The 2025 Paradox: What Kazhydromet’s Data Leaves Out

Almaty Air Initiative analyzed Kazhydromet’s official reports and the publicly available raw data from its automated air quality monitoring stations for 2025.

The analysis reveals a contradictory picture of air quality in Almaty in 2025.

Key Findings from the Annual Report

• Average annual pollution levels decreased by 31%. The Air Pollution Index (API) fell from 5.8 to 4.0, formally placing the city in the “low pollution” category.
• Extreme pollution episodes became more severe. The Standard Index (SI) increased from 7.9 to 9.6 (+21.5%). This means that, during one episode, the highest recorded concentration of a pollutant exceeded its maximum permissible concentration by a factor of approximately 10.
• The annual Maximum Frequency Index (MFI) increased. The index rose from 21% to 28% (+7 percentage points).
• The number of exceedances of maximum permissible concentrations changed significantly. Exceedances of the maximum one-time permissible concentration for ozone fell by approximately 100 times (−99%). Meanwhile, nitrogen dioxide (NO₂) exceedances increased to 12,309 cases (+11.6%), and nitric oxide (NO) exceedances rose to 2,749 cases (+50.2%).

The citywide annual average concentrations calculated from the raw data of automated monitoring stations, excluding manual monitoring posts, were*:

• PM2.5: 14.27 µg/m³
• PM10: 23.7 µg/m³
• SO₂: 25.0 µg/m³
• CO: 484.4 µg/m³
• NO₂: 39.8 µg/m³
• NO: 32.7 µg/m³
• O₃: 4.9 µg/m³

Measurements vary by tens of times between stations. According to automated monitoring data, annual average PM2.5 concentrations at individual stations range from less than 1 µg/m³ to 43.6 µg/m³. This calls for a careful examination of the monitoring network’s structure.

The year was warmer and considerably drier than 2024. In some months, precipitation was several times below normal, while weather conditions were frequently dominated by anticyclonic periods with weak winds. Such conditions limit the mixing of air masses and reduce the atmosphere’s natural cleansing processes, contributing to the accumulation of pollutants near the ground.

The 2025 Paradox: Lower Background Levels, Higher Peaks

According to official data, Almaty’s air became “cleaner” in 2025. However, a more detailed analysis reveals a more complex picture.

Comparison of the SI, MFI, and API in Almaty, 2021–2025

Source: “Environmental Bulletin on the State of the Environment in Almaty City, Almaty Region, and Zhetysu Region, 2025.”

Kazhydromet uses the Air Pollution Index (API) to assess overall background pollution. Additional indicators are used to provide a more complete picture:

• SI (Standard Index) reflects the most severe pollution episode and indicates how many times the maximum concentration exceeded the established limit.
• MFI (Maximum Frequency Index) indicates how frequently pollutant concentrations exceeded permissible levels. The report uses the MFI calculated for a single monitoring station.

According to the agency’s annual report, average background pollution in Almaty decreased by 31% in 2025 compared with 2024 (API: 5.8 → 4.1), formally placing the city in the low-pollution category under the API scale.

At the same time, the report records substantial decreases across several indicators. The number of registered exceedances of maximum one-time permissible concentrations fell by 99% for ozone (from 8,036 to 77 episodes), by 79.6% for carbon monoxide (CO; from 4,406 to 898), by 34% for PM2.5 (from 3,236 to 2,131), by 29% for PM10 (from 1,091 to 772), and by 25% for sulfur dioxide (from 2,271 to 1,704).

For reference:

API (Air Pollution Index) is an indicator reflecting average annual pollution levels across a group of priority pollutants.

SI (Standard Index) represents the most extreme single pollutant concentration recorded during the year.

MFI (Maximum Frequency Index) indicates how frequently pollutant concentrations exceed sanitary limits at an individual monitoring station.

At first glance, these figures suggest an overall improvement in air quality. However, the composition of pollution has changed considerably. While ozone was among the leading pollutants in terms of exceedances in 2024 (8,036 cases), its contribution had almost disappeared by 2025 (77 cases).

At the same time, nitrogen compounds became more prominent: nitrogen dioxide (NO₂) exceedances increased from 11,024 to 12,309 cases (+11.6%), while nitric oxide (NO) exceedances rose from 1,830 to 2,749 (+50.2%). However, comparing these figures with the raw data reveals certain discrepancies.

Comparison of the Number of Maximum One-Time Permissible Concentration Exceedances for Major Pollutants, According to Kazhydromet (2024–2025)

Such substantial changes may reflect not only actual changes in air quality but also differences in measurement procedures, including the number of measurements taken and the methods used to process the data.

Of these, 12 are automated stations that conduct continuous measurements every 20 minutes. All stations undergo annual verification and are officially certified as measuring instruments. Another four monitoring posts use manual sampling, with measurements taken three times a day.

This analysis considers only data from automated stations because they provide continuous time series that make it possible to assess air pollution trends throughout the year.

To assess how comprehensively automated monitoring stations measure concentrations of different pollutants and how the data is distributed geographically across the city, a map of automated air quality monitoring station coverage was developed. Each point represents an individual monitoring station and is labeled with its identifier, allowing its geographical location to be compared with the statistical indicators presented in the table.

The summary table to the left of the map contains key statistical characteristics of pollutant concentrations measured at the stations in 2025. For each pollutant (PM2.5, PM10, SO₂, CO, NO₂, NO, and O₃), it presents:

• Annual average concentration;
• Data coverage (the percentage of time during which the station transmitted valid measurements);
• Minimum and maximum recorded values.

The colors used in the table indicate data completeness:

• Green: Consistent and sufficient coverage (≥80% of days with valid data);
• Yellow: Moderate data completeness (≥50%);
• Red: Low data completeness (<50%);
• Gray: No measurements or unavailable data for the pollutant in question.

The map therefore helps reveal the actual basis of the calculations: which monitoring locations contribute to the city’s “average air quality” and how representative the resulting values are.

Map of Kazhydromet’s Automated Air Quality Monitoring Station Coverage and Pollution Statistics

Link to the detailed interactive map

Important Considerations:

• Incomplete list of measured pollutants: Not all automated stations are equipped with analyzers for every pollutant. As a result, seasonal charts may contain no measurements for certain substances throughout the year. Importantly, even when the official report lists a wide range of monitored pollutants, the datasets actually provided for some stations contain no measurements for particular substances. For example, data for PM2.5 and PM10 are entirely missing at certain stations (Skat-2, Skat-5, and Station 27), as are SO₂ data at Skat-4 and O₃ data at Station 29, despite these pollutants being listed as monitored substances in the report. This means that citywide annual average pollutant concentrations are effectively calculated using a limited number of monitoring locations rather than the entire monitoring network.
• Differences in data completeness: Some stations provide almost continuous measurements (more than 98–100% of days with valid data), while others have substantial gaps. These may be associated with equipment maintenance, instrument calibration, interruptions in data transmission, or the operating characteristics of individual analyzers. Consequently, the contribution of individual stations to the final citywide indicators may vary depending on the pollutant.
• Zero values in annual data require cautious interpretation: In some cases, annual average concentrations at individual stations are reported as 0.00 µg/m³ (for example, Station 27 for NO₂ and NO), sometimes over extended periods. Such concentrations are virtually nonexistent in a major city because even very clean air contains background concentrations of aerosols and gases. These values generally indicate an absence of recorded measurements rather than an absence of pollution, potentially due to prolonged equipment downtime or data export procedures.
• Abnormal data discreteness and detection thresholds (the Skat-2 case): Analysis of the raw data reveals technical anomalies that cast doubt on whether some monitoring stations accurately reflect actual environmental conditions. A striking example is the Skat-2 station’s nitrogen dioxide (NO₂) measurements. The dataset shows unnatural discreteness: values “jump” between 0 and 100 µg/m³, while numerous readings exceed 200 µg/m³, where variability resumes. This distribution may indicate a software or hardware limitation (a lower detection threshold) that causes the instrument to ignore actual concentration fluctuations at lower levels.
• Geographical distortion of indicators (the Skat-2 case): One of the automated monitoring stations, Skat-2, is located outside Almaty’s administrative boundaries. Including its data in citywide calculations distorts the actual pollution indicators because the station measures suburban background conditions, which may differ from smog levels in urban residential areas. On the one hand, air pollution does not follow administrative boundaries, and pollutant transport between the city and its suburbs is a natural process. On the other hand, including a suburban station can affect the final citywide indicators, especially when pollution levels at that location differ severalfold from those within the city.

Why the City May Look Cleaner (or Dirtier) on Paper

To assess the consistency of the results, we compared indicators calculated from the raw data of 12 automated monitoring stations, covering only seven major pollutants (PM2.5, PM10, SO₂, CO, NO₂, NO, and O₃), with the aggregated data presented in Kazhydromet’s official annual report for 2025. This means that differences between the calculated and official values may result from differences in the stations included, the incorporation of manual monitoring posts into official statistics, data processing methods, and other factors.

Comparison of Kazhydromet’s Official Bulletin Data with Automated Monitoring Station Data

The comparison shows that the indicators calculated from the raw data of automated monitoring stations generally reproduce Kazhydromet’s official results. Minor discrepancies for individual pollutants are primarily associated with differences in the stations included and data aggregation methods. For most pollutants, the differences are minimal and fall within the statistical margin of error associated with aggregation.

The most notable discrepancies were recorded for:

PM10: The difference in the number of exceedances (+116 cases) may be associated with data aggregation methods and differences in the monitoring stations included.

NO, CO, and PM2.5: Some discrepancies were observed in the number of exceedances of the maximum one-time permissible concentration, although the annual average concentrations are almost identical.

Nitrogen dioxide (NO) requires particular attention, especially the Skat-2 station, which is located outside Almaty’s administrative boundaries.

The most significant difference concerns the number of exceedances. When Skat-2, which is located outside Almaty’s administrative boundaries, is included, the number of exceedances (12,118) is almost identical to the official figure (12,309). However, excluding this station reduces the number of exceedances to 4,817, which is 7,492 fewer than reported in the official statistics. This indicates that a single monitoring station makes a substantial contribution to the final citywide indicator.

Comparison of Kazhydromet’s Official Data with Automated Monitoring Station Data (Excluding Skat-2)

At the same time, the maximum one-time concentrations are almost identical, confirming that extreme values have been processed correctly.

Analysis of the raw data from individual monitoring stations reveals technical anomalies that cast doubt on whether the measurements accurately reflect actual environmental conditions. A striking example is the Skat-2 station’s nitrogen dioxide (NO₂) measurements. The dataset shows unnatural discreteness: values “jump” between 0 and 100 µg/m³, while numerous readings exceed 200 µg/m³, where variability resumes.

This distribution suggests a possible software or hardware limitation (a lower detection threshold) that causes the instrument to ignore actual concentration fluctuations at lower levels. The figure illustrates the NO₂ time series recorded by the Skat-2 station in 2025. According to Kazhydromet’s 2025 report, measurements from Skat-2 were responsible for the city’s high Maximum Frequency Index (MFI = 28%). Examining the visualization suggests that the elevated concentrations reported by Kazhydromet in 2025 may be associated less with actual environmental conditions than with the operating characteristics of the measuring instruments or the way their data is recorded.

Hourly average nitrogen dioxide measurements from Automated Monitoring Station No. 2 (Skat-2), Turksib District, Burunday Motor Depot, Aerodromnaya Street.

Link to the interactive chart

A similar situation can be observed at Automated Monitoring Station No. 5 for nitric oxide (NO). Kazhydromet classified air pollution levels in Almaty in 2025 as high based on the Standard Index (SI = 9.6), driven by a nitric oxide (NO) concentration of 1,960 µg/m³ recorded at Station No. 5 near Halyk Arena. The figure below shows the dynamics of nitric oxide concentrations recorded at Station No. 5 (Skat-5).

Hourly average nitric oxide measurements from Automated Monitoring Station No. 5 (Skat-5), Medeu District, Halyk Arena Ice Rink, Duman Microdistrict.

Link to the interactive chart

At Monitoring Station No. 5, nitric oxide (NO) concentrations remained relatively stable at around 7–9 µg/m³ for more than half of the year, until October 14. After that, a noticeable increase occurred, with concentrations stabilizing at approximately 22 µg/m³. Between December 16 and 24, sharp spikes of up to 2,963 µg/m³ were recorded, after which concentrations returned to approximately 22 µg/m³ and remained at that level until the end of the year. This pattern is inconsistent with nitrogen oxide concentration trends described in scientific literature and with data from the independent monitoring network for 2023–2024, as presented in the Almaty Air Initiative report. It may also be attributable to the operating characteristics of the measuring instruments or the way their data is recorded.

Such anomalies in the data suggest that the city’s official classification as having a “low level of pollution” may, in some cases, be based on incorrect operation of individual analyzers or errors in data recording. The analysis shows that the Skat-2 and Skat-5 cases are not isolated incidents. Examination of 2025 time series from several other monitoring stations, including those in other regions of Kazakhstan, confirms that critically high (or low) readings used to calculate a city’s final pollution indices often result not from natural or human-caused environmental factors, but from hardware or software issues.

Trends in PM2.5 Particulate Matter Concentrations

Despite the pronounced seasonal variation in PM2.5 concentrations, clearly visible in the chart of daily average values for 2025, data from individual monitoring stations show patterns that require further analysis.

Daily average PM2.5 measurements from Kazhydromet’s automated monitoring stations.

Link to the interactive chart

The chart of daily average concentrations shows a typical seasonal pattern of air pollution in Almaty:

• Winter months (January–February) are characterized by high concentrations associated with the heating season and unfavorable meteorological conditions;
• In spring and summer, pollution levels decrease significantly due to increased atmospheric circulation;
• In autumn and early winter, concentrations begin to rise again.

For example, Station 28 recorded extremely high daily average concentrations in January, exceeding 250 µg/m³. These values correspond to severe urban smog episodes also observed by sensors in independent monitoring networks.

Overall, this pattern is consistent with the well-established seasonal characteristics of air pollution in Almaty. However, a closer examination of the chart reveals that not all stations exhibit comparable variations in pollutant concentrations.

Against the pronounced seasonal fluctuations recorded by several stations, nearly horizontal lines stand out in some time series. This is particularly noticeable at Skat-3 and Skat-4. For a significant part of the year, readings at these stations remain close to zero or virtually unchanged, while neighboring stations record noticeable fluctuations in concentrations.

Such a pattern is unusual for an urban atmosphere. Under real-world conditions, even when air quality is relatively good, PM2.5 concentrations:

• respond to changes in weather conditions,
• fluctuate during periods of peak economic activity,
• increase during temperature inversions,
• exhibit pronounced seasonal variation.

Therefore, prolonged periods of nearly constant readings may indicate not exceptionally clean air, but rather the operating characteristics of the measuring instruments or the way their data is recorded.

This inconsistency is also clearly visible in the chart of annual average PM2.5 concentrations across Kazhydromet’s monitoring stations. The chart shows that concentrations vary by an order of magnitude between stations.

Average PM2.5 Concentrations Recorded by Kazhydromet’s Automated Monitoring Stations

Such a wide variation within a single urban monitoring network is excessive.

For comparison, according to the Almaty Air Initiative annual report, based on data from an independent sensor network, the annual average PM2.5 concentration in Almaty was 31.2 µg/m³ in 2025. This value is close to the readings recorded at Skat-6 and Station 28 and significantly higher than those reported by most other stations in the national air quality monitoring network.

Skat-3 (near Almaty Arena) and Skat-4 (near the Ili Highway) stand out in particular. Their annual average concentrations were below 1 µg/m³ (!!!).

Based on these readings alone, Almaty could be considered one of the cleanest cities in the world, as such concentrations are lower than those recorded in many remote natural areas.

In a major city where coal is used, traffic is heavy, and road dust is widespread, annual PM2.5 concentrations below 1 µg/m³ appear atypical compared with other available data and may require additional technical verification. For comparison, according to IQAir’s 2024–2025 reports, even the cleanest locations on the planet record concentrations above 1 µg/m³. For example, Mayagüez (Puerto Rico), an island city exposed to ocean winds, has an annual average concentration of 1.1 µg/m³.

Trends in Ground-Level Ozone (O₃) Concentrations

Trends in ground-level ozone (O₃) deserve particular attention. According to Kazhydromet’s official report, the number of ozone-related exceedances decreased by 99% in 2025 compared with 2024.

At first glance, this might suggest that the problem of photochemical air pollution has almost completely disappeared. However, an analysis of daily average ozone concentrations also reveals substantial differences between monitoring stations, not only in the concentrations recorded but also in their patterns over time.

Average Ozone Concentrations Recorded by Automated Monitoring Stations No. 30, No. 31, and AQM1*

Link to the interactive chart

Ozone is a secondary pollutant formed in the atmosphere through photochemical reactions between nitrogen oxides and volatile organic compounds under the influence of sunlight. As a result, ozone typically exhibits a relatively consistent seasonal pattern:

• Higher concentrations during summer;
• Lower concentrations during winter.

However, the chart of daily average concentrations shows that ozone levels behave quite differently across monitoring stations, with some stations exhibiting patterns that require cautious interpretation.

• Station No. 30 — Incomplete Time Series: Ozone data completeness at Station No. 30 is approximately 48%.

In practice, measurements are available only for the first half of the year, primarily until June. Moreover, a substantial proportion of the readings during this period are recorded as zero. The time series is almost entirely missing for the second half of the year. Consequently, the ozone data from this station cannot be used to assess the characteristic summer peaks that typically occur during the warmer months.

• Station AQM1 — Atypical Seasonal Pattern: This station exhibits an unusual seasonal pattern. The chart shows that ozone concentrations remain relatively low and stable throughout the first half of the year, followed by sharp, short-lived peaks reaching significantly higher levels during the second half.

This pattern differs from the typical seasonal behavior of ozone, where concentrations generally increase gradually as solar radiation and air temperatures rise. Several factors may explain this pattern, including the station’s location, the influence of local sources of ozone precursors, or technical characteristics of data recording.

• Station No. 31 — Abrupt Change in Concentrations During the Year: The chart shows that ozone readings at this station are largely absent or remain at zero during the first half of the year. However, beginning around the second half of the year, concentrations rise sharply and start reaching significantly higher levels.

Why This Matters for Citywide Averages

Citywide air quality indicators are typically calculated by averaging measurements across a network of monitoring stations.

If some stations report abnormally low or high concentrations, show prolonged periods of identical readings, or provide incomplete time series, this can significantly distort the final annual average pollution estimate.

In other words, averaging such data may produce a citywide indicator that is considerably lower than the actual air pollution levels recorded in the city’s most heavily polluted areas.

To understand how air pollution changes throughout the year, monthly average pollutant concentrations recorded by automated monitoring stations were analyzed.

Seasonal Trends in Monthly Average Pollutant Concentrations Recorded by Kazhydromet’s Automated Monitoring Stations (Almaty, 2025)

Link to the interactive chart

Seasonal trends for most pollutants follow the expected pattern: higher concentrations during colder months and lower concentrations during warmer periods. However, for certain pollutants and monitoring stations, apparent drops or abrupt changes may be associated not with actual changes in air quality, but with the way data is collected and recorded.

As a result, seasonal charts for individual pollutants reflect not only actual changes in concentrations but also the structure and completeness of the available measurements.

The monthly chart shows that, for several pollutants, the lines remain almost flat throughout the year.

Under real-world conditions in a major city with our climate, such patterns are extremely rare. Pollutant concentrations typically vary significantly from month to month due to:

• Temperature changes and the heating season;
• Wind patterns;
• Traffic intensity;
• Photochemical processes in the atmosphere.

Therefore, prolonged flat sections on the charts may indicate several possible factors:

• Extended periods of analyzer downtime;
• Repeated reporting of identical values when no new measurements are available;
• Data processing or export procedures;
• Technical malfunctions at individual stations or calibration-related issues.

These characteristics call for cautious interpretation of annual indicators, as even a small number of stations reporting incorrect or identical values can noticeably affect the final citywide average.

Annual Average Pollutant Concentrations by District

Annual average pollutant concentrations were calculated for each of the city’s administrative districts using data from 12 automated monitoring stations.

Annual Average Pollutant Concentrations by District in Almaty, Based on Data from Kazhydromet’s Automated Monitoring Stations (2025)

It is important to note that some districts of the city have no automated monitoring stations, with observations conducted only at manual monitoring posts, whose data were not included in this analysis. Therefore, the spatial analysis covers only districts for which automated monitoring station data are available.

The highest concentrations of carbon monoxide (CO) were recorded in Alatau District, at approximately 710 µg/m³, and Medeu District, at around 714 µg/m³. Concentrations in Bostandyk District and near the Burunday Motor Depot were approximately 537–526 µg/m³. The lowest CO concentrations were recorded in Turksib District, at around 181 µg/m³.

For nitrogen dioxide (NO₂), the highest concentrations were observed near the Burunday Motor Depot and in Zhetysu District, at approximately 92–94 µg/m³, while concentrations in Alatau and Medeu Districts were significantly lower, at around 17–22 µg/m³.

Nitric oxide (NO) also shows substantial variation. Its annual average concentration exceeds 120 µg/m³ near the Burunday Motor Depot and reaches approximately 42 µg/m³ in Zhetysu District, while in Bostandyk District it is close to 5 µg/m³.

For particulate matter, the differences are less pronounced but consistent. Annual average PM2.5 concentrations range from 1.4 µg/m³ in Alatau District to 17 µg/m³ in Turksib District and 32 µg/m³ in Zhetysu District. For PM10, the highest concentrations reach 39 µg/m³, while the lowest are approximately 1.5 µg/m³.

Sulfur dioxide (SO₂) concentrations in most districts range from 5 to 55 µg/m³, with elevated levels in Alatau and Zhetysu Districts.

Annual average ozone (O₃) concentrations range from 3 to 5 µg/m³, showing relatively moderate spatial variation.

A comparison of Kazhydromet’s PM2.5 data with the Almaty Air Initiative (AAI) 2025 report reveals substantial discrepancies:

While Kazhydromet reports a downward trend in average background pollution levels, AAI data indicate that pollution levels have remained high or even increased (by 30% compared with 2024). Possible reasons for these discrepancies include differences in the density and structure of the monitoring networks, the stations included in the calculations, the accuracy of data recording, data completeness, and aggregation methods, all of which can affect the final annual average values differently.

Both sources confirm that Turksib District is among the areas experiencing elevated pollution levels. However, their estimates differ: 17.0 µg/m³ according to Kazhydromet versus 46.8 µg/m³ according to AAI. The figure reported by Kazhydromet is comparable to pollution levels in several European capitals, whereas the AAI value reflects higher pollution levels recorded by a dense sensor network. The difference may be explained by the density and placement of monitoring stations: the AAI network includes more measurement locations and is not limited to sites with more favorable air circulation conditions.

The substantial discrepancy in Alatau District (1.4 versus 37.7 µg/m³) is likely associated with technical factors affecting government monitoring stations, such as periods of analyzer malfunction or calibration-related issues. A concentration of 1.4 µg/m³ appears exceptionally low for an urban district and requires additional verification.

For comparison, concentrations of around 1 µg/m³ are characteristic of remote, sparsely urbanized areas, such as oceanic islands. In Alatau District, where traffic, urban development, and solid-fuel heating are present, such a low concentration appears atypical. At the same time, the AAI figure of 37.7 µg/m³ is consistent with pollution levels observed in major cities with complex terrain and a significant share of households using solid-fuel heating.

These findings highlight the importance of combined monitoring approaches: government monitoring stations provide certified data, while independent monitoring networks offer denser coverage across the city.

Distribution of Pollutant Concentrations (Based on Daily Values)

Additionally, the distribution of daily average concentrations for each pollutant was analyzed using data from automated monitoring stations.

Distribution of Daily Average Pollutant Concentrations Based on Data from Kazhydromet’s Automated Monitoring Stations (Almaty, 2025)

Unlike average values, concentration distributions help explain how the final indicator is formed: whether it reflects a consistent background pollution level or rare but significant concentration peaks.

Key Observations:

• PM2.5 and PM10: The distributions are strongly right-skewed: most days are characterized by moderate concentrations, but there are periods with substantially elevated values. For example, during winter, some stations recorded daily PM2.5 concentrations exceeding 200–250 µg/m³, tens of times higher than World Health Organization guidelines. In other words, air quality may appear relatively “average” for much of the year, while several dozen days account for a substantial share of the city’s actual pollution burden.
• Nitrogen dioxide (NO₂): The distribution is wider, with a noticeable tail of high concentrations. This means that nitrogen dioxide exceedances do not occur only as isolated spikes but may persist for several consecutive days. Such distributions contribute to relatively high annual average concentrations in certain districts. The Skat-2 station, located near the Burunday Motor Depot, plays a particularly significant role in the city’s pollution statistics. Analysis of the raw data shows that a substantial proportion of recorded NO₂ exceedances comes from this monitoring location. This pattern requires cautious interpretation, as the significant contribution of a single station to citywide statistics may considerably influence the final air pollution assessment, particularly for nitrogen compounds.
• Nitric oxide (NO): Several stations exhibit substantial variability, ranging from numerous days with low concentrations to occasional pronounced peaks. This reflects the indicator’s sensitivity to local conditions and short-term emission sources. Particular attention should be paid to the time series from Skat-5, located near Halyk Arena. In 2025, this station recorded the maximum one-time nitric oxide concentration that determined the city’s Standard Index (SI = 9.6). However, the station’s readings show a distinctive pattern throughout the year: concentrations remain relatively stable for an extended period, followed by an abrupt change in levels and the appearance of pronounced peaks.
• Carbon monoxide (CO): The distributions show a shift toward elevated concentrations at certain stations, while readings on most days remain within a relatively stable range. This suggests a comparatively consistent background concentration with periodic increases.
• Sulfur dioxide (SO₂): In most cases, the distribution is concentrated within a relatively narrow range, without pronounced extreme spikes, resulting in a smoother annual pattern. At the same time, the time series shows that such peaks are episodic: after brief periods of elevated concentrations, readings return to low levels. At most other stations across the city, SO₂ concentrations remain significantly lower and also exhibit more stable patterns throughout the year.
• Ozone (O₃): The distributions are compact, without significant extreme values, suggesting a more uniform pattern of ozone formation throughout the year. However, closer examination of the raw data reveals substantial measurement gaps and inconsistencies in the time series at several stations. Some monitoring locations report prolonged periods of zero readings or incomplete time series, while others show abrupt changes in concentration levels during the year. As a result, these distributions may reflect the availability and completeness of monitoring data rather than actual changes in atmospheric ozone concentrations.

The analysis of concentration distributions shows that differences between districts and monitoring stations are determined not only by average pollution levels but also by how concentrations vary throughout the year.

For certain pollutants, particularly nitrogen oxides and fine particulate matter, annual indicators depend significantly on the number of days with elevated concentrations, whereas other pollutants exhibit a more uniform pattern.

Meteorological Conditions

Meteorological conditions significantly influence the formation and dispersion of air pollutants. Temperature patterns, precipitation, wind conditions, and prevailing atmospheric circulation determine how intensely pollutants accumulate and how air pollution develops in urban environments.

The following section presents a comparative analysis of meteorological conditions in Almaty in 2024 and 2025, based on Kazhydromet’s annual reports.

Precipitation

In 2025, precipitation was significantly lower than in 2024, negatively affecting the atmosphere’s natural cleansing processes:

• February was predominantly dry.
• In April 2025, precipitation was approximately half the normal amount (53.3 mm compared with the normal 112 mm), whereas April 2024 was close to the climatic norm (110 mm).
• During summer (June–August) 2025, precipitation was 2–20 times below normal (2.3 mm in July), whereas in July 2024, precipitation was almost twice the normal amount (82 mm compared with the normal 43 mm).
• Overall, 2025 was considerably drier than 2024 in certain months, particularly during spring and summer.

Air Temperature

In 2025, temperatures tended to be higher than in 2024, as confirmed by data for several key periods:

• April: Temperatures in 2025 were 3°C above normal, whereas in 2024 they remained within the climatic norm.
• Summer: July 2025 was classified as “very hot,” with temperatures exceeding the norm, while July 2024 was close to normal. In August 2025, temperatures were 1°C above normal. By comparison, the first half of August 2024 was hot, but the second half was cooler, with daytime temperatures dropping to +21°C.
• December: December 2025 was significantly warmer, with the monthly average temperature 3.6°C above normal. In contrast, December 2024 was 1°C below normal.

Atmospheric Pressure Patterns and Anticyclone Influence

A key distinguishing feature of 2025 was increased anticyclonic activity, which directly contributes to stagnant air conditions and smog accumulation.

• February 2025: Unlike the unstable weather conditions of February 2024, Almaty was frequently influenced by an anticyclone in 2025. This resulted in predominantly dry weather and a 20.4 mm reduction in precipitation (2.2 times less than in 2024).
• November 2025: Periods of anticyclonic conditions were recorded, accompanied by nighttime temperatures falling to −2°C to −7°C amid an overall precipitation deficit (26.4 mm less than in November 2024).
• Atmospheric stability: Weather conditions in June and July 2025 were characterized as “stable.” Under prevailing anticyclonic conditions, this limits the vertical mixing of air masses.

Wind Conditions

Wind activity, which helps ventilate the city, remained predominantly low in both years. However, several differences were observed:

• In May 2024, wind gusts reached 18 m/s, whereas in May 2025, the maximum wind speed did not exceed 9 m/s (a twofold decrease in intensity).
• In June and July 2025, maximum wind speeds were slightly higher (up to 16–17 m/s) than in 2024 (up to 6–15 m/s).
• During winter and autumn months (January, February, September, and October), light winds prevailed in both years, with speeds generally not exceeding 4–9 m/s. These conditions contribute to the accumulation of harmful pollutants in the near-surface atmospheric layer.

Summary

Compared with 2024, 2025 was characterized by:

• Higher temperatures, particularly during spring and summer and toward the end of the year.
• A significant precipitation deficit during most months (particularly in spring and midsummer, when precipitation was 2–35 times lower), reducing the removal of pollutants from the atmosphere through precipitation.
• Persistently low wind speeds, which, combined with the precipitation deficit in 2025, created less favorable conditions for pollutant dispersion than in 2024.

Main Sources of Air Pollution (Almaty and Almaty Region)

According to Kazhydromet’s annual report, the regional emissions situation remains challenging, while traffic-related pressure continues to increase.

According to the statistics presented in the report, the region had a total of 8,974 stationary emission sources in 2025, including:

• Organized emission sources — 5,581.
• Sources equipped with emission control systems — 1,078.

This means that approximately 88% of facilities are not equipped with emission control systems. However, the document does not provide an estimate of the total pollution generated by these facilities, making it difficult to assess their actual contribution to overall air pollution levels.

According to data from Almaty’s Green Economy Department cited in the report, the city had a total of 151,059 private houses in 2025, including:

• Houses using gas heating — 149,341.

The report also states that the gasification rate of residential houses is 98.8%. However, this calculation may not fully account for buildings in informal development areas, temporary structures, and private bathhouses, where coal or other fuels may be used for heating.

Furthermore, the analysis provides only limited consideration of the Almaty metropolitan area, from which air currents may transport emissions from suburban settlements into the city’s basin.

According to Almaty Police Department data cited by Kazhydromet, a total of 643,470 motor vehicles were registered, representing an increase of 12,745 vehicles compared with 2024. These included:

• Passenger cars — 578,022 (33,955 more than the previous year).
• Buses — 11,208 (862 more than in 2024).
• Trucks — 43,648 (2,746 more than in 2024).
• Special-purpose vehicles — 1,258 (89 more than in 2024).
• Motorcycles and other motorized two-wheelers — 9,334 (1,014 more than in 2024).

The report also notes an increase in the vehicle fleet compared with the previous year, amounting to 41,734 vehicles in 2025.

At the same time, the number of exceedances increased for nitric oxide (+50.2%) and nitrogen dioxide (+11.6%). The report does not provide an assessment of the contribution of each emission source to overall pollution levels.

Conclusions

An analysis of official air quality monitoring data and raw measurements shows that the air quality situation in Almaty in 2025 presents a mixed picture. Formally, several indicators suggest improvement: the average pollution level decreased according to official statistics, and the Air Pollution Index moved into the “low pollution” category.

However, a more detailed analysis of time series, concentration distributions, and the structure of the monitoring network reveals systemic problems in the national air quality monitoring network in Almaty.

First, average indicators may depend significantly on the structure of the monitoring network and data completeness. Several stations report prolonged periods of identical or zero readings, while data for certain pollutants are missing or available only as incomplete time series.

Under these conditions, a decline in citywide pollution indicators may reflect not only improvements in air quality but also the characteristics of the measurement process. In other words, the city may appear to be “getting cleaner” in official statistics faster than it actually is.

Second, the composition of air pollution has changed. Despite declines in several indicators, the number of nitrogen dioxide exceedances remains very high, pointing to the growing influence of transport.

At the same time, a substantial proportion of these exceedances comes from a single monitoring station (Skat-2) located outside the city’s administrative boundaries. This raises further questions about how the final citywide pollution assessment is calculated.

Third, meteorological conditions in 2025 were themselves unfavorable for pollutant dispersion. Warmer weather, insufficient precipitation, frequent anticyclones, and weak winds created conditions in which pollutants remained in the near-surface atmospheric layer for longer periods.

Another important aspect, often overlooked in public discussions, must also be considered. Decisions to classify pollution levels as high and, in particular, to declare periods of adverse meteorological conditions (AMC), are based not only on regulatory standards but also on actual monitoring data.

If the monitoring system has data gaps, is sensitive to technical limitations, or fails to capture certain concentration fluctuations, some pollution episodes may simply be absent from the official picture.

As a result, the city may appear to have “better” air quality on dashboards than what people actually experience at street level.

This does not necessarily indicate a problem with the regulatory standards themselves. Rather, it highlights that the accuracy of air quality assessments directly depends on the quality and completeness of measurements.

The main conclusion is straightforward: average indicators may create an impression of improvement, but they do not always reflect the actual environmental burden on the city.

To assess air quality objectively, it is important to consider not only the final figures but also how they are produced: which monitoring stations are included in the calculations, how accurate and complete the data are, and which pollution episodes fall outside the averages.

In this context, improving the monitoring system itself — increasing data completeness, ensuring the comparability of measurements, and making methodologies more transparent — is just as important as discussing regulatory standards.

Ultimately, this determines how accurately the city can “see” the air its residents breathe.