Next Article in Journal
Rethinking Accessibility as a Cultural Generative Ecosystem
Previous Article in Journal
Urban Tree Canopy, Asthma Prevalence, and Health Equity in Los Angeles: A Geospatial Analysis with Implications for Urban Forestry
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Urban Science
Volume 9
Issue 11
10.3390/urbansci9110477
urbansci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Monitoring Air Pollution in Wartime Kyiv (Ukraine): PM2.5 Spikes During Russian Missile and Drone Attacks

by
Kseniia Bondar
1,*,
Iryna Tsiupa
2 and
Mykhailo Virshylo
3
1
Institute of Geophysics, Polish Academy of Sciences, Księcia Janusza 64, 01-452 Warsaw, Poland
2
Institute of Geology, Taras Shevchenko National University of Kyiv, Vasylkivska St. 90, 03022 Kyiv, Ukraine
3
Department of Economics, Kyiv School of Economics, Mykoly Shpaka St. 3, 03113 Kyiv, Ukraine
*
Author to whom correspondence should be addressed.
Urban Sci. 2025, 9(11), 477; https://doi.org/10.3390/urbansci9110477
Submission received: 9 October 2025 / Revised: 8 November 2025 / Accepted: 10 November 2025 / Published: 14 November 2025
(This article belongs to the Section Urban Environment and Sustainability)
Browse Figures
Versions Notes

Abstract

This study investigates the environmental impact of combined missile and drone attacks on Kyiv, the capital of Ukraine, with a focus on the release of particulate matter (PM) into the urban atmosphere. These military strikes frequently result in the destruction of residential and industrial infrastructure, as well as fires, leading to acute increases in ambient concentrations of fine particulate matter (PM2.5). Observational data were collected between 1 and 30 June 2025 using a distributed network of low-cost air quality monitoring stations aggregated by the SaveEcoBot platform. The optical particle counters, based on light scattering technology, enable real-time monitoring of airborne particulate fractions of PM2.5 along with meteorological parameters and gas pollutants. The study period included two significant attacks (10 and 17 June), during which the temporal and spatial dynamics of PM2.5 concentrations were analyzed in comparison to baseline levels observed under non-attack conditions. Raw concentrations of PM2.5 up to 241 μg/m3 were observed in the epicenters of air-strike-induced fires, while smog plumes covered half of the city area. Elevated PM2.5 concentrations were recorded during and for several hours following the attacks and corresponding air raid alerts. The findings show days of PM2.5 exceedances above the World Health Organization (WHO) daily threshold of 15 μg/m3. These results underscore the acute environmental and public health hazards posed by military assaults on urban centers. Furthermore, this research highlights the role of citizen-driven environmental monitoring as a valuable tool for both scientific documentation and potential evidentiary support in assessing the environmental impacts of warfare.

1. Introduction

Air quality studies widely use data on particulate matter (PM) concentrations to evaluate PM’s harmful effects [1,2,3,4]. Recent findings emphasize correlations between PM concentrations and various metals [5,6], ferromagnetic particles [7,8,9], microplastics and polycyclic aromatic hydrocarbons [10,11], and black carbon [12,13]. Frequent co-occurrence of high PM concentrations with various harmful substances implies shared transport or source mechanisms. In urban environments, PM of different aerodynamic diameters originates from a combination of natural sources (e.g., resuspended soil dust, pollen, biogenic particles) and anthropogenic activities, such as vehicular exhaust, brake and tire wear, industrial emissions, construction activities, residential heating, and secondary aerosol formation through atmospheric photochemical reactions [14,15,16,17,18]. The smaller the aerodynamic diameter of PM, the greater its potential danger to human health. Ultrafine (PM1) and fine particles (PM2.5) can penetrate deeply into the respiratory tract, reaching the alveolar region, where they may translocate into the bloodstream, causing, in the long run, cardiovascular and respiratory disease, lung cancer, and premature mortality [19,20,21]. PM2.5 is classified as carcinogenic to humans (Group 1) by IARC [22].
Ambient concentrations of PM2.5 exhibit substantial variability depending on the temporal scale of observation. In the short run (hourly or diurnal cycles), PM2.5 levels are strongly influenced by short-term events associated with both local and broader emission sources, while monthly average concentrations reflect seasonality and long-lasting trends of both natural and anthropogenic origin. For instance, PM2.5 concentrations often peak during morning and evening rush hours or nighttime stagnation events when atmospheric mixing is minimal [23]. The highest PM2.5 concentrations are systematically recorded on the days of celebrations when a huge number of firecrackers, sparklers, and aerial fireworks are burnt [24,25]. In winter months, PM2.5 concentrations tend to rise due to increased heating demand and stable atmospheric conditions, whereas summer episodes may be dominated by photochemical secondary aerosol formation [26]. Measurements of magnetic susceptibility and heavy metal contents conducted on air filters across Kyiv revealed monthly pollution patterns, where road dust resuspension largely drives temporal fluctuations in airborne particulates, especially in high-traffic areas [27].
The recent Russian-Ukrainian war is a significant factor in air deterioration, as recognized both from satellite-based remote sensing and on-ground measurements in the short run. Notably, tropospheric nitrogen dioxide (NO2) levels over major Ukrainian cities—including Kyiv—declined by 15–60% in early 2022 compared to prior years [28,29,30]. This reduction is attributed to the sharp contraction in transportation, industrial activity, and energy production caused by the war. PM2.5 levels also initially decreased—by approximately 38% nationally and 31% in Kyiv—before later rising due to war-related fires and explosions [28]. A general improvement in air quality in the near-frontline Ukrainian cities has been recognized from decreases in PM magnetic susceptibility and metal concentrations, attributable to the broader impacts of the ongoing war. A sharp decline in PM magnetic susceptibility was caused by the destruction of industrial facilities in Kharkiv and a significant reduction in traffic intensity due to a 5-fold depopulation in Kherson [31]. Contrasting with the general decline in anthropogenic emissions, localized spikes in pollutants were recorded during intense military operations. Early in the conflict, PM2.5 concentrations in Kyiv surged to 27.8 times the WHO-recommended 24-h guideline levels due to widespread bombing and resulting structural fires [32]. Frontline regions also experienced heightened levels of CO, O3, SO2, and NO2, driven by artillery, explosions, and burning infrastructure [33,34,35]. For instance, PM2.5 peaked at 24.2 µg/m3, while NO2 reached 139.7 µmol/m2, both significantly exceeding recommended health thresholds [28].
Although long-range trends show short-term reductions in emissions from paused industrial activity, new sources—such as fires and explosive-related releases of toxic substances—pose severe localized threats to air quality and human health [36]. The short-term improvements in baseline air pollutants reflect disrupted socioeconomic functions rather than genuine environmental recovery. Meanwhile, toxic exposure events exacerbate risks for the population in Kyiv and may have long-term health implications.
This research seeks to monitor and assess variations in PM2.5 concentrations in Kyiv (Ukraine) during and after the episodes of intense combined drone and missile attacks in June 2025, using low-cost particulate matter sensors. Russian forces have been carrying out between two and five such assaults per month on the capital, yet little is known about the spatial and temporal distribution of air quality impacts across the urban environment. By addressing this gap, the study aims to advance understanding of how large-scale military strikes influence urban air pollution under conditions of active warfare and to provide evidence that may inform evaluations of the broader health and environmental consequences.

2. Materials and Methods

2.1. Study Area and War Situation Characterization

This study was conducted in the Kyiv metropolitan area (50.45° N; 30.53° E) (Figure 1), the capital city of Ukraine, with a total population of 2,952,301 in 2022 [37] and a total area of 835.6 km2, located on both banks of the Dnipro River. Kyiv urban area is situated at the boundary between the Forest and the Forest-Steppe geographical zones, which have a temperate climate.
The average annual air temperature is +8.9–+11.9 °C, the average annual rainfall is 600–700 mm, the predominant direction of the wind in summer is west, and in winter, it is northwest [38]. A defining aspect of the urban microclimate is the pronounced temperature gradient between the flat eastern bank and the topographically elevated western bank. The thermal regime is further shaped by anthropogenic influences, particularly heat fluxes originating from district heating infrastructure, building surfaces, and thermal power plants [39]. These factors collectively intensify the urban heat island effect, resulting in systematically higher air temperatures within the city compared to its surrounding areas. 21.4% or 78.0 km2 of the total built-up land of the city is occupied by industrial and transport facilities, represented by power plants, construction industry, machine-building, metal and woodworking industry, chemical-pharmaceutical factories, motorways, and railway. Industrial activities together with intensive vehicular traffic constitute major contributors to atmospheric pollution, with mobile emission sources accounting for over 70% of the total pollutant load in the city’s ambient air [40].
Following the onset of Russia’s large-scale invasion of Ukraine in 2022, profound and dynamic transformations have been observed in the urban life of Kyiv metropolitan area. Kyiv continues to experience combined drone and missile attacks, resulting in casualties, fires, and destruction in residential areas and infrastructure. Strikes targeting the energy system have led to unplanned outages and emergency operating modes [41]. According to the United Nations, the 2024 campaign of strikes caused more extensive damage to energy infrastructure than during the winter of 2022–2023, increasing the risk of prolonged disruptions to essential urban services [42]. The World Bank’s RDNA4 report documents operational disruptions in urban life, including frequent air raid alerts, interruptions to commercial and service operations, logistical delays, and reduced productivity [43]. Summer 2025 marked one of the most devastating periods in terms of civilian casualties and displacement. Intensified attacks worsened access to services and increased the need for shelters and humanitarian support in Kyiv and the surrounding region [44].

2.2. Air Quality Monitoring Methodology and PM2.5 Mapping

Figure 1b shows 94 sites, from where in situ measurements were collected and aggregated within the Air Quality Monitoring platform SaveEcoBot [45] during the period of 1–30 June 2025. The air quality monitoring stations are presented only if they reported data at least once during June 2025, although many exhibit gaps in data acquisition for various reasons. The platform consolidates real-time data on air quality provided by several networks [46]. This study analyses data collected from the SaveDnipro network—22 stations (https://www.savednipro.org (accessed on 1 August 2025)), Sensor.Community—18 stations (https://luftdaten.info/ (accessed on 1 August 2025)), AirVisual—3 stations (https://www.iqair.com/ (accessed on 1 August 2025)), LUN Misto—18 stations (https://lun.ua/misto/air (accessed on 1 August 2025)), Taras Shevchenko National University of Kyiv (KNU)—1 station (https://ecoimpact.knu.ua/ (accessed on 1 August 2025)), Kyiv City State Administration (KCSA)—25 stations (https://kyivcity.gov.ua/ (accessed on 1 August 2025)), and Airly—7 stations (https://airly.org/ (accessed on 1 August 2025)). Using WiFi, the collected data are transmitted to data processing, storage, and visualization platforms.
The providers of air quality data use several types of optical particle counters using light scattering.
SaveDnipro and Sensor.Community stations are equipped with Nova SDS011 optical sensors (Nova Fitness Co., Ltd., Jinan, China) enabling real-time monitoring of airborne particulate fractions of 2.5 µm and 10 µm collected every 2.5 min. The Plantower PMS series optical sensors (models PMS3003, PMS5003, PMS7003, and PMSA003; Plantower Co., Ltd., Beijing, China) for PM1, PM2.5, and PM10 are used by LUN Misto, KNU, KCSA, and Airly at 2 min (LUN and KNU), 15 min (Airly), and 20 min (KCSA) resolution. The integrated temperature-humidity-pressure sensors allow automatic adjustment of the obtained PM data depending on weather conditions, while the heating module minimizes interference during fog, precipitation, and subzero temperatures. SaveDnipro and Sensor.Community stations provide data on PM2.5, PM10, temperature (T), relative humidity (H), and atmospheric pressure (P). LUN Misto, KNU, KCSA, and Airly stations, in addition to PM indicators, also measure T, H, P, nitric oxide (NO), nitrogen dioxide (NO2), carbon dioxide (CO2), ozone (O3), hydrogen sulfide (H2S), sulfur dioxide (SO2), and formaldehyde (CH2O). Studies on particle-size selectivity of optical low-cost particulate matter sensors have shown that the SDS011 and PMS sensors both can be reasonably accurate in the measurements of PM2.5, but are not suitable for the measurement of coarse-mode particles, and the measurements of PM10 can be grossly inaccurate [47,48]. It is also pointed out that the ability to measure PM2.5 depends on the stability of the ambient air size distribution, humidity, and aerosol composition [49,50]. The study on reproducibility and evaluation of low-cost sensors for ambient PM2.5 monitoring [47] claims that values of the coefficient of variation were lower than 7% in the case of SDS011 and PMS sensors, which allows their joint use in monitoring. However, quite a large dispersion of data and high relative errors of PM2.5 estimations were observed for concentration ranges below 20–30 μg/m3.
The stations of the AirVisual network are equipped with an optical sensor AVPM25b (proprietary sensor co-developed by AirVisual/IQAir; manufacturer not publicly disclosed) for aerosol concentration measurements, which allows determining the concentrations of PM1, PM2.5, and PM10 in the range from 0 to 1000 µg/m3, with measuring accuracy of ±10% and 1-h resolution [51]. AirVisual stations also measure T, H, and P.
LUN Misto and AirVisual data are reported to be justified against measurements of PM2.5 mass concentrations performed using the HORIBA APDA-371 Air Pollution Dust Analyzer, HORIBA, Ltd., Kyoto, Japan [35,52,53] installed in the Popudrenka station, which operates in accordance with European Union and United States Environmental Protection Agency (EPA) regulations and belongs to the O.M. Marzeiev Institute for Public Health of the National Academy of Medical Sciences of Ukraine (50.459° N, 30.634° E).
The performance of low-cost air quality sensors is subject to disruptions that can compromise data quality, such as electrical instabilities and data transmission failures, creating gaps in time series. Sensors are usually mounted outdoors, 1.5–3 m above the ground, 3–5 m away from the busy streets in well-airable places.
Although the accuracy of data produced by low-cost sensors remains a subject of discussion, particularly when benchmarked against high-precision reference instruments, numerous studies concur that these sensors perform well in tracking spatial heterogeneity and capturing short-term variations and diurnal trends of PM2.5 in urban areas, which is in line with the purpose of our study.
This study focuses on PM2.5, continuously monitored over a one-month period from 1 to 30 June 2025. For PM2.5, raw mass concentration data (µg/m3) were extracted from each operating sensor in the city of Kyiv and near the outskirts, with temporal resolutions ranging from 2 min to 1 h in the original time series.
Each Russian attack on Kyiv, which happened in June 2025, is documented as a time series starting from the moment of air raid alert announcement at each of the operating stations till the time the PM2.5 concentrations reach background values. In this study, we demonstrate two strike episodes of 10 and 17 June 2025.
The collected data were used to simulate time-lapse spatial distributions of PM2.5 for each strike episode, covering the period “from background to background” through spatial interpolation. To determine the most reasonable and effective method for estimating PM concentrations at unsampled locations, cross-validation analysis was performed for the most commonly used conventional interpolation techniques, based on the relative positions of monitoring points (e.g., Inverse Distance Weighting (IDW), kriging, Radial Basis Function (RBF), Natural Neighbor (NN)). The evaluation was based on the RMSE (Root Mean Square Error) and MAPE (Mean Absolute Percentage Error) of estimated PM2.5 concentrations [54] and implemented using the Spatial Analyst Tools in ArcGIS Desktop 10.8 and ArcGIS Pro 3.0.2.

2.3. Timeline of Air Strikes

In June 2025, Kyiv was subjected to a sustained campaign of combined ballistic missile and drone strikes, demonstrating both increased frequency and lethality. According to the monitoring mission Armed Conflict Location & Event Data (ACLED), the first three weeks of June alone saw at least 80–90 long-range aerial attacks, with more than 21 strikes confirmed within Kyiv [55,56,57,58].
On the night of 5–6 June, Russia executed one of its largest aerial blitzes, deploying hundreds of drones and missiles across Kyiv and other regions. This assault resulted in at least four civilian deaths and widespread infrastructural devastation across multiple districts. A second major bombardment occurred on 10 June, during which Russia launched 315 drones and seven missiles. This strike caused fatalities—including at least three civilian deaths—and inflicted injuries across seven of Kyiv’s ten districts, with extensive fires and structural damage reported. The deadliest attack occurred overnight on 16–17 June, when 175 drones and 16 missiles struck Kyiv, resulting in 14 to 28 civilian deaths and over 100 injuries. The attack razed more than 50 residential buildings, educational and railway infrastructure, and collapsed a nine-story apartment block in the Solomianskyi district. This event was characterized by both ACLED and the United Nations Human Rights Monitoring Mission in Ukraine [59] as the most lethal attack on Kyiv in nearly a year. Finally, on 23 June, another combined drone-missile strike killed approximately nine to ten civilians and wounded over 30, with multiple city districts suffering damage.
In June 2025, the total duration of air raid alerts reached 71 h and 27 min (Figure 2g). The longest individual episodes occurred on 1 June (9 h 32 min), 8–9 June (6 h 29 min), and 17 June (8 h 59 min). Not every strike resulted in a large fire or visible smoke plume; however, contemporary reporting from Kyiv documented extensive smoke when major urban fires followed strikes on infrastructure and fuel-storage facilities. Large, coordinated barrages increased the likelihood of multiple simultaneous fires merging into city-scale smoke events. Reuters and other outlets described the 10 June and 17 June assaults on Kyiv as particularly heavy barrages. Eyewitness accounts and photographic evidence from the city show both uncontrolled fires and rapid emergency responses in different cases. Dense smoke over Kyiv was reported on 10 and 17 June 2025.

3. Results

3.1. Cross-Network Mutual Verification of PM2.5 Sensor Performance

Cross-network mutual verification was conducted as a quality assurance step ensuring that PM2.5 data from different sensors and networks can be reliably compared, merged and used jointly for spatial interpolation [53].
Figure 3 presents graphs comparing hourly averaged PM2.5 data from selected stations of different monitoring networks collected between 1 and 5 June 2025. Each pair of stations presented in the graphs is located within 50 m of one another (Figure 4). The consistency of the particulate matter concentration data is confirmed by linear regression analysis, with Pearson correlation coefficients of r = 0.89 for the pair of Science.Community station #24340 and KCSA station #24181 (Figure 3a), r = 0.94 for KCSA station #24181 and Airly station #17112 (Figure 3b), and r = 0.90 for KCSA station #22643 and KCSA station #24459 (Figure 3c).
It should be noted that the issue of systematic offset visible in the graphs remains unresolved, as it may reflect both varying urban conditions and calibration differences. However, relative temporal variations in PM2.5 concentration can still be reliably captured, while care should be taken to filter out stations with large systematic offsets from the dataset prior to spatial interpolation of PM2.5 concentrations.

3.2. Temporal Variations of PM2.5

Figure 2 shows temporal variations of raw PM2.5 concentrations at six selected sites, representing different urban conditions and data providers (Figure 4).
AirVisual station #3498, located at 57 Sichovyh Striltsiv St., Airly station #17114, located at 17b Holosiivskij Av. and KCSA station #14634, located at 26 Verbytskoho St. represent yard areas protected by a green barrier from the busy roads with PM2.5 values averaging below 20 μg/m3.
LUN Misto station #12950, located at 23B Urlivska St., Science. Community station #24340, located at 37 Kniaziv Ostrozkyh St., and SaveDnipro station #23976, located at 33 Chokolivskyi Av. represent green residential area with PM2.5 values averaging below 10 μg/m3.
Temporal variation patterns at the selected sites exhibit noticeable correlations, despite the sites being widely separated within the city. During the first two decades of June, PM2.5 concentrations were moderately higher than in the last decade, an effect that may be attributed to meteorologically induced stagnation of near-surface traffic emissions. Alternatively, it may be associated with a deterioration in air quality in Kyiv due to fires in the near outskirts [46,53].
The PM2.5 concentration time series exhibits short-duration spikes, some of which can be attributed to drone and missile strikes and the combustion processes that followed. While not every strike resulted in fire, certain PM2.5 peaks can be clearly linked to large-scale attacks that produced extensive damage and city-wide smog. The ability of an air quality monitoring station to record such events depends on its proximity to the strike site and on favorable meteorological conditions. Wind speed and direction play a crucial role in the dispersion of aerosol pollutants.
Among the selected monitoring stations, the overnight attack of 9–10 June is distinctly reflected in the records of sensors #14634 and #12950, which registered peaks of 80 and 70 μg/m3, respectively. The duration of the corresponding anomalies exceeded the official air raid alert period, as multiple fires were extinguished only 2–4 h later. Similarly, the PM2.5 concentration patterns associated with the combined attack of 17 June—one of the most destructive of the entire war—were captured by the stations #14634, #12950, #23976, and #17114. At the station #17114, located along the smoke plume trajectory, a peak concentration exceeding 90 μg/m3 was recorded. The temporal characteristics of these anomalies more closely reflect fire dynamics than the duration of the attacks themselves, generally extending beyond the time frame of the strikes.

3.3. Relationship Between PM2.5, Meteorological Parameters, and Gaseous Pollutants

Figure 5 presents the hourly averaged concentrations of PM2.5 at the Station #14634, in comparison with the meteorological parameters and gas pollutants obtained for the period of intense Russian attacks on Kyiv in June 2025.
The temperature data display very similar monthly behavior to the PM2.5 data, suggesting that the thermal trend shapes PM2.5 concentrations in the city, as stagnant air and thermal inversions can trap pollutants near the surface, preventing dispersion in the warm season. However, neither temperature nor humidity correlates with peak values in PM2.5 concentrations, showing no relation between them and meteorological factors.
The NO2 and CO data exhibit similar daily patterns to the PM2.5 data; however, they do not follow the long-term monthly trend observed in PM2.5, suggesting a common local source and a relatively short atmospheric residence time for both gases. These gases also show a sharp increase in the hours of heavy strike and fires on 17 June 2025, but not on 10 June 2025.

3.4. Determination of the Optimal PM2.5 Interpolation Method

According to the literature, Inverse Distance Weighting (IDW) and kriging are the most commonly used conventional interpolation methods for PM estimation [60], although the applicability of each approach is site-specific and remains a subject of considerable debate, e.g., [54,61]. Less commonly used but still advantageous for handling irregularly distributed data are the Radial Basis Function (RBF) and Natural Neighbor (NN) interpolation methods [62,63]. Cross-validation–based comparative evaluations were conducted to determine the optimal technique for spatial estimation of PM2.5 concentrations.
The peak episode of PM2.5 contamination, associated with the strike on 17 June and observed at 7:00 a.m., was used to calculate statistical error metrics required for cross-validation of the interpolation methods. PM2.5 concentration data from 52 monitoring stations (Figure 4), recorded within 15 min of the modeled time (7:00 a.m.), were used for further analysis. Among all stations, observations at 41 (black dots in Figure 4) were used as reference data, whereas observations at the remaining 11 stations (green dots in Figure 4) were used for accuracy validation. The performance of the method was evaluated with the RMSE and MAPE of estimated PM concentrations.
The IDW showed better performance than other methods for the PM2.5 interpolation (Table 1). The results of IDW achieve the lowest RMSE and MAPE. The accuracy of the interpolation methods in terms of RMSE decreased in the order of the Ordinary Kriging (OK), Natural Neighbor, Radial Basis Function and IDW. MAPE values reached extremely high levels for the OK and RBF methods due to a combination of unfavorable conditions, including the sparse and uneven distribution of monitoring stations and sharp spikes in PM2.5 concentrations during air strikes, which may exceed background levels by up to order of magnitude. These factors make the OK and RBF methods unsuitable for spatial prediction of PM2.5 concentrations.

3.5. Spatio-Temporal Distributions of PM2.5 Following Air Strike Episodes

This subsection examines two air strikes in Kyiv, on 10 and 17 June 2025, both of which resulted in fires and extensive smoke episodes.
Meteorological conditions for the days of the strikes were obtained from the database of the Central Geophysical Observatory named after Borys Sreznevskyi (http://cgo-sreznevskyi.kyiv.ua/ (accessed on 4 August 2025)). No precipitation was recorded during either episode. On 10 June, the air temperature ranged from 13.4 to 20.6 °C, while on 17 June it varied between 15.3 and 30.0 °C. Wind speeds on both days remained below 5 m/s. Wind direction shifted from northwest to west during the first episode and remained predominantly southwest during the second (Figure 6).
The first episode extended from the onset of the air raid alert at 00:02 on 10 June to the suppression of fires at 11:00 on 10 June, approximately six hours after the alert had ended (Figure 7). It was recorded by 53 of the 94 monitoring stations operational in June 2025. The explosion that occurred after 3:00 caused fire in the city center. Significant air quality deterioration was first detected at 04:00 on 10 June, when a plume of approximately 30 km2 developed over the western-bank city center with maximal PM2.5 concentrations of up to 39 µg/m3. Fire spread in the central area and the emergence of additional ignition sites by 05:00 produced a plume with a pronounced southeastward trajectory, consistent with prevailing winds. Maximal PM2.5 value of 173 µg/m3 was reached in Holosiivskyi district at 6:00. By 08:00, fires on the western bank were largely extinguished and those on the eastern bank were substantially reduced; however, renewed fire activity at 9:00 in the Darnytskyi district resulted in maximum PM2.5 concentrations of up to 105 µg/m3. Background levels were restored by 11:00 following the implementation of firefighting measures.
The air raid alert corresponding to the second episode (Figure 8) began at 21:06 on 16 June 2025; however, no strikes were recorded in Kyiv until 02:30 on 17 June. The subsequent intensive missile barrage, combined with a drone attack, ignited large fires that produced extensive smoke over the entire southern part of the western-bank city. The highest PM2.5 concentrations, reaching 241 µg/m3 between 05:00 and 07:00, were observed in the Solomianskyi and Holosiivskyi districts. The southern wind facilitated the propagation of the smoke plume along the Dnipro River. On the eastern bank, localized fire outbreaks were recorded, with PM2.5 concentrations ranging from 39 to 105 µg/m3. Firefighting operations continued on the eastern bank between 08:00 and 10:00, and by 11:00, PM2.5 levels across the city had returned to background values. This episode was captured by 52 of the 94 monitoring stations working in June 2025.

4. Discussion

As noted by Wei et al., 2025 [35], monthly average PM2.5 concentrations derived from the AirVisual network in Kyiv were lower in 2022 compared to 2020 and 2021, indicating an overall improvement in air quality despite the anticipated negative impacts of bombardments, large-scale fires, increased use of generators during power outages, and greater reliance on wood heating in the autumn and winter months. This suggests that war-related humanitarian crisis factors, such as population decline and industrial disruption, played a dominant role in shaping air quality. Such dynamics may generate seemingly contradictory patterns: short-term measurements capture acute pollution episodes linked to local sources, whereas long-term averages tend to smooth these fluctuations and instead reveal broader seasonal or interannual trends. This complexity highlights the necessity of multi-temporal monitoring approaches to adequately characterize PM2.5 exposure and the associated health risks [64].
Repeated airstrikes involving large numbers of drones (and sometimes ballistic/cruise missiles) targeting Kyiv have become more frequent in 2025 compared to previous years (2022–2024). The trend is especially clear for drone attacks, both in terms of total number and in the use of massive coordinated barrages. As demonstrated by the air strikes of 10 and 17 June 2025, each lasting up to nine hours, peak moments were characterized by a tenfold increase in raw PM2.5 concentrations recorded by air quality monitoring stations (Figure 2, Figure 7 and Figure 8). These episodes led to a marked elevation of daily average PM2.5 levels, both within the most affected urban districts and in downwind areas located along the trajectories of smoke plumes generated by the fires (Figure 9).
Daily PM2.5 concentrations are subject to regulatory standards in both Europe and the United States. Under the U.S. National Ambient Air Quality Standards (NAAQS) for fine particulate matter, the 24-h (daily) standard established in 2013 is 35 µg/m3 [65]. In the European Union, the new Directive (EU) 2024/2881 introduces a 24-h limit value for PM2.5 of 25 µg/m3, not to be exceeded more than 18 times per calendar year [66]. The World Health Organization’s Air Quality Guidelines (2021) recommend a more stringent 24-h mean of 15 µg/m3, with exceedances limited to no more than 3–4 days annually [18].
Strike episodes can substantially elevate diurnal PM2.5 concentrations, at times exceeding WHO and EU standards (Figure 9), although their impact is diluted in monthly and annual averages. It is important to note that routine urban activities alone may cause exceedances, while strikes can exacerbate this situation. Not all strikes, however, have such effects: those impacting fuel depots, petrochemical facilities, warehouses with combustible materials, or densely built structures tend to generate large, sustained fires and consequently far greater particulate emissions (PM2.5) than strikes on non-combustible targets or open areas.

5. Conclusions

PM2.5 has proven to be a robust indicator of air quality, capturing long-term tendencies and short-term episodes that are not adequately reflected by gaseous pollutants. The deployment of low-cost air quality sensors by citizen scientists in Kyiv provides an effective means of documenting the environmental consequences of armed conflict. Despite incomplete datasets, the existing PM monitoring network enables the mapping of individual pollution episodes, estimation of affected areas, and modeling of the impacts of mass bombardments and subsequent fires.
During the Russian missile and drone attacks of June 2025, the sensors recorded sharp spikes in PM2.5 concentrations, offering quantitative evidence of pollution directly associated with military activity. The prevailing winds were shown to play a decisive role in the spatial development of fire-related smoke plumes. Although the configuration of air defense systems is likely relevant, available data are insufficient to assess its influence. Short-term episodes (up to 8–9 h) were found to elevate daily average PM2.5 concentrations beyond WHO guideline values.
In this context, citizen-led monitoring not only advances scientific understanding of conflict-related air pollution but may also serve as supplementary evidence in assessments of potential environmental impacts and their broader public health implications.

Author Contributions

K.B.—Conceptualization, methodology, writing—original draft preparation; I.T.—validation, formal analysis, writing—review and editing, visualization; M.V.—software, data curation. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported within the framework of the project PAN.BFB.S.BWZ.394.022.2023, being implemented through the Long-term program of support of the Ukrainian research teams at the Polish Academy of Sciences, carried out in collaboration with the U.S. National Academy of Sciences, with the financial support of external partners.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data will be made available on request.

Acknowledgments

This work was supported by the Ministry of Education and Science of Ukraine (grant for the prospective development of the scientific direction “Mathematical sciences and natural sciences” at the Taras Shevchenko National University of Kyiv).

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Pope, C.A.; Dockery, D.W. Health Effects of Fine Particulate Air Pollution: Lines that Connect. J. Air Waste Manag. Assoc. 2006, 56, 709–742. [Google Scholar] [CrossRef]
  2. Rogula-Kozłowska, W.; Błaszczak, B.; Szopa, S.; Klejnowski, K.; Sówka, I.; Zwoździak, A.; Jabłońska, M.; Mathews, B. PM2.5 in the central part of Upper Silesia, Poland: Concentrations, elemental composition, and mobility of components. Environ. Monit. Assess. 2013, 185, 581–601. [Google Scholar] [CrossRef]
  3. Majewski, G.; Przewozniczuk, W. Study of particulate matter pollution in the Warsaw area. Pol. J. Environ. Stud. 2009, 18, 293–300. [Google Scholar]
  4. Jordanova, D.; Jordanova, N.; Lanos, P.; Petrov, P.; Tsacheva, T. Magnetism of outdoor and indoor settled dust and its utilization as a tool for revealing the effect of elevated particulate air pollution on cardiovascular mortality. Geochem. Geophys. Geosyst. 2012, 13, Q08Z49. [Google Scholar] [CrossRef]
  5. Hofman, J.; Castanheiro, A.; Nuyts, G.; Joosen, S.; Spassov, S.; Blust, R.; De Wael, K.; Lenaerts, S.; Samson, R. Impact of urban street canyon architecture on local atmospheric pollutant levels and magneto-chemical PM10 composition: An experimental study in Antwerp, Belgium. Sci. Total Environ. 2020, 712, 135534. [Google Scholar] [CrossRef]
  6. Matei, E.; Râpă, M.; Mateș, I.M.; Popescu, A.-F.; Bădiceanu, A.; Balint, A.I.; Covaliu-Mierlă, C.I. Heavy Metals in Particulate Matter—Trends and Impacts on Environment. Molecules 2025, 30, 1455. [Google Scholar] [CrossRef] [PubMed]
  7. Jeleńska, M.; Górka-Kostrubiec, B.; Werner, T.; Kądziałko-Hofmokl, M.; Szczepaniak-Wnuk, I.; Gonet, T.; Szwarczewski, P. Evaluation of indoor/outdoor urban air pollution by magnetic, chemical and microscopic studies. Atmos. Pollut. Res. 2017, 8, 754–766. [Google Scholar] [CrossRef]
  8. Muxworthy, A.R.; David Green, C.L.; Cowan, A.; Maher, B.A.; Gonet, T. Magnetic characterisation of London’s airborne nanoparticulate matter. Atmos. Environ. 2022, 287, 119292. [Google Scholar] [CrossRef]
  9. Winkler, A.; Amoroso, A.; Di Giosa, A.; Marchegiani, G. The effect of COVID-19 lockdown on airborne particulate matter in Rome, Italy: A magnetic point of view. Environ. Pollut. 2021, 291, 118191. [Google Scholar] [CrossRef] [PubMed]
  10. Dytłow, S.; Kida, M.; Ziembowicz, S. Examining Magnetic Susceptibility as a Proxy for Microplastic Pollution in Granulometric Fractions of Road Dust: A Case Study in Warsaw, Poland. J. Hazard. Mater. 2025, 494, 138493. [Google Scholar] [CrossRef]
  11. Akhbarizadeh, R.; Dobaradaran, S.; Amouei Torkmahalleh, M.; Saeedi, R.; Aibaghi, R.; Faraji Ghasemi, F. Suspended fine particulate matter (PM2.5), microplastics (MPs), and polycyclic aromatic hydrocarbons (PAHs) in air: Their possible relationships and health implications. Environ. Res. 2021, 192, 110339. [Google Scholar] [CrossRef]
  12. Reche, C.; Querol, X.; Alastuey, A.; Viana, M.; Pey, J.; Moreno, T.; Rodríguez, S.; González, Y.; Fernández-Camacho, R.; de la Rosa, J.; et al. New considerations for PM, black carbon and particle number concentration for air quality monitoring across different European cities. Atmos. Chem. Phys. 2011, 11, 6207–6227. [Google Scholar] [CrossRef]
  13. Milford, C.; Fernández-Camacho, R.; Sánchez de la Campa, A.M.; Rodríguez, S.; Castell, N.; Marrero, C.; Bustos, J.J.; de la Rosa, J.D.; Stein, A.F. Black carbon aerosol measurements and simulation in two cities in south-west Spain. Atmos. Environ. 2016, 126, 55–65. [Google Scholar] [CrossRef][Green Version]
  14. Baensch-Baltruschat, B.; Kocher, B.; Stock, F.; Reifferscheid, G. Tyre and road wear particles (TRWP): A review of generation, properties, emissions, human health risk, ecotoxicity, and fate in the environment. Sci. Total Environ. 2020, 733, 137823. [Google Scholar] [CrossRef]
  15. Kelly, F.J.; Fussell, J.C. Size, source and chemical composition as determinants of toxicity attributable to ambient particulate matter. Atmos. Environ. 2012, 60, 504–526. [Google Scholar] [CrossRef]
  16. Dytłow, S.; Karasiński, G. Exploring the particle size effect, land use classification, and magnetic characteristics of street dust in urbanized areas in Poland. Sci. Rep. 2025, 15, 12844. [Google Scholar] [CrossRef]
  17. Tang, L.; Feng, Z.; Shang, D.; Zeng, L.; Wu, Z.; Wang, H.; Chen, S.; Li, X.; Zeng, L.; Hu, J.; et al. Ongoing uncoordinated anthropogenic emission abatement promotes atmospheric new particle growth in a Chinese megacity. Nat. Commun. 2025, 16, 6720. [Google Scholar] [CrossRef]
  18. WHO Global Air Quality Guidelines. Available online: https://www.eea.europa.eu/en/analysis/publications/air-quality-in-europe-2022/europes-air-quality-status-2022 (accessed on 30 June 2025).
  19. Lequy, E.; Siemiatycki, J.; Leblond, S.; Meyer, C.; Zhivin, S.; Vienneau, D.; Hoogh, K.; Goldberg, M.; Zins, M.; Jacquemin, B. Long-term exposure to atmospheric metals assessed by mosses and mortality in France. Environ. Int. 2019, 129, 145–153. [Google Scholar] [CrossRef]
  20. Sangkham, S.; Phairuang, W.; Sherchan, S.P.; Pansakun, N.; Munkong, N.; Sarndhong, K.; Islam, M.A.; Sakunkoo, P. An update on adverse health effects from exposure to PM2.5. Environ. Adv. 2024, 18, 100603. [Google Scholar] [CrossRef]
  21. Alves, C.; Evtyugina, M.; Vicente, E.; Vicente, A.; Rienda, I.C.; de la Campa, A.S.; Tom’e, M.; Duarte, I. PM2.5 chemical composition and health risks by inhalation near a chemical complex. J. Environ. Sci. 2023, 124, 860–874. [Google Scholar] [CrossRef]
  22. Loomis, D.; Huang, W.; Chen, G. The international agency for research on cancer (IARC) evaluation of the carcinogenicity of outdoor air pollution: Focus on China. Chin. J. Cancer 2014, 33, 189–196. [Google Scholar] [CrossRef]
  23. Tian, F.; Qi, J.; Qian, Z.; Li, H.; Wang, L.; Wang, C.; Geiger, S.D.; McMillin, S.E.; Yin, P.; Lin, H.; et al. Differentiating the effects of air pollution on daily mortality counts and years of life lost in six Chinese megacities. Sci. Total Environ. 2022, 827, 154037. [Google Scholar] [CrossRef]
  24. Rodríguez-Trejo, A.; Bohnel, H.N.; Ibarra-Ortega, H.E.; Salcedo, D.; González-Guzmán, R.; Castaneda-Miranda, A.G.; Sánchez-Ramos, L.E.; Chaparro, M.A.E.; Chaparro, M.A.E. Air Quality Monitoring with Low-Cost Sensors: A Record of the Increase of PM2.5 during Christmas and New Year’s Eve Celebrations in the City of Queretaro, Mexico. Atmosphere 2024, 15, 879. [Google Scholar] [CrossRef]
  25. Khobragade, P.P.; Vikram Ahirwar, A. Chemical and morphological characterization of PM2.5 samples collected over an urban industrial region Raipur, Chhattisgarh. Acta Geophys. 2023, 71, 3057–3076. [Google Scholar] [CrossRef]
  26. Tai, A.P.K.; Mickley, L.J.; Jacob, D.J. Correlations between fine particulate matter (PM2.5) and meteorological variables in the United States: Implications for the sensitivity of PM2.5 to climate change. Atmos. Environ. 2010, 44, 3976–3984. [Google Scholar] [CrossRef]
  27. Bondar, K.M.; Tsiupa, I.V. Long- and short-term pollution effect in megapolis assessed from magnetic and geochemical measurements on soils, tree trunk bark, and air filters. Environ. Monit. Assess. 2024, 196, 1041. [Google Scholar] [CrossRef]
  28. Zalakeviciute, R.; Mejia, D.; Alvarez, H.; Bermeo, X.; Bonilla-Bedoya, S.; Rybarczyk, Y.; Lamb, B. War impact on air quality in Ukraine. Sustainability 2022, 14, 13832. [Google Scholar] [CrossRef]
  29. Ialongo, I.; Bun, R.; Hakkarainen, J.; Virta, H.; Oda, T. Satellites capture socioeconomic disruptions during the 2022 full-scale war in Ukraine. Sci. Rep. 2023, 13, 14954. [Google Scholar] [CrossRef]
  30. Zhang, C.; Hu, Q.; Su, W.; Xing, C.; Liu, C. Satellite spectroscopy reveals the atmospheric consequences of the 2022 Russia–Ukraine war. Sci. Total Environ. 2023, 869, 161759. [Google Scholar] [CrossRef]
  31. Bondar, K.M.; Tsiupa, I.V.; Menshov, O. War-time changes in air pollution across Ukrainian cities were assessed through magnetic susceptibility and heavy metal contents in particulate matter collected on air filters. Environ. Pollut. 2025, 385, 127076. [Google Scholar] [CrossRef]
  32. Hryhorczuk, D.; Levy, B.S.; Prodanchuk, M.; Kravchuk, O.; Bubalo, N.; Hryhorczuk, A.; Erickson, T.B. The environmental health impacts of Russia’s war on Ukraine. J. Occup. Med. Toxicol. 2024, 19, 1. [Google Scholar] [CrossRef]
  33. Meng, X.; Lu, B.; Liu, C.; Zhang, Z.; Chen, J.; Herrmann, H.; Li, X. Abrupt exacerbation in air quality over Europe after the outbreak of Russia–Ukraine war. Environ. Int. 2023, 178, 108120. [Google Scholar] [CrossRef]
  34. Malytska, L.; Galytska, E.; Savenets, M.; Ladstätter-Weißenmayer, A.; Krajčovičová, J. The Impact of Fires on Air Quality in Ukraine During Two Years of Military Conflict (2022–2023): Analyzing Satellite, Ground-Based Observations of NO2, CO, and Aerosols. 2025. Available online: https://ssrn.com/abstract=5185870 (accessed on 10 August 2025).
  35. Wei, X.; Yukhymchuk, Y.; Danylevsky, V.; Milinevsky, G.; Goloub, P.; Fesianov, I.; Syniavskyi, I.; Turos, O.; Maremukha, T.; Petrosian, A.; et al. Impact of military activity on atmospheric aerosol characteristics in Ukraine and Kyiv city. Atmos. Pollut. Res. 2025, 16, 102496. [Google Scholar] [CrossRef]
  36. War Worsens Climate and Environmental Challenges in Ukraine. Joint Research Centre (11 April 2025). Available online: https://joint-research-centre.ec.europa.eu/jrc-news-and-updates/war-worsens-climate-and-environmental-challenges-ukraine-2025-04-11_en (accessed on 7 June 2025).
  37. Ecological Passport City of Kyiv in 2022 Year (2023) Assessment Report. Available online: https://ecodep.kyivcity.gov.ua/ekolohichnyi-pasport-ta-rehionalna-dopovid/ekolohichnyi-pasport (accessed on 9 August 2025). (In Ukrainian)
  38. Vyshnievskyi, V.I.; Donich, O.A.; Kutsyi, A.V. The Climate of Kyiv and Its Outskirts; Naukova Dumka: Kyiv, Ukraine, 2023; p. 124. (In Ukrainian) [Google Scholar]
  39. Shevchenko, O.G.; Snizhko, S.I.; Matviienko, M.O. Simulation of the thermal comfort conditions of urban areas: A case study in Kyiv. Visnyk V. N. Karazin Kharkiv Natl. Univ. Ser. Geol. Geogr. Ecol. 2019, 51, 186–198. [Google Scholar] [CrossRef]
  40. Regional Report on the Condition of Environment in the Cty of Kyiv in 2023. IEA, 2024. Available online: https://mepr.gov.ua/wp-content/uploads/2024/10/Regionalna-dopovid-2023.zip (accessed on 15 August 2025). (In Ukrainian)
  41. IEA (2024), Ukraine’s Energy Security and the Coming Winter, IEA, Paris. Licence: CC BY 4.0. Available online: https://www.iea.org/reports/ukraines-energy-security-and-the-coming-winter (accessed on 12 September 2025).
  42. Office of the United Nations High Commissioner for Human Rights (19 September 2024). Attacks on Ukraine’s Electricity inFrastructure Threaten Key Aspects of Life as Winter Approaches. Available online: https://ukraine.ohchr.org/en/Attacks-On-Ukraines-Electricity-Infrastructure (accessed on 23 September 2025).
  43. World Bank. 2025. Ukraine—Fourth Rapid Damage and Needs Assessment (RDNA4). PRESS RELEASE NO: 2025/ECA/079. Available online: https://www.worldbank.org/en/news/press-release/2025/02/25/updated-ukraine-recovery-and-reconstruction-needs-assessment-released (accessed on 23 September 2025).
  44. United Nations Office for the Coordination of Humanitarian Affairs. (2025). Ukraine: Humanitarian Situation Snapshot (June–July 2025). Available online: https://www.unocha.org/publications/report/ukraine/ukraine-humanitarian-situation-snapshot-june-july-2025 (accessed on 25 September 2025).
  45. Air Quality in the City Kyiv. Available online: https://www.saveecobot.com/ (accessed on 28 June 2025).
  46. Shelestov, A.; Yailymova, H.; Yailymov, B.; Kussul, N. Air Quality Estimation in Ukraine Using SDG 11.6.2 Indicator Assessment. Remote Sens. 2021, 13, 4769. [Google Scholar] [CrossRef]
  47. Badura, M.; Batog, P.; Drzeniecka-Osiadacz, A.; Modzel, P. Evaluation of low-cost sensors for ambient PM2.5 monitoring. J. Sens. 2018, 2018, 5096540. [Google Scholar] [CrossRef]
  48. Kuula, J.; Makela, T.; Aurela, M.; Teinila, K.; Varjonen, S.; González, Ó.; Timonen, H. Laboratory evaluation of particle-size selectivity of optical low-cost particulate matter sensors. Atmos. Meas. Tech. 2020, 13, 2413–2423. [Google Scholar] [CrossRef]
  49. Sayahi, T.; Butterfield, A.; Kelly, K.E. Long-term field evaluation of the Plantower PMS low-cost particulate matter sensors. Environ. Pollut. 2019, 245, 932–940. [Google Scholar] [CrossRef]
  50. Zusman, M.; Schumacher, C.S.; Gassett, A.J.; Spalt, E.W.; Austin, E.; Larson, T.V.; Carvlin, G.; Seto, E.; Kaufman, J.D.; Sheppard, L. Calibration of low-cost particulate matter sensors: Model development for a multi-city epidemiological study. Environ. Int. 2020, 134, 105329. [Google Scholar] [CrossRef]
  51. IQAir AirVisual Series. Available online: https://www.iqair.com/air-quality-monitors/airvisual-pro (accessed on 2 August 2025).
  52. Air Quality Map of Ukraine. Available online: https://lun.ua/misto/air (accessed on 2 August 2025).
  53. Zhang, C.; Shulga, V.; Milinevsky, G.; Danylevsky, V.; Yukhymchuk, Y.; Kyslyi, V.; Syniavsky, I.; Sosonkin, M.; Goloub, P.; Turos, O.; et al. Spring 2020 Atmospheric Aerosol Contamination over Kyiv City. Atmosphere 2022, 13, 687. [Google Scholar] [CrossRef]
  54. Choi, K.; Chong, K. Modified Inverse Distance Weighting Interpolation for Particulate Matter Estimation and Mapping. Atmosphere 2022, 13, 846. [Google Scholar] [CrossRef]
  55. Ukraine War Situation Update: 7–13 June 2025. ACLED. Available online: https://acleddata.com/update/ukraine-war-situation-update-7-13-june-2025 (accessed on 25 June 2025).
  56. Ukraine War Situation Update: 14–20 June 2025. ACLED. Available online: https://acleddata.com/update/ukraine-war-situation-update-14-20-june-2025 (accessed on 7 July 2025).
  57. Ukraine War Situation Update: 21–27 June 2025. ACLED. Available online: https://acleddata.com/update/ukraine-war-situation-update-21-27-june-2025 (accessed on 7 July 2025).
  58. Ukraine War Situation Update: 31 May–6 June 2025. ACLED. Available online: https://acleddata.com/update/ukraine-war-situation-update-31-may-6-june-2025 (accessed on 18 June 2025).
  59. Kyiv Suffers Deadliest Attack in Almost a Year, UN Human Rights Monitors Say. UN Human Rights Monitoring Mission in Ukraine. The United Nations in Ukraine. Available online: https://ukraine.un.org/en/296381-kyiv-suffers-deadliest-attack-almost-year-un-human-rights-monitors-say (accessed on 17 June 2025).
  60. Li, J.; Heap, A.D. A review of comparative studies of spatial interpolation methods in environmental sciences: Performance and impact factors. Ecol. Inform. 2011, 6, 228–241. [Google Scholar] [CrossRef]
  61. Shukla, K.; Kumar, P.; Mann, G.S.; Khare, M. Mapping spatial distribution of particulate matter using kriging and inverse distance weighting at supersites of megacity Delhi. Sustain. Cities Soc. 2020, 54, 101997. [Google Scholar] [CrossRef]
  62. Tang, Y.; Xie, S.; Huang, L.; Liu, L.; Wei, P.; Zhang, Y.; Meng, C. Spatial Estimation of Regional PM2.5 Concentrations with GWR Models Using PCA and RBF Interpolation Optimization. Remote Sens. 2022, 14, 5626. [Google Scholar] [CrossRef]
  63. Etherington, T.R. Discrete natural neighbour interpolation with uncertainty using cross-validation error distance fields. PeerJ Comput. Sci. 2020, 6, e282. [Google Scholar] [CrossRef]
  64. van Donkelaar, A.; Martin, R.V.; Li, C.; Burnett, R.T. Regional estimates of chemical composition of fine particulate matter using a combined geoscience-statistical method with information from satellites, models, and monitors. Environ. Sci. Technol. 2021, 55, 2874–2883. [Google Scholar] [CrossRef]
  65. NAAQS Table. United States Environmental Pollution Agency. NAAQS. Available online: https://www.epa.gov/criteria-air-pollutants/naaqs-table (accessed on 21 July 2025).
  66. Directive (EU) 2024/2881 of the European Parliament and of the Council of 23 October 2024 on Ambient Air Quality and Cleaner Air for Europe (Recast). Official Journal of the European Union, L 2024/2881, 20.11.2024. Available online: http://data.europa.eu/eli/dir/2024/2881/oj (accessed on 5 August 2025).
Urbansci 09 00477 g001
Figure 1. Map of Ukraine with capital city of Kyiv in the context of war situation: (a) Areas, occupied by Russia (red), and recaptured by Ukrainian military forces (blue) as of June 2025; (b) Map of Kyiv with labeled stations operated in June 2025 from the air quality monitoring platform SaveEcoBot.
Figure 1. Map of Ukraine with capital city of Kyiv in the context of war situation: (a) Areas, occupied by Russia (red), and recaptured by Ukrainian military forces (blue) as of June 2025; (b) Map of Kyiv with labeled stations operated in June 2025 from the air quality monitoring platform SaveEcoBot.
Urbansci 09 00477 g001
Urbansci 09 00477 g002
Figure 2. Temporal variations of raw PM2.5 concentrations at the selected stations in Kyiv in June 2025 (a–f) and the timeline of air raid alerts (g). Air strikes of 10 and 17 June are marked.
Figure 2. Temporal variations of raw PM2.5 concentrations at the selected stations in Kyiv in June 2025 (a–f) and the timeline of air raid alerts (g). Air strikes of 10 and 17 June are marked.
Urbansci 09 00477 g002
Urbansci 09 00477 g003
Figure 3. Comparison of hourly averaged PM2.5 data from the stations of different networks obtained on 1–5 June 2025: (a) Science.Community station #24340 vs. KCSA station #24181; (b) KCSA station #24181 vs. Airly station 17112; (c) KCSA station #22643 vs. KCSA station #24459.
Figure 3. Comparison of hourly averaged PM2.5 data from the stations of different networks obtained on 1–5 June 2025: (a) Science.Community station #24340 vs. KCSA station #24181; (b) KCSA station #24181 vs. Airly station 17112; (c) KCSA station #22643 vs. KCSA station #24459.
Urbansci 09 00477 g003
Urbansci 09 00477 g004
Figure 4. Map of Kyiv with labeled air monitoring stations used for cross-network verification (red dots); representative of PM2.5 temporal variations in different city areas (blue dots); used for interpolation testing reference stations (black dots) and validation stations (green dots).
Figure 4. Map of Kyiv with labeled air monitoring stations used for cross-network verification (red dots); representative of PM2.5 temporal variations in different city areas (blue dots); used for interpolation testing reference stations (black dots) and validation stations (green dots).
Urbansci 09 00477 g004
Urbansci 09 00477 g005
Figure 5. Concentrations of PM2.5 at the Station #14634, in comparison with: (a) temperature, (b) relative humidity, (c) CO concentrations, and (d) NO2 concentrations.
Figure 5. Concentrations of PM2.5 at the Station #14634, in comparison with: (a) temperature, (b) relative humidity, (c) CO concentrations, and (d) NO2 concentrations.
Urbansci 09 00477 g005
Urbansci 09 00477 g006
Figure 6. Wind parameters on the dates of air strikes: (a,b) wind speed and direction on 10 June 2025 with strike and fire interval indicated; (c,d) wind speed and direction on 17 June 2025 with strike and fire interval indicated.
Figure 6. Wind parameters on the dates of air strikes: (a,b) wind speed and direction on 10 June 2025 with strike and fire interval indicated; (c,d) wind speed and direction on 17 June 2025 with strike and fire interval indicated.
Urbansci 09 00477 g006
Urbansci 09 00477 g007
Figure 7. Time-slice maps of PM2.5 concentrations following the 10 June 2025 air strike, recorded at: (a) 3:00; (b) 4:00; (c) 5:00; (d) 6:00; (e) 7:00; (f) 8:00; (g) 9:00; (h) 10:00; (i) 11:00.
Figure 7. Time-slice maps of PM2.5 concentrations following the 10 June 2025 air strike, recorded at: (a) 3:00; (b) 4:00; (c) 5:00; (d) 6:00; (e) 7:00; (f) 8:00; (g) 9:00; (h) 10:00; (i) 11:00.
Urbansci 09 00477 g007
Urbansci 09 00477 g008
Figure 8. Time-slice maps of PM2.5 concentrations following the 17 June 2025 air strike, recorded at: (a) 00:00; (b) 2:00; (c) 4:00; (d) 5:00; (e) 7:00; (f) 8:00; (g) 9:00; (h) 10:00; (i) 11:00.
Figure 8. Time-slice maps of PM2.5 concentrations following the 17 June 2025 air strike, recorded at: (a) 00:00; (b) 2:00; (c) 4:00; (d) 5:00; (e) 7:00; (f) 8:00; (g) 9:00; (h) 10:00; (i) 11:00.
Urbansci 09 00477 g008
Urbansci 09 00477 g009
Figure 9. Average diurnal concentrations of PM2.5 at the selected stations in June 2025 (WHO [18], EU [66]).
Figure 9. Average diurnal concentrations of PM2.5 at the selected stations in June 2025 (WHO [18], EU [66]).
Urbansci 09 00477 g009
Table 1. PM2.5 interpolation errors for different methods.
Table 1. PM2.5 interpolation errors for different methods.
Inverse Distance WeightingOrdinary KrigingRadial Basis FunctionNatural Neighbor
RMSE (μg/m3)4.1612.369.4314.18
MAPE (%)26.18Extremely highExtremely high33.42
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Bondar, K.; Tsiupa, I.; Virshylo, M. Monitoring Air Pollution in Wartime Kyiv (Ukraine): PM2.5 Spikes During Russian Missile and Drone Attacks. Urban Sci. 2025, 9, 477. https://doi.org/10.3390/urbansci9110477

AMA Style

Bondar K, Tsiupa I, Virshylo M. Monitoring Air Pollution in Wartime Kyiv (Ukraine): PM2.5 Spikes During Russian Missile and Drone Attacks. Urban Science. 2025; 9(11):477. https://doi.org/10.3390/urbansci9110477

Chicago/Turabian Style

Bondar, Kseniia, Iryna Tsiupa, and Mykhailo Virshylo. 2025. "Monitoring Air Pollution in Wartime Kyiv (Ukraine): PM2.5 Spikes During Russian Missile and Drone Attacks" Urban Science 9, no. 11: 477. https://doi.org/10.3390/urbansci9110477

APA Style

Bondar, K., Tsiupa, I., & Virshylo, M. (2025). Monitoring Air Pollution in Wartime Kyiv (Ukraine): PM2.5 Spikes During Russian Missile and Drone Attacks. Urban Science, 9(11), 477. https://doi.org/10.3390/urbansci9110477

Article Metrics

Back to TopTop