PM2.5 Trajectory Classes and Spatial Investment Patterns in Portugal

PM2.5 Trajectory Classes and Spatial Investment Patterns in Portugal | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article PM2.5 Trajectory Classes and Spatial Investment Patterns in Portugal Raimundo Elias GOMEZ, Maria Gabriela MIÑO This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7790529/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The article examines temporal trajectories of fine particulate matter (PM2.5) across Portugal’s 308 municipalities (2019–2025) in relation to seasonal wind patterns and municipal economic and cultural investment. Annual PM2.5 concentrations and wind statistics were obtained from CAMS NRT/ECMWF satellite data. After standardising and clustering the PM2.5 distribution with k-means, twenty-nine municipalities closest to cluster centroids were compared using 2024 fiscal data. Results reveal marked spatial disparities in PM2.5 exposure linked to seasonal regimes and investment profiles. High economic-cultural investment areas often align with high-pollution trajectories, whereas low investment correlates with lower levels, both reflecting seasonal environmental and social patterns. PM2.5 Municipal Investment Spatial Inequality Trajectory Analysis Air Pollution Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Highlights - PM2.5 trajectories reveal three distinct pollution classes across Portugal - Dense economic-cultural zones show 2× higher PM2.5 than peripheral areas - Low pollution areas have limited economic and cultural infrastructure - Capital externalities concentrate where investments are highest - Seasonal wind patterns redistribute PM2.5 across municipal boundaries 1. Introduction Fine particulate matter (PM2.5) it’s an environmental health risks in Portugal and contemporary societies, with consequences that extend well beyond urban cores (Barbosa et al., 2024 ; Corda et al., 2024 ; Rodriguez Avellaneda et al., 2025 ). Most studies addressing air pollution in Portugal have adopted national or metropolitan perspectives, often giving limited attention to related dimensions at the municipal scale. This point is important because pollution inequalities are shaped by the uneven distribution of resources, capacities, and priorities among localities, raising questions about the regional configurations of capital externalities (Bille & Honoré, 2025 ; Işıkara, 2023 ; Yu et al., 2025 ). Yet, municipalities represent a crucial locus for the articulation of both economic and cultural capital investments—factors that can profoundly influence the production and persistence of environmental risks. The differentiated accumulation of investments, the legacy of spatial development, and the orientation of local policies contribute to a mosaic of PM2.5 exposures that is neither accidental nor easily explained by macro-level trends alone. This article aims to advance the understanding of these dynamics by: (1) mapping the spatial distribution and temporal evolution of PM2.5 across all 308 Portuguese municipalities between 2019 and 2025; (2) identifying and classifying distinct PM2.5 trajectories at the municipal level; and (3) examining how these pollution trajectories correspond with municipal profiles of economic and cultural investment, focusing on the most recent consolidated year with complete fiscal data (2024). The study is grounded in a relational perspective, which posits that spatial inequalities in environmental exposure are intrinsically linked to the uneven configuration of capital competences and investment regimes (Bourdieu, 2018b ; Gomez, Miño, Hojman, et al., 2025 ; Gomez & Miño, 2025 ; Marom, 2014 ; Siblot et al., 2024 ). By integrating geospatial air quality data and environmental estimates with detailed fiscal and cultural indicators, the analysis exposes how municipal capital allocation aligns with differentiated vulnerabilities to air pollution, thereby providing an empirical basis for more targeted decoupling strategies and policy design. Following this approach, capital is understood not only as an economic dimension but also as cultural investments. Both forms exist in distinct states: embodied as personal dispositions, institutionalised as formalised investments, and objectified as infrastructures, pollution footprints, or the spatial distribution of populations and services. In its objectified state, capital represents the durable inscription of resources and power but also the consolidation of inequality, visible not only in demographic and cultural facilities but also in capital externalities such as PM2.5 (Bourdieu, 2020 ; Gomez, Miño, Pereira, et al., 2025). The analysis focuses on the evolution and distribution of PM2.5 as an externality—arising when the objectification of industrial activities, through material manifestations and infrastructural outcomes, generates unintended/uncontrolled effects beyond the intentions or interests of the original investors. While externalities may manifest as either positive or negative consequences, emissions of PM2.5 are globally recognized as harmful to health and disproportionately affect specific places and populations. From a relational perspective, these emissions are not random but are structurally linked to the spatial differentiation of investment regimes. The forms, intensities, and spatial configurations of economic and cultural investment correlate directly with the objectified landscape, generating new inequalities, as well as heightened vulnerabilities to environmental risks. Accordingly, the differentiated trajectories of PM2.5 across Portuguese municipalities are analysed as objectified outcomes of situated strategies and capital allocation. Accordingly, the article addresses the following question: How are the temporal trajectories of PM2.5 exposure (2019–2025) distributed across Portuguese municipalities, and to what extent do these patterns correspond with major economic and cultural investments from the most recent consolidated fiscal year (2024) ? This structural approach expands the analytical scope of conventional studies on capital externalities. Elucidating the links between PM2.5 patterns and municipal investment profiles is essential for guiding policy interventions aimed at reducing environmental spatial inequalities, protecting public health, and deepening understanding of the social cost associated with fine particulate emissions. 2. Methodology 2.1. Area of Study The analysis covers the entire territory of Portugal, encompassing all 308 municipalities, including those located in the autonomous regions of Madeira and the Azores. The municipal scale offers an appropriate level of granularity for investigating the spatial differentiation of both air pollution and investment dynamics. 2.2. Data Sources and Variables PM2.5 monthly and annual mean concentrations for the period from January 2019 to March 2025 were obtained from the Copernicus Atmosphere Monitoring Service (CAMS) Near-Real-Time reanalysis, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) (CAMS, 2021 ). This dataset integrates satellite and ground-based measurements, providing harmonised estimates of air quality (EU/ESA/Copernicus, 2024 ). All scenes from 2019 to 2025 were accessed and processed via Google Earth Engine. Monthly and annual means were spatially averaged within the official boundaries of Portugal’s 308 municipalities, yielding a consistent panel dataset suitable for temporal and spatial comparison. To complement the PM2.5 concentrations, wind fields were derived from the ERA5 reanalysis produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) and distributed through the Copernicus Climate Data Store (CAMS, 2021 ; Kouyate et al., 2024 ). The ERA5 dataset provides hourly estimates of meteorological variables, including the horizontal components of wind at 10 m above ground level: one oriented east–west (u) and the other north–south (v). These two dimensions together describe wind speed and direction and are generated by assimilating satellite and in-situ observations into a global numerical weather prediction model (C3S, 2018 ). Monthly and annual averages were calculated with Google Earth Engine from the complete series (from January 2019 to March 2025) and spatially aggregated across the 308 Portuguese municipalities. Both the PM2.5 and wind databases can be accessed online, together with an analysis of their spatial and temporal distribution (Gomez, 2025 c, 2025 b). All municipal-level socioeconomic data used in this study were obtained from statistical sheets published by the National Statistics Institute of Portugal (Instituto Nacional de Estatística, INE)—specifically the Municipal Sheet – Socioeconomic and Environmental Context ( Ficha Municipal, Contexto Socioeconómico e Ambiental , CSE) and its companion series Information Return to Respondents ( Retorno de Informação aos Respondentes )—available through the INE Municipal Data Portal ( Portal de Dados Municipais do INE ) (INE-PT, 2025 ). This year was selected as it provides the most recent, fully consolidated, and publicly available municipal account data. Such temporal delimitation reflects the institutional assumption that municipal economic and cultural investments are subject to structural inertia and shaped by programmatic planning cycles rather than short-term fluctuations (CFP, 2025 ; CNC, 2025 ). The complete database, together with an analytical dashboard, can be accessed online (Gomez, 2025 a). Table 1 Considered variables, abbreviations, investment dimensions, and definitions. Variable Abbrev. Dimension INE definition Gross value added (million €) GVA Economic Value of goods + services produced minus intermediate consumption (official “Valor acrescentado bruto”). Business turnover (million €) BizVol Economic Total value of sales of goods and services (“Volume de negócios”) realised during the reference year. Number of firms (registered units) Firms Economic Count of active enterprises recorded in the Statistical Business Register. Newly registered legal entities (annual) NewEnt Economic Annual number of newly incorporated collective persons (companies, cooperatives, etc.). Tourism revenue (thousand €) TourRev Economic Receipts from accommodation and related tourist services reported by municipalities. Median housing price (€/m²) HousePrice Economic Median €/m² of dwellings transacted, based on notarised deeds ( Estatísticas de preços da habitação ). Per-capita cultural expenditure (€/inh.) CultExp Cultural Municipal budget outlay on culture divided by resident population. Library & archives expenditure (thousand €) LibExp Cultural Municipal spending on libraries and archival services. Performing-arts expenditure (thousand €) ArtsExp Cultural Municipal spending on music, theatre, dance and related live-performance activities. Gross secondary-education enrolment rate (%) SecEnroll Cultural Ratio of students enrolled in secondary education to the resident population of the official age group. Secondary-education completion rate (%) SecCompl Cultural Share of students completing secondary schooling within the expected period. Inverse basic-education dropout rate (%) InvDrop Cultural 100 – (dropout + retention rate) in compulsory basic education. In line with a relational approach, both economic and cultural investments are understood as key structuring forces shaping the spatial distribution of resources, power, and opportunity (Bille & Honoré, 2025 ; Bourdieu, 2018a ; Gomez, 2021 ). These dimensions are operationalised using composite indicators that capture municipal investment strategies. Variable selection was guided by criteria of completeness, discriminative power, and conceptual relevance. The final set of indicators represents economic dimensions (gross value added, business turnover, number of firms, newly registered legal entities, tourism revenue, median housing price) and cultural dimensions (per-capita cultural expenditure, library and archives spending, performing-arts spending, secondary education enrolment and completion rates, and inverse dropout rate) (Table 1 ). All variables were standardised as z-scores to ensure comparability across municipalities. 2.3. Analytical Procedures Temporal clustering methods with k-means are widely adopted in environmental research for detecting latent patterns and grouping cases within longitudinal pollution datasets (Choi et al., 2024 ; Espinoza-Guillen et al., 2025 ). By partitioning municipalities according to the similarity of their PM2.5 time series, k-means provides a robust preliminary classification that facilitates the investigation of how distinct pollution trajectories relate to wind patterns and underlying municipal investment profiles. In a second stage, to enable a focused and interpretable comparison, a subsample of 29 municipalities was selected to represent the diversity of economic and cultural capital allocation on each trajectory class. Within each class, municipalities were ranked according to their Euclidean distance from the class centroid in the multidimensional PM2.5 space, ensuring that the closest decile—those most representative of their respective class—was retained (Class 1: 9 municipalities; Class 2: 10 municipalities; Class 3: 10 municipalities). This selection strategy not only preserves the internal coherence of each class but also accounts for district-level diversity by limiting inclusion to one municipality per administrative district within each class. The slight deviation from a theoretical sample of 30 reflects this balancing criterion. The resulting representative sample forms the basis for a detailed comparative analysis of investment structures across distinct PM2.5 trajectory classes. 3. Results 3.1 How are the annual trajectories of PM2.5 exposure (2019—2025) spatially distributed across Portugal? Between 2019 and 2020, PM2.5 concentrations fell sharply across all trajectory classes in Portugal, mirroring the decline in industrial and transport activities that occurred in many cities of the globe during the COVID-19 lockdown period and the associated, albeit temporary, reduction in anthropogenic emissions (Ashraf et al., 2025 ; Prapassonpithaya & Jinsart, 2025 ) (Fig. 1 ). From 2020 onward, PM2.5 levels increased again, with pronounced fluctuations, particularly in the high pollution trajectory (Traj 3), as economic activities and emissions resumed. Throughout the observation period, all classes displayed closely aligned seasonal cycles, underscoring the significant influence of meteorological and climatic drivers—including temperature inversions, humidity, and wind—on PM2.5 dispersal and accumulation at the national scale (Alam et al., 2025 ). After the k-means clustering, the three identified trajectory classes correspond to distinct pollution regimes, each characterised by specific mean concentrations and variability. Trajectory 1 – Low Pollution Trajectories . Municipalities within this class exhibit mean PM2.5 levels ranging from approximately 5.44 to 6.85 µg/m³ and coefficients of variation (CV) between 0.28 and 0.43, indicative of moderate variability. Examples are Trancoso (Guarda) with 5.75 µg/m³ (CV 0.39), Moura (Beja) at 6.00 µg/m³ (CV 0.32), and Melgaço (Viana do Castelo) with 6.27 µg/m³ (CV 0.43) (Fig. 1 , Table 2 ). These locations typically correspond to areas with limited urbanisation or industrial activity, reflected in their lower PM2.5 levels and moderate internal dispersion. Trajectory 2 – Intermediate Pollution Trajectories. This group is defined by municipalities where average PM2.5 concentrations lie between 6.70 and 8.10 µg/m³, and CVs fall between 0.24 and 0.33. Notable examples are Batalha (Leiria) with 7.78 µg/m³ (CV 0.27), Cantanhede (Coimbra) at 8.10 µg/m³ (CV 0.30), and Montijo (Setúbal) at 7.50 µg/m³ (CV 0.25) (Fig. 1 , Table 2 ). This class represents an intermediate pollution scenario, with lower internal variability compared to Class 1. Trajectory 3 – High Pollution Trajectories . Municipalities in this category register PM2.5 concentrations from about 8.63 to 11.07 µg/m³ and CVs between 0.25 and 0.37. Examples include Amadora (Lisboa) at 9.97 µg/m³ (CV 0.26), Esposende (Braga) at 10.64 µg/m³ (CV 0.25), and Vila do Conde (Porto) at 11.07 µg/m³ (CV 0.25) (Fig. 1 , Table 2 ). These municipalities maintain persistently high PM2.5 levels with moderate internal variation, typically associated with more urbanised or industrialised contexts. The seasonal progression of air pollution between 2019 and 2025, as indicated by average monthly PM2.5 concentrations, highlights both commonalities and distinctions among the three trajectory classes (Fig. 2 ). Trajectory 1 , representing the lowest pollution levels, follows a regular cyclical trend, reaching a moderate peak in March (7.2 µg/m³) before decreasing through late autumn and early winter, with a trough in November (4.5 µg/m³). This suggests a marked responsiveness to seasonal emission sources, potentially linked to heating or biomass burning activities. Trajectory 2 , characterised by intermediate pollution, records its highest value in February (9.0 µg/m³), then gradually declines until August (7.1 µg/m³), followed by a secondary maximum in September (9.8 µg/m³). This pattern may be attributed to atmospheric inversions after summer or the lagged effect of emissions. Trajectory 3 , with consistently elevated pollution, sustains high PM2.5 levels, peaking in February (11.7 µg/m³) and again in September (10.0 µg/m³). This trajectory reflects enduring structural sources of pollution, such as industry, transport, and dense populated areas. The reduction in PM2.5 for Classes 2 and 3 from May to August likely corresponds to decreased emissions during the summer holiday period and more favourable meteorological conditions for pollutant dispersal. Overall, seasonality is apparent across all trajectories, but the magnitude and timing of peaks differ, emphasising the importance of regionally and temporally adapted mitigation measures. The cartographic rendering of the three latent trajectory classes (Fig. 3 ) discloses the tripartite geography of PM2.5 dynamics that cuts across Portugal’s administrative mosaic. A broad, quasi-continuous belt of Class 1 trajectory (blue) extends from the high-relief Minho-Trás-os-Montes frontier through the Beira and Alto Alentejo interiors to the Guadiana valley, epitomising trajectories characterised by persistently low background concentrations and muted seasonal amplitudes. Flanking this inland core, a discontinuous Class 2 littoral corridor trajectory (orange) follows the main Atlantic urban–industrial axis from the Aveiro lagoon to the Sado and lower Guadiana estuaries; here, moderate baseline levels combine with pronounced winter-summer contrasts that mirror the rhythm of coastal traffic, port activity and heating demand. Finally, Class 3 enclaves trajectories (green) punctuate the map in three strategic settings: (i) the autonomous archipelagos of Madeira and the Azores, where marine aerosol, shipping plumes and episodic Saharan dust generate sharp but short-lived peaks; (ii) selected tourism-intensive tracts of the Algarve and western Alentejo littoral; and (iii) port-industrial conurbations such as Porto-Leixões and Setúbal. The spatial distribution of PM is a complex phenomenon shaped by multiple environmental and social drivers. Among the unavoidable environmental factors in any spatial approach are the direction, strength, and frequency of winds during specific periods, particularly in certain seasons (C3S, 2018 ; CAMS, 2021 ). Yet beyond their meteorological function, winds operate as redistributive forces that dissolve the apparent boundaries between polluters and those affected. They disperse PM2.5 into zones far removed from its industrial or urban points of origin, thereby transforming what could be perceived as a localised burden into a diffuse, collective externality. This diffusion not only reconfigures the geography of exposure but also the sociology of responsibility: the costs of certain industrial and urban productions are ultimately absorbed by the broader community. Research in atmospheric sciences confirms that wind-mediated transport of fine particles can extend over hundreds of kilometres, redistributing health risks and environmental burdens (Lelieveld et al., 2015 ; Seinfeld & Pandis, 2016 ). Moreover, winds affect urban heat islands, modulate odour corridors, and influence the propagation of noise, further framing the lived experience of environmental externalities in dense urban areas (Oke et al., 2017 ). In this sense, wind functions simultaneously as an ecological and a social driver: it spreads the by-products of production and consumption across populations unequally positioned to mitigate or resist their effects. To illustrate these dynamics, seasonal wind rose diagrams are presented for Portugal (Fig. 3 ). They display the frequency distribution of wind direction and speed, with radial spokes indicating the proportion of time the wind blows from each compass point and concentric circles representing cumulative frequency levels. Colour gradients along the spokes correspond to PM2.5 concentrations, enabling the simultaneous interpretation of prevailing wind patterns and the associated particulate matter intensities across different sectors. At the national scale, Portugal displays pronounced seasonal shifts in wind patterns, each with its own association to PM2.5 levels (Fig. 4 ). Winter presents the most balanced directional distribution, with notable frequency peaks towards the northeast (~ 45°) and southwest (~ 225–240°), while the northwest (~ 315°) remains minimal. In this season, PM2.5 concentrations tend to be higher along the N–NE and WSW–SW sectors, and lower towards the southeast and northwest. Summer is characterised by a strong predominance of E–SE flows (≈ 90–150°), which occur with high persistence but are generally associated with lower PM2.5 intensities. Autumn exhibits a more variable regime, dominated by E–SE winds with secondary S–SW components; during this season, the most pronounced PM2.5 peaks are observed along E–SE and S sectors. Spring represents a transitional stage, with an increasing dominance of E–SE winds and moderate directional variability; PM2.5 concentrations are more evenly spread across sectors but show slightly higher values under N–NE flows. These seasonal distinctions highlight how both wind direction and intensity relate to the spatial distribution of PM2.5, underscoring the importance of meteorological context in interpreting pollution dynamics. The combined spatial patterns of wind and PM2.5 reflect distinct temporal signatures across Portugal rather than a simple gradient of magnitude. Each municipality exhibits its own distinctive characteristics (that can be explored interactively in (Gomez, 2025 c). 3.2 What is the relationship between PM2.5 exposure trends (2019—2025) and the 2024 economic and cultural investments in key Portuguese municipalities? To examine the relationship between PM2.5 exposure trajectories (2019–2025) and 2024 economic and cultural investments in key Portuguese municipalities, the study identified a representative subset for detailed analysis. Approximately 10% of the municipalities closest to each trajectory class centroid in PM2.5 space were selected, ensuring that no two municipalities belonged to the same administrative district (NAME_1). This procedure preserved geographic diversity while accurately reflecting the typical exposure patterns of each trajectory class (Table 2 ). Table 2 Selected representatives municipalities of PM2.5 by trajectory classes. MUNICIPIO_ID NAME_1 class mean_pm25 cv_pm25 Trancoso (Guarda) Guarda 1 5.753121 0.385091 Penedono (Viseu) Viseu 1 5.949848 0.413845 Fundão (Castelo Branco) Castelo Branco 1 6.205403 0.359353 Oliveira do Hospital (Coimbra) Coimbra 1 6.543970 0.349714 Valpaços (Vila Real) Vila Real 1 5.809820 0.401057 Carrazeda de Ansiães (Bragança) Bragança 1 5.441534 0.368988 Nisa (Portalegre) Portalegre 1 6.420625 0.290337 Estremoz (Évora) Évora 1 6.282494 0.288777 Mação (Santarém) Santarém 1 6.848363 0.283500 Moura (Beja) Beja 1 5.996860 0.316643 Melgaço (Viana do Castelo) Viana do Castelo 1 6.272007 0.432637 Alcoutim (Faro) Faro 1 6.516323 0.294728 Batalha (Leiria) Leiria 2 7.778669 0.267184 Santarém (Santarém) Santarém 2 7.624499 0.243763 Cadaval (Lisboa) Lisboa 2 8.091333 0.237724 Mortágua (Viseu) Viseu 2 7.868129 0.314529 Cantanhede (Coimbra) Coimbra 2 8.104945 0.302071 Mealhada (Aveiro) Aveiro 2 7.601800 0.327078 Montijo (Setúbal) Setúbal 2 7.501571 0.253616 Ponte de Sôr (Portalegre) Portalegre 2 6.703244 0.264534 Odemira (Beja) Beja 2 7.205290 0.265516 Vendas Novas (Évora) Évora 2 6.745696 0.266541 Moita (Setúbal) Setúbal 3 9.394950 0.252603 Amadora (Lisboa) Lisboa 3 9.966470 0.256918 Portimão (Faro) Faro 3 9.181091 0.267180 Esposende (Braga) Braga 3 10.643280 0.246694 Nordeste (Azores) Azores 3 9.423650 0.347062 Santa Cruz (Madeira) Madeira 3 8.626807 0.365447 Vila do Conde (Porto) Porto 3 11.071163 0.252238 Municipalities selected according to PM2.5 trajectory classes display distinct annual mean concentrations (mean_pm25, in µg/m³) together with their relative temporal variability, expressed as the coefficient of variation (cv_pm25). Class 1 municipalities record the lowest average PM2.5 levels (≈ 5.4–6.8 µg/m³) with comparatively higher variability (CV 0.28–0.43). Class 2 falls within an intermediate range (≈ 6.7–8.1 µg/m³; CV 0.24–0.33). Class 3 concentrates the highest pollution levels (≈ 8.6–11.1 µg/m³), with stable annual profiles and moderate variability (CV 0.25–0.37). The 2024 investment profiles of these municipalities, standardised as z-scores, reveal sharp contrasts across the three trajectories (Fig. 5 ). Trajectory 1—the lowest-pollution group—consistently records negative scores in most cultural indicators and shows weak performance in economic variables. This profile indicates that low emissions are partly sustained by limited economic density, but at the cost of reduced cultural infrastructure. Trajectory 2 occupies a transitional position: some indicators rise modestly above the national mean, while others present mixed results. Trajectory 3 municipalities—those with the highest and most stable PM2.5 levels—display the strongest overall profiles, combining dense economic activity with above-average cultural expenditures. In other words, there are clear contrasts in how economic and cultural investments correspond to the three PM2.5 trajectory classes (Fig. 6 ). Trajectory 1 (low pollution) shows a restrained investment profile: economic indicators such as turnover, employees, and GVA remain well below the national mean, while cultural investments are uneven. The only above-average values appear in inverse dropout rates and secondary enrolment, whereas most cultural expenditure variables are depressed. This profile suggests municipalities that sustain low emissions partly through limited economic density, but at the cost of reduced cultural infrastructure. Trajectory 2 (intermediate pollution) presents a more balanced pattern. Economic indicators like turnover, employees, and hotel capacity rise above the national mean, signalling moderate economic dynamism. Culturally, the most distinctive feature is the sharp peak in secondary education completion, accompanied by positive if modest values in library and arts expenditure. These municipalities exemplify transitional contexts, where medium-level pollution accompanies middling economic and cultural endowments. Trajectory 3 (high pollution) reveals the most expansive investment structure. Economic indicators—including GVA, turnover, employment, tourism, and hotel capacity—register consistently high positive scores, confirming the concentration of productive and service activities. On the cultural side, expenditure on libraries, performing arts, and per-capita culture stand clearly above average, reinforcing the image of urbanised and culturally active municipalities. Correlation analysis further emphasizes these contrasts (Fig. 7 and Table 3 ). In Trajectory 1, economic variables—gross value added, business turnover, and employment—are almost perfectly correlated, forming a tight but isolated productive block. Cultural linkages are limited, with strong correlations only between heritage and creative spending, suggesting fragmented cultural integration. Trajectory 2 shows a more hybrid pattern: classical economic associations are reinforced by a strong tourism–hospitality coupling and by consistent links between creative and interdisciplinary cultural expenditure, signalling selective but diversified investments. Trajectory 3 reveals the densest and most interwoven structure: economic indicators are tightly connected to one another and simultaneously to cultural expenditures, with near-perfect correlations between employment, creative spending, and heritage funding. Yet this dense accumulation of both economic and cultural capital coincides with persistently high PM2.5 exposure, underscoring the relational argument that environmental externalities concentrate where capital investments are highest. Underlying these investment patterns lie distinct population structures—of age, gender, and settlement density—that condition local demand, mobility, and emissions (Table 4 ). Table 3 Top absolute correlations (Pearson) Variable A Variable B r N Gross Value Added (million €) Number of Employees 0.98 29 Gross Value Added (million €) Business Turnover (million €) 0.96 29 Business Turnover (million €) Number of Employees 0.96 29 Gross Value Added (million €) Cultural and Creative Activities Expenditure (thousand €) 0.92 29 Number of Employees Cultural and Creative Activities Expenditure (thousand €) 0.91 29 Gross Value Added (million €) Heritage Expenditure (thousand €) 0.91 29 Business Turnover (million €) Cultural and Creative Activities Expenditure (thousand €) 0.89 29 Cultural and Creative Activities Expenditure (thousand €) Heritage Expenditure (thousand €) 0.89 29 Number of Employees Heritage Expenditure (thousand €) 0.87 29 Business Turnover (million €) Heritage Expenditure (thousand €) 0.81 29 Cultural and Creative Activities Expenditure (thousand €) Library and Archives Expenditure (thousand €) 0.77 29 Tourism Revenue (thousand €) Hotel Establishments (n) 0.68 29 Gross Value Added (million €) Library and Archives Expenditure (thousand €) 0.66 29 Number of Employees Library and Archives Expenditure (thousand €) 0.66 29 Business Turnover (million €) Library and Archives Expenditure (thousand €) 0.64 29 Heritage Expenditure (thousand €) Library and Archives Expenditure (thousand €) 0.64 29 Cultural and Creative Activities Expenditure (thousand €) Interdisciplinary Activities Expenditure (thousand €) 0.56 29 Cultural and Creative Activities Expenditure (thousand €) Performing Arts Expenditure (thousand €) 0.56 29 Business Turnover (million €) Interdisciplinary Activities Expenditure (thousand €) 0.53 29 Number of Employees Interdisciplinary Activities Expenditure (thousand €) 0.50 29 The demographic composition of municipalities mirrors the gradient of capital competences and PM2.5 trajectories; percentages are population-weighted by class and density is a simple mean across municipalities (Table 4 ): Trajectory 1 (low PM2.5). This group (n = 13) combines a small youth share (11.0% <15) with a high older-adult share (31.7% ≥65) and very low density (32.5 inh./km²). The implied working-age share is ≈ 57.3%, indicating a thin labour pool and limited demographic renewal. The female share (52.2%) is modestly above parity, consistent with ageing patterns. Demographically, these municipalities align with the investment profiles described earlier: weaker economic throughput and fragmented cultural infrastructure coincide with low exposure, suggesting that reduced capital intensity and demographic shrinkage jointly dampen PM2.5 levels. Trajectory 2 (intermediate PM2.5). Municipalities in this class (n = 9) present a more balanced age structure (12.8% <15; 25.1% ≥65; working-age ≈ 62.1%) and moderate density (89.6 inh./km²). The female share (50.4%) sits near parity. This configuration matches the “transitional” investment profile: moderate economic dynamism and selective cultural spending co-exist with middling exposure. Demographically, the presence of families and school-age cohorts suggests active local demand for services and education—features that can tip municipalities toward either cleaner growth or increased emissions depending on how investments are instrumented. Trajectory 3 (high PM2.5). The high-exposure group (n = 7) is markedly urban: very high density (1,519.4 inh./km²), larger youth presence (14.4% <15), smaller older-adult share (21.6% ≥65), and a working-age share near 64.0%. The female share (52.6%) is slightly above parity. These demographics align with the strongest economic and cultural profiles: large employee bases, higher turnover and GVA, and intensive cultural infrastructures. The same demographic dynamism that underpins capital competences (workforce depth, consumer bases, cultural demand) also amplifies activity levels and mobility, reinforcing the persistence of high PM2.5 exposure in these places. Table 4 Demographic profile by PM2.5 trajectory class Class N (municipalities) Population < 15 (%) Population ≥ 65 (%) Female (%) Density (inh./km²) Trajectory 1 (low PM2.5) 13 11.0% 31.7% 52.2% 32.5 Trajectory 2 (intermediate) 9 12.8% 25.1% 50.4% 89.6 Trajectory 3 (high PM2.5) 7 14.4% 21.6% 52.6% 1519.4 By placing these results alongside the temporal trajectory analysis, the study exposes that PM2.5 pollution in Portugal is not only a matter of meteorology or geography but is intimately tied to the socio-spatial configuration of capital competences. At the intersection of these findings lies a structural paradox in Portugal: the very municipalities that accumulate economic and cultural resources—fostering productivity, employment, and cultural infrastructures—are also those where PM2.5 concentrations remain consistently above the national average. Conversely, structurally peripheral municipalities sustain lower emissions largely because of limited capital endowments, though at the cost of weaker cultural and economic infrastructures. These differentiated trajectories confirm that fine particulate exposure is not randomly distributed but systematically embedded in capital configurations. This pattern underscores an inconsistency of many development processes—where investments that enhance local capacities also reinforce capital externalities—and highlights the need for territorially differentiated policies capable of balancing capital development with the mitigation of pollution burdens. 4. Conclusion In sociological terms, objectified capital designates the material crystallisation of social and economic investments into durable forms such as infrastructures, cultural facilities, and built environments, which inscribe hierarchies of power and resources into the physical landscape (Bourdieu, 2018a; Gomez, Miño, Pereira, et al., 2025). At the national scale of Portugal, such objectifications are related with the fiscal capacities and economic and cultural expenditures of municipalities, which condense into patterned competences for development, employment, and social reproduction. When juxtaposed with the temporal trajectory classes of PM2.5, the analysis reveals how differentiated capital endowments become inscribed into spatial regimes of environmental burden. Municipalities with higher levels of economic and cultural investment tend to anchor more intense trajectories of particulate exposure, while those with weaker investment profiles register lower exposures but at the expense of curtailed opportunities and thinner infrastructures of provision. In summary, these dynamics can be read as objectified capital externalities . Positive externalities emerge where investment in infrastructures and cultural facilities generates spill-overs in the form of economic opportunities, knowledge circulation, and improved services. Yet the same investments frequently yield negative outputs, such as heightened air pollution, congestion, or environmental degradation, whose costs are externalised onto local populations and ecosystems. The Portuguese analysis thus reveals the paradox of capital allocation: the very municipal investments that consolidate social and economic capacities also entrench ecological externalities in the form of persistent PM2.5 exposures. Effective policy strategies must therefore confront this duality, developing territorially differentiated approaches that reconcile the productivity and cultural benefits of municipal investment with the mitigation of their environmental costs. Capital competences today must encompass structured capacities to produce, but also to absorb, and mitigate spillovers through the mobilisation of economic and cultural resources. By integrating collectively shared externalities into the analysis of the objectified state of capital, the study reframes municipal PM2.5 trajectories not only as environmental outcomes but as indicators of how competences crystallise in space, interact with meteorological regimes, and structure inequalities. This conceptual move enables comparison between municipalities not only in terms of the capital they hold but in terms of how much they generate, displace, or internalise externalities. It thereby shifts the focus beyond a binary of ‘capital versus environment,’ showing instead that environmental burdens and benefits are intrinsic to the very processes by which capital is allocated, mobilised, and contested at the subnational scale. Indeed, capital externalities are often invoked yet seldom specified to localised levels. In this article they were treated not as an abstract residue of market failure, but as the objectified by-products of capital development—effects that persist in space and time because they are embedded in infrastructures, settlement patterns, and investment regimes. Conceptually, this reframing matters because it links externalities to the composition and volumes of capital at subnational scales, where economic and cultural resources take material form in firms, transport and port systems, cultural venues, educational pipelines, and housing markets. Empirically, the study operationalised this idea by using PM2.5 as a negative capital externality and by situating its trajectories (2019–2025) within the differentiated municipal endowments of economic and cultural investments (2024). Measuring these processes is challenging: externalities are multi-source and path-dependent; exposure varies with meteorology; and capital competences evolve under institutional inertia. The article addressed these problems by: (i) modelling trajectories rather than single-year snapshots of PM2.5 and wind configurations; and (ii) treating municipal investment as structured, slow-moving capacity rather than short-term volatility, thereby matching the temporalities of accumulation and exposure in a relational design. 4.1. Contributions. First, the paper demonstrates that PM2.3 exposure in Portugal can be organised into relatively stable trajectory classes, each characterised by distinctive means and variability, and with clear seasonal signatures linked to national wind regimes. These are not mere gradients of magnitude: they are spatial–temporal types—with a low-pollution inland belt, an intermediate littoral corridor, and high-pollution enclaves in archipelagos, tourism nodes, and port-industrial conurbations—that jointly expose how meteorology and spatial development intertwine. This typology cautions against explanations relying solely on geography or solely on activity levels; it shows instead that PM2.5 emissions reflects the co-production of environmental and socio-economic structures. Second, the study links these trajectories to municipal investment profiles —a composite of economic (e.g., GVA, turnover, employment, tourism receipts) and cultural (e.g., per-capita cultural expenditure, libraries, heritage, performing arts, education) dimensions, standardised to enable comparison. By selecting a representative subset of municipalities nearest to class centroids—while ensuring district diversity—the analysis avoids cherry-picking outliers and instead reads typical cases for each trajectory. This design strengthens internal validity by directing the analysis toward patterns that typify classes rather than idiosyncrasies of single places. In summary, the juxtaposition of trajectory classes with z-scored investment profiles yields a reasoned argument: where economic and cultural capital accumulate most intensely—in port-industrial zones, tourism nodes, and archipelagos with concentrated flows—PM2.5 exposure is higher and more persistent; where economic density is limited and cultural infrastructures are thinner, exposure remains lower but at the cost of curtailed opportunities and weaker public investments. Between these poles lie transitional municipalities that register moderate emissions alongside selective cultural linkages and tourism–hospitality couplings. The correlation network consolidates this contribution: high centrality of GVA, turnover, and employment; robust ties to cultural–creative and heritage expenditures; and strong tourism–hotel links—all consistent with an integrated regime of production, services, and cultural objectifications. In short, fine particulates map onto capital arrangements. 4.2. Implications for policy—targeting the subnational scale. If exposure is structurally embedded, mitigation must be spatially differentiated . The evidence suggests at least four families of possible interventions: Place-specific emission reduction in high-exposure regimes (Trajectory 3). Prioritising port-industrial corridors and archipelagos where dense economy–culture linkages coincide with persistent peaks. Practical levers include clean-fuel standards for shipping and logistics, low-emission zones linked to freight scheduling, and electrification incentives for tourism and cultural venues with high visitor throughput. The goal is not to de-accumulate capital, but to re-instrument it so that objectified capacities do not reproduce objectified harms. Transitional packages for intermediate regimes (Trajectory 2). Here the diagnostic is a hybrid coupling of productive activity, tourism, and selective cultural investment. Policy can amplify co-benefits: mobility demand management during seasonal peaks; performance-based grants for cultural facilities that adopt low-emission operations; and targeted support for firms aligning productivity upgrades with emissions abatement. Opportunity-enhancing investments for low-exposure regimes (Trajectory 1). In Portugal, lower pollution partly reflects limited economic density and more fragmented cultural infrastructures. Equity requires that decoupling does not come at the price of durable under-investment. Place-based cultural and educational programmes (libraries, heritage, secondary-education completion) can be scaled while preserving low exposure through stringent siting rules, small-scale distributed generation, and clean mobility links to regional centres. Meteorology-aware planning everywhere. The monthly signatures and seasonal wind roses highlight when and where dispersion is weakest. Subnational authorities should align emission caps, traffic restrictions, port operations, and biomass-burning controls with forecast windows of adverse dispersion (e.g., late winter inversions, autumn peaks), making mitigation temporal as well as spatial. These suggestions share a common premise: because externalities are co-produced with capital competences, interventions must treat economic and cultural infrastructures as levers for mitigation, not as background conditions. Policies that merely relocate emissions risk displacing harm rather than transforming it. 4.3. Limitations and directions for future work. Two types of limitations are salient. Data and scale. The CAMS reanalysis provides harmonised coverage but at ~0.4° resolution; while municipal averages mitigate this constraint, sub-municipal gradients and micro-environments (street canyons, industrial plumes) remain unresolved. Integrating higher-resolution products, validated monitors, and local emission inventories would refine attribution. Temporal alignment and endogeneity. Investment indicators refer to 2024 and are assumed to exhibit structural inertia. This is theoretically consistent with capital accumulation, but longitudinal municipal financial series would improve causal inference and help distinguish cyclical shocks from trend capacities. Finally, the approach is portable: any country with municipal boundaries, basic fiscal/cultural statistics, and access to reanalysis products can replicate and improve the workflow. The key is to preserve the trajectory logic (time-series clustering), the representative sub-sample (centroid proximity with territorial diversity), and the cross-domain pairing (environmental trajectories × capital indicators). This enables comparative research on how different development models (port economies, tourism regions, inland agricultural belts) produce distinctive externality regimes and how cultural investments could be mobilised as mitigation partners. Placed alongside the temporal analysis, the investment profiles and correlation structures lead to a simple but incommode conclusion: PM2.5 in Portugal is not only weather and not only geography—it is a social cartography of capital investments. Municipalities that most successfully develop economic and cultural competences also face the hardest task of decoupling those competences from particulate exposure; those with thinner endowments face the opposite challenge of building opportunity without importing risk. Recognising this paradox of development shifts the policy question from “How much growth?” to “ Which investments, where and when , and with what environmental approach?” The answer, the article results suggest, lies in territorially differentiated strategies that treat cultural and economic possibilities as protagonists of mitigation, aligning cycles of investment with cycles of air quality so that the gains of development do not come with the costs of persistent PM2.5 burdens. Declarations 5. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. All authors have approved the submitted version of the manuscript. 7. CRediT authorship contribution statement Raimundo Elias Gomez : Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing - Original draft, Writing - Review & editing, Visualization, Project administration, Funding acquisition. Maria Gabriela Miño : Conceptualization, Methodology, Investigation, Writing - Original draft, Writing - Review & editing, Validation. 9. Ethical Statement The authors confirm that this work is original, has not been published elsewhere, and is not currently under consideration by another journal. All authors have read, understood, and have complied as applicable with the statement on "Ethical responsibilities of Authors" as found in the “Instructions for Authors”. 6. Funding sources This project has received funding from the European Union’s Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 101102978 — SSpaceGX: "Social Space and Nature Conservation in the Transboundary Biosphere Reserve Gerês-Xurés (Portugal/Spain). Author Contribution *Raimundo Elias Gomez:* Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing - Original draft, Writing - Review & editing, Visualization, Project administration, Funding acquisition. *Maria Gabriela Miño:* Conceptualization, Methodology, Investigation, Writing - Original draft, Writing - Review & editing, Validation. Data Availability PM2.5 and wind data are publicly available through Copernicus Atmosphere Monitoring Service (CAMS) and ERA5 reanalysis. The complete database and analytical dashboards are accessible at: https://doi.org/10.5281/ZENODO.16981588 (economic and cultural capital data), https://doi.org/10.5281/ZENODO.16981258 (PM2.5 spatial distribution), and https://doi.org/10.5281/ZENODO.16980568 (wind roses analysis). Municipal socioeconomic data are available from Instituto Nacional de Estatística Portugal (INE-PT) portal. References Alam, M. J., Karim, I., & Zaman, S. U. (2025). Seasonal dynamics and trends in air pollutants: A comprehensive analysis of PM2.5, NO2, CO, SO2 and O3 in Houston, USA. Air Quality, Atmosphere, & Health . https://doi.org/10.1007/s11869-025-01790-9 Ashraf, S., Pausata, F. S. R., Leroyer, S., Stevens, R., & Munoz-Alpizar, R. (2025). Impact of reduced anthropogenic emissions associated with COVID-19 lockdown on PM2.5 concentration and canopy urban heat island in Canada. GeoHealth , 9 (2), e2023GH000975. https://doi.org/10.1029/2023GH000975 Barbosa, J. V., Nunes, R. A. O., Alvim-Ferraz, M. C. M., Martins, F. G., & Sousa, S. I. V. (2024). Health and economic burden of wildland fires PM2.5-related pollution in Portugal - A longitudinal study. Environmental Research , 240 (Pt 1), 117490. https://doi.org/10.1016/j.envres.2023.117490 Bille, T., & Honoré, S. (2025). Cultural capital externalities: Causal evidence from a Danish ticket scheme for theatres. Kyklos: International Review for Social Sciences . https://doi.org/10.1111/kykl.12469 Bourdieu, P. (2018a). The forms of capital. In The sociology of economic life (pp. 78–92). Routledge. https://www.taylorfrancis.com/chapters/edit/10.4324/9780429494338-6/forms-capital-pierre-bourdieu Bourdieu, P. (2020). The field of power and the division of the labour of domination: Handwritten notes for the 1985-1986 collège de France lectures. In Researching Elites and Power (pp. 33–44). Springer International Publishing. https://doi.org/10.1007/978-3-030-45175-2_3 Bourdieu, P. (2018b). Social Space and the Genesis of Appropriated Physical Space: FORUM. International Journal of Urban and Regional Research , 42 (1), 106–114. https://doi.org/10.1111/1468-2427.12534 C3S. (2018). ERA5 hourly data on single levels from 1940 to present [Dataset]. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). https://doi.org/10.24381/CDS.ADBB2D47 CAMS. (2021). Copernicus Atmosphere Monitoring Service: Global atmospheric composition forecasts [Dataset]. ECMWF. https://doi.org/10.24381/04A0B097 CFP. (2025). Conselho das Finanças Públicas, Portugal. https://www.cfp.pt/pt Choi, W., Ho, C.-H., & Lee, Y. (2024). Temporal pattern classification of PM2.5 chemical compositions in Seoul, Korea using K-means clustering analysis. The Science of the Total Environment , 927 (172157), 172157. https://doi.org/10.1016/j.scitotenv.2024.172157 CNC. (2025). Comissão de Normalização Contabilística, Portugal. https://www.cnc.min-financas.pt/sncap2017.html Corda, M. O., Charalampous, P., Haagsma, J. A., Assunção, R., & Martins, C. (2024). Mortality burden of cardiovascular disease attributable to ambient PM2.5 exposure in Portugal, 2011 to 2021. BMC Public Health , 24 (1), 1188. https://doi.org/10.1186/s12889-024-18572-0 Espinoza-Guillen, J. A., Alderete-Malpartida, M. B., Roncal-Romero, F. D., & Vilcanqui-Sarmiento, J. C. (2025). Identification of particulate matter (PM10 and PM2.5) sources using bivariate polar plots and k-means clustering in a South American megacity: Metropolitan Area of Lima-Callao, Peru. Environmental Monitoring and Assessment , 197 (3), 226. https://doi.org/10.1007/s10661-025-13696-1 EU/ESA/Copernicus. (2024). Sentinel-5P NRTI AER AI: Near Real-Time UV Aerosol Index . Google for Developers. https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S5P_NRTI_L3_AER_AI Gomez, R. E. (2021). Espacio social y clases sociales en Posadas, Argentina. Estudios Demograficos y Urbanos , 36 (3), 865–889. https://doi.org/10.24201/edu.v36i3.1983 Gomez, R. E. (2025a). MCA1. Economic and cultural capital by PM2.5 trajectory classes . Zenodo. https://doi.org/10.5281/ZENODO.16981588 Gomez, R. E. (2025b). Spatial Distribution of Fine Particulate Matter in Portugal (2019–2025) . Zenodo. https://doi.org/10.5281/ZENODO.16981258 Gomez, R. E. (2025c). Wind Roses II (2019–2025): Interactive analysis of PM2.5 and wind direction in Portuguese municipalities . Zenodo. https://doi.org/10.5281/ZENODO.16980568 Gomez, R. E., & Miño, M. G. (2025). Extensive objectified footprints: A multidimensional approach to spatial inequalities. City and Environment Interactions , 28 (100226), 100226. https://doi.org/10.1016/j.cacint.2025.100226 Gomez, R. E., Miño, M. G., Hojman, G. D., Arellano, D. M., Cossi, C. A., & Sosa, L. (2025). Tracing Emplaced Capital Footprints in Arcos de Valdevez, Portugal. Comparative Sociology , 24 (3), 410–444. https://doi.org/10.1163/15691330-bja10137 Gomez, R. E., Miño, M. G., Pereira, V. B., & Jardón, C. M. (2025). Spatial distribution and classification of objectified capital in the Gerês-Xurés Transboundary Biosphere Reserve. Frontiers in Environmental Economics , 4 , 1463694. https://doi.org/10.3389/frevc.2025.1463694 INE-PT. (2025). Instituto Nacional de Estatística, Portugal. Caraterização sócio – económica dos municípios. https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_doc_municipios_cse&xlang=pt Işıkara, G. (2023). Capitalism, economics, and externalities: What are externalities external to? Capitalism Nature Socialism , 34 (2), 40–56. https://doi.org/10.1080/10455752.2023.2192954 Kouyate, M., Arola, A., Benedictow, A., Bennouna, Y., Blake, L., Bouarar, I., Cuevas, E., Errera, Q., Eskes, H. J., Griesfeller, J., Basart, S., Kapsomenakis, J., Langerock, B., Mortier, A., Pitkänen, M. R. A., Pison, I., Ramonet, M., Richter, A., Schoenhardt, A., … Zerefos, C. (2024). Validation report of the CAMS near-real-time global atmospheric composition service Period June – August 2023 . Copernicus Atmosphere Monitoring Service. https://doi.org/10.24380/A7AS-QAEG Lelieveld, J., Evans, J. S., Fnais, M., Giannadaki, D., & Pozzer, A. (2015). The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature , 525 (7569), 367–371. https://doi.org/10.1038/nature15371 Marom, N. (2014). Relating a city’s history and geography with Bourdieu: One hundred years of spatial distinction inTelAviv. International Journal of Urban and Regional Research , 38 (4), 1344–1362. https://doi.org/10.1111/1468-2427.12027 Oke, T. R., Mills, G., Christen, A., & Voogt, J. A. (2017). Air Pollution. In Urban Climates (pp. 294–331). Cambridge University Press. https://doi.org/10.1017/9781139016476.012 Prapassonpithaya, P., & Jinsart, W. (2025). The relationship between PM2.5 levels and COVID-19 after major outbreak waves: A Bangkok metropolitan area study. Environmental Challenges (Amsterdam, Netherlands) , 101255 , 101255. https://doi.org/10.1016/j.envc.2025.101255 Rodriguez Avellaneda, F., Chacón-Montalván, E. A., & Moraga, P. (2025). Multivariate disaggregation modeling of air pollutants: a case-study of PM2.5, PM10 and ozone prediction in Portugal and Italy. The American Statistician , 1–21. https://doi.org/10.1080/00031305.2025.2537055 Seinfeld, J. H., & Pandis, S. N. (2016). Atmospheric chemistry and physics (3rd ed.). Wiley-Blackwell. Siblot, Y., Hugrée, C., & Pereira, V. B. (2024). Comparing countries, exporting classifications, surpassing methodological nationalism: Class, gender, and education gaps in and between France and Portugal. The Sociological Review . https://doi.org/10.1177/00380261241299991 Yu, Y. T., Zhang, S., Xiang, S., & Wu, Y. (2025). Socioeconomic inequalities in PM2.5 exposure and local source contributions at community scales using hyper-localized taxi-based mobile monitoring in Xi’an, China. Environmental Science & Technology , 59 (14), 7222–7234. https://doi.org/10.1021/acs.est.4c11385 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7790529","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":532554555,"identity":"3d0d2b02-be0d-4358-88ee-3622d9a1786a","order_by":0,"name":"Raimundo Elias GOMEZ","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYPACGxmStaTxMLDBOcxEaTlMghbz/sVHN3z4c56HX7734ccfDIfzGKT7D+DVInPjWdrNGTy3eSTb2I2leRgOFzPIHMZvi4TEGbPbPBK3eQyOsTFIA12Y2CCRTISWPwbneOyPsTH//EGUFv4es9sMCQd4DNjY2CR4iLOFLe1mz4FkHoljaWzWPAbpxWwyhw0I2HL42I0ff+zk+JuPMd/8UWGdxy/d+ICANQnIPAOGBDYJ/BoYGPgPoPITGAhqGQWjYBSMgpEGAFfWPhsmRMgvAAAAAElFTkSuQmCC","orcid":"","institution":"University of Porto (Portugal) / National Scientific and Technical Research Council","correspondingAuthor":true,"prefix":"","firstName":"Raimundo","middleName":"Elias","lastName":"GOMEZ","suffix":""},{"id":532554556,"identity":"1980452f-d283-41b4-bd93-47689540cfe6","order_by":1,"name":"Maria Gabriela MIÑO","email":"","orcid":"","institution":"National Scientific and Technical Research Council","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Gabriela","lastName":"MIÑO","suffix":""}],"badges":[],"createdAt":"2025-10-06 10:38:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7790529/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7790529/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97019730,"identity":"23e00a81-3368-4e78-a436-fde7b1cc5e57","added_by":"auto","created_at":"2025-11-28 18:00:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":159207,"visible":true,"origin":"","legend":"\u003cp\u003eTypology of Average Annual PM2.5 Trajectories. January 2019 to April 2025.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/ddd938499925942a4bcfed69.png"},{"id":97139738,"identity":"2eeb0c10-dee1-4fb6-a47b-7fe9322c598a","added_by":"auto","created_at":"2025-12-01 10:02:14","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":194393,"visible":true,"origin":"","legend":"\u003cp\u003eTypology of Average Monthly PM2.5. January 2019 to April 2025\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/08aa3a49ca78275d8993c4f1.png"},{"id":97138050,"identity":"349d22f3-38bc-4507-adea-0d31efdc1b92","added_by":"auto","created_at":"2025-12-01 09:58:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":291480,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of PM2.5 trajectory classes in Portugal (2019—2025).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/bbab141eb041832fe4809417.png"},{"id":97019734,"identity":"f570cea8-900d-49e6-b579-4a0f99928858","added_by":"auto","created_at":"2025-11-28 18:00:12","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":307883,"visible":true,"origin":"","legend":"\u003cp\u003eSeasonal wind roses of PM2.5 concentrations in Portugal (2019–2025).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/b4eaa17ce958a81ccb332492.png"},{"id":97019729,"identity":"a9815a60-fc51-4b8e-95fd-09956fbf3fd4","added_by":"auto","created_at":"2025-11-28 18:00:12","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":179940,"visible":true,"origin":"","legend":"\u003cp\u003eEconomic and cultural investment profiles by PM2.5 trajectory class (z-scores)\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/b9e76bf76b1c5ac1bf7652c2.png"},{"id":97019736,"identity":"2a4e0502-42a7-45da-a5d6-467c1481124e","added_by":"auto","created_at":"2025-11-28 18:00:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":250499,"visible":true,"origin":"","legend":"\u003cp\u003eRadar plots of economic and cultural investment indicators across three PM2.5 trajectory classes (2019–2025)\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/3a7c63e3738f880669db3704.png"},{"id":97138337,"identity":"353418f6-8299-450e-b425-06d818cbd170","added_by":"auto","created_at":"2025-12-01 09:58:46","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":187703,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation network between economic and cultural investment indicators (|r| ≥ 0.5).\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/0dab527ece209826bb85014d.png"},{"id":100152901,"identity":"437d8aad-33cd-451f-83be-b9f5587b466e","added_by":"auto","created_at":"2026-01-13 13:40:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2506233,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7790529/v1/c5e9cb20-6612-4ea2-a10d-fb11763a11e5.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"PM2.5 Trajectory Classes and Spatial Investment Patterns in Portugal","fulltext":[{"header":"Highlights","content":"\u003cp\u003e- PM2.5 trajectories reveal three distinct pollution classes across Portugal\u003c/p\u003e\u003cp\u003e- Dense economic-cultural zones show 2\u0026times; higher PM2.5 than peripheral areas\u003c/p\u003e\u003cp\u003e- Low pollution areas have limited economic and cultural infrastructure\u003c/p\u003e\u003cp\u003e- Capital externalities concentrate where investments are highest\u003c/p\u003e\u003cp\u003e- Seasonal wind patterns redistribute PM2.5 across municipal boundaries\u003c/p\u003e"},{"header":"1. Introduction","content":"\u003cp\u003eFine particulate matter (PM2.5) it\u0026rsquo;s an environmental health risks in Portugal and contemporary societies, with consequences that extend well beyond urban cores (Barbosa et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Corda et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Rodriguez Avellaneda et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Most studies addressing air pollution in Portugal have adopted national or metropolitan perspectives, often giving limited attention to related dimensions at the municipal scale. This point is important because pollution inequalities are shaped by the uneven distribution of resources, capacities, and priorities among localities, raising questions about the regional configurations of capital externalities (Bille \u0026amp; Honor\u0026eacute;, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Işıkara, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Yu et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Yet, municipalities represent a crucial locus for the articulation of both economic and cultural capital investments\u0026mdash;factors that can profoundly influence the production and persistence of environmental risks. The differentiated accumulation of investments, the legacy of spatial development, and the orientation of local policies contribute to a mosaic of PM2.5 exposures that is neither accidental nor easily explained by macro-level trends alone.\u003c/p\u003e\u003cp\u003eThis article aims to advance the understanding of these dynamics by: (1) mapping the spatial distribution and temporal evolution of PM2.5 across all 308 Portuguese municipalities between 2019 and 2025; (2) identifying and classifying distinct PM2.5 trajectories at the municipal level; and (3) examining how these pollution trajectories correspond with municipal profiles of economic and cultural investment, focusing on the most recent consolidated year with complete fiscal data (2024). The study is grounded in a relational perspective, which posits that spatial inequalities in environmental exposure are intrinsically linked to the uneven configuration of capital competences and investment regimes (Bourdieu, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e; Gomez, Mi\u0026ntilde;o, Hojman, et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Gomez \u0026amp; Mi\u0026ntilde;o, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Marom, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Siblot et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). By integrating geospatial air quality data and environmental estimates with detailed fiscal and cultural indicators, the analysis exposes how municipal capital allocation aligns with differentiated vulnerabilities to air pollution, thereby providing an empirical basis for more targeted decoupling strategies and policy design. Following this approach, capital is understood not only as an economic dimension but also as cultural investments. Both forms exist in distinct states: embodied as personal dispositions, institutionalised as formalised investments, and objectified as infrastructures, pollution footprints, or the spatial distribution of populations and services. In its objectified state, capital represents the durable inscription of resources and power but also the consolidation of inequality, visible not only in demographic and cultural facilities but also in capital externalities such as PM2.5 (Bourdieu, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Gomez, Mi\u0026ntilde;o, Pereira, et al., 2025).\u003c/p\u003e\u003cp\u003eThe analysis focuses on the evolution and distribution of PM2.5 as an externality\u0026mdash;arising when the objectification of industrial activities, through material manifestations and infrastructural outcomes, generates unintended/uncontrolled effects beyond the intentions or interests of the original investors. While externalities may manifest as either positive or negative consequences, emissions of PM2.5 are globally recognized as harmful to health and disproportionately affect specific places and populations. From a relational perspective, these emissions are not random but are structurally linked to the spatial differentiation of investment regimes. The forms, intensities, and spatial configurations of economic and cultural investment correlate directly with the objectified landscape, generating new inequalities, as well as heightened vulnerabilities to environmental risks. Accordingly, the differentiated trajectories of PM2.5 across Portuguese municipalities are analysed as objectified outcomes of situated strategies and capital allocation. Accordingly, the article addresses the following question: \u003cem\u003eHow are the temporal trajectories of PM2.5 exposure (2019\u0026ndash;2025) distributed across Portuguese municipalities, and to what extent do these patterns correspond with major economic and cultural investments from the most recent consolidated fiscal year (2024)\u003c/em\u003e? This structural approach expands the analytical scope of conventional studies on capital externalities. Elucidating the links between PM2.5 patterns and municipal investment profiles is essential for guiding policy interventions aimed at reducing environmental spatial inequalities, protecting public health, and deepening understanding of the social cost associated with fine particulate emissions.\u003c/p\u003e"},{"header":"2. Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Area of Study\u003c/h2\u003e\u003cp\u003eThe analysis covers the entire territory of Portugal, encompassing all 308 municipalities, including those located in the autonomous regions of Madeira and the Azores. The municipal scale offers an appropriate level of granularity for investigating the spatial differentiation of both air pollution and investment dynamics.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Data Sources and Variables\u003c/h2\u003e\u003cp\u003ePM2.5 monthly and annual mean concentrations for the period from January 2019 to March 2025 were obtained from the Copernicus Atmosphere Monitoring Service (CAMS) Near-Real-Time reanalysis, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) (CAMS, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This dataset integrates satellite and ground-based measurements, providing harmonised estimates of air quality (EU/ESA/Copernicus, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). All scenes from 2019 to 2025 were accessed and processed via Google Earth Engine. Monthly and annual means were spatially averaged within the official boundaries of Portugal\u0026rsquo;s 308 municipalities, yielding a consistent panel dataset suitable for temporal and spatial comparison.\u003c/p\u003e\u003cp\u003eTo complement the PM2.5 concentrations, wind fields were derived from the ERA5 reanalysis produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) and distributed through the Copernicus Climate Data Store (CAMS, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Kouyate et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The ERA5 dataset provides hourly estimates of meteorological variables, including the horizontal components of wind at 10 m above ground level: one oriented east\u0026ndash;west (u) and the other north\u0026ndash;south (v). These two dimensions together describe wind speed and direction and are generated by assimilating satellite and in-situ observations into a global numerical weather prediction model (C3S, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Monthly and annual averages were calculated with Google Earth Engine from the complete series (from January 2019 to March 2025) and spatially aggregated across the 308 Portuguese municipalities. Both the PM2.5 and wind databases can be accessed online, together with an analysis of their spatial and temporal distribution (Gomez, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003ec, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003eAll municipal-level socioeconomic data used in this study were obtained from statistical sheets published by the National Statistics Institute of Portugal (Instituto Nacional de Estat\u0026iacute;stica, INE)\u0026mdash;specifically the Municipal Sheet \u0026ndash; Socioeconomic and Environmental Context (\u003cem\u003eFicha Municipal, Contexto Socioecon\u0026oacute;mico e Ambiental\u003c/em\u003e, CSE) and its companion series Information Return to Respondents (\u003cem\u003eRetorno de Informa\u0026ccedil;\u0026atilde;o aos Respondentes\u003c/em\u003e)\u0026mdash;available through the INE Municipal Data Portal (\u003cem\u003ePortal de Dados Municipais do INE\u003c/em\u003e) (INE-PT, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This year was selected as it provides the most recent, fully consolidated, and publicly available municipal account data. Such temporal delimitation reflects the institutional assumption that municipal economic and cultural investments are subject to structural inertia and shaped by programmatic planning cycles rather than short-term fluctuations (CFP, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; CNC, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The complete database, together with an analytical dashboard, can be accessed online (Gomez, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003ea).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eConsidered variables, abbreviations, investment dimensions, and definitions.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAbbrev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDimension\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eINE definition\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross value added (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGVA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEconomic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eValue of goods\u0026thinsp;+\u0026thinsp;services produced minus intermediate consumption (official \u0026ldquo;Valor acrescentado bruto\u0026rdquo;).\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBizVol\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEconomic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTotal value of sales of goods and services (\u0026ldquo;Volume de neg\u0026oacute;cios\u0026rdquo;) realised during the reference year.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of firms (registered units)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFirms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEconomic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eCount of active enterprises recorded in the Statistical Business Register.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNewly registered legal entities (annual)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNewEnt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEconomic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAnnual number of newly incorporated collective persons (companies, cooperatives, etc.).\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTourism revenue (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTourRev\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEconomic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eReceipts from accommodation and related tourist services reported by municipalities.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMedian housing price (\u0026euro;/m\u0026sup2;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHousePrice\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEconomic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMedian \u0026euro;/m\u0026sup2; of dwellings transacted, based on notarised deeds (\u003cem\u003eEstat\u0026iacute;sticas de pre\u0026ccedil;os da habita\u0026ccedil;\u0026atilde;o\u003c/em\u003e).\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePer-capita cultural expenditure (\u0026euro;/inh.)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCultExp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCultural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMunicipal budget outlay on culture divided by resident population.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLibrary \u0026amp; archives expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLibExp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCultural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMunicipal spending on libraries and archival services.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePerforming-arts expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eArtsExp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCultural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMunicipal spending on music, theatre, dance and related live-performance activities.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross secondary-education enrolment rate (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSecEnroll\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCultural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRatio of students enrolled in secondary education to the resident population of the official age group.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSecondary-education completion rate (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSecCompl\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCultural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eShare of students completing secondary schooling within the expected period.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInverse basic-education dropout rate (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInvDrop\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCultural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e100 \u0026ndash; (dropout\u0026thinsp;+\u0026thinsp;retention rate) in compulsory basic education.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn line with a relational approach, both economic and cultural investments are understood as key structuring forces shaping the spatial distribution of resources, power, and opportunity (Bille \u0026amp; Honor\u0026eacute;, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Bourdieu, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Gomez, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These dimensions are operationalised using composite indicators that capture municipal investment strategies. Variable selection was guided by criteria of completeness, discriminative power, and conceptual relevance. The final set of indicators represents economic dimensions (gross value added, business turnover, number of firms, newly registered legal entities, tourism revenue, median housing price) and cultural dimensions (per-capita cultural expenditure, library and archives spending, performing-arts spending, secondary education enrolment and completion rates, and inverse dropout rate) (Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). All variables were standardised as z-scores to ensure comparability across municipalities.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Analytical Procedures\u003c/h2\u003e\u003cp\u003eTemporal clustering methods with k-means are widely adopted in environmental research for detecting latent patterns and grouping cases within longitudinal pollution datasets (Choi et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Espinoza-Guillen et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). By partitioning municipalities according to the similarity of their PM2.5 time series, k-means provides a robust preliminary classification that facilitates the investigation of how distinct pollution trajectories relate to wind patterns and underlying municipal investment profiles. In a second stage, to enable a focused and interpretable comparison, a subsample of 29 municipalities was selected to represent the diversity of economic and cultural capital allocation on each trajectory class. Within each class, municipalities were ranked according to their Euclidean distance from the class centroid in the multidimensional PM2.5 space, ensuring that the closest decile\u0026mdash;those most representative of their respective class\u0026mdash;was retained (Class 1: 9 municipalities; Class 2: 10 municipalities; Class 3: 10 municipalities). This selection strategy not only preserves the internal coherence of each class but also accounts for district-level diversity by limiting inclusion to one municipality per administrative district within each class. The slight deviation from a theoretical sample of 30 reflects this balancing criterion. The resulting representative sample forms the basis for a detailed comparative analysis of investment structures across distinct PM2.5 trajectory classes.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e3.1 How are the annual trajectories of PM2.5 exposure (2019\u0026mdash;2025) spatially distributed across Portugal?\u003c/h2\u003e\u003cp\u003eBetween 2019 and 2020, PM2.5 concentrations fell sharply across all trajectory classes in Portugal, mirroring the decline in industrial and transport activities that occurred in many cities of the globe during the COVID-19 lockdown period and the associated, albeit temporary, reduction in anthropogenic emissions (Ashraf et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Prapassonpithaya \u0026amp; Jinsart, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). From 2020 onward, PM2.5 levels increased again, with pronounced fluctuations, particularly in the high pollution trajectory (Traj 3), as economic activities and emissions resumed. Throughout the observation period, all classes displayed closely aligned seasonal cycles, underscoring the significant influence of meteorological and climatic drivers\u0026mdash;including temperature inversions, humidity, and wind\u0026mdash;on PM2.5 dispersal and accumulation at the national scale (Alam et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAfter the k-means clustering, the three identified trajectory classes correspond to distinct pollution regimes, each characterised by specific mean concentrations and variability.\u003c/p\u003e\u003cp\u003e\u003cb\u003eTrajectory\u003c/b\u003e \u003cb\u003e1 \u0026ndash; Low Pollution Trajectories\u003c/b\u003e. Municipalities within this class exhibit mean PM2.5 levels ranging from approximately 5.44 to 6.85 \u0026micro;g/m\u0026sup3; and coefficients of variation (CV) between 0.28 and 0.43, indicative of moderate variability. Examples are Trancoso (Guarda) with 5.75 \u0026micro;g/m\u0026sup3; (CV 0.39), Moura (Beja) at 6.00 \u0026micro;g/m\u0026sup3; (CV 0.32), and Melga\u0026ccedil;o (Viana do Castelo) with 6.27 \u0026micro;g/m\u0026sup3; (CV 0.43) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These locations typically correspond to areas with limited urbanisation or industrial activity, reflected in their lower PM2.5 levels and moderate internal dispersion. \u003cb\u003eTrajectory\u003c/b\u003e \u003cb\u003e2 \u0026ndash; Intermediate Pollution Trajectories.\u003c/b\u003e This group is defined by municipalities where average PM2.5 concentrations lie between 6.70 and 8.10 \u0026micro;g/m\u0026sup3;, and CVs fall between 0.24 and 0.33. Notable examples are Batalha (Leiria) with 7.78 \u0026micro;g/m\u0026sup3; (CV 0.27), Cantanhede (Coimbra) at 8.10 \u0026micro;g/m\u0026sup3; (CV 0.30), and Montijo (Set\u0026uacute;bal) at 7.50 \u0026micro;g/m\u0026sup3; (CV 0.25) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This class represents an intermediate pollution scenario, with lower internal variability compared to Class 1. \u003cb\u003eTrajectory\u003c/b\u003e \u003cb\u003e3 \u0026ndash; High Pollution Trajectories\u003c/b\u003e. Municipalities in this category register PM2.5 concentrations from about 8.63 to 11.07 \u0026micro;g/m\u0026sup3; and CVs between 0.25 and 0.37. Examples include Amadora (Lisboa) at 9.97 \u0026micro;g/m\u0026sup3; (CV 0.26), Esposende (Braga) at 10.64 \u0026micro;g/m\u0026sup3; (CV 0.25), and Vila do Conde (Porto) at 11.07 \u0026micro;g/m\u0026sup3; (CV 0.25) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These municipalities maintain persistently high PM2.5 levels with moderate internal variation, typically associated with more urbanised or industrialised contexts.\u003c/p\u003e\u003cp\u003eThe seasonal progression of air pollution between 2019 and 2025, as indicated by average monthly PM2.5 concentrations, highlights both commonalities and distinctions among the three trajectory classes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). \u003cem\u003eTrajectory 1\u003c/em\u003e, representing the lowest pollution levels, follows a regular cyclical trend, reaching a moderate peak in March (7.2 \u0026micro;g/m\u0026sup3;) before decreasing through late autumn and early winter, with a trough in November (4.5 \u0026micro;g/m\u0026sup3;). This suggests a marked responsiveness to seasonal emission sources, potentially linked to heating or biomass burning activities. \u003cem\u003eTrajectory 2\u003c/em\u003e, characterised by intermediate pollution, records its highest value in February (9.0 \u0026micro;g/m\u0026sup3;), then gradually declines until August (7.1 \u0026micro;g/m\u0026sup3;), followed by a secondary maximum in September (9.8 \u0026micro;g/m\u0026sup3;). This pattern may be attributed to atmospheric inversions after summer or the lagged effect of emissions. \u003cem\u003eTrajectory 3\u003c/em\u003e, with consistently elevated pollution, sustains high PM2.5 levels, peaking in February (11.7 \u0026micro;g/m\u0026sup3;) and again in September (10.0 \u0026micro;g/m\u0026sup3;). This trajectory reflects enduring structural sources of pollution, such as industry, transport, and dense populated areas. The reduction in PM2.5 for Classes 2 and 3 from May to August likely corresponds to decreased emissions during the summer holiday period and more favourable meteorological conditions for pollutant dispersal. Overall, seasonality is apparent across all trajectories, but the magnitude and timing of peaks differ, emphasising the importance of regionally and temporally adapted mitigation measures.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe cartographic rendering of the three latent trajectory classes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) discloses the tripartite geography of PM2.5 dynamics that cuts across Portugal\u0026rsquo;s administrative mosaic. A broad, quasi-continuous belt of Class 1 trajectory (blue) extends from the high-relief Minho-Tr\u0026aacute;s-os-Montes frontier through the Beira and Alto Alentejo interiors to the Guadiana valley, epitomising trajectories characterised by persistently low background concentrations and muted seasonal amplitudes. Flanking this inland core, a discontinuous Class 2 littoral corridor trajectory (orange) follows the main Atlantic urban\u0026ndash;industrial axis from the Aveiro lagoon to the Sado and lower Guadiana estuaries; here, moderate baseline levels combine with pronounced winter-summer contrasts that mirror the rhythm of coastal traffic, port activity and heating demand. Finally, Class 3 enclaves trajectories (green) punctuate the map in three strategic settings: (i) the autonomous archipelagos of Madeira and the Azores, where marine aerosol, shipping plumes and episodic Saharan dust generate sharp but short-lived peaks; (ii) selected tourism-intensive tracts of the Algarve and western Alentejo littoral; and (iii) port-industrial conurbations such as Porto-Leix\u0026otilde;es and Set\u0026uacute;bal.\u003c/p\u003e\u003cp\u003eThe spatial distribution of PM is a complex phenomenon shaped by multiple environmental and social drivers. Among the unavoidable environmental factors in any spatial approach are the direction, strength, and frequency of winds during specific periods, particularly in certain seasons (C3S, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; CAMS, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Yet beyond their meteorological function, winds operate as redistributive forces that dissolve the apparent boundaries between polluters and those affected. They disperse PM2.5 into zones far removed from its industrial or urban points of origin, thereby transforming what could be perceived as a localised burden into a diffuse, collective externality. This diffusion not only reconfigures the geography of exposure but also the sociology of responsibility: the costs of certain industrial and urban productions are ultimately absorbed by the broader community. Research in atmospheric sciences confirms that wind-mediated transport of fine particles can extend over hundreds of kilometres, redistributing health risks and environmental burdens (Lelieveld et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Seinfeld \u0026amp; Pandis, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Moreover, winds affect urban heat islands, modulate odour corridors, and influence the propagation of noise, further framing the lived experience of environmental externalities in dense urban areas (Oke et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). In this sense, wind functions simultaneously as an ecological and a social driver: it spreads the by-products of production and consumption across populations unequally positioned to mitigate or resist their effects. To illustrate these dynamics, seasonal wind rose diagrams are presented for Portugal (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). They display the frequency distribution of wind direction and speed, with radial spokes indicating the proportion of time the wind blows from each compass point and concentric circles representing cumulative frequency levels. Colour gradients along the spokes correspond to PM2.5 concentrations, enabling the simultaneous interpretation of prevailing wind patterns and the associated particulate matter intensities across different sectors.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAt the national scale, Portugal displays pronounced seasonal shifts in wind patterns, each with its own association to PM2.5 levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). \u003cb\u003eWinter\u003c/b\u003e presents the most balanced directional distribution, with notable frequency peaks towards the northeast (~\u0026thinsp;45\u0026deg;) and southwest (~\u0026thinsp;225\u0026ndash;240\u0026deg;), while the northwest (~\u0026thinsp;315\u0026deg;) remains minimal. In this season, PM2.5 concentrations tend to be higher along the N\u0026ndash;NE and WSW\u0026ndash;SW sectors, and lower towards the southeast and northwest. \u003cb\u003eSummer\u003c/b\u003e is characterised by a strong predominance of E\u0026ndash;SE flows (\u0026asymp;\u0026thinsp;90\u0026ndash;150\u0026deg;), which occur with high persistence but are generally associated with lower PM2.5 intensities. \u003cb\u003eAutumn\u003c/b\u003e exhibits a more variable regime, dominated by E\u0026ndash;SE winds with secondary S\u0026ndash;SW components; during this season, the most pronounced PM2.5 peaks are observed along E\u0026ndash;SE and S sectors. \u003cb\u003eSpring\u003c/b\u003e represents a transitional stage, with an increasing dominance of E\u0026ndash;SE winds and moderate directional variability; PM2.5 concentrations are more evenly spread across sectors but show slightly higher values under N\u0026ndash;NE flows. These seasonal distinctions highlight how both wind direction and intensity relate to the spatial distribution of PM2.5, underscoring the importance of meteorological context in interpreting pollution dynamics. The combined spatial patterns of wind and PM2.5 reflect distinct temporal signatures across Portugal rather than a simple gradient of magnitude. Each municipality exhibits its own distinctive characteristics (that can be explored interactively in (Gomez, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003ec).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e3.2 What is the relationship between PM2.5 exposure trends (2019\u0026mdash;2025) and the 2024 economic and cultural investments in key Portuguese municipalities?\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo examine the relationship between PM2.5 exposure trajectories (2019\u0026ndash;2025) and 2024 economic and cultural investments in key Portuguese municipalities, the study identified a representative subset for detailed analysis. Approximately 10% of the municipalities closest to each trajectory class centroid in PM2.5 space were selected, ensuring that no two municipalities belonged to the same administrative district (NAME_1). This procedure preserved geographic diversity while accurately reflecting the typical exposure patterns of each trajectory class (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSelected representatives municipalities of PM2.5 by trajectory classes.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMUNICIPIO_ID\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNAME_1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eclass\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003emean_pm25\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ecv_pm25\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrancoso (Guarda)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGuarda\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.753121\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.385091\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePenedono (Viseu)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eViseu\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.949848\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.413845\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFund\u0026atilde;o (Castelo Branco)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCastelo Branco\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.205403\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.359353\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOliveira do Hospital (Coimbra)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoimbra\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.543970\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.349714\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eValpa\u0026ccedil;os (Vila Real)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eVila Real\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.809820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.401057\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCarrazeda de Ansi\u0026atilde;es (Bragan\u0026ccedil;a)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBragan\u0026ccedil;a\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.441534\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.368988\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNisa (Portalegre)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePortalegre\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.420625\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.290337\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEstremoz (\u0026Eacute;vora)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026Eacute;vora\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.282494\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.288777\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMa\u0026ccedil;\u0026atilde;o (Santar\u0026eacute;m)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSantar\u0026eacute;m\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.848363\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.283500\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMoura (Beja)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBeja\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.996860\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.316643\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMelga\u0026ccedil;o (Viana do Castelo)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eViana do Castelo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.272007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.432637\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAlcoutim (Faro)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFaro\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.516323\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.294728\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBatalha (Leiria)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLeiria\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.778669\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.267184\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSantar\u0026eacute;m (Santar\u0026eacute;m)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSantar\u0026eacute;m\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.624499\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.243763\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCadaval (Lisboa)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLisboa\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e8.091333\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.237724\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMort\u0026aacute;gua (Viseu)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eViseu\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.868129\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.314529\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCantanhede (Coimbra)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoimbra\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e8.104945\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.302071\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMealhada (Aveiro)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAveiro\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.601800\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.327078\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMontijo (Set\u0026uacute;bal)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSet\u0026uacute;bal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.501571\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.253616\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePonte de S\u0026ocirc;r (Portalegre)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePortalegre\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.703244\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.264534\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOdemira (Beja)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBeja\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.205290\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.265516\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVendas Novas (\u0026Eacute;vora)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026Eacute;vora\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6.745696\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.266541\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMoita (Set\u0026uacute;bal)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSet\u0026uacute;bal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e9.394950\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.252603\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmadora (Lisboa)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLisboa\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e9.966470\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.256918\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePortim\u0026atilde;o (Faro)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFaro\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e9.181091\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.267180\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEsposende (Braga)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBraga\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e10.643280\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.246694\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNordeste (Azores)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAzores\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e9.423650\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.347062\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSanta Cruz (Madeira)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMadeira\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e8.626807\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.365447\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVila do Conde (Porto)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePorto\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e11.071163\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.252238\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eMunicipalities selected according to PM2.5 trajectory classes display distinct annual mean concentrations (mean_pm25, in \u0026micro;g/m\u0026sup3;) together with their relative temporal variability, expressed as the coefficient of variation (cv_pm25). Class 1 municipalities record the lowest average PM2.5 levels (\u0026asymp;\u0026thinsp;5.4\u0026ndash;6.8 \u0026micro;g/m\u0026sup3;) with comparatively higher variability (CV 0.28\u0026ndash;0.43). Class 2 falls within an intermediate range (\u0026asymp;\u0026thinsp;6.7\u0026ndash;8.1 \u0026micro;g/m\u0026sup3;; CV 0.24\u0026ndash;0.33). Class 3 concentrates the highest pollution levels (\u0026asymp;\u0026thinsp;8.6\u0026ndash;11.1 \u0026micro;g/m\u0026sup3;), with stable annual profiles and moderate variability (CV 0.25\u0026ndash;0.37).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe 2024 investment profiles of these municipalities, standardised as z-scores, reveal sharp contrasts across the three trajectories (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Trajectory 1\u0026mdash;the lowest-pollution group\u0026mdash;consistently records negative scores in most cultural indicators and shows weak performance in economic variables. This profile indicates that low emissions are partly sustained by limited economic density, but at the cost of reduced cultural infrastructure. Trajectory 2 occupies a transitional position: some indicators rise modestly above the national mean, while others present mixed results. Trajectory 3 municipalities\u0026mdash;those with the highest and most stable PM2.5 levels\u0026mdash;display the strongest overall profiles, combining dense economic activity with above-average cultural expenditures.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn other words, there are clear contrasts in how economic and cultural investments correspond to the three PM2.5 trajectory classes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). \u003cb\u003eTrajectory 1 (low pollution)\u003c/b\u003e shows a restrained investment profile: economic indicators such as turnover, employees, and GVA remain well below the national mean, while cultural investments are uneven. The only above-average values appear in inverse dropout rates and secondary enrolment, whereas most cultural expenditure variables are depressed. This profile suggests municipalities that sustain low emissions partly through limited economic density, but at the cost of reduced cultural infrastructure. \u003cb\u003eTrajectory 2 (intermediate pollution)\u003c/b\u003e presents a more balanced pattern. Economic indicators like turnover, employees, and hotel capacity rise above the national mean, signalling moderate economic dynamism. Culturally, the most distinctive feature is the sharp peak in secondary education completion, accompanied by positive if modest values in library and arts expenditure. These municipalities exemplify transitional contexts, where medium-level pollution accompanies middling economic and cultural endowments. \u003cb\u003eTrajectory 3 (high pollution)\u003c/b\u003e reveals the most expansive investment structure. Economic indicators\u0026mdash;including GVA, turnover, employment, tourism, and hotel capacity\u0026mdash;register consistently high positive scores, confirming the concentration of productive and service activities. On the cultural side, expenditure on libraries, performing arts, and per-capita culture stand clearly above average, reinforcing the image of urbanised and culturally active municipalities.\u003c/p\u003e\u003cp\u003eCorrelation analysis further emphasizes these contrasts (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In Trajectory 1, economic variables\u0026mdash;gross value added, business turnover, and employment\u0026mdash;are almost perfectly correlated, forming a tight but isolated productive block. Cultural linkages are limited, with strong correlations only between heritage and creative spending, suggesting fragmented cultural integration. Trajectory 2 shows a more hybrid pattern: classical economic associations are reinforced by a strong tourism\u0026ndash;hospitality coupling and by consistent links between creative and interdisciplinary cultural expenditure, signalling selective but diversified investments. Trajectory 3 reveals the densest and most interwoven structure: economic indicators are tightly connected to one another and simultaneously to cultural expenditures, with near-perfect correlations between employment, creative spending, and heritage funding. Yet this dense accumulation of both economic and cultural capital coincides with persistently high PM2.5 exposure, underscoring the relational argument that environmental externalities concentrate where capital investments are highest. Underlying these investment patterns lie distinct population structures\u0026mdash;of age, gender, and settlement density\u0026mdash;that condition local demand, mobility, and emissions (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTop absolute correlations (Pearson)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable A\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eVariable B\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003er\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross Value Added (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNumber of Employees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.98\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross Value Added (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBusiness Turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness Turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNumber of Employees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross Value Added (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of Employees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross Value Added (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHeritage Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness Turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHeritage Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of Employees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHeritage Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness Turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHeritage Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLibrary and Archives Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTourism Revenue (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHotel Establishments (n)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGross Value Added (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLibrary and Archives Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of Employees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLibrary and Archives Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness Turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLibrary and Archives Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeritage Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLibrary and Archives Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInterdisciplinary Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCultural and Creative Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePerforming Arts Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness Turnover (million \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInterdisciplinary Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of Employees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInterdisciplinary Activities Expenditure (thousand \u0026euro;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe demographic composition of municipalities mirrors the gradient of capital competences and PM2.5 trajectories; percentages are population-weighted by class and density is a simple mean across municipalities (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e): \u003cb\u003eTrajectory 1 (low PM2.5).\u003c/b\u003e This group (n\u0026thinsp;=\u0026thinsp;13) combines a small youth share (11.0% \u0026lt;15) with a high older-adult share (31.7% \u0026ge;65) and very low density (32.5 inh./km\u0026sup2;). The implied working-age share is \u0026asymp;\u0026thinsp;57.3%, indicating a thin labour pool and limited demographic renewal. The female share (52.2%) is modestly above parity, consistent with ageing patterns. Demographically, these municipalities align with the investment profiles described earlier: weaker economic throughput and fragmented cultural infrastructure coincide with low exposure, suggesting that reduced capital intensity and demographic shrinkage jointly dampen PM2.5 levels. \u003cb\u003eTrajectory 2 (intermediate PM2.5).\u003c/b\u003e Municipalities in this class (n\u0026thinsp;=\u0026thinsp;9) present a more balanced age structure (12.8% \u0026lt;15; 25.1% \u0026ge;65; working-age\u0026thinsp;\u0026asymp;\u0026thinsp;62.1%) and moderate density (89.6 inh./km\u0026sup2;). The female share (50.4%) sits near parity. This configuration matches the \u0026ldquo;transitional\u0026rdquo; investment profile: moderate economic dynamism and selective cultural spending co-exist with middling exposure. Demographically, the presence of families and school-age cohorts suggests active local demand for services and education\u0026mdash;features that can tip municipalities toward either cleaner growth or increased emissions depending on how investments are instrumented. \u003cb\u003eTrajectory 3 (high PM2.5).\u003c/b\u003e The high-exposure group (n\u0026thinsp;=\u0026thinsp;7) is markedly urban: very high density (1,519.4 inh./km\u0026sup2;), larger youth presence (14.4% \u0026lt;15), smaller older-adult share (21.6% \u0026ge;65), and a working-age share near 64.0%. The female share (52.6%) is slightly above parity. These demographics align with the strongest economic and cultural profiles: large employee bases, higher turnover and GVA, and intensive cultural infrastructures. The same demographic dynamism that underpins capital competences (workforce depth, consumer bases, cultural demand) also amplifies activity levels and mobility, reinforcing the persistence of high PM2.5 exposure in these places.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDemographic profile by PM2.5 trajectory class\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClass\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN (municipalities)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePopulation\u0026thinsp;\u0026lt;\u0026thinsp;15 (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePopulation\u0026thinsp;\u0026ge;\u0026thinsp;65 (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eFemale (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDensity (inh./km\u0026sup2;)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrajectory 1 (low PM2.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e11.0%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e31.7%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e52.2%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e32.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrajectory 2 (intermediate)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e12.8%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e25.1%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e50.4%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e89.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrajectory 3 (high PM2.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e14.4%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e21.6%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e52.6%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1519.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eBy placing these results alongside the temporal trajectory analysis, the study exposes that PM2.5 pollution in Portugal is not only a matter of meteorology or geography but is intimately tied to the socio-spatial configuration of capital competences. At the intersection of these findings lies a structural paradox in Portugal: the very municipalities that accumulate economic and cultural resources\u0026mdash;fostering productivity, employment, and cultural infrastructures\u0026mdash;are also those where PM2.5 concentrations remain consistently above the national average. Conversely, structurally peripheral municipalities sustain lower emissions largely because of limited capital endowments, though at the cost of weaker cultural and economic infrastructures. These differentiated trajectories confirm that fine particulate exposure is not randomly distributed but systematically embedded in capital configurations. This pattern underscores an inconsistency of many development processes\u0026mdash;where investments that enhance local capacities also reinforce capital externalities\u0026mdash;and highlights the need for territorially differentiated policies capable of balancing capital development with the mitigation of pollution burdens.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eIn sociological terms, \u003cem\u003eobjectified capital\u003c/em\u003e designates the material crystallisation of social and economic investments into durable forms such as infrastructures, cultural facilities, and built environments, which inscribe hierarchies of power and resources into the physical landscape (Bourdieu, 2018a; Gomez, Mi\u0026ntilde;o, Pereira, et al., 2025). At the national scale of Portugal, such objectifications are related with the fiscal capacities and economic and cultural expenditures of municipalities, which condense into patterned competences for development, employment, and social reproduction. When juxtaposed with the temporal trajectory classes of PM2.5, the analysis reveals how differentiated capital endowments become inscribed into spatial regimes of environmental burden. Municipalities with higher levels of economic and cultural investment tend to anchor more intense trajectories of particulate exposure, while those with weaker investment profiles register lower exposures but at the expense of curtailed opportunities and thinner infrastructures of provision. In summary, these dynamics can be read as \u003cem\u003eobjectified\u003c/em\u003e \u003cem\u003ecapital externalities\u003c/em\u003e. Positive externalities emerge where investment in infrastructures and cultural facilities generates spill-overs in the form of economic opportunities, knowledge circulation, and improved services. Yet the same investments frequently yield negative outputs, such as heightened air pollution, congestion, or environmental degradation, whose costs are externalised onto local populations and ecosystems. The Portuguese analysis thus reveals the paradox of capital allocation: the very municipal investments that consolidate social and economic capacities also entrench ecological externalities in the form of persistent PM2.5 exposures. Effective policy strategies must therefore confront this duality, developing territorially differentiated approaches that reconcile the productivity and cultural benefits of municipal investment with the mitigation of their environmental costs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCapital competences today must encompass structured capacities to produce, but also to absorb, and mitigate spillovers through the mobilisation of economic and cultural resources. By integrating collectively shared externalities into the analysis of the objectified state of capital, the study reframes municipal PM2.5 trajectories not only as environmental outcomes but as indicators of how competences crystallise in space, interact with meteorological regimes, and structure inequalities. This conceptual move enables comparison between municipalities not only in terms of the capital they hold but in terms of how much they generate, displace, or internalise externalities. It thereby shifts the focus beyond a binary of \u0026lsquo;capital versus environment,\u0026rsquo; showing instead that environmental burdens and benefits are intrinsic to the very processes by which capital is allocated, mobilised, and contested at the subnational scale. Indeed, capital externalities are often invoked yet seldom specified to localised levels. In this article they were treated not as an abstract residue of market failure, but as the \u003cem\u003eobjectified\u003c/em\u003e by-products of capital development\u0026mdash;effects that persist in space and time because they are embedded in infrastructures, settlement patterns, and investment regimes. Conceptually, this reframing matters because it links externalities to the \u003cem\u003ecomposition and volumes of capital\u003c/em\u003e at subnational scales, where economic and cultural resources take material form in firms, transport and port systems, cultural venues, educational pipelines, and housing markets. Empirically, the study operationalised this idea by using PM2.5 as a negative capital externality and by situating its trajectories (2019\u0026ndash;2025) within the differentiated municipal endowments of economic and cultural investments (2024). Measuring these processes is challenging: externalities are multi-source and path-dependent; exposure varies with meteorology; and capital competences evolve under institutional inertia. The article addressed these problems by: (i) modelling \u003cem\u003etrajectories\u003c/em\u003e rather than single-year snapshots of PM2.5 and wind configurations; and (ii) treating municipal investment as structured, slow-moving capacity rather than short-term volatility, thereby matching the temporalities of accumulation and exposure in a relational design.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.1. Contributions.\u003c/strong\u003e First, the paper demonstrates that PM2.3 exposure in Portugal can be organised into relatively stable trajectory classes, each characterised by distinctive means and variability, and with clear seasonal signatures linked to national wind regimes. These are not mere gradients of magnitude: they are spatial\u0026ndash;temporal types\u0026mdash;with a low-pollution inland belt, an intermediate littoral corridor, and high-pollution enclaves in archipelagos, tourism nodes, and port-industrial conurbations\u0026mdash;that jointly expose how meteorology and spatial development intertwine. This typology cautions against explanations relying solely on geography or solely on activity levels; it shows instead that PM2.5 emissions reflects the \u003cem\u003eco-production\u003c/em\u003e of environmental and socio-economic structures. Second, the study links these trajectories to municipal \u003cem\u003einvestment profiles\u003c/em\u003e\u0026mdash;a composite of economic (e.g., GVA, turnover, employment, tourism receipts) and cultural (e.g., per-capita cultural expenditure, libraries, heritage, performing arts, education) dimensions, standardised to enable comparison. By selecting a representative subset of municipalities nearest to class centroids\u0026mdash;while ensuring district diversity\u0026mdash;the analysis avoids cherry-picking outliers and instead reads typical cases for each trajectory. This design strengthens internal validity by directing the analysis toward patterns that typify classes rather than idiosyncrasies of single places. In summary, the juxtaposition of trajectory classes with z-scored investment profiles yields a reasoned argument: where economic and cultural capital accumulate most intensely\u0026mdash;in port-industrial zones, tourism nodes, and archipelagos with concentrated flows\u0026mdash;PM2.5 exposure is higher and more persistent; where economic density is limited and cultural infrastructures are thinner, exposure remains lower but at the cost of curtailed opportunities and weaker public investments. Between these poles lie transitional municipalities that register moderate emissions alongside selective cultural linkages and tourism\u0026ndash;hospitality couplings. The correlation network consolidates this contribution: high centrality of GVA, turnover, and employment; robust ties to cultural\u0026ndash;creative and heritage expenditures; and strong tourism\u0026ndash;hotel links\u0026mdash;all consistent with an integrated regime of production, services, and cultural objectifications. In short, fine particulates \u003cem\u003emap onto\u003c/em\u003e capital arrangements.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2. Implications for policy\u0026mdash;targeting the subnational scale.\u003c/strong\u003e If exposure is structurally embedded, mitigation must be \u003cem\u003espatially differentiated\u003c/em\u003e. The evidence suggests at least four families of possible interventions:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003e\u003cstrong\u003ePlace-specific emission reduction in high-exposure regimes (Trajectory 3).\u003c/strong\u003e Prioritising port-industrial corridors and archipelagos where dense economy\u0026ndash;culture linkages coincide with persistent peaks. Practical levers include clean-fuel standards for shipping and logistics, low-emission zones linked to freight scheduling, and electrification incentives for tourism and cultural venues with high visitor throughput. The goal is not to de-accumulate capital, but to \u003cem\u003ere-instrument\u003c/em\u003e it so that objectified capacities do not reproduce objectified harms.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eTransitional packages for intermediate regimes (Trajectory 2).\u003c/strong\u003e Here the diagnostic is a \u003cem\u003ehybrid\u003c/em\u003e coupling of productive activity, tourism, and selective cultural investment. Policy can amplify co-benefits: mobility demand management during seasonal peaks; performance-based grants for cultural facilities that adopt low-emission operations; and targeted support for firms aligning productivity upgrades with emissions abatement.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eOpportunity-enhancing investments for low-exposure regimes (Trajectory 1).\u003c/strong\u003e In Portugal, lower pollution partly reflects limited economic density and more fragmented cultural infrastructures. Equity requires that decoupling \u003cem\u003edoes not\u003c/em\u003e come at the price of durable under-investment. Place-based cultural and educational programmes (libraries, heritage, secondary-education completion) can be scaled while preserving low exposure through stringent siting rules, small-scale distributed generation, and clean mobility links to regional centres.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eMeteorology-aware planning everywhere.\u003c/strong\u003e The monthly signatures and seasonal wind roses highlight when and where dispersion is weakest. Subnational authorities should align emission caps, traffic restrictions, port operations, and biomass-burning controls with forecast windows of adverse dispersion (e.g., late winter inversions, autumn peaks), making mitigation \u003cem\u003etemporal\u003c/em\u003e as well as spatial.\u0026nbsp;\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThese suggestions share a common premise: because externalities are \u003cem\u003eco-produced\u003c/em\u003e with capital competences, interventions must treat economic and cultural infrastructures as \u003cem\u003elevers\u003c/em\u003e for mitigation, not as background conditions. Policies that merely relocate emissions risk displacing harm rather than transforming it.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.3. Limitations and directions for future work.\u003c/strong\u003e Two types of limitations are salient. \u003cem\u003eData and scale.\u003c/em\u003e The CAMS reanalysis provides harmonised coverage but at ~0.4\u0026deg; resolution; while municipal averages mitigate this constraint, sub-municipal gradients and micro-environments (street canyons, industrial plumes) remain unresolved. Integrating higher-resolution products, validated monitors, and local emission inventories would refine attribution. \u003cem\u003eTemporal alignment and endogeneity.\u003c/em\u003e Investment indicators refer to 2024 and are assumed to exhibit structural inertia. This is theoretically consistent with capital accumulation, but longitudinal municipal financial series would improve causal inference and help distinguish cyclical shocks from trend capacities. Finally, the approach is portable: any country with municipal boundaries, basic fiscal/cultural statistics, and access to reanalysis products can replicate and improve the workflow. The key is to preserve the \u003cem\u003etrajectory logic\u003c/em\u003e (time-series clustering), the \u003cem\u003erepresentative sub-sample\u003c/em\u003e (centroid proximity with territorial diversity), and the \u003cem\u003ecross-domain pairing\u003c/em\u003e (environmental trajectories \u0026times; capital indicators). This enables comparative research on how different development models (port economies, tourism regions, inland agricultural belts) produce distinctive externality regimes and how cultural investments could be mobilised as mitigation partners.\u003c/p\u003e\n\u003cp\u003ePlaced alongside the temporal analysis, the investment profiles and correlation structures lead to a simple but incommode conclusion: PM2.5 in Portugal is not only weather and not only geography\u0026mdash;it is a social cartography of capital investments. Municipalities that most successfully develop economic and cultural competences also face the hardest task of decoupling those competences from particulate exposure; those with thinner endowments face the opposite challenge of building opportunity without importing risk. Recognising this \u003cem\u003eparadox of development\u003c/em\u003e shifts the policy question from \u0026ldquo;How much growth?\u0026rdquo; to \u0026ldquo;\u003cem\u003eWhich\u003c/em\u003e investments, \u003cem\u003ewhere\u003c/em\u003e and \u003cem\u003ewhen\u003c/em\u003e, and with \u003cem\u003ewhat\u003c/em\u003e environmental approach?\u0026rdquo; The answer, the article results suggest, lies in territorially differentiated strategies that treat cultural and economic possibilities as protagonists of mitigation, aligning cycles of investment with cycles of air quality so that the gains of development do not come with the costs of persistent PM2.5 burdens.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003e5. Declaration of Competing Interest\u003c/h2\u003e\u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. All authors have approved the submitted version of the manuscript.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003e7. CRediT authorship contribution statement\u003c/h2\u003e\u003cp\u003e\u003cem\u003eRaimundo Elias Gomez\u003c/em\u003e: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing - Original draft, Writing - Review \u0026amp; editing, Visualization, Project administration, Funding acquisition. \u003cem\u003eMaria Gabriela Mi\u0026ntilde;o\u003c/em\u003e: Conceptualization, Methodology, Investigation, Writing - Original draft, Writing - Review \u0026amp; editing, Validation.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003e9. Ethical Statement\u003c/h2\u003e\u003cp\u003eThe authors confirm that this work is original, has not been published elsewhere, and is not currently under consideration by another journal. All authors have read, understood, and have complied as applicable with the statement on \"Ethical responsibilities of Authors\" as found in the \u0026ldquo;Instructions for Authors\u0026rdquo;.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003e6. Funding sources\u003c/h2\u003e\u003cp\u003eThis project has received funding from the European Union\u0026rsquo;s Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 101102978 \u0026mdash; SSpaceGX: \"Social Space and Nature Conservation in the Transboundary Biosphere Reserve Ger\u0026ecirc;s-Xur\u0026eacute;s (Portugal/Spain).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003e*Raimundo Elias Gomez:* Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing - Original draft, Writing - Review \u0026amp; editing, Visualization, Project administration, Funding acquisition. *Maria Gabriela Mi\u0026ntilde;o:* Conceptualization, Methodology, Investigation, Writing - Original draft, Writing - Review \u0026amp; editing, Validation.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003ePM2.5 and wind data are publicly available through Copernicus Atmosphere Monitoring Service (CAMS) and ERA5 reanalysis. The complete database and analytical dashboards are accessible at: https://doi.org/10.5281/ZENODO.16981588 (economic and cultural capital data), https://doi.org/10.5281/ZENODO.16981258 (PM2.5 spatial distribution), and https://doi.org/10.5281/ZENODO.16980568 (wind roses analysis). Municipal socioeconomic data are available from Instituto Nacional de Estat\u0026iacute;stica Portugal (INE-PT) portal.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAlam, M. J., Karim, I., \u0026amp; Zaman, S. U. (2025). Seasonal dynamics and trends in air pollutants: A comprehensive analysis of PM2.5, NO2, CO, SO2 and O3 in Houston, USA. \u003cem\u003eAir Quality, Atmosphere, \u0026amp; Health\u003c/em\u003e. https://doi.org/10.1007/s11869-025-01790-9\u003c/li\u003e\n\u003cli\u003eAshraf, S., Pausata, F. S. R., Leroyer, S., Stevens, R., \u0026amp; Munoz-Alpizar, R. (2025). Impact of reduced anthropogenic emissions associated with COVID-19 lockdown on PM2.5 concentration and canopy urban heat island in Canada. \u003cem\u003eGeoHealth\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(2), e2023GH000975. https://doi.org/10.1029/2023GH000975\u003c/li\u003e\n\u003cli\u003eBarbosa, J. V., Nunes, R. A. O., Alvim-Ferraz, M. C. M., Martins, F. G., \u0026amp; Sousa, S. I. V. (2024). Health and economic burden of wildland fires PM2.5-related pollution in Portugal - A longitudinal study. \u003cem\u003eEnvironmental Research\u003c/em\u003e, \u003cem\u003e240\u003c/em\u003e(Pt 1), 117490. https://doi.org/10.1016/j.envres.2023.117490\u003c/li\u003e\n\u003cli\u003eBille, T., \u0026amp; Honor\u0026eacute;, S. (2025). Cultural capital externalities: Causal evidence from a Danish ticket scheme for theatres. \u003cem\u003eKyklos: International Review for Social Sciences\u003c/em\u003e. https://doi.org/10.1111/kykl.12469\u003c/li\u003e\n\u003cli\u003eBourdieu, P. (2018a). The forms of capital. In \u003cem\u003eThe sociology of economic life\u003c/em\u003e (pp. 78\u0026ndash;92). Routledge. https://www.taylorfrancis.com/chapters/edit/10.4324/9780429494338-6/forms-capital-pierre-bourdieu\u003c/li\u003e\n\u003cli\u003eBourdieu, P. (2020). The field of power and the division of the labour of domination: Handwritten notes for the 1985-1986 coll\u0026egrave;ge de France lectures. In \u003cem\u003eResearching Elites and Power\u003c/em\u003e (pp. 33\u0026ndash;44). Springer International Publishing. https://doi.org/10.1007/978-3-030-45175-2_3\u003c/li\u003e\n\u003cli\u003eBourdieu, P. (2018b). Social Space and the Genesis of Appropriated Physical Space: FORUM. \u003cem\u003eInternational Journal of Urban and Regional Research\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(1), 106\u0026ndash;114. https://doi.org/10.1111/1468-2427.12534\u003c/li\u003e\n\u003cli\u003eC3S. (2018). \u003cem\u003eERA5 hourly data on single levels from 1940 to present\u003c/em\u003e [Dataset]. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). https://doi.org/10.24381/CDS.ADBB2D47\u003c/li\u003e\n\u003cli\u003eCAMS. (2021). \u003cem\u003eCopernicus Atmosphere Monitoring Service: Global atmospheric composition forecasts\u003c/em\u003e [Dataset]. ECMWF. https://doi.org/10.24381/04A0B097\u003c/li\u003e\n\u003cli\u003eCFP. (2025). Conselho das Finan\u0026ccedil;as P\u0026uacute;blicas, Portugal. https://www.cfp.pt/pt\u003c/li\u003e\n\u003cli\u003eChoi, W., Ho, C.-H., \u0026amp; Lee, Y. (2024). Temporal pattern classification of PM2.5 chemical compositions in Seoul, Korea using K-means clustering analysis. \u003cem\u003eThe Science of the Total Environment\u003c/em\u003e, \u003cem\u003e927\u003c/em\u003e(172157), 172157. https://doi.org/10.1016/j.scitotenv.2024.172157\u003c/li\u003e\n\u003cli\u003eCNC. (2025). Comiss\u0026atilde;o de Normaliza\u0026ccedil;\u0026atilde;o Contabil\u0026iacute;stica, Portugal. https://www.cnc.min-financas.pt/sncap2017.html\u003c/li\u003e\n\u003cli\u003eCorda, M. O., Charalampous, P., Haagsma, J. A., Assun\u0026ccedil;\u0026atilde;o, R., \u0026amp; Martins, C. (2024). Mortality burden of cardiovascular disease attributable to ambient PM2.5 exposure in Portugal, 2011 to 2021. \u003cem\u003eBMC Public Health\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(1), 1188. https://doi.org/10.1186/s12889-024-18572-0\u003c/li\u003e\n\u003cli\u003eEspinoza-Guillen, J. A., Alderete-Malpartida, M. B., Roncal-Romero, F. D., \u0026amp; Vilcanqui-Sarmiento, J. C. (2025). Identification of particulate matter (PM10 and PM2.5) sources using bivariate polar plots and k-means clustering in a South American megacity: Metropolitan Area of Lima-Callao, Peru. \u003cem\u003eEnvironmental Monitoring and Assessment\u003c/em\u003e, \u003cem\u003e197\u003c/em\u003e(3), 226. https://doi.org/10.1007/s10661-025-13696-1\u003c/li\u003e\n\u003cli\u003eEU/ESA/Copernicus. (2024). \u003cem\u003eSentinel-5P NRTI AER AI: Near Real-Time UV Aerosol Index\u003c/em\u003e. Google for Developers. https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S5P_NRTI_L3_AER_AI\u003c/li\u003e\n\u003cli\u003eGomez, R. E. (2021). Espacio social y clases sociales en Posadas, Argentina. \u003cem\u003eEstudios Demograficos y Urbanos\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e(3), 865\u0026ndash;889. https://doi.org/10.24201/edu.v36i3.1983\u003c/li\u003e\n\u003cli\u003eGomez, R. E. (2025a). \u003cem\u003eMCA1. Economic and cultural capital by PM2.5 trajectory classes\u003c/em\u003e. Zenodo. https://doi.org/10.5281/ZENODO.16981588\u003c/li\u003e\n\u003cli\u003eGomez, R. E. (2025b). \u003cem\u003eSpatial Distribution of Fine Particulate Matter in Portugal (2019\u0026ndash;2025)\u003c/em\u003e. Zenodo. https://doi.org/10.5281/ZENODO.16981258\u003c/li\u003e\n\u003cli\u003eGomez, R. E. (2025c). \u003cem\u003eWind Roses II (2019\u0026ndash;2025): Interactive analysis of PM2.5 and wind direction in Portuguese municipalities\u003c/em\u003e. Zenodo. https://doi.org/10.5281/ZENODO.16980568\u003c/li\u003e\n\u003cli\u003eGomez, R. E., \u0026amp; Mi\u0026ntilde;o, M. G. (2025). Extensive objectified footprints: A multidimensional approach to spatial inequalities. \u003cem\u003eCity and Environment Interactions\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(100226), 100226. https://doi.org/10.1016/j.cacint.2025.100226\u003c/li\u003e\n\u003cli\u003eGomez, R. E., Mi\u0026ntilde;o, M. G., Hojman, G. D., Arellano, D. M., Cossi, C. A., \u0026amp; Sosa, L. (2025). Tracing Emplaced Capital Footprints in Arcos de Valdevez, Portugal. \u003cem\u003eComparative Sociology\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(3), 410\u0026ndash;444. https://doi.org/10.1163/15691330-bja10137\u003c/li\u003e\n\u003cli\u003eGomez, R. E., Mi\u0026ntilde;o, M. G., Pereira, V. B., \u0026amp; Jard\u0026oacute;n, C. M. (2025). Spatial distribution and classification of objectified capital in the Ger\u0026ecirc;s-Xur\u0026eacute;s Transboundary Biosphere Reserve. \u003cem\u003eFrontiers in Environmental Economics\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e, 1463694. https://doi.org/10.3389/frevc.2025.1463694\u003c/li\u003e\n\u003cli\u003eINE-PT. (2025). Instituto Nacional de Estat\u0026iacute;stica, Portugal. Carateriza\u0026ccedil;\u0026atilde;o s\u0026oacute;cio \u0026ndash; econ\u0026oacute;mica dos munic\u0026iacute;pios. https://www.ine.pt/xportal/xmain?xpid=INE\u0026amp;xpgid=ine_doc_municipios_cse\u0026amp;xlang=pt\u003c/li\u003e\n\u003cli\u003eIşıkara, G. (2023). Capitalism, economics, and externalities: What are externalities external to? \u003cem\u003eCapitalism Nature Socialism\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(2), 40\u0026ndash;56. https://doi.org/10.1080/10455752.2023.2192954\u003c/li\u003e\n\u003cli\u003eKouyate, M., Arola, A., Benedictow, A., Bennouna, Y., Blake, L., Bouarar, I., Cuevas, E., Errera, Q., Eskes, H. J., Griesfeller, J., Basart, S., Kapsomenakis, J., Langerock, B., Mortier, A., Pitk\u0026auml;nen, M. R. A., Pison, I., Ramonet, M., Richter, A., Schoenhardt, A., \u0026hellip; Zerefos, C. (2024). \u003cem\u003eValidation report of the CAMS near-real-time global atmospheric composition service Period June \u0026ndash; August 2023\u003c/em\u003e. Copernicus Atmosphere Monitoring Service. https://doi.org/10.24380/A7AS-QAEG\u003c/li\u003e\n\u003cli\u003eLelieveld, J., Evans, J. S., Fnais, M., Giannadaki, D., \u0026amp; Pozzer, A. (2015). The contribution of outdoor air pollution sources to premature mortality on a global scale. \u003cem\u003eNature\u003c/em\u003e, \u003cem\u003e525\u003c/em\u003e(7569), 367\u0026ndash;371. https://doi.org/10.1038/nature15371\u003c/li\u003e\n\u003cli\u003eMarom, N. (2014). Relating a city\u0026rsquo;s history and geography with Bourdieu: One hundred years of spatial distinction inTelAviv. \u003cem\u003eInternational Journal of Urban and Regional Research\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(4), 1344\u0026ndash;1362. https://doi.org/10.1111/1468-2427.12027\u003c/li\u003e\n\u003cli\u003eOke, T. R., Mills, G., Christen, A., \u0026amp; Voogt, J. A. (2017). Air Pollution. In \u003cem\u003eUrban Climates\u003c/em\u003e (pp. 294\u0026ndash;331). Cambridge University Press. https://doi.org/10.1017/9781139016476.012\u003c/li\u003e\n\u003cli\u003ePrapassonpithaya, P., \u0026amp; Jinsart, W. (2025). The relationship between PM2.5 levels and COVID-19 after major outbreak waves: A Bangkok metropolitan area study. \u003cem\u003eEnvironmental Challenges (Amsterdam, Netherlands)\u003c/em\u003e, \u003cem\u003e101255\u003c/em\u003e, 101255. https://doi.org/10.1016/j.envc.2025.101255\u003c/li\u003e\n\u003cli\u003eRodriguez Avellaneda, F., Chac\u0026oacute;n-Montalv\u0026aacute;n, E. A., \u0026amp; Moraga, P. (2025). Multivariate disaggregation modeling of air pollutants: a case-study of PM2.5, PM10 and ozone prediction in Portugal and Italy. \u003cem\u003eThe American Statistician\u003c/em\u003e, 1\u0026ndash;21. https://doi.org/10.1080/00031305.2025.2537055\u003c/li\u003e\n\u003cli\u003eSeinfeld, J. H., \u0026amp; Pandis, S. N. (2016). \u003cem\u003eAtmospheric chemistry and physics\u003c/em\u003e (3rd ed.). Wiley-Blackwell.\u003c/li\u003e\n\u003cli\u003eSiblot, Y., Hugr\u0026eacute;e, C., \u0026amp; Pereira, V. B. (2024). Comparing countries, exporting classifications, surpassing methodological nationalism: Class, gender, and education gaps in and between France and Portugal. \u003cem\u003eThe Sociological Review\u003c/em\u003e. https://doi.org/10.1177/00380261241299991\u003c/li\u003e\n\u003cli\u003eYu, Y. T., Zhang, S., Xiang, S., \u0026amp; Wu, Y. (2025). Socioeconomic inequalities in PM2.5 exposure and local source contributions at community scales using hyper-localized taxi-based mobile monitoring in Xi\u0026rsquo;an, China. \u003cem\u003eEnvironmental Science \u0026amp; Technology\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(14), 7222\u0026ndash;7234. https://doi.org/10.1021/acs.est.4c11385\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"PM2.5, Municipal Investment, Spatial Inequality, Trajectory Analysis, Air Pollution","lastPublishedDoi":"10.21203/rs.3.rs-7790529/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7790529/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe article examines temporal trajectories of fine particulate matter (PM2.5) across Portugal’s 308 municipalities (2019–2025) in relation to seasonal wind patterns and municipal economic and cultural investment. Annual PM2.5 concentrations and wind statistics were obtained from CAMS NRT/ECMWF satellite data. After standardising and clustering the PM2.5 distribution with k-means, twenty-nine municipalities closest to cluster centroids were compared using 2024 fiscal data. Results reveal marked spatial disparities in PM2.5 exposure linked to seasonal regimes and investment profiles. High economic-cultural investment areas often align with high-pollution trajectories, whereas low investment correlates with lower levels, both reflecting seasonal environmental and social patterns.\u003c/p\u003e","manuscriptTitle":"PM2.5 Trajectory Classes and Spatial Investment Patterns in Portugal","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-28 18:00:08","doi":"10.21203/rs.3.rs-7790529/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e500d567-71b9-4aa3-b8f8-672c7fe340e3","owner":[],"postedDate":"November 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-13T13:39:08+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-28 18:00:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7790529","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7790529","identity":"rs-7790529","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}