conducted in the Theatre were examined, with interviews carried out with managers and, when necessary, the workers involved.

The analysis phase was based on two complementary census methods, in agreement with Italian workplace safety legislation. Specifically, Legislative Decree 81/2008 [38] mandates that a safety expert prepare a risk assessment document, which includes all information on potential hazards to which workers may be exposed, along with their analysis and quantitative evaluation. Additionally, at the time of the measurements, Legislative Decree 230/1995 [39], supplemented by Legislative Decree 241/2000 [40], required that radon risk assessments be conducted by a qualified expert in radiation protection, a professional with specific qualifications defined by the same law.

A comprehensive analysis of work activities and tasks across the various operational phases was undertaken. This included gathering statements on professional roles, conducting interviews with department heads, and involving workers directly. Targeted inspections followed, aimed at assessing workplaces, systems, equipment, substances, preparations, and agents in use, to verify their compliance with current legislation.

Identified issues and non-conformities were then examined through a semi-quantitative assessment of exposure levels, considering operational factors such as task frequency and duration, agent characteristics, and the efficacy of collective and individual protection measures. Additionally, an estimation was made regarding the likelihood of occurrence and potential severity of associated damage.

The inspections also revealed that the site’s particular configuration poses significant challenges for interventions on the monument. This is not only for remediation purposes but also for broader improvements to environmental and safety conditions.

During the expert assessment phase, environmental parameters were measured and the potential risk of radon exposure to individuals was assessed, as detailed below.

Dose Assessment Model and Monitoring Features

Monitoring of maintenance workers, initiated in 2018, has been conducted in accordance with Italian legislation (Figure 6).

According to the Italian legislative workflow summarised in Figure 6, radon measurements must be representative of the mean annual activity concentration, explicitly accounting for seasonal (typically higher in summer and lower in winter) and diurnal fluctuations. Should the mean annual radon concentration exceed 500 Bq m⁻³, remedial actions are required, and a qualified expert must verify that the effective dose for workers does not exceed 3 mSv year⁻¹. In the present study, worker exposure to radon is assessed using Equation (4), expressed in its conventional form.

$$E=T\cdot F\cdot C$$

(4)

where E is the effective dose expressed in mSv yr⁻¹, C is the radon activity concentration in Bq m⁻³, T is the occupancy time (i.e., the actual annual duration of stay) in h, and F is the conventional conversion factor in mSv m³ Bq⁻¹ h⁻¹.

For the monitoring period, the conventional conversion factor adopted in the standard regulatory practice cited in the manuscript is 3·10⁻⁶ mSv m³ Bq⁻¹ h⁻¹ [41]; the manuscript also notes the currently revised value of 6.7·10⁻⁶ mSv m³ Bq⁻¹ h⁻¹ [24,42,43].

The practical implementation of Equation (4) requires: (i) representative C values; and (ii) documented occupancy times, T. In this case study, a qualified expert designated 20 monitoring points to capture the pronounced spatial variability within the monument, distinguishing street-level access areas (points 1–4 in Figure 7) from deeper underground areas (points 5–20 in Figure 7). Annual means are utilised when translating exposure into management constraints. Worker-specific T values are derived from recorded working hours (time tracking), as described in the monitoring procedure. Finally, Equation (4) is applied here primarily for worker exposure management, as worker presence is recurrent and cumulative. Conversely, as visitor presence is typically short and occasional (limited to the duration of a guided visit), visitor protection is addressed operationally by limiting stay durations and controlling access conditions. Meanwhile, quantitative annual dose assessment is used to define and verify staff exposure constraints within the aforementioned regulatory framework.

Consequently, INAIL conducted individual worker monitoring from 28 May to 22 June 2018. Nine workers were equipped with CR-39 passive dosimeters6 featuring nuclear trace detectors. These devices are robust against high relative humidity, such as that observed in the Theatre (

Table 4. Temperature and humidity values recorded in 2008.

Quantity	Range of Values	Average Value
Air temperature (°C)	13.9–15.1	14.3
Relative humidity (%)	96.5–99.0	98.0

).

The procedure followed the following steps:

• Initial technical inspection.
• Assignment to the 9 workers of dosimeters equipped with appropriate identification codes.
• Recording the working hours of each person involved in the monitoring.
• Analysis of the radon gas activity concentration data.
• Conversion factors described in Legislative Decree 241/2000 [40].
• Assessment of personnel exposure.

A year-long monitoring program was implemented to enhance understanding of radon concentration levels within the Theatre. The objective was to characterize seasonal variations and confirm the accuracy of prior monitoring results. Following an on-site inspection, the qualified expert designated 20 measurement points, illustrated in Figure 7, for the placement of passive nuclear trace dosimeters Radout (Mi.am Srl, Piacenza, Italy).

2.3.5. Monitoring Campaign Started in 2024

A second monitoring campaign is currently underway to ensure the annual average radon concentration remains below the 300 Bq m⁻³ limit, in compliance with current Italian legislation [38]. Concurrently, a multi-parameter monitoring campaign has been initiated at 46 different positions to enhance the understanding of the Theatre’s internal environmental conditions. This monitoring employs shielded dataloggers to record temperature, relative humidity, air velocity, and CO₂ concentration, alongside passive nuclear trace dosimeters. Additionally, surface temperature and relative humidity sensors are being utilized. Should radon concentration levels exceed the limit, mitigation measures must be implemented to reduce concentration. If reduction below the new limit is not feasible, the annual effective dose must be evaluated using currently applicable reference values.

3. Results and Discussion

Results from the monitoring data collected over the years will be shown and discussed by following the timeline of the different campaigns still in progress.

3.1. The Ventilation Study Carried out in 1998

Figure 8 summarizes the measurement results, demonstrating effective tracer mixing in the air due to fan operation and placement.

Particularly, the trends of the measured concentrations quite overlap (Figure 8a). Unfortunately, considering the boundary conditions, an effective steady-state condition was not reached, except approximately and for short intervals of time. In the first stationarity interval, identified between 2:30 pm and 3:30 pm, an average value of n = 0.4 h⁻¹ was calculated (Figure 8b); in the second interval, between 4:30 pm and 5:00 pm, n = 1.9 h⁻¹ was obtained (Figure 8c), values that were decidedly too low to guarantee the dilution of the radon. A more accurate assessment of air exchange rates would have required continuous, multi-day monitoring conducted during both winter and summer periods. Nevertheless, the values derived were adequate to conclude that ventilation alone would be insufficient to achieve safe conditions. This determination is further underscored by the fact that measurements were acquired within one of the lower-risk areas, attributed to its direct well connection.

3.2. The Ventilation Study Carried out in 1998 and 2008

The 1999 monitoring data, summarized in

Table 5. Results from the monitoring campaign carried out in 1999 and 2008 in positions showed in Figure 5. Concentration measurements were performed at breathing zone height (1.5 m above floor level).

Point	Height Above Sea Level (m)	Mean Radon Concentration (Bq m−3)	SD (Bq m−3)
1999
Higher media cavea (A)	34.38	239	121
Big well (B)	28.86	2384	120
Lower media cavea (C)	23.34	2562	128
Orchestra (D)	17.82	2313	116
Mean value	-	2406	61
2008
Proscaenium (I)	17.82	371	312

, substantiated earlier findings regarding the insufficient ventilation of the Theatre. Specifically, the average radon concentration measured at positions A through D was approximately 2500 Bq m⁻³. This value was five times greater than the 500 Bq m⁻³ limit established by Italian legislation at that time for radon exposure in underground environments.

Furthermore, the concentration documented in 2008 was even higher. This increase is mainly due to the placement of the measuring point (I in Figure 5) within a remote, poorly ventilated tunnel of the theatre.

3.3. Measurement Campaign Carried out over the Period 2018–2019

3.3.1. Monitoring on Workers

The 2018 monthly monitoring results for workers, detailed in

Table 6. Results from INAIL’s monitoring ¹ on nine workers over the period from 28 May 2018 and 22 June 2018.

Dosimeter ID	Worker	Total Exposure Time (h)	Dose (mSv)
57139	A	98	0.01
57183	B	97	0.01
57222	C	194	0.6
57259	D	103	0.1
57296	E	106	0.01
57092	F	203	0.4
57135	G	75	0.0
57294	H	200	0.2
57301	I	36	0.01

¹ In bold maximum absorbed dose and the minimum absorbed dose in relation to biological damage.

, revealed dose values significantly below the regulatory limit of 3 mSv yr⁻¹. Worker C exhibited the highest absorbed dose, corresponding to 194 h of exposure within the Theatre. Consequently, as previously indicated, the annual monitoring protocol was implemented at the 20 measurement points specified in Figure 7. The outcomes are summarized in Table 6 and discussed in the next section.

3.3.2. Half-Yearly and Yearly Monitoring

According to the data summarized in

Table 7. Results from the radon concentration yearly monitoring over the period from 28.05.2018 to 28.05.2019 in the positions indicated in Figure 7 with SD values.

Measurement Point	Height Above Sea Level (m)	1st Semester Average Concentration (Bq m−3)	2nd Semester Average Concentration (Bq m−3)	Yearly Average Concentration (Bq m−3)
1	39.65	212 ± 23	325 ± 35	269 ± 42
2	34.38	242 ± 26	381 ± 40	312 ± 48
3	34.38	291 ± 31	522 ± 54	407 ± 62
4	34.38	333 ± 36	549 ± 57	442 ± 67
5	34.38	4195 ± 423	1747 ± 177	2965 ± 459
6	34.38	1972 ± 200	1442 ± 147	1706 ± 248
7	17.82	3575 ± 361	1190 ± 121	2376 ± 381
8	17.82	4018 ± 405	1343 ± 137	2673 ± 428
9	17.82	3585 ± 362	1300 ± 132	2436 ± 385
10	17.82	3207 ± 324	1247 ± 127	2222 ± 348
11	23.34	3084 ± 311	1409 ± 143	2242 ± 342
12	28.86	3475 ± 350	1771 ± 180	2619 ± 394
13	32.31	3880 ± 391	1505 ± 153	2686 ± 420
14	17.82	3944 ± 398	1358 ± 138	2644 ± 421
15	28.86	3042 ± 307	1488 ± 151	2261 ± 342
16	29.90	2241 ± 227	1338 ± 136	1787 ± 265
17	17.82	1717 ± 174	1446 ± 147	1581 ± 228
18	27.48	3429 ± 346	1552 ± 158	2486 ± 380
19	17.82	2927 ± 295	1203 ± 123	2060 ± 320
20	17.82	3945 ± 398	1282 ± 131	2606 ± 419

, during the first semester of measurements from 28 May to 3 December 2018, over a total of 4536 h of exposure, the concentration values recorded at the entrance of the monument at street level (points 1 to 4 in Figure 7) were below the threshold, ranging from 212 to 333 Bq m⁻³. Conversely, at the deeper measurement points, the concentration values were higher, ranging from 1717 to 4195 Bq m⁻³, with an average concentration of 3265 Bq m⁻³, exceeding the limit value.

During the second semester, the radon concentration in the underground hypogeum varied between 1190 and 1771 Bq m⁻³, showing a consistent reduction compared to the summer. This phenomenon is due to the temperature difference between the interior and exterior. More specifically, during the hot period (1St Semester), the indoor-outdoor thermo-hygrometric conditions result in an average internal temperature that is much lower than the external one. This causes air stratification, which hinders natural ventilation. In contrast, during the winter (2nd Semester), the Theatre average temperature is significantly higher than the external temperature. This encourages the air to rise naturally through the building, improving ventilation and lowering average pollutant levels.

This is strongly influenced by the morphology of the hypogeum and the position of its openings. The average yearly values inside the Theatre also vary significantly depending on the depth and location. At street-level access areas, the radon concentration ranges from 269 to 442 Bq m⁻³, while in the underground areas, it varies between 1581 and 2965 Bq m⁻³.

3.3.3. Definition of the Allowable Exposure Time for Workers

According to the flowchart in Figure 6, to apply Equation (4) and calculate the maximum exposure time, it would have been necessary to close the monument for an estimated period of 18 months, distribute environmental dosimeters at strategic points with the support of the qualified expert, wait for the conclusion of the annual monitoring to determine the yearly radon concentration, and subsequently estimate the annual effective dose based on the actual hours of workers’ presence. While awaiting the results, the decision was made to reopen to the public based on the 1999 data (Table 5). The expert cautiously defined maximum stay times for guides and security personnel, ensuring that the calculated effective dose remained within current legal limits. In particular, the qualified expert assumed that the effective dose was equal to 0.6 mSv yr⁻¹ and the radon concentration was equal to 1030.93 Bq m⁻³; assuming the conventional conversion factor to be 3·10⁻⁶ mSv m³ Bq⁻¹ h⁻¹, he therefore obtained a precautionary monthly value of the residence time equal to 16 h.

From the last column in Table 7, it is evident that the average radon concentration for the archaeological area (points 5 to 20 in Figure 7) is 2334 Bq m⁻³. By applying Equation (4) with an exposure threshold of 3 mSv/year, the maximum residence time would be 428 h yr⁻¹, equivalent to 36 h month⁻¹. This value is significantly higher than the previously established 16 h month⁻¹, indicating a broad safety margin. However, for enhanced safety, the qualified expert has set a maximum annual residence duration inside the Theatre at 350 h yr⁻¹ for workers, corresponding to 30 h month⁻¹.

This decision is motivated by the following factors:

• The variability of radon concentration throughout the year, which, as mentioned, varies significantly depending on the depth and location of the environments.
• The occasional presence of workers inside the site, which depends on their tasks and visiting shifts.
• Monitoring conditions.

To ensure compliance with this limit, it was decided to record access and duration of stay at the Theatre using entry badges. Additionally, a specific information program regarding the risks of ionizing radiation was organized for both tourist guides and workers to increase awareness of radon risk among all personnel working inside the Theatre.

3.3.4. Further Investigations on the Causes of the Presence of Radon

Radon levels within the Herculaneum Theatre are affected by two primary mechanisms: direct gas emission from tuffaceous rock and the emergence of radon-laden groundwater that permeates these geological formations. Under conditions of limited aeration and water-saturated tuff, the egress of water from the theatre walls can significantly contribute to the accumulation of gas within the environment. Consequently, a comprehensive analysis of the surrounding geological context was deemed essential.

Table 8. Result of the continuous core drilling geotechnical survey in Piazza Fontana, commissioned by the Municipality of Ercolano, which highlights the presence of groundwater 3.5 metres from ground level, altitudes above sea level 39 m, probe 40 mm in diameter.

Depth Form Ground Level (m)	Soil Description
−1.50	Pyroclastic fill material
−3.50	Water table
−4.50	Ash and lapilli—ash and ejecta
−6.50	Fine sand
−9.00	Sand with grey pumice and lava inclusions
−11.00	Coarse sand and with grey pumice
−12.50	Coarse sand with lenses of tuffaceous conglomerate
−14.00	Layers of tuffaceous conglomerate
−16.20	Gray-green tuffaceous conglomerate (massive bed)
−18.00	Tuffaceous conglomerate with white pumice—mosaic
−19.50	Lapilli and Herculaneum bricks—humic layer
−20.00	Heterogeneous sand

summarizes the results of a continuous core drilling survey carried out in the 1970s near the Theatre. It is clear that at just 3.5 m from the ground level, there is groundwater that can penetrate the Theatre, favouring the transport of radon inside the monument.

Figure 9 shows an excerpt from the hydrogeological map of the area where the Theatre is located, which highlights the intersection between the monument and the isopiestic. The map demonstrates the importance of full knowledge of the underground water regime to plan the management of activities aimed at mitigating the effects related to the release of radon into the Theatre. More particularly, in porous and granular soils such as pyroclastic ones, the water present in the interstices between the granules is in hygrometric equilibrium with that present in the air, giving rise to substantial hygrometric inertia that characterizes the Theatre. Moreover, where saturation is only partial, the pores between soil grains are not filled with water but contain enough of it to generate capillary forces that give the soil an apparent cohesion, due to the surface tension resulting from the formation of menisci.

This contributes to maintaining temporary stability; this phenomenon is, for example, observed in pyroclastic soils on the pulpit. Conversely, an increase in the degree of saturation to a fully saturated state would lead to a rise in neutral pressure (interstitial water pressure) which would nullify the capillary forces, resulting in a significant reduction in mechanical strength. Under these conditions, the material faces a risk of localized instability or collapse. On the other hand, if the soil dries excessively, the capillary tension crucial for maintaining apparent cohesion dissipates and this reversion to a granular state can again lead to potential instability.

The observations regarding the pulpitum extend to all excavated areas of the Theatre made by loose pyroclastic material; in contrast, sections in cemented rocky material do not exhibit these issues. The substantial dimensions of the pulpitum vault (13 m × 10 m in plan and up to 7 m high) introduce significantly more severe stress conditions. This configuration inherently increases the probability of collapse if apparent cohesion is lost, especially when compared to tunnels with smaller cross-sections and vaults (even if a lowered arch) of 1.00 m to 1.50 m span.

3.4. Possible Mitigation Solutions Under Discussion

The unique characteristics of the Herculaneum Theatre preclude the application of conventional radon mitigation strategies. The sole viable solution is the reopening of the Enzechetta well (Figure 2). Given its cross-section and height, this well would function as a natural ventilation shaft, effectively reducing radon concentrations in the deeper sections of the monument where the highest values were recorded (Table 7). In 2024, inspections by the Fire Department, utilizing geophones and proximity sensors, confirmed the feasibility of reopening the well. Its apex was identified within a dwelling on Corso Resina. However, before proceeding, a thorough study is necessary. This study must quantify the benefits of reopening the well in terms of radon reduction and assess the risk that a potential decrease in relative humidity could disrupt the hygrometric balance between the walls and the air, potentially leading to stability issues. In essence, a delicate balance exists between reducing radon concentrations and preserving the monument’s structural integrity.

3.5. The Fruition of the Theatre Today

The implemented radon risk assessment strategies have made the Herculaneum Theatre accessible while simultaneously necessitating restoration interventions. These interventions extend beyond mere asset maintenance, encompassing aspects related to the Theatre’s fruition, which aligns with the overarching objective of its preservation.

From the eighteenth century to the present, the concept of restoration has evolved. It is no longer understood as an activity solely focused on “restoring efficiency” but rather as a comprehensive set of actions aimed at the conservation of pre-existing structures to which the community has attributed significant cultural value. In Western culture, restoration is not about “remaking”. Instead, it aligns with the Latin concept of “restauratio est renovata creatio” [44], which means preserving the unique and unrepeatable authenticity of something created in its time. This philosophy echoes Giambattista Vico’s thought. Specifically, Italian legislation explicitly identifies protection and valorisation as fundamental to preserving national memory and promoting cultural development, with both activities ultimately aimed at public enjoyment.

To improve visitor access and safety, the Archaeological Park of Herculaneum has introduced specific interventions for the Theatre’s new route. These measures adhere to plant design rules and restoration guidelines, ensuring reversibility, minimal intervention, clear recognizability, an interdisciplinary approach, and the use of compatible materials. This is especially important given the Theatre’s underground location, which holds significant geological, speleological, and historical–archaeological value.

3.5.1. The Restoration

Following a later intervention, the restoration project addressed the collapsing plasterwork in the 18th- and 19th-century rooms. This was achieved by injecting NHL 3.5 hydraulic lime mortar to re-establish cohesion with the wall support, supplemented by ethyl silicate impregnation in areas like entrance gaps and under the vault. The team planned similar work for the Roman-era plasterwork on the lower floor. For the rooms opening onto Via Mare, plasterwork gaps were repaired using NHL 3.5 hydraulic lime mortar, pozzolana with river sand, and lime putty.

3.5.2. The Lighting System

A new lighting system, installed in 2018, now illuminates the Theatre. Designed to comply with regulations and respect the site’s historical significance [45,46,47,48,49], it specifically aims to recreate the atmospheric charm of 18th and 19th-century Grand Tour explorations. This “emotional” experience is achieved through warm, low-intensity, dimmable LED lighting, reminiscent of torchlight, which also provides significant energy savings. The light strategically accentuates the Roman-era decorations and the unique, centuries-old karst phenomena. The new system highlighted the Theatre’s plaster graffiti (Figure 10), a compelling testament to its enduring history.

These inscriptions range from the Roman era through the 18th and 19th centuries, extending to markings made when the building functioned as a World War II air-raid shelter. Many are located in the vast vaulted ambulatory connecting the summa cavea and media caves, where various signatures and writings in languages like Italian, French, English, and Russian are visible amidst the initial excavators’ pickaxe marks.

Tunnels now feature a wide-beam light cone that accentuates their height and width, while above-ground areas utilize diffused light. The marble fragments within the Theatre are illuminated with accent or grazing light to best showcase them. The previous lighting system, dating from the war period, remains in place. It stands as a historical record of the building’s diverse uses and a stark reminder of a difficult period for the people of Resina, the city’s name until 1969 when it became Ercolano.

The pipes and electrical ducts remain visible and above ground, ensuring full respect for all existing historical layers, from the Roman underground elements to the 18th and 19th-century entrance areas. For improved safety and to complement the new lighting, visitors proceed in predefined groups with Park staff. They are required to wear raincoats and helmets with headlamps provided by the Park and are advised to come with appropriate footwear and clothing.

At present, a disclaimer containing a clear specification of all risks associated with the activities carried out within the archaeological park, is mandatory solely for external technicians and scholars operating in the Herculaneum site. With regard to visitors, they are duly informed of all risks connected with access to the site—including radon exposure—by means of dedicated signage positioned throughout the park. The Herculaneum Archaeological Park staff is also developing an appropriate all-inclusive liability waiver for visitors, designed to be easily administered.

3.5.3. Other Interventions

Beyond the physical route, visitor safety was a high priority. Among the other measures implemented to improve safety along the route was the enhancement of telephone communication with the outside, signage was designed to inform visitors about potential difficulties and the safety precautions in place. Crucially, a defibrillator was installed for first aid emergencies, with staff specially trained by the Italian Red Cross to use it.

4. Limitations

The proposed strategy necessarily reflects the practical constraints of an entirely underground cultural heritage site. Key limitations include the heterogeneity of monitoring evidence across campaigns (in terms of duration and spatial coverage), the difficulty of fully characterising air-exchange and ventilation conditions under real operating constraints, and unavoidable simplifications related to complex underground geometry and spatial variability. Moreover, mitigation and operational choices are constrained by conservation requirements, which limit intrusive interventions and require careful trade-offs between risk reduction and preservation needs. Future work should therefore prioritise extended multi-season monitoring to consolidate representative annual exposure conditions, targeted ventilation characterisation under comparable winter and summer boundary conditions, and a systematic evaluation of conservation-compatible, non-invasive operational measures to further strengthen transferability to other underground heritage sites.

5. Conclusions

The investigation presented herein offers a novel strategy specifically designed to evaluate radon risk and improve fruition in the unique setting of the Roman Theatre of Herculaneum. This site, which holds a rich history and many meanings which characterized its use over the centuries, achieved the significant milestone of reopening in 2019 after 20 years of being closed to the public, driven by the determination and extensive research of the Archaeological Park. The methodology developed was crucial in providing the radon risk analysis and other safety measures necessary to offer immersive archaeological and historical experience.

The approach was inspired by the four-level SOBANE methodology, which is ideally suited to this complex challenge. This tiered methodology perfectly fits to the evolution of the microclimatic conditions of the archaeological site, transitioning from closed-off stability to the dynamic environment created by public access. By providing a scalable framework, this strategy allows for the formulation of the best intervention solutions tailored to the site’s shifting conditions—a critical factor given the exciting new scenarios rapidly emerging for the monument’s future and its visitor experience [50].

The effectiveness of this strategy in the Roman Theatre of Herculaneum—which is characterized as a sole theatre of the Roman age, totally underground—demonstrates its robust potential. Its success is rooted in a multidisciplinary approach that skilfully integrates the expertise of archaeologists, architects, physicians, and engineers. This holistic, collaborative framework is a key strength that allows the methodology to be adopted by other underground sites characterized by fewer geological and architectural constraints. The ultimate goal remains to evoke, stimulate, and dramatically increase participation, while simultaneously multiplying interpretations, broadening readings, and profoundly deepening knowledge, a challenge where enabling technologies must always be firmly rooted in the robust methodological foundations and foundational cultural matrices we’ve briefly outlined here for scientific discussion.