Thesis title: Evaluation of the contribution of biomass for domestic heating to air quality and health for analysis of the effectiveness of the Air Quality Remediation Plan of the Lazio Region
Introduction
In recent decades, air pollution caused by particulate matter (PM), especially PM10, has emerged as a critical environmental and public health concern worldwide. Numerous studies have demonstrated that exposure to PM can lead to a wide range of health issues, including cardiovascular and respiratory diseases, neurological disorders, and even increased mortality rates (Kim et al., 2015; Kioumourtzoglou et al., 2016; Miri et al., 2017). The World Health Organization (WHO) and the European Union (EU) have established air quality standards to protect public health, yet many urban populations continue to be exposed to pollutant levels exceeding these guidelines. For instance, in 2021, approximately 76% of the urban population in the EU was exposed to PM10 concentrations above the WHO annual guideline, highlighting the persistent challenge of air quality management (EEA, 2023).
Among the various sources of PM, biomass burning for residential heating has been identified as a significant contributor, especially in regions where traditional heating practices are prevalent. Biomass combustion releases a complex mixture of pollutants, including polycyclic aromatic hydrocarbons (PAHs), potentially toxic elements (PTEs), organic compounds, and soot particles, which pose serious health risks (Belis et al., 2011; Piazzalunga et al., 2013; Butt et al., 2016; Kukkonen et al., 2020). Recent meta-analyses have confirmed that short-term exposure to biomass burning emissions is associated with increased mortality and adverse health outcomes (Karanasiou et al., 2021). In Italy, the reliance on biomass for home heating remains substantial, with approximately 24.3% of households using wood or pellets. Despite the availability of modern, cleaner technologies, a significant proportion (about 70%) still employ traditional, less efficient devices such as open fireplaces and old stoves (ISTAT, 2018). These older appliances tend to burn fuel less completely, resulting in higher emissions of PM and associated pollutants. Factors influencing emission characteristics include the type of device, fuel moisture content, combustion temperature, and maintenance practices (Gonçalves et al., 2011; Alves et al., 2011; Vicente et al., 2015a, 2015b). For example, wood combustion can produce a different chemical profile compared to pellet burning, with variations in VOCs, PAHs, and PTEs (Zosima et al., 2016; Frasca et al., 2018).
Understanding the chemical composition and sources of PM from biomass burning is essential for assessing its environmental and health impacts. Receptor modeling techniques like Positive Matrix Factorization (PMF) enable the identification and quantification of different emission sources based on chemical fingerprints (Paatero & Tapper, 1994; Hopke, 2000). However, there remains a gap in integrating real-world field data with laboratory emission studies, which is crucial for developing effective mitigation strategies. Given the increasing push towards renewable energy sources like biomass, it is vital to evaluate the actual emissions and their chemical characteristics in real-world scenarios.
Research objectives
The main objective is to evaluate how biomass combustion devices contribute to PM10 production in two contexts within the Lazio region: Rome and the Sacco Valley. Rome, a large urban center, is largely characterized by the use of modern pellet stoves, whereas in the Sacco Valley traditional, less efficient devices prevail. Geographical and meteorological differences between the areas influence pollutant dispersion and accumulation, making them exemplary cases for understanding the impact of residential biomass on local air quality.
As a first specific objective, we proceeded with chemical characterization and source identification: comprehensive chemical analyses of PM10 samples collected during dedicated campaigns in both regions, including elemental and organic carbon, inorganic ions, trace elements, and biomass-related tracers such as levoglucosan. By applying receptor models such as PMF, sources were differentiated with particular emphasis on biomass combustion from different types of household appliances. Dispersion conditions were also evaluated based on measured data to frame the atmospheric dispersion capacity of pollutants and to understand how meteorological and topographic factors influence PM10 levels and source contributions.
Second, we compared real-scenario data with laboratory data, specifically those from the AIRUSE LIFE project (http://doi.org/10.5281/zenodo.5095807), to validate the outcome of source attribution and to understand how operational conditions and appliance types influence emissions. Additional data from questionnaires on biomass use in the two areas allowed assessment of the role of older, less efficient Sacco Valley devices compared with the modern solutions in Rome, providing a scientific basis for targeted policies to reduce residential biomass pollutants. The PM10 campaign data supported inhalation risk estimation for carcinogenic and non-carcinogenic substances, using EPA algorithms, to evaluate potential health risks associated with particulate matter inhalation. This multidimensional approach aims to provide a comprehensive understanding of the role of domestic biomass combustion in particulate pollution and to contribute to effective mitigation strategies for environment and public health.
Materials and Methods
The study was conducted in two areas of the Lazio Region: the Rome metropolitan area, where most heating plants use pellets, and the Sacco Valley (about 100 km southeast of Rome), characterized by a higher share of traditional wood fireplaces and stoves. In Rome, 169 PM10 samples were collected at two sites: Sapienza (inside the Experiment Botanical Garden, near busy roads, defined as traffic-oriented urban background) and Via Saredo (traditionally urban background in a residential area). In the Sacco Valley, 115 PM10 samples were collected at seven ARPA Lazio stations (Ceccano (CE), Ferentino (FE), Frosinone (FR), Alatri (AL), Anagni (AN), Cassino (CA), and Fontechiari (FO)). The topographic context of the Sacco Valley, with valleys and mountainous relief, promotes pollutant stagnation, amplified by temperature inversions; industrialization and intensive agriculture have further degraded the environment, making this valley one of the most polluted areas in Italy. To optimize representativeness, data from multiple sites were aggregated to produce two datasets per area: one for Rome and one for the Sacco Valley, allowing the inclusion of different emission sources and environmental conditions. In Rome, two relatively homogeneous stations were chosen to capture significant variations within a uniform urban context. In the Sacco Valley, instead, a broader and more dispersed monitoring network was adopted to reflect the heterogeneous territorial characteristics of the area.
In Rome, daily PM10 samples were collected with two flow-through sampling lines at 2.3 m3 h-1 on quartz and PTFE filters. In the Sacco Valley, at CE and FE, two HSRS in parallel at 2 L min-1 were used for 13 monitoring periods of 3 weeks each (one line with quartz, one with PTFE). At sites FR, AL, AN, CA, and FO, PM10 were sampled with Smart Samplers (SS) equipped with PTFE filters at 0.5 L min-1 for 7 monitoring periods, each lasting 6 weeks, except the last which lasted 3 weeks. At the CE site, 24-hour PM10 filters were also collected from March 28 to May 21, 2020. Quartz filters were analyzed for elemental carbon (EC) and organic carbon (OC) using thermo-optical analysis.
PTFE filters were employed for gravimetric PM10 mass determination using an automatic microbalance. Subsequently, macroelements (Al, Ca, Cl, Fe, K, Mg, Na, S, Si, Ti, Zn) were quantified by X-ray fluorescence. Filters were then extracted with deionized water, and the resulting extract was analyzed for anions (Cl⁻, NO₃⁻, SO₄²⁻) and cations (Ca²⁺, K⁺, Mg²⁺, Na⁺, NH₄⁺) by ion chromatography. Levoglucosan (LVGSN) was quantified by high-performance anion exchange chromatography with pulsed amperometric detection. The remaining water extract was filtered (0.45 μm) and analyzed for 26 elements (As, Ba, Bi, Cd, Ce, Co, Cs, Cu, Fe, La, Li, Mn, Mo, Ni, Pb, Rb, Sb, Sn, Sr, Ti, Tl, U, V, W, Zn, Zr) by inductively coupled plasma mass spectrometry. The residual filters were subjected to microwave-assisted acid digestion (HNO₃/H₂O₂ 2:1), filtered (0.45 μm), and analyzed by ICP-MS for the same 26 elements.
The analysis of Positive Matrix Factorization (PMF) was applied separately to the Rome and Sacco Valley datasets to highlight the characteristic chemical compositions of the sources. The set of chemical variables used for source apportionment includes EC, OC, LVGSN, SO4, Cl-, Na+, K+, NH4+, NO3-, Al, Cs, Cu, Fe, Rb, Sb, and Zn (selected based on S/N and the percentage of detections above the MDL). The input for the Sacco Valley comprises 115 samples (60 at 3–6 week resolution and 55 daily); for Rome, 169 daily samples. It was possible to aggregate data with different resolutions because PMF does not require temporal homogeneity. Different numbers of factors were explored; the final solution was chosen by evaluating the objective function Q and the stability of the solutions, as suggested by Paatero et al. and Ridolfo et al. The analysis coherently considered the same chemical elements across the two areas to ensure comparability of the results. The ventilation coefficient (VC) is a key metric for air-pollutant dispersion, which serves as an index of the dilution capacity per unit time within the planetary boundary layer (PBL). In this study, VC was estimated for two areas to assess atmospheric stability and dispersion efficiency. Wind data came from ARPA Lazio micrometeorological stations (Rome area: Rome Tor Vergata; Sacco Valley: Frosinone), using June–August and December–February 2020 for Rome, and 2015 for Sacco Valley. The boundary layer height was averaged between summer and winter periods; Sacco Valley data were drawn from ISAC/CNR (2018), while Rome data were sourced from the Alicenet network (Bellini et al., 2024).
To estimate the weighted emission factors (EF) for PM10 from biomass combustion, literature data from the LIFE AIRUSE project were combined with the results of the ARPA Lazio survey (2019) on domestic biomass use. The categories considered include traditional fireplaces, enclosed fireplaces, wood stoves, and pellet stoves. The EF for each category were selected from the literature and combined to obtain EFarea = Σ wi × EFi, with specific assumptions (e.g., equivalence between open/closed fireplace and cold-start conditions, holm oak as wood, average EF for pellet stove). The chemical profiles of measured PM10 were compared with those of the combustion sources using common tracer ratios (e.g., Cs/LVGSN, Cs/EC) for the wood stove and pellet stove categories.
The assessment of non-carcinogenic and carcinogenic inhalation risk was conducted following US-EPA guidelines. For non-carcinogenic risks, the Average Daily Dose over a Lifetime (ADDinh) is estimated using a formula that incorporates the environmental concentration, inhalation rates, correction factors, exposure frequency, exposure duration, body weight, and averaging time. The non-carcinogenic risk is expressed through the Hazard Quotient (HQ) and Hazard Index (HI, sum of HQs). An HQ > 1 indicates potential adverse effects. For carcinogenic risks, the Incremental Lifetime Cancer Risk (ILCR) is estimated; the formula considers the average concentration of the species, exposure time, exposure frequency, exposure duration, Inhalation Unit Risk, and averaging time; the protection goal recommended by the U.S. EPA is 10-6. Specific carcinogenic elements (As, Pb, Cr, Ni) were considered in this assessment.
Results and discussion
The results indicate a clear difference between the two areas in terms of PM10 levels, biomass profiles, and dispersion potential. In the Sacco Valley, mean PM10 values and typical biomass markers (levoglucosan, K+, Cs, Rb, LVGSN; OC and EC) are significantly higher than in Rome. Elements such as Cu, Fe, Sb, and Zn, associated with traffic and Saharan or terrestrial dust, show similar values in the two contexts, suggesting a background component linked to traffic, soil, and diffuse emissions, while traces of Cl− and Na+, attributable to marine exchange, are more abundant in Rome, consistent with greater proximity to the sea. The PMF analysis identified biomass patterns in both regions, with profiles consistent with biomass associated with wood and pellet fuel use but with significant hardware differences. In Rome, a unique biomass component “BB mixed” emerged, contributing substantially to total PM10, characterized by high fractions of OC, EC, LVGNS, K+, Cs, and Rb, associated with a mix of technologies, with a prevalence of pellet stoves. In the Sacco Valley, conversely, two distinct biomass profiles were extracted: “BB pellet” and “BB woodstove.” The BB pellet profile is strongly marked by tracers of K+, Cs, and Rb, consistent with pellet devices, while BB woodstove shows dominant markers OC and LVGSN, interpretable as predominant use of traditional wood stoves or open fireplaces. This is particularly relevant because it shows that, even within the same region, device type drastically influences the emission signature. In terms of dispersion consequences, the VC shows dispersion conditions that are markedly better in Rome than in the Sacco Valley, attributable to different atmospheric dynamics and valley orography, which promote stagnation and confinement of pollutants. The correlation analysis between biomass profiles identified by PMF and laboratory emission data of the three device categories (chimneys, traditional stoves, pellet stoves) indicates a significant concordance between the observed chemical fingerprints and their respective emission profiles. The area-weighted emission factors (EFs), derived from pairing survey weights with laboratory data, show different levels between Rome and Sacco Valley, with EF for Sacco Valley generally higher, consistent with the prevalent use of traditional devices with lower efficiency. In particular, the EF for Rome lies intermediate between the pellet stove and traditional fireplace categories, in line with the presence of a “BB mixed” component, while Sacco Valley shows EFs closer to the fireplace and woodstove categories, consistent with the predominant “BB woodstove” profile. The analysis of the reasons for these differences indicates that not only the domestic use composition of various devices but also atmospheric dispersion conditions play a crucial role in determining the observed PM10 concentrations. In this context, the importance of the number of samples and their temporal variability should be highlighted: despite methodological differences between datasets (temporal resolution, sampling period), the consistent application of PMF enabled robust solutions, with profiles that are compatible between the two areas despite different sampling configurations. The analysis of chemical ratios such as Cs/LVGSN and Cs/EC between biomass profiles and emission datasets provided further interpretive elements: in the Sacco Valley woodstove profile, for example, Cs/LVGSN ratios are significantly lower than in the pellet stove, concordant with a higher levoglucosan signature in emissions from untreated wood at high temperatures compared to pellet, where combustion temperatures reach higher levels. This finding is supported by recent studies showing that levoglucosan is less favorable as a marker of modern automated stoves, where combustion temperatures tend to be higher. Furthermore, the dispersion analysis shows that, even after normalizing biomass contributions per VC, the Sacco Valley remains significantly more critical than Rome, reinforcing the hypothesis that orography and stagnation-prone meteorological conditions play an essential role in determining local PM10 peaks, in addition to the impact of the combustion technologies employed. In summary, the results show that the Sacco Valley, with a greater prevalence of traditional devices, exhibits higher biomass emissions and less favorable atmospheric conditions for dispersion, compared with Rome where the spread of pellet and high-efficiency devices reduces emissions and improves dispersion. These results underscore the need for targeted policies to promote technological upgrading and maintenance of domestic systems, within the context of region-specific characteristics, in order to reduce the impact of biomass on PM10, improve air quality, and protect public health.
The inhalation risk analysis, conducted for the Sacco Valley, benefited from greater data availability and robustness over a longer period, allowing a consistent estimation of exposure parameters and concentration levels. In this area, the carcinogenic risk was below the protective threshold of 10-6, indicating a non significant expected cancer incidence at the population level in the long term. Conversely, the non-carcinogenic risk exceeded reference limits for both children and adults, with particular attention to Co, Cr and Mn, observing the contributions of these species to the total HI.
Health-wise, HQs above 1 for Co, Cr and Mn suggest a possible increased risk of respiratory effects, especially in children, who have different exposure pathways and tolerance mechanisms. Therefore, in addition to developing emission control and mitigation plans, periodic monitoring of vulnerable populations and promotion of public health campaigns are advisable.
Conclusions
In conclusion, this study enables assessing, through an integrated approach, how PM10 emissions from biomass for domestic use in two territorial contexts of Lazio contribute to air quality. The results indicate that the Sacco Valley records biomass-derived concentration levels significantly higher than those in Rome, partly due to greater diffusion of traditional and less efficient devices, but also because of less favorable dispersion conditions, owing to the valley’s topography and characteristic atmospheric dynamics. Rome shows emission levels and profiles that are less intense, thanks to the prevalence of pellet stoves and modern appliances, with generally more efficient atmospheric dispersion.
The PMF analysis allowed distinguishing distinct biomass profiles between the two areas: in Rome a “BB mixed” profile emerges, reflecting heterogeneity in the use of various devices, while in the Sacco Valley separate profiles are distinguished for pellet stoves and wood ovens, confirming a more differentiated impact of device type on emissions. The comparison between field-observed profiles and laboratory data reinforces the interpretation: combustion conditions and biomass types substantially influence the amount and composition of PM10. Weighted EF estimates, supported by an investigation of device usage, indicate higher EFs in the Sacco Valley, in line with the use of less efficient devices.
Dispersion, measured by the ventilation coefficient, shows an atmospheric environment less favorable to contaminant dilution in the Sacco Valley, with an increased potential for population exposure. The innovation of the study lies in the integration of field PM10 data, an emission dataset, PMF models, and a regional survey on device usage, offering an analytical framework to define mitigation strategies based on concrete territorial evidence.
Within the Sacco Valley, the cancer risk is below the 10-6 threshold, indicating a non-significant long-term incidence. By contrast, the non-cancer risk exceeds reference limits for children and adults, with Co, Cr, and Mn among the main contributors. The proposed framework integrates emission profiles, dispersion dynamics, and health risks according to international criteria, offering a comprehensive view to align real mitigation measures, continuous monitoring, and integrated risk assessments across energy needs, public health, and environmental quality in the different regional contexts of Lazio. Future studies could assess long-term health effects, the effectiveness of targeted interventions, and refine the integration between regional inventories and monitoring events.
References
Kim, K. H., Kabir, E., Kabir, S., 2015. A review on the human health impact of airborne particulate matter. Environment international, 74, 136-143. https://doi.org/10.1016/j.envint.2014.10.005.
Kioumourtzoglou, M. A., Schwartz, J. D., Weisskopf, M. G., Melly, S. J., Wang, Y., Dominici, F., Zanobetti, A., 2016. Long-term PM2.5 exposure and neurological hospital admissions in the northeastern United States. Environmental health perspectives, 124(1), 23-29. https://doi.org/10.1289/ehp.1408973.
Miri, M., Ebrahimi Aval, H., Ehrampoush, M.H., Mohammadi, A., Toolabi, A., Nikonahad, A., Derakhshan, Z., Abdollahnejad, A., 2017. Human health impact assessment of exposure to particulate matter: an AirQ software modeling. Environ. Sci. Pollut. Res. 24 (19), 16513–16519. https://doi.org/10.1007/s11356-017-9189-9.
EEA, European Environment Agency, 2023. Europe’s air quality status 2023.
Belis, C. A., Cancelinha, J., Duane, M., Forcina, V., Pedroni, V., Passarella, R., G. Tanet, K. Douglas, A. Piazzalunga, E. Bolzacchini, G. Sangiorgi, M.-G. Perrone, L. Ferrero, P. Fermo, Larsen, B. R., 2011. Sources for PM air pollution in the Po Plain, Italy: I. Critical comparison of methods for estimating biomass burning contributions to benzo (a) pyrene. Atmospheric Environment, 45(39), 7266-7275. https://doi.org/10.1016/j.atmosenv.2011.08.061.
Piazzalunga, A., Anzano, M., Collina, E., Lasagni, M., Lollobrigida, F., Pannocchia, A., Fermo, P., Pitea, D., 2013. Contribution of wood combustion to PAH and PCDD/F concentrations in two urban sites in Northern Italy. Journal of Aerosol Science, 56, 30-40. https://doi.org/10.1016/j.jaerosci.2012.07.005.
Butt, E. W., Rap, A., Schmidt, A., Scott, C. E., Pringle, K. J., Reddington, C. L., Richards, N. A. D., Woodhouse, M. T., Ramirez-Villegas, J., Yang, H., Vakkari, V., Stone, E. A., Rupakheti, M., S. Praveen, P., G. van Zyl, P., P. Beukes, J., Josipovic, M., Mitchell, E. J. S., Sallu, S. M., Forster, P. M., Spracklen, D. V., 2016. The impact of residential combustion emissions on atmospheric aerosol, human health, and climate, Atmos. Chem. Phys., 16, 873–905. https://doi.org/10.5194/acp-16-873-2016.
Kukkonen, J., López-Aparicio, S., Segersson, D., Geels, C., Kangas, L., Kauhaniemi, M., Maragkidou, A., Jensen, A., Assmuth, T., Karppinen, A., Sofiev, M., Hellén, H., Riikonen, K., Nikmo, J., Kousa, A., Niemi, J. V., Karvosenoja, N., Santos, G. S., Sundvor, I., Im, U., Christensen, J. H., Nielsen, O.-K., Plejdrup, M. S., Nøjgaard, J. K., Omstedt, G., Andersson, C., Forsberg, B., Brandt, J., 2020. The influence of residential wood combustion on the concentrations of PM2.5 in four Nordic cities, Atmos. Chem. Phys., 20, 4333–4365. https://doi.org/10.5194/acp-20-4333-2020.
Karanasiou, A., Alastuey, A., Amato, F., Renzi, M., Stafoggia, M., Tobias, A., Reche, C., Forastiere, F., Gumy, S., Mudu, P., Querol, X., 2021. Short-term health effects from outdoor exposure to biomass burning emissions: A review. Science of The Total Environment, 781, 146739. https://doi.org/10.1016/j.scitotenv.2021.146739.
ISTAT, Italian National Institute of Statistics, 2018. DATI STATISTICI PER IL TERRITORIO- Regione Lazio.
¬Gonçalves, C., Alves, C., Fernandes, A.P., Monteiro, C., Tarelho, L., Evtyugina, M., Pio, C., 2011. Organic compounds in PM2.5 emitted from fireplace and woodstove combustion of typical Portuguese wood species. Atmos. Environ. 45, 4533–4545. https://doi.org/10.1016/j.atmosenv.2011.05.071.
Alves, Gonçalves, C., Fernandes, A.P., Tarelho, L., Pio, C., 2011. Fireplace and woodstove fine particle emissions from combustion of western Mediterranean wood types. Atmos. Res. 101, 692–700. https://doi.org/10.1016/j.atmosres.2011.04.015.
Vicente E.D., Duarte, M.A., Calvo, A.I., Nunes, T.F., Tarelho, L., Alves, C.A., 2015a. Emission of carbon monoxide, total hydrocarbons and particulate matter during wood combustion in a stove operating under distinct conditions. Fuel Process. Technol. 131. https://doi.org/10.1016/j.fuproc.2014.11.021.
Vicente, E.D., Duarte, M.A., Calvo, A.I., Nunes, T.F., Tarelho, L.A.C., Custódio, D., Colombi, C., Gianelle, V., Sanchez de la Campa, A., Alves, C.A., 2015b. Influence of operating conditions on chemical composition of particulate matter emissions from residential combustion. Atmos. Res. 166, 92–100. https://doi.org/10.1016/j.atmosres.2015.06.016.
Zosima, A. T., Tzimou-Tsitouridou, R. D., Nikolaki, S., Zikopoulos, D., Ochsenkühn-Petropoulou, M. Th., 2016. PM10 emissions and PAHs: The importance of biomass type and combustion conditions. Journal of Environmental Science and Health, Part A, 51(4), 341–347. https://doi.org/10.1080/10934529.2015.1109409.
Frasca, D., Marcoccia, M., Tofful, L., Simonetti, G., Perrino, C., Canepari, S., 2018. Influence of advanced wood-fired appliances for residential heating on indoor air quality. Chemosphere, 211, 62-71. https://doi.org/10.1016/j.chemosphere.2018.07.102.
Hopke, P. K. (2000). A guide to positive matrix factorization, Workshop on UNMIX and PMF as Applied to PM2. 5. Edited by Willis, RD, RTP, NC. (EPA 600/A-00/048).
Paatero, P., Tapper, U., 1994. Positive matrix factorization: a non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 5, 111–126. https://doi.org/10.1002/env.3170050203.
Bellini, A., Diémoz, H., Di Liberto, L., Gobbi, G. P., Bracci, A., Pasqualini, F., Barnaba, F., 2024. ALICENET–an Italian network of automated lidar ceilometers for four-dimensional aerosol monitoring: infrastructure, data processing, and applications. Atmospheric Measurement Techniques, 17(20), 6119-6144. https://doi.org/10.5194/amt-17-6119-2024.
CNR-ISAC, Institute of Atmospheric Sciences and Climate, Contribution to the meteorological characterization of the Sacco Valley through use of active remote sensing instrumentation, 2018.
ARPA Lazio, Regional Environmental Protection Agency, 2019. Sample survey on domestic consumption of woody biomass in Lazio.