Integrated Assessment of Indoor Air Quality, Fungal Contamination and Visitor Perception in Museum Environments

Abstract

The indoor microclimate of museums plays an essential role in preserving priceless cultural heritage for future generations and in ensuring visitors’ comfort and health. In this context, the present study aimed to evaluate indoor air quality, the degree of fungal contamination, and visitors’ perceptions in a museum environment through an integrated, interdependent approach. Measurements of the physicochemical parameters of air quality (temperature, relative humidity, CO₂, TVOC, HCHO, PM_2.5 and PM₁₀, negative and positive ions and brightness) were carried out in three exhibition halls within a museum in Oradea, Romania, during the period January–August 2024. Fungal contamination was assessed using surface and air samples, with classical isolation and microscopic identification methods. Visitors’ perceptions were analysed using a standardised questionnaire that focused on perceived comfort and visit duration. The results showed that the parameters defining indoor air quality generally fell within the limits set by the international standards in force, with occasional exceedances. These conditions are associated with the presence of fungi of the genera Cladosporium, Penicillium, and Aspergillus in the air and on museum exhibits, which pose risks to human health and the deterioration of the exhibited materials. The statistical decision-making model determined the critical thresholds above which visitor behaviour changed visibly. The results highlighted the importance of maintaining a stable microclimate in museum spaces, not only for the protection of exhibits, but also for optimising the cultural experience. Indoor air quality indicators and fungal microflora can only affect vulnerable people or those with pre-existing conditions. Occasional visitors do not present a significant risk of developing new conditions, considering the limited duration of exposure.

1. Introduction

Ensuring adequate indoor air quality (IAQ) in museum spaces is essential for heritage conservation, health protection, and the provision of a quality recreational experience. This involves the simultaneous control of air parameters, such as temperature (T), relative humidity (RH), carbon dioxide (CO₂), particulate matter (PM), total volatile organic compounds (TVOC, etc.) and bacterial and fungal microflora, determinant factors both individually and through their cumulative action on exhibited objects and visitors [1,2,3,4,5,6]. Furthermore, specific IAQ microclimatic conditions can generate the proliferation of fungi and bacteria, which increases both the biodeterioration processes of materials and the health risks for museum visitors and employees [7,8,9]. In this context, research on the indoor microclimate of heritage buildings has advanced significantly in recent years, highlighting the importance of maintaining stable environmental conditions for the conservation of cultural heritage housed in museums [10,11,12,13,14].

Another influential factor in indoor microclimate dynamics is the presence of visitors. Since the 1990s, it has been shown that they contribute to increases in CO₂ levels, the resuspension of PM, and RH variations [15], and more recent studies have confirmed their potential negative impact on exhibits [16,17]. Thus, visitors are perceived as beneficiaries of the heritage and as sources of “pollution” [18]. However, museums exist primarily for the public, which makes it essential to maintain a clean and healthy indoor environment for visitors, not only to protect the heritage. For this reason, recent research has begun to pay increased attention to visitors’ perceptions of air quality, showing that IAQ parameters directly influence perceived comfort, well-being and even the duration of the visit [19,20,21,22,23,24,25]. The stability of microclimate parameters, therefore, becomes essential for preventing physical and chemical degradation of materials, which explains why rigorous assessment of environmental conditions is increasingly carried out through specific performance indices [26,27,28,29], and long-term monitoring and identification of critical risk periods have become priorities in contemporary research [30,31]. In this sense, heating, ventilation and air conditioning (HVAC) systems are considered essential for achieving this objective [1,11,12], but the specialised literature also highlights secondary effects, such as fluctuations in T and RH, air currents, and the accumulation of pollutants in indoor spaces [1].

A distinct and increasingly visible area of contemporary research is the microbiological contamination of museums, focusing on the impacts of fungi and bacteria on human health and cultural objects. Fluctuations in hygrothermal conditions, especially RH and T, favour the proliferation of microorganisms, leading to accelerated biodeterioration of materials [7,22,32,33]. Microbial load affects the exhibited heritage and poses risks comparable to those of Sick Building Syndrome (SBS), with direct implications for visitors and museum staff [7]. Assessing these risks usually involves aerobiological analysis, microbiological sampling, and correlating fungal and bacterial loads with microclimate parameters [7,34]. The solutions proposed in various studies include air purification and controlled ventilation [17,35], as well as innovative methods, such as natural treatments based on essential oils and nanoparticles, to reduce biological contamination without chemical impact on exhibits [36,37,38]. At the same time, recent research highlights the role of advanced molecular techniques in profiling microbial communities in built environments, allowing for more precise identification of biodeteriogenic microorganisms and underpinning less invasive preventive interventions [39].

Despite the growing body of research on indoor air quality, microclimatic dynamics, and microbiological contamination in museum environments, most studies address these components separately, focusing on environmental monitoring, microbial characterisation, or visitor perception. Only limited research has adopted an integrated approach that simultaneously analyses microclimatic parameters, fungal contamination (both airborne and surface-associated), and their potential implications for both heritage conservation and visitor experience.

Building on this identified research gap, the hypothesis of this study was that IAQ parameters can influence the development of fungal microflora and that their combined action can affect both the well-being of visitors and the material integrity of the exhibits. Starting from this premise, the study aimed to highlight the extent to which IAQ can be considered a risk factor for the museum environment, whether the monitored parameters favor the growth and diversification of fungi, as well as whether the fungal genera identified in the air and on the surfaces of the exhibits present pathogenic potential for visitors or can contribute to the biodeterioration processes of the heritage exhibited in the largest museum in Oradea Municipality, Romania. At the same time, the approach also included an analysis of visitors’ perceptions of air quality in the monitored rooms, to obtain an integrated perspective on the interaction among the built environment, microbial communities, and the public’s experience in the analysed museum space.

2. Materials and Methods

2.1. Study Object

Crişurilor County Museum is located in Oradea Municipality on Armatei Române Street no. 1A and is one of Romania’s largest museums (Figure 1). Founded in 1971, the museum today holds a heritage of approximately 450,000 archaeological, historical, ethnographic and art pieces, illustrating the cultural diversity specific to the western region of Romania, marked by an ethnic and cultural mosaic. Since 2006, the museum has been operating in the building of the former Cadet School, a historical monument built in 1897 to serve military education.

The ethnographic collection of the Crişurilor County Museum was selected for the present study. The choice is justified by the high sensitivity of the exhibited objects (textiles, furniture and artefacts made of organic materials) to variations in T, RH, other pollutants and microbiological contamination, factors that can accelerate degradation processes. At the same time, this collection is among the most visited by the public in this museum, making understanding visitors’ perceptions of microclimate conditions an essential objective of the research.

Within this museum collection, three exhibition halls were monitored. All three spaces share common characteristics: they are dedicated to the exhibition of traditional clothing and ethnographic furniture and are integrated into the same visitor route, ensuring a comparable frequency of visitors. At the same time, the interconnection of the halls favours an active exchange of air between the spaces, which results in uneven air circulation and, implicitly, a redistribution of PM, pollutants, and microbiological load. Exhibition hall 1 has an area of 40.95 m² and a total air volume of 192.47 m³, housing exhibits such as traditional clothing, household textiles and traditional furniture. Exhibition hall 2 has an area of 37.76 m² and an air volume of 177.47 m³, mainly dedicated to presenting traditional clothing. Exhibition Hall 3, with an area of 37.17 m² and a total volume of 174.7 m³, houses a collection of household textiles, traditional furniture, and household objects, made mainly of wood (

Table 1. Spatial characteristics of the monitored exhibition halls and types of exhibits.

Hall	Length (m)	Width (m)	Height (m)	Area (m2)	Air Volume (m3)	Exhibits
Exhibition hall 1	6.5	6.3	4.7	40.95	192.47	traditional clothing, household textiles, traditional furniture
Exhibition hall 2	6.4	5.9	4.7	37.76	177.47	traditional clothing
Exhibition hall 3	6.3	5.9	4.7	37.17	174.7	household textiles, traditional furniture, wooden household items

). The analysed rooms lack windows with direct exposure to the outdoor environment, and the indoor microclimate is fully controlled through mechanical ventilation systems. The exhibition spaces are equipped with HVAC units: three systems in Exhibition Halls 2 and 3, and two in Exhibition Hall 1. These systems provide both heating and cooling, ensuring precise control over indoor environmental conditions and contributing to the overall stability of the monitored microclimate.

Ilies et al. [40] also studied the ethnography section of the Crişurilor County Museum, but the aim focused mainly on evaluating the interaction between microclimate parameters and the conservation processes of museum heritage. They focused particularly on the statistical dimension, using IAQ data integrated into impact indices for visitors’ and employees’ health and the conservation status of exhibits. Unlike that research, the present approach extends the analysis by including the microbiological dimension and visitors’ perceptions of IAQ.

2.2. Determination of IAQ and Fungal Contamination Indicator Values

In this paper, an integrated approach was pursued, aiming to analyse the leading indicators of IAQ, fungal microflora, and visitors’ perceptions of microclimate conditions in museum spaces (Figure 2). These three components are deeply interconnected, each exerting a direct influence on the others. IAQ parameters condition both the conservation processes of the exhibited heritage and the development and diversity of fungal microflora. At the same time, IAQ and the presence of biological agents significantly affect visitors’ health, comfort and perceptions.

For indoor air monitoring, data was obtained between January and August 2024, for a set of ten indicators: T, RH, CO₂, positive ions (I⁺), negative ions (I⁻), formaldehyde (HCHO), TVOC, suspended particles with a diameter of ≤2.5 µm (PM_2.5), particles with a diameter of ≤10 µm (PM₁₀) and artificial illumination level (AL).

Data collection was carried out using datalogger devices for the indicators T, RH, CO₂, HCHO, TVOC, PM_2.5, and PM₁₀, set to record and store data at hourly intervals, and placed at strategic points in the three monitored rooms. Regarding the indicators I⁺, I⁻ and AL, in the absence of datalogger devices, the data were collected manually three times a day (in the morning, at midday, and in the evening) from different points of the exhibition spaces with the help of dedicated detectors.

To monitor T and RH, recognised as essential indicators for heritage conservation and for the development of bacterial and fungal microflora, twenty-two KlimaloggPro thermo-hygrometers (TFA, Ottersberg, Germany) were used in conjunction with HOBO U23 Pro v2 datalogger sensors (Onset Computer Corporation, Bourne, MA, USA). CO₂, HCHO and TVOC concentrations (including the additional parameters T and RH) were determined using three Evikontroll Gas detection and control system devices (Evikontroll Gas, Tartu, Estonia). PM_2.5 and PM₁₀ were recorded with three DeltaOHM HD50PM devices (Senseca Italy Srl, Padova, Italy). For the indicators I⁺ and I⁻, measurements were made using the NKMH-103 device (Ion Trading, Tokyo, Japan), and for AL, the Extech SDL400 (Extech Instruments, Nashua, NH, USA) was used (Figure 3). The values of these parameters were obtained from a dense spatial distribution of collection points, and the resulting data were averaged to yield values as representative as possible for each indicator [40].

All this equipment was arranged in a dense network of monitoring points, configured to ensure the most uniform coverage of the exhibition spaces. This approach allows the average characterisation of the microclimate and the identification of minor fluctuations in the parameters, generated by the presence of visitors, local pollutant emissions or the peculiarities of air circulation between interconnected rooms. Thus, the implemented monitoring system offers high spatial and temporal resolution, essential for the precise assessment of risks to heritage and human health.

After the monitoring period ended, all data were analysed against the international standards in force for assessing IAQ, with consideration of both human health protection and the conservation of the exhibited heritage. In addition, the monitored indicators were interpreted in relation to their influence on the growth and development of fungal microflora, the well-being of visitors, and their perception in the exhibition spaces.

Regarding microbiological contamination, six air samples and six surface samples (two per room) were collected in July 2024, under consistent indoor conditions (Figure 3). The sampling date was selected during the warm season, when fungal growth potential is generally higher, and all measurements were conducted on the same day to ensure comparability and minimise temporal variability.

The Koch sedimentation method was used to quantify fungal contamination in the air [41]. The selected sampling points correspond to exhibit locations where only personnel can change the exhibits, or to locations near walking passages where visitors have access. Three Petri plates were set up and left open for half an hour at each sampling point. The Petri plates contained Sabouraud agar medium with 40 g/L glucose, 10 g/L peptone, 15 g/L agar, and 50 µg/mL chloramphenicol, adjusted to pH 5.6 [42]. The plates used in the Koch method were incubated at 20 °C for 10 days. With Omelianski’s formula, the fungal colony-forming units (CFUs) in the air were estimated:

$$CFU/{\mathrm{m}}^3={\displaystyle \frac{n\times \mathrm{10,000}}{S\times k}}$$

where n represents the number of colonies on the plate, S is the surface area of the Petri dish (∅90 mm), and k is the air exposure time coefficient: k = 1 for 5 min, k = 2 for 10 min, k = 3 for 15 min, and so on [43].

The degree of contamination on the surfaces was assessed using sterile swabs. The swabbed surfaces included the textile materials of the adjacent clothing present on the exposed mannequins (Figure 4). Swabs were rinsed with 1 mL of saline solution and vigorously vortexed before 150 µL of the resuspended cells was plated on Sabouraud agar. For each of the six swabs, the resuspended cells were plated onto three plates, which were then incubated for 10 days at 20 °C.

Figure 3 and

Table 2. Description of the samples collected from the inside of the three exhibition halls of the analysed museum.

Sample Code	Collection Point	Height Above Ground	Sampling
AS1	Traditional bed with covers	~1.5 m	Koch method
AS2	Floor lever	~0 m	Koch method
AS3	Traditional bed with covers	~1.5 m	Koch method
AS4	Floor lever	~0 m	Koch method
AS5	Floor lever	~0 m	Koch method
AS6	Floor lever	~0 m	Koch method
SS1	Traditional bed cover	1.5 m	Swab
SS2	Traditional blouse	1.5 m	Swab
SS3	Covers on the traditional bed	1.5 m	Swab
SS4	Sleeve of the traditional fur coat	1.5 m	Swab
SS5	Traditional blouse	1.5 m	Swab
SS6	Traditional blouse	1.5 m	Swab

indicate the sampling locations for air and surface samples, their placement within the museum exhibit, the elevation above the ground, and the sampling technique.

After being parafilm-sealed, all samples were transferred and further examined using standardised practices for handling microbiological samples [44] at the Molecular Biology Centre of the Institute of Interdisciplinary Research on Bio-Nano-Sciences of Babeş-Bolyai University in Cluj-Napoca, Romania. Following the manufacturer’s instructions, DNA was extracted from each independent colony cultured on Sabouraud agar plates for ten days using the Animal and Fungi DNA Preparation Kit^® (Jena Bioscience, Jena, Germany). The polymerase chain reaction (PCR) amplification reactions took place in a total volume of 25 µL, with 1× MyTaq Reaction Buffer (Meridian Bioscience^®, London, UK), 0.5 mM of each primer (Macrogen Inc., Seoul, Republic of Korea), 1.25 U of MyTaq Red DNA Polymerase (Meridian Bioscience^®, London, UK) and 50–100 ng of template DNA. The applied PCR program consisted of an initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s, hybridisation at 56 °C for 30 s, and elongation at 72 °C for 30 s, with a final elongation step at 72 °C for 5 min. The internal transcribed spacer (ITS) region was used for fungal identification with the primers ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′). The DNA fragments generated by PCR were purified from agarose gels using the NucleoSpin™ Gel and PCR Clean-up kit (Macherey-Nagel, Ping-Tung, Taiwan), and sequencing was performed by the commercial Sanger sequencing service at Macrogen (MacrogenEurope, Amsterdam, The Netherlands). The DNA sequences were submitted to GenBank under the accession numbers PV916258-PV916276.

2.3. The Determination of the Visitors’ Perceptions

In the study, to complement the information obtained through microclimatic and microbiological monitoring, a standardised questionnaire was administered to museum visitors throughout the monitoring period (Supplementary Materials). The instrument included 29 items and was designed to capture, in an integrated manner, visitors’ medical conditions, their perception of IAQ, and basic socio-demographic information. The questionnaire comprised a combination of closed-ended questions (including dichotomous yes/no items and multiple-choice formats) and a limited number of open-ended questions, allowing both structured quantitative analysis and the capture of subjective perceptions.

The questionnaire was applied to a sample of 94 visitors throughout the entire monitoring period (January–August 2024). The sample was selected through a convenience sampling methodology, which involved randomly contacting people in the exhibition spaces and asking about their willingness to participate in the research. At the same time, a balanced distribution of respondents by gender and age group was sought to ensure a minimum level of representativeness of the museum’s visiting public. The final sample included 41 male and 53 female respondents, with a mean age of 39.5 years, a median of 38 years, and an age range between 16 and 75 years. No significant seasonal differences were observed in the distribution or completion of the questionnaires across the monitoring period.

The health-related questions targeted symptoms such as frequent cough, nasal congestion, headaches, nausea, eye irritation, or severe fatigue, as well as aspects of chronic conditions, allergies, medications, smoking, or contact lens use, to identify possible interactions between indoor exposure and individual vulnerability. The component dedicated to the perception of air quality included items regarding the level of comfort felt in relation to microclimatic parameters (high or low T, excessively high or low RH values, unpleasant odours, dust or stagnant air) and the general assessment of IAQ, allowing the correlation of objective monitoring data with the subjective perceptions of visitors. The final section focused on personal and demographic information (gender, age) necessary for comparative analyses and statistical interpretations. The dependent variable was the time spent in the museum, while the independent variables included both medical factors and perceptions of the indoor microclimate.

During the study, a series of statistical methods were applied to test the formulated hypothesis, but the most appropriate was ordinary multiple regression. This statistical approach is based on a theoretical model of multiple linear regression, which allows quantification of the relationships between a dependent variable (Y) and a set of independent variables (X₁, X₂, …, X₂₀), each with its own weight, represented by the coefficient β_i. The model used can be formalized as follows:

$$Y=\alpha +{\displaystyle \sum _{i=1}^{20}}{\beta }_i{X}_i+u$$

where $Y$—dependent variable, ${\displaystyle {\sum }_{i=1}^{20}}{\beta }_i{X}_i$—corresponding variables and their coefficients, and u—Model’s residual.

The applicability of this model in the present work lies in its ability to simultaneously analyse the influence of several IAQ factors on a synthesis indicator. Through this approach, it is possible to evaluate not only the individual impact of each parameter, but also their cumulative effects and potential interactions, thus providing a much more complex and rigorous perspective on the determinants and visitors’ perceptions.

To predict the time spent in the museum, we also used Decision Tree modelling. In particular, we classified the dependent variable (time spent in the museum) into two classes: class 0—less than 4 h; 1—more than 4 h. The Gini index, based on impurity, was employed to classify nodes. Gini impurity quantifies the likelihood that a randomly selected element from a set would be misclassified if assigned a label according to the set’s label distribution. It achieves its lowest value (zero) when all instances within a node belong to the same target category. The Gini index is expressed as follows [45]:

$$Gini=1-{\displaystyle \sum _{i=1}^{i=n}}(p_i^2)$$

where p_i is the proportion of class i in the node and n is the number of nodes.

To formulate the accuracy, precision, and recall used in the reporting of decision trees, first of all, it is crucial to identify the following:

• True Positives (TP) → Correctly predicted positive cases.
• True Negatives (TN) → Correctly predicted negative cases.
• False Positives (FP) → Incorrectly predicted positive cases (Type I error).
• False Negatives (FN) → Incorrectly predicted negative cases (Type II error).

Accuracy measures the overall correctness of the model and can be formulated as follows:

$$Accuracy= {\displaystyle \frac{TP+FN}{TP+TN+FP+FN}}$$

Precision is used to measure how many predicted positives are correct and is formulated as follows:

$$Precision= {\displaystyle \frac{TP}{TP+FP}}$$

Recall measures how well the model finds all actual positives. Its formula is as follows:

$$Recall= {\displaystyle \frac{TP}{TP+FN}}$$

By using the model’s performance indicators, we aim to assess how accurately and completely the questionnaire responses explain and predict the time visitors spend in the museum. This gives us an objective picture of how perceptions of air quality and individual factors influence their behaviour. In this way, we can identify the questionnaire variables that have the strongest impact on visit duration, as well as the microclimate aspects perceived as limiting the visitor experience.

3. Results

3.1. The Analysis of the IAQ Indicator Values

T in the three monitored rooms of the museum presented an average value of 19.7 °C, with a standard deviation of 1.4 °C, varying between a minimum of 17.3 °C and a maximum of 23.8 °C, which suggests a general thermal stability, but with point fluctuations relevant for the comfort of visitors. In parallel, RH recorded an average of 62.4%, with a standard deviation of 7%, with extreme values between 40% and 75% (Figure 5), a range that reflects significant and potentially damaging variations for sensitive materials, especially textiles and objects on organic supports, where maintaining stable conditions is essential to prevent degradation processes [40]. The peaks exceeding 70% RH were identified throughout the entire monitoring period (January–August 2024), occurring across multiple months rather than being confined to a specific season. Therefore, these elevated values cannot be attributed exclusively to summer conditions or particular rainy periods. Instead, they reflect short-term fluctuations in indoor microclimate conditions, likely influenced by internal factors such as ventilation patterns, visitor presence, and HVAC system dynamics, rather than clear seasonal variability.

These values are to a certain extend within the limits recommended by ASHRAE standards for the conservation of exhibits (average T of approximately 20 °C ± 1–2 °C and ideal RH of 45–60%) [1,46] and correspond to the thermal comfort requirements of people, where the optimal recommended T range is between 20 °C and 24 °C, with RH between 30% and 60% [47]. This is particularly evident for T, where only 11.9% of the measurements exceed the recommended range. In contrast, RH more frequently falls outside the prescribed limits (78.8% of the values), although these deviations are generally minor and remain close to the acceptable range. Regarding the possibility of active fungal germination under the measured conditions, this aspect is particularly relevant, as fungal proliferation is known to intensify at RH values between 60% and 70%. Temperature also plays a key role in fungal development, with most indoor fungi showing optimal growth in the range of 20–30 °C, indicating that conditions near the lower limit of this favourable interval are common [48,49,50].

The determinations carried out on the chemical and particulate parameters revealed for TVOC an average value of 0.55 mg/m³, with a standard deviation of 0.06 mg/m³ and limits between 0.45 and 0.69 mg/m³, while for HCHO an average of 0.07 mg/m³ was obtained, the standard deviation being 0.04 mg/m³, and the variation in the values falling between 0 and 0.29 mg/m³. CO₂ concentrations had an average of 506 ppm and a standard deviation of 70 ppm, with extreme values of 441 and 1314 ppm. For PM_2.5, a mean of 15.5 µg/m³, a standard deviation of 5.1 µg/m³ and a range of 8 to 30 µg/m³ were recorded, while PM₁₀ had a mean value of 20.7 µg/m³, with a standard deviation of 6.3 µg/m³, with minimum and maximum limits of 10 and 38 µg/m³, respectively [40] (Figure 5).

The values determined are mainly within the limits set by international standards, but some exceedances may raise concerns with prolonged exposure. CO₂ concentrations were below the safety threshold of 1000 ppm [51], while the average for HCHO (0.07 mg/m³) exceeded the recommended multi-annual value of 0.04 mg/m³ [52]. The elevated HCHO levels are likely attributable to off-gassing from indoor materials such as wooden furniture, textiles, adhesives, and display cases, with cleaning products representing a potential secondary source. The average TVOC level (0.55 mg/m³) was below the WHO threshold of 1 mg/m³ [53], and the averages for PM_2.5 and PM₁₀ exceeded the multi-annual value of 12 µg/m³ but did not exceed the daily threshold of 35 µg/m³ [54]. High levels of CO₂ do not constitute a nutrient for fungi but indicate poor ventilation and moisture accumulation, which favour fungal development [3]. TVOCs can inhibit fungal growth at high concentrations, but they do not prevent it at low levels and may even originate from microbial processes [55]. PM can transport spores and provide a nutrient substrate, amplifying the risk of colonisation [56], with cumulative effects on visitors’ well-being and health and on heritage conservation.

The values determined for air ionisation indicate for I⁺ an average of 1400 ions/m³ (maximum 300 m³, minimum 3300 m³, standard deviation 503 m³) and for I⁻ an average of 1100 ions/m³ (maximum 200 m³, minimum 3000 m³, standard deviation 454 m³). Although there are no universally accepted international standards, the literature recommends that I⁺ be kept below 1000 ions/m³ and I⁻ above 1000 ions/m³ [57,58]; in this context, the measured values frequently exceed the limit for I⁺, and those for I⁻ are at the lower limit of the optimal range, suggesting insufficient negative ionisation. Regarding AL, the average of 71.3 lux (minimum 6 lux, maximum 314 lux, standard deviation 53.5 lux) indicates a heterogeneous distribution but is generally consistent with the 50–200 lux range recommended by the British Standards Institution [59] for heritage conservation [40]. However, extreme values can affect both the protection of exhibits and the visual comfort of visitors. According to the literature, ions can help reduce microbial load. However, efficiency depends on concentration and microclimate [60]. AL can inhibit some sensitive fungal species but mainly affects them indirectly by modifying T and RH [61].

3.2. Fungal Contamination Analysis

Using IAQ assessment criteria, the fungal CFU levels and contamination levels in the air samples were calculated (

Table 3. Fungal loads in indoor air. Fungal CFU/m³ values represent the average of the triplicate measurements and the corresponding SD.

Sample	Number of Fungal CFU/m3 of Air	Degree of Contamination
AS1	236.4 ± 0.57	Medium
AS2	1004.72 ± 1.52	High
AS3	52.11 ± 0.57	Low
AS4	236.4 ± 0.57	Medium
AS5	236.4 ± 2.3	Medium
AS6	59.11 ± 0.57	Low

) [62]. According to the standards, fungal loads were classified as high for CFU values between 500 and 1000 CFU/m³ in the case of AS2 sample, medium for CFU values between 100 and 500 CFU/m³ in the case of AS1, AS4 and AS5 samples and as low for CFU values between 50 and 100 CFU/m³ in the case of AS3 and AS6 sample.

A total of seventy-seven fungal isolates were examined and classified into seven different genera: Alternaria, Cytospora, Penicillium, Naganishia, Cladosporium, Aspergillus, and Nerospora, along with two uncultured fungi. The most predominant genera included Cladosporium (46.75%), Penicillium (14.28%), and Nerospora (14.28%). At the phylum level, the identified fungi belonged to Ascomycota (88.31%) and Basidiomycota (11.69%). Concerning the distribution of fungi in the studied environment, 59.75% of the isolated species were on the analyzed surfaces (SS1: 14.28%; SS2: 9.09%; SS3: 14.28%; SS4: 6.49%; SS5: 9.09%; and SS6: 6.49%); in contrast, the fungi retrieved from the air comprised 40.25% (5.19%—AS1; 22.07%—AS2; 12.98%—AS3; 5.19%—AS4; 5.19—AS5; and 1.29%—AS6) (Figure 6 and Figure 7).

The identified culturable fungi are similar to those described in previous studies on their presence in museums [2,39]. Various elements, including the influx of outside air (suggested by the presence of the frequent plant pathogen species), human presence and actions leading to resuspension of PM due to visitors walking, bioaerosol release from human actions such as speaking or coughing, indoor microclimate factors, the materials used in construction of the structure itself, along with the arrangement of the cultural heritage items can influence the levels of airborne fungi within museums. As fungi are more frequently present on the exhibited items, curators can apply cleaning strategies to decrease the fungal load [63].

Various fungal genera in indoor air, such as Alternaria, Aspergillus, Penicillium and Cladosporium, are recognised as etiological agents of multiple diseases, from mild respiratory manifestations to severe systemic diseases. The analysed figure highlights the distribution of these fungi across the main categories of human pathologies: Alternaria is consistently implicated in allergies, hypersensitivity syndromes, and respiratory symptoms, whereas Aspergillus is significantly associated with respiratory, systemic, and neuropsychiatric infections. Penicillium and Cladosporium are frequently involved in respiratory and skin infections, as well as in rheumatological diseases and hypersensitivity reactions (Figure 8).

3.3. Analysis of Visitors’ Perceptions of IAQ

The analysis of self-reported health-related symptoms indicates a generally low prevalence among respondents. Most reported conditions were observed in a small proportion of participants, including repeated sneezing (12.5%), dizziness or fainting (11.8%), and eye irritation (10.5%), while other symptoms such as headaches (7.7%), nausea (7.1%), frequent cough (6.7%), and dry throat (6.3%) were reported even less frequently. The remaining conditions, including nasal congestion (5.0%), skin irritation (5.0%), increased fatigue (4.2%), frequent colds (3.3%), and wrist or bone pain (1.3%), were infrequent.

Overall, these results suggest that most visitors did not experience significant adverse health effects during their time in the museum. The relatively low frequency of reported symptoms is consistent with the short duration of exposure and the generally acceptable indoor environmental conditions. However, the presence of such symptoms, even at low levels, may be relevant for sensitive or vulnerable individuals, supporting the importance of maintaining optimal indoor air quality in museum environments.

The analysis of the distribution of the variables included in the study highlights notable differences in their manifestation. The dependent variable, time spent in the museum, is approximately one hour, indicating a dominant behavioural pattern among most visitors. However, there are also extreme responses, with a small number of participants reporting visits lasting four hours or more, suggesting distinct visitor typologies.

Regarding perceptions of environmental conditions, responses on air temperature, relative humidity, and dry air indicate that a large majority of respondents (78.1%) did not experience extreme conditions, reflecting a generally positive evaluation of the indoor microclimate. At the same time, responses concerning personal behaviours, such as smoking, show that only a small proportion of participants (25%) fall into this category, thereby limiting the potential influence of this factor on the dependent variable (Figure 9).

A crucial preliminary step in implementing OLS regression is analysing the correlation matrix, which allows one to identify relationships among the independent variables. The results indicate that some variables are strongly correlated, a situation that warrants careful consideration to ensure the robustness of the regression model (Figure 10).

Because of strong correlations among some variables, we excluded 8 independent variables from the OLS regression and conducted the analysis in IBM SPSS 30 (IBM, Armonk, NY, USA).

Table 4. Summary of the regression model parameters.

Model	R	R-Squared	Adjusted R Squared	Standard Error
1	0.750	0.563	0.498	1.0847

presents statistics on the model’s R-squared values.

The R-squared value is 0.56, indicating that the independent variables explain 56% of the variation in the dependent variable. Also, the F-value indicates that the variables have a significant effect on the dependent variable (

Table 5. Analysis of variance (ANOVA) results for the regression model indicate the overall statistical significance of the predictors and their joint explanatory contribution to the dependent variable’s variance.

Model		Sum of Squares	Df	Mean Square	F	Significance
1	Regression	122.806	12	10,234	8.698	0.000
	Residual	95.303	81	1.177
	Total	218.109	93

).

Table 6. The estimated coefficients of the OLS regression model and their statistical significance in explaining the time spent in the museum.

Model		Non-Standard Coefficients		Standardized Coefficients	t	Significance
		B	Standard Error	Beta
1	Intercept	−3.954	1.300		−3.042	0.003
	air_temp_high	0.128	0.286	0.049	0.449	0.655
	air_temp_low	−0.056	0.322	−0.014	−0.174	0.862
	humid_air	5.100	0.785	0.589	6.494	0.000
	unpleasant_odor	−0.646	0.377	−0.171	−1.711	0.091
	dust	0.603	0.638	0.119	0.945	0.347
	air_quality	−0.171	0.311	−0.054	−0.550	0.584
	smoking	−0.342	0.603	−0.045	−0.566	0.573
	cont_lense	3.237	0.786	0.374	4.121	0.000
	follow_med	1.610	1.203	0.213	1.338	0.185
	health_prob	−1.069	1.378	−0.123	−0.776	0.440
	gender	0.079	0.248	0.026	0.321	0.749
	age	0.018	0.007	0.194	2.490	0.015

shows the coefficients of the OLS regression. The significance column clearly shows that humidity, contact lens use, and age affect the time spent in museums.

The regression coefficients indicate that perceived humid air (B = 5.1), wearing contact lenses (B = 3.237), and age (B = 0.018) are positively associated with the time spent in the museum, while the presence of unpleasant odour (B = −0.646) is negatively associated with visit duration. These coefficients represent the estimated effects within the regression model and should be interpreted as statistical relationships rather than direct or absolute changes in time spent. Mathematically, we can form the above-given relationship as follows:

$$Y=5.1X_4-{0.646X}_6+{3.24X}_{16}+0.018X_{20}-3.95$$

In other words, the results of the OLS regression indicate positive associations between perceived humid air, contact lens use, and age with time spent in the museum, whereas the presence of an unpleasant odour is associated with a decrease in visit duration.

In addition to the regression approach, a Decision Tree model was applied to classify respondents as having spent less than or more than 4 h in the museum. This method allowed for the identification of key discriminating variables and provided a complementary perspective on the factors that differentiate short-term from long-term visitors, thereby reinforcing the multidimensional character of the determinants of museum attendance.

The Decision Tree Classification Model predicts whether the time spent is above or below the median (4 h) based on input features. The model accuracy is 52.63%. The results show an imbalance between the two classes (Class 0—less than 4 h; Class 1—more than 4 h). The precision and recall for Class 0 (below median) are relatively higher (64% and 69%), but Class 1 is poorly predicted (20% precision, 17% recall), indicating the model struggles to identify instances where time spent is above the median.

Overall, the study indicates that relative humidity, contact lens use, and age are significantly associated with time spent in the museum, whereas the presence of an unpleasant odour is associated with a decrease in visit duration (Figure 11). However, these relationships should be interpreted with caution, as they reflect statistical associations rather than direct causal effects. In particular, relative humidity is not necessarily consciously perceived by visitors and may serve as a proxy for broader indoor environmental conditions. One possible explanation is that contact lens wearers may experience greater visual comfort compared to glasses users, particularly in exhibition environments where reflections, lighting variability, or peripheral vision may influence the viewing experience. This could facilitate a more immersive interaction with exhibits and, consequently, longer visit times.

4. Discussion

Overall, the IAQ in the analysed spaces remained within the limits set by current international standards, with no frequent or significant exceedances of the monitored parameters. The values recorded for T, RH, CO₂, TVOC and PM showed a moderate variation, without sudden or persistent oscillations, indicating significant imbalances in the microclimate. However, the values obtained also indicate potential risks to the conservation of heritage and to visitors’ health [40]. RH exceeded the 60% threshold allowed by the international standards in force in some situations, which may affect visitor comfort [14] and favour the development of Cladosporium and Penicillium fungi, known for their ability to damage organic materials and cause respiratory allergies [64,65]. This aspect is particularly relevant, as fungal proliferation is known to intensify at RH values between 60% and 70%, whereas the present study recorded an average RH of 62.4% [66]. T also plays a key role in fungal development, with most indoor fungi showing optimal growth in the range of 20–30 °C. In the present study, the average temperature was 19.7 °C [67], indicating conditions close to the lower limit of this favourable interval. PM_2.5 and PM₁₀, observed in high concentrations (although not exceeding the limit of 35 µg/m³), can act as vectors for fungal spores, facilitating the dispersion of the dominant genera identified in this study, such as Aspergillus, Cladosporium and Penicillium [68,69]; the same being indicated in the present study.

Usually, fungi are not a health threat to people, but they can cause several health problems. Most commonly, immunocompromised individuals or those with underlying medical conditions, such as diabetes, are more susceptible to fungal infections [70]. The identified fungal loads in this study (~300 CFU/m³) are consistent with values reported in similar museum environments, where airborne fungal concentrations typically range between approximately 100 and 800 CFU/m³ [2,5,23]. According to earlier studies, fungi are more common on the surfaces of museum exhibits than in the air because the objects provide a superior source of nutrients for fungal growth [71,72,73].

Alternaria, Penicillium, Cladosporium, and Aspergillus are frequently reported on a variety of cultural heritage materials (e.g., wood, textiles, and painting canvas) in museum exhibition and storage environments [74,75]. These genera are among the most commonly associated with museum settings, being detected both on exposed objects and in indoor air [2,39]. Among them, Cladosporium was identified as the dominant genus in the Crişurilor County Museum, with known associations to allergies, respiratory infections, and superficial skin conditions [76]. Penicillium can cause various diseases and is most frequently associated with respiratory tract infections [77]. Although extremely rare, Neurospora can cause human infections of the eyes or lungs [78,79]. Alternaria and Aspergillus can cause especially respiratory infections, allergies, and systemic infections [80]. Typically, Naganishia is not a human pathogen but can cause skin infections and, in immunocompromised patients, potentially fatal infections [81]. Cytospora and the uncultured fungi (identified by the Access Numbers EF504762.1 and AM711465.1 in the NCBI database) have never been described as human pathogens, as they are plant pathogens, plant endophytic fungi, or found in compost [82].

Given the limited duration of exposure (1–2 h on average) and the microclimate parameters and fungal contamination levels that are generally within international standards, it is unlikely that museum visitors will develop new conditions associated with these factors, in the absence of pre-existing conditions [40]. These aspects are also supported by the specialised literature highlighting the low risk for the general population under short-term exposure to moderate levels of bioaerosols [7,23].

Regarding visitor perception, it was observed that odours perceived as unpleasant, correlated with increased TVOC levels or high RH, led to a reduction in the duration of stay in certain rooms. This behaviour is attributable to sensory discomfort and the subconscious perception of an unhealthy environment, as supported by Wargocki et al. [83], who highlighted the negative impact of olfactory pollutants on the cognitive evaluation of indoor spaces. The applied linear regression model confirmed that factors such as RH, T, and odour perception are significant predictors of visit duration, in agreement with Frontczak and Wargocki [84] and Bessoudo et al. [85], who demonstrated that thermal discomfort and poor air quality reduce user satisfaction in public environments. The critical thresholds identified in the decision analysis correlate with a significant decrease in visit duration, a similar phenomenon reported by Tamer et al. [86] in studies on thermal comfort in museums. Similarly, Ilieș et al. [87] showed that a negative perception of indoor air affects the length of stay and the likelihood of recommending the cultural space to other visitors.

5. Conclusions

The microclimate parameters measured in the museum spaces were generally close to the recommended values for human comfort and heritage conservation. Still, some occasional exceedances of RH and HCHO were identified. These conditions can cause slight discomfort for visitors and favour the growth of fungal colonies, especially in poorly ventilated areas. High RH, combined with poor ventilation and the presence of PM, creates an environment conducive to spore accumulation and the establishment of fungal species with allergic or pathogenic potential. Also, ionic imbalance (I⁺ prevails in the museum, to the detriment of I⁻) can contribute to fatigue and concentration problems in sensitive individuals; however, this effect can be mitigated by using air ionisation devices that increase negative ion concentrations.

The analysis of fungal contamination in the investigated museum spaces revealed variable airborne loads, with high contamination levels in some areas. Of the 77 fungal isolates identified, the genera Cladosporium (46.75%), Penicillium (14.28%), Nerospora (14.28%) and Aspergillus proved to be predominant, these being recognized both for their biodeteriorative potential on organic materials (textiles, wood), and for their ability to induce allergic reactions or infections in vulnerable individuals, especially those with pre-existing conditions. These fungi were found in both air samples and on exhibit surfaces, with a higher frequency on exhibit surfaces. This distribution emphasises the multiple risks that fungal contamination can generate in the museum environment, potentially affecting the integrity of cultural heritage.

Statistical analysis of visitor perception showed that certain microclimate conditions and individual factors significantly influenced the duration of visitors’ stay in the museum. Slightly higher RH, older age, and contact lens wear were associated with longer visit times. Conversely, unpleasant odours reduced the duration of stay, indicating discomfort that may affect the cultural experience. The regression model explained a significant proportion of the variation in visitor behaviour, and the decision-making model confirmed that perceptions of RH and odours are essential for time spent inside the museum.

6. Study Limitations

The study has several limitations that should be considered in future research. The monitoring period, conducted between January and August 2024, did not capture the full seasonal cycle, and fungal identification relied on classical methods, which could be complemented by molecular techniques in future investigations. In addition, the relatively small sample size relative to the number of analysed variables poses a methodological constraint that may affect the stability and generalizability of the statistical models. However, appropriate steps were taken to reduce multicollinearity and to ensure model consistency, supporting the reliability of the identified relationships within the analysed dataset.

Furthermore, while the applied statistical models provide valuable insights into the relationships between variables, their performance could be further enhanced by expanding the sample size and incorporating additional variables related to visitor behaviour and environmental perception.