Trends in Atmospheric Emissions in Central Asian Countries Since 1990 in the Context of Regional Development

Abstract

In Central Asian countries (CACs) atmosphere pollution is increasing due to population growth, economic growth, agricultural development, energy consumption and climate change. The countries of the region developed climate change adaptation strategies—Nationally Determined Contributions (NDCs) under the UN Framework Convention on Climate Change (UNFCCC). At the same time, regional integration, which should be a necessary condition for achieving the Sustainable Development Goals (SDGs) in the solving of general environmental problems, is not involved. This article shows the importance of a comprehensive analysis of greenhouse gas (GHG) and non-greenhouse emissions into the atmosphere for the entire Central Asian region as a single ecosystem. The energy intensity of national economies structure was chosen as the main factor determining the level of pollution. The analysis shows that over the past 30 years, the main part of the commodity exports (73.6–81.4%) of Kazakhstan and Turkmenistan has been fossil natural resources. There is a strong economic dependence on coal and other types of fuel, which leads to atmospheric emissions. The analysis shows that limited financial resources, lack of effective systemic monitoring and control of air quality that meets modern international requirements and standards, leads to absence of tangible changes in practice yet. Over 30 years in CACs, the share of CO₂ emissions associated with fuel combustion has not decreased and amounts to 78%. The key mechanisms for reducing atmospheric emissions are significantly increase investments in the transformation of the economies in the context of regional development, interstate cooperation, the introduction of environmental norms, standards harmonized with world ones, green technologies based on alternative energy, sustainable transport and logistics infrastructure.

1. Introduction

The most important aspect of environmental balance and sustainable development of any region in the world is air quality. A significant deterioration in the quality of air breathed by more than 50% of the world’s population has been observed in the last decade [1]. The global economic losses due to premature mortality caused by fine particulate matter (PM) and ground-level ozone air pollution are estimated by the Organization for Economic Co-operation and Development (OECD) to exceed USD 1.7 trillion per year. It is equivalent to about 3.5% of global gross domestic product (GDP) [2].

Central Asia, which includes five states—Kazakhstan, Kyrgyzstan, Uzbekistan, Tajikistan and Turkmenistan, is certainly not a single economic space. Each of the CACs have its own development paradigm, its own national currency, its own vision of entering the global economy, its own model of building an economy, and a combination of market and state. These states participate in international and regional trade, cooperation and interaction to different levels. The multi-vector nature of economic policy, the diversity of economic models and social reforms, and competition for foreign investment determine the region’s indicators heterogeneity [3]. At the same time, these states have much in common. They are geographical location, previously established economic and transport links, the availability of common resources, cultural proximity, as well as a set of unresolved transboundary environmental problems. In recent years, interstate relations within the region have also undergone dramatic changes at all levels, and mutually beneficial cooperation is gradually developing [4].

The benefits of globalization for the region are linked to the international division of labor in the context of the region’s special geographical position—between Europe and China. The position of CACs at the transport corridors crossroads creates opportunities for the transit potential development both in terms of goods transportation and for the involvement of the regional countries in cross-border production chains. The importance of Asian states in the foreign trade is increasing [5]. The turnover of foreign trade in goods of CACs with other countries amounted to USD 165.5 billion in 2021 and has grown six-fold since 2000. During the same period, the volume of trade between countries within the Central Asia region exceeded 11 billion dollars in 2024. Important factors include the strengthening of China’s position and the CACs involvement in the “One Belt, One Road” initiative implementation. UN experts suggest that the development of cooperative ties in Central Asia will allow the regional GDP to double within 10 years, ensuring the efficient use of the regions’ resource potential and their sustainable development [6]. Thus, strengthening multifaceted cooperation has made it possible to increase the combined GDP of CACs from USD 297 billion in 2018 to USD 450 billion in 2023 (https://kisi.kz/ru/ekonomicheskij-i-kommunikachionnyj-potenchial-chentralnoj-azii/, accessed on 12 October 2024). The region has stepped up its integration into the global economy. It maintains trade and economic relations with almost all countries in the world. The EU, Russia and China account for about 60% of the volume of CACs foreign trade (https://capsunlock.org/?page_id=41, accessed on 18 August 2025). Moreover, foreign trade with China exceeds both the EU (European Union) and Russia. Thus, in 1992, the volume of trade between China and Central Asia was only USD 527 million, and in 2024—about USD 89 billion.

The raw materials-exporting economic model that has developed in the countries of Central Asia leads to large socio-ecological-economic problems [7]. This, in turn, maintains a high dependence on coal and other fossil fuels in the economies of CACs. The use of extensive technologies and energy sources in the industrial and domestic sectors causes atmospheric emissions. Thus, the fossil fuel-based industry and energy sector in these countries accounts for the bulk of GHG emissions [8]. Research and knowledge on air quality changes in this region are very limited. It is mostly focused on individual countries, cities or economic sectors. The data that allows us to study and assess risks is not systematic and not homogeneous. This significantly complicates the analysis and adoption of management decisions [7,8,9,10,11]. It is necessary to jointly solve the air pollution problem. The need for an integrated approach to solving this issue lies in the fact that the deterioration of CACs air quality is multifactorial.

In the world energy security system, Central Asia is of great importance because it has huge reserves of oil, gas, coal, hydro resources, etc. The region has large deposits of fuel resources, such as Kashagan on the Caspian shelf of Kazakhstan—the world’s largest coastal field, the Karachaganak oil and gas condensate field in the West Kazakhstan region; South Yoloten in Turkmenistan, one of the three largest gas fields on the planet. The complexity of risk management in the region is determined by three common, interrelated risk factors: global warming and environmental destruction; linked and complicated economies, communities and population changes [12]. During the last half-century, the regional climate became noticeably warmer. The general properties of the climate of the CACs are expressed in a combination of moderate and subtropical elements, sharp continentality, aridity and climatic risks. CACs face serious challenges related to climate risks. The data from the Climate Risk Profile for Central Asia [13] and the results of the “Analysis of Climate Change in Central Asia” indicate that the environment in Central Asia has undergone transformations due to global warming [14,15].

Global warming accelerates natural resources depletion, as production and consumption under these conditions, against the backdrop of continuing atmospheric pollution, require more energy and materials. The areas of energy, water resources, and food supply are particularly sensitive to climate challenges. This hinders economic growth, deteriorates the environmental conditions, adversely affects human health and exacerbates regional disparities.

Climate changes in Central Asia are caused by the following natural and anthropogenic regional factors: drying up of the Aral Sea, increased wind erosion of the dried-up seabed surface, glaciers thawing, snow cover decrease and desertification [16]. Rising extreme temperatures and extreme weather events present significant risks not only to human health, but also to natural ecosystems, agricultural systems, water management, energy resources and infrastructure.

One of the most significant global warming consequences in Central Asia is the thawing of glaciers and the resulting creation of glacial lakes. The rate of decrease in glacier volume in Central Asia is 0.2–1% per year. Overall, during the past 50–60 years, due to changing climate conditions, the surface area of glaciers in Central Asia, which are the main source of freshwater, has decreased by 30% [17]. Drought has always been a typical phenomenon for Central Asia, but recently drought cycles have become more frequent. As a result, the Central Asia region is classified as a water stressed region [18]. The observed decrease in river flow (the estimated decrease in the Syr Darya and Amu Darya basins water volumes by 2050 is 10–15%) will have a negative impact in the near future [19]. According to international experts, climate changes will affect all CACs and will lead to variations in precipitation levels and temperature background of the entire region [20].

Against this background, climate change is becoming an increasingly acute problem for agricultural growth in CA. The agro-ecological zones boundaries are predicted to shift, and the region as a whole will become hotter and less suitable for traditional farming. According to the international scientific journal Nature, the region consumes 2.5 cubic meters of water per dollar of GDP, which places the CACs among the world leaders in the inefficient water resources usage [21].

The presence of manufacturing plants in the region, including metallurgical, oil and gas, coal mining, mineral extraction, as well as transport vehicles, especially the multiple increase in passenger cars and trucks, has led to a rise in atmospheric pollutants, especially in urbanized areas. Unlike EU countries and China, where significant investments are made in environmental technologies and monitoring programs, in Central Asia insufficient attention is paid to systematic monitoring of air quality. Limited financial resources and lack of effective regulation prevent the implementation of modern air purification technologies. However, the problem is not limited to just a technological solution to the issue of cleaning the air from industrial pollutants. The use of pesticides, fertilizers and irrigation methods in agriculture in the arid climate of the region require more and more energy resources, which also affects the environmental situation, in particular the growth of atmospheric emissions. The growing internal migration of rural populations to cities in search of work has many consequences. Among other things, it leads to a disproportionate increase in industrial activity, housing and infrastructure construction, automobile traffic, increased energy consumption, accumulation of household waste and other negative factors. In the context of increasing urbanization, the lack of modern public transport and the old vehicle fleet also contribute to the deterioration of air quality [5].

To keep temperatures below 1.5 °C, a central objective of global initiatives aimed at addressing climate change is to decrease emissions of CO₂ and other GHGs by 45% by the year 2030, relative to levels recorded in 2010 [22,23]. In this context, leading countries, with a 25% share of global GDP, have established legal obligations to achieve net-zero emissions targets under the Paris Agreement (https://www.un.org/ru/climatechange/paris-agreement, accessed 12 October 2024). All CACs have signed the Paris Agreement. Moreover, within the framework of its implementation, Kazakhstan has undertaken obligations to achieve a reduction in GHG emissions of 15% by 2030 compared to the 1990 baseline level with the possibility of increasing this target to 25% under specific conditions [24]. The target indicators for 2050 for Central Asian countries are currently under development. In particular, Kazakhstan has presented its Voluntary National Review 2025, outlining its target orientation up to 2030.

Currently, Kazakhstan, Turkmenistan and Uzbekistan have defined national targets as part of their green economy strategies, while Kyrgyzstan and Tajikistan have chosen to concentrate on programs for climate change mitigation and adaptation, likely because of the significant contribution of hydropower in the country’s energy balance. CACs are gradually including in their national programs and policies the development of renewable energy sources (RES), stimulation of sustainable agricultural development and implementation of a circular approach to the use of natural resources [25]. The World Bank projects that global growth will increase to an average of 2.7% in 2025–2026 (https://www.imf.org/en/Publications/WEO/Issues/2024/10/22/world-economic-outlook-october-2024, accessed 2 November 2024). Based on the World Economic Outlook report from the International Monetary Fund (IMF), among the CACs, Uzbekistan has the highest projected real GDP growth for 2026 (5.5%), followed by Kazakhstan (4.4%), Kyrgyzstan, Tajikistan (4% each) and Turkmenistan (3.7%) (https://www.imf.org/en/Publications/WEO/Issues/2024/10/22/world-economic-outlook-october-2024, accessed 2 November 2024).

All of the above allowed us to formulate the research problem to identify the key factors that determine the level of air pollution in CACs, analyze their relationship with the economic structure of the region and develop recommendations for reducing emissions in the context of regional development. The problem of assessing air quality is becoming especially relevant in the region, where various sources of pollution cause significant damage to the environment and have a negative impact on public health [26]. The relevance and validity of this approach is explained not only by the climatic and geographical similarity of the ecosystem in question, but also, first of all, by the commonality of unresolved transboundary environmental problems and, to a significant extent, by the limited knowledge of the state of air quality in the region.

This research addresses the scientific gaps concerning the air pollution in CACs. Previously, studies of air pollution in Central Asia focused on individual aspects. Global emission reports (Emissions Database for Global Atmospheric Research (EDGAR), World Bank, International Energy Agency (IEA)) analyze the dynamics of CO₂, but without detailed consideration of economic and regional factors. Work on air quality (UNECE (United Nations Economic Commission for Europe), United Nations Environment Programme (UNEP), KazHydromet) considers the level of PM_2.5 and SO₂, but does not link them to industry and the energy intensity of the economy. It is especially important to study the impact of economic structure on atmospheric emissions in Central Asia as a single ecosystem. Energy sector research (IEA, Enerdata) focuses on the structure of energy consumption but does not assess its impact on emissions in Central Asia. NDCs, Paris Agreement of Central Asian countries declare intentions to reduce emissions, but do not contain an objective analysis of their implementation.

The novelty of our study lies in several key aspects. For the first time, a unified comparative analysis of the dynamics of CO₂, PM_2.5 and SO₂ emissions in Central Asian countries for the period 1990–2024 has been conducted, which makes it possible to identify long-term pollution trends. Unlike previous studies, the study examines the impact of the energy intensity of the economy and the carbon intensity of GDP on pollution levels, which allows us to establish a direct relationship between economic structure and emissions. Another important element is to take into account the role of economic structure in shaping emissions: previous studies have not detailed the dependence of emissions on dominant economic sectors such as oil and gas, coal and metallurgy. Thus, the novelty of the research idea is based on the fact that the impact of climate change and environmental degradation in the region must be considered in relation to the region’s economy.

In this context, the uniqueness of this study lies in the following aspects. First, the energy intensity of national economies was selected as the primary determinant of pollution levels. Second, unlike many existing studies that focus solely on GHG or non-GHG pollutants, this research examines the effects of both types of emissions within the same regional framework. Third, the hypothesis will be tested using a combination of analytical and descriptive methods, providing a comprehensive review-based approach to the investigation.

The idea underlying the study and the problems raised allowed us to formulate the hypothesis of this study. H1: The energy-intensive structure of the Central Asian economy, focused on the extraction and export of hydrocarbons, is the main factor behind the high level of atmospheric emissions. The transition to a low-carbon economy is possible with a comprehensive approach that includes: the development of renewable energy sources (RES), improved energy efficiency, tighter environmental standards and deeper regional cooperation. According to this hypothesis, as a null hypothesis H0, the study assumes that the energy-intensive structure of the national economies of Central Asia does not in any way affect the maintenance of an elevated level and the growth of atmospheric emissions. The high level of emissions is due to other factors such as climate conditions, population growth and increasing energy consumption.

Goal of this research is to comprehensively analyze the dynamics of atmospheric emissions since 1990, identify key factors in their formation and assess possible mechanisms for their reduction.in the context of dependence on the energy-intensive structure of the national economy in CACs. The study performs the following:

• Compares CO₂, PM_2.5 and SO₂ emissions in CACs;
• Assesses the impact of economic structure on pollution levels;
• Examines current strategies for reducing emissions and provides recommendations for adapting them to regional conditions.

The results obtained will make it possible to formulate scientifically based recommendations for the development of regional environmental strategies and measures to decarbonize the Central Asian economy.

2. Materials and Methods

2.1. Research Methodology

This study is interdisciplinary in nature and combines cartographic analysis, econometric modeling, comparative analysis of economic and environmental indicators, and a systems analysis of policies to reduce emissions in Central Asia. The choice of methodology is determined by several factors:

• The heterogeneity of data across countries in the region, which requires a comprehensive approach to their analysis; the dynamic nature of emissions, which must be assessed over time (1990–2024);
• The relationship between economic structure and pollution levels, which requires the use of econometric indicators;
• International commitments of CACs to reduce emissions, which makes a comparative analysis of strategies and policies necessary.

The rationale for the choice of methodology lies in the combined use of quantitative and geospatial methods, which allows us to identify patterns and determine the dynamics of emissions at the regional level. The use of statistical and econometric indicators ensures the objectivity of the analysis and reveals long-term trends. A comparative analysis of strategies makes it possible to assess the effectiveness of existing measures and propose sound recommendations. The main limitations of the study are related to incomplete data for some countries, such as Turkmenistan, which requires the use of alternative sources, including modeling and indirect estimates. Differences in emission accounting methods between countries may cause discrepancies in the data, which is also taken into account in the interpretation of the results.

The methodology for executing this research is structured around addressing the following tasks, for which the stages are outlined:

• Scientometric analysis of publications. It will be the study of publications devoted to the problem of air pollution.
• Identification of the Central Asia economy sectors features. It will identify the answer to the question: What economic sectors and factors have the most significant impact on air quality? To assess the relationship between the economic structure of countries and the level of pollution, methods of descriptive statistics, correlation analysis and regression modeling are used. Then it is necessary to estimate problems and challenges. To evaluate the energy efficiency of the CACs economies, indicators such as the following:
  • Energy Intensity of CACs for the period 1990–2022;
  • Specific CO₂ Emissions per capita;
  • Specific CO₂ Emissions per GDP;
  • GDP per unit of energy consumption, quoted in dollars per kilogram of oil equivalent were used.
• Assessment of the ecological-climatic changes impact on air quality of the region. To study the pollution dynamics, a statistical analysis of time series (1990–2024) of CO₂, PM_2.5 and SO₂ emissions is carried out. To achieve this goal, indicators, such as the following, were used:
  • The volume of GHGs in CACs for the period 1990–2021;
  • Average CO₂ emission (carbon factor) during 1990–2023;
  • Dynamics of CO₂ emissions by burning fossil fuels of CACs for the period 1990–2022;
  • Dynamics of PM_2.5, SO₂ in CACs for the period 1990–2024.
• Econometric modeling in the R program was used to identify relationships between CO₂ emissions from fuel combustion and the factors that determine it, taking into account the specifics of country development:
  • Dependent and independent variables were identified for all five countries;
  • A correlation analysis of the dependence of the indicators was carried out, a thermal matrix was constructed;
  • A fixed effects (FE) model and a random effects (RE) model were constructed;
  • For comparison of fixed effects and random effects models, We used both approaches to estimate the fixed effects model—the classical Wallace–Hussain method, as well as the more robust Amemiya method, recommended for analyzing small panel samples (n = 5 countries);
  • A forecast of indicators based on the constructed model for all five countries to identify the achievability of the climate parameters declared by the countries by 2030 was built.
• Strategies in Central Asia for implementing measures to reduce air pollution. Development of Recommendations for improvement. To provide a comparative analysis of emission reduction strategies, the environmental strategies of the CACs are examined in comparison with successful cases from other regions, including EU policies (ETS, European Green Deal), the Chinese air pollution strategy and the US experience in emission regulation.

2.2. Data Sources

The study is based on three groups of data. The first group is environmental indicators, including emissions of CO₂, PM_2.5 and SO₂. The data were obtained from EDGAR, World Bank Open Data, UNEP, national statistical services of Kazakhstan, Uzbekistan, Kyrgyzstan, Tajikistan and Turkmenistan and national reports of CACs under the UNFCCC.

The second group is economic and energy indicators extracted from TheGlobalEconomy.com (generally based on World Bank Open Data), World Energy & Climate Statistics (Enerdata), IEA.

The third group includes regional and international strategies to reduce emissions, such as The Paris Agreement and NDCs of CACs, the EU Green Deal policies, the Chinese Blue Sky Defense program, as well as the UN Sustainable Development Programs and UNECE recommendations.

• Scientometric analysis of publications from Scopus and WebScience databases was carried out using VoSViewer: by keywords, by country, by author [27].
• Identification of the Central Asia economy sectors features was carried out using environmental indicators, economic and energy indicators that make it possible to understand the relationship between economic activity and the level of atmospheric pollution [17,23,28,29,30,31,32,33,34]. Data regarding energy, fuel consumption and air emissions are gathered and systematized by national statistical offices, energy departments and environmental protection agencies. The work used information from the National Statistics Bureau of the CACs; however, it should be noted that they are systematized mainly in Kazakhstan and Kyrgyzstan. Thus, only in Kazakhstan and Kyrgyzstan are the National Ambient Air Quality Standards (NAAQS) enshrined in laws available for review [29,35]. The following are used as key indicators:

Energy intensity, koe/USD15p—is the ratio of the energy consumed by the system to the results of its operation. Energy intensity is expressed in tons of conventional fuel (tons of oil equivalent) per GDP of the country, (USD1000 in 2015 prices at purchasing power parity), data are presented in https://yearbook.enerdata.net/ (accessed on 10 October 2024).

Specific CO₂ emissions per capita are calculated as the total amount of CO₂ emissions divided by the population. This indicator allows us to assess how efficiently a country uses its resources and what its environmental impacts are per capita. All of the data presented on the website are taken from https://www.worldbank.org/ (accessed 12 October 2024).

Specific CO₂ emissions per unit of GDP are calculated as the total amount of CO₂ emissions divided by the gross domestic product expressed in constant prices. This indicator provides an understanding of how much CO₂ is emitted per unit of economic activity, which is critical for analyzing the sustainability and environmental performance of countries. This allows us to estimate the efficiency of resource use in the economy and the level of pollution per unit of economic activity, data are taken from https://data.worldbank.org/indicator/EN.GHG.CO2.WA.MT.CE.AR5 (accessed 12 October 2024).

GDP per unit of energy consumption, expressed in dollars per kilogram of oil equivalent, is an important indicator reflecting the energy of country’s economy. This indicator reflects how much GDP is generated for each unit of energy consumed in oil equivalent, data is presented at https://data.worldbank.org/assets/images/placeholder.png (accessed 12 October 2024).

• Assessment of the ecological-climatic changes impact on air quality of the region: It was used the National Communications and the Biennial Reports of the CACs to the UNFCCC [36,37,38,39,40]. The use of GHG inventory data in the analysis of atmospheric emissions in Central Asia is justified, because they are based on international standardized methodologies Intergovernmental Panel on Climate Change (IPCC) Guidelines 2006, mainly Tier 1, (Tier 2 Method: for waste, agriculture sectors), which allows for the comparison of analogous data, the assessment of emissions and their sources [41]. CH₄ emissions mainly occur in agricultural production were obtained from GHG inventory data, supplemented by data from the Bureau of National Statistics and TheGlobalEconomy.com [28,35,42]. The lack of data on other pollutants from industry, transport, and energy that are not related to GHGs, such as PM_2.5, was filled in from relevant sources [29,30,31,32,33,34].

In this regard, analytical reports of the EDB, 2022; UNECE, 2019, data from the World Bank, World Energy Climate Statistics-Yearbook, 2024; The U.S. Energy Information Administration and GHG emissions were used, which include environmental and economic indicators that make it possible to understand the relationship between economic activity and the level of atmospheric pollution [17,23,28,29,30,31,32,33,34].

The average CO₂ emission factor, also known as the carbon factor, is determined by calculating the ratio of CO₂ emissions to primary energy consumption. It shows the amount of CO₂ emitted per unit of energy, such as per kilowatt hour (kWh) or per ton of fuel burned. This indicator is vital for evaluating the effects of various energy sources on climate change. The data is presented at https://yearbook.enerdata.net/ (accessed on 10 October 2024).

• The ratio of specific CO₂ emissions from fuel combustion to GDP is expressed in kilograms of CO₂ per unit of GDP and serves as a significant measure for assessing the carbon intensity of an economy. The data are presented at UNECE, Data Portal https://w3.unece.org/SDG/ru/Indicator?id=28 (accessed 10 October 2024). [38].

Geospatial analysis of emissions is performed using mapping modeling in ArcGIS 10.8 based on EDGAR data and Sentinel-5P satellite measurements. The analysis produced maps of the geographic distribution of CO₂, PM_2.5 and SO₂ in 2020, heat maps of emissions changes from 1990 to 2024, and a mapping analysis of urban areas with the highest emissions. ArcMap 10.8 software was used to create maps of CO₂ emissions and average monthly concentrations of PM_2.5 air pollution in CACs in 2020. Emissions data were taken from public sources such as global environmental databases (e.g., EDGAR, Global Carbon Project) and national environmental monitoring agencies. ArcMap 10.8 introduced basemaps that visualize data using color gradations, from light for low outliers to dark for high outliers. Data of PM_2.5 concentrations in individual cities, taken from [9] are presented as colored circles of different sizes and colors symbolizing a certain level of air pollution. Data on PM_2.5 concentrations collected from ground stations or satellite sources [9]. These data were grouped into ranges (27, 74, 99, 110 and 112 μg/m³) to reflect the degree of pollution in each city represented.

• Strategies in Central Asia for implementing measures to reduce air pollution. Development of Recommendations for improvement.

The following were used: the results of the work performed; analytical reports of Eurasian Development Bank (EDB), 2022; the UNECE, 2019; OECD, 2019; WMO, 2024, UNEP, 2021; data from the World Bank; TheGlobalEconomy.com; and the U.S. Energy Information Administration and national reports of the CACs to the UNFCCC [17,20,23,29,31,33,34,36,37,38,39,40].

3. Results

3.1. Scientometric Analysis of Publications

Our study was conducted using scientometric analysis of data from the Scopus and Web&Science databases by country (Figure 1), keywords and authors for the period from 1990 to 2025. An analysis of the problem knowledge showed that in this period 51,267 articles were published, which in one way or another revealed the issues of the atmospheric emissions impact and environmental pollution. In research on atmospheric emissions, an increase in publications has been observed since 2016, when publication activity increased several times [43]. In our opinion, this is linked to the adoption of the Sustainable Development Goals (SDGs) by the UN in 2015, as a significant portion of the SDGs are either directly or indirectly related to addressing air pollution. As part of this study, a scientometric analysis was guided using some keywords: “climate change in Central Asia”, “air pollutants reduction”, “emissions into the atmosphere”, “reduction in air quality”, “atmospheric emissions in Central Asia”, “decarbonization of the economy”. It is worth noting that the majority of the research, specifically 51%, was conducted by scientists from China and the United States. The next group includes scientists from the UK, Germany, Italy and India—just over 22%. Moreover, according to the heat map in Appendix A (https://data.mendeley.com/preview/9v4h2cxw9b?a=e01eb9a7-5e93-4759-b716-cdd6283eb6d6, accessed on 10 April 2025), the majority of publications, more than 40% since 2018 are focused on the use of keywords such as: “air quality”, “particulate matters”, “emissions control”, “carbon dioxide”, and since 2020: “decarbonization”, “sustainability”, “energy policy”, “electric vehicles”. On the one hand, this reflects ongoing changes in global policies related to clean air, as well as the implementation of programs aimed at reducing atmospheric emissions and combating global warming. Conversely, it indicates a shift in research focus from pollution processes to their consequences and potential solutions.

The bibliometric analysis of author contributions (Appendix A, https://data.mendeley.com/preview/9v4h2cxw9b?a=e01eb9a7-5e93-4759-b716-cdd6283eb6d6, accessed on 10 April 2025) reveals that Chinese researchers are among the leading authors publishing articles on air pollution. Notable contributors include Zhang, Y.; Yang, W.; Wang, J.; Zhang, K.; Hao, Y.; Li, Q.; Jiang, J.H.; Wu, D.L., and others [15,16,44]. This trend underscores a significant increase in publication activity in recent years, focusing on air pollution and its potential adverse impacts on human health.

There is a noticeable geographical shift in publication activity, with fewer publications from Europe and the US, while contributions from Southeast Asia, especially China, are increasing. However, there is a lack of systematic air quality monitoring studies in the Central Asian region. The literature does not cover how economic structure affects emissions in this region. Existing studies are often fragmented and conducted within the framework of individual international grant projects, and air pollution data remain limited [44].

Despite these gaps, several high-impact articles address climate change and related greenhouse gas (GHG) emissions in the Central Asian region (CACs) [14,16,19,23,45]. Additionally, some studies focus on atmospheric pollution caused by non-greenhouse substances, particularly fine particulate matter (PM_2.5) [8,9,26,44]. There are no studies that combine the analysis of GHG and non-GHG emissions. This would allow for a more objective assessment of atmospheric pollution in the region.

The need for a comprehensive approach to addressing this issue is driven by the importance of studying the region as a single, interconnected ecosystem. This necessity arises from several key factors: the geographical proximity of the countries, the similarity of their climatic and environmental conditions, and the mutual influence of atmospheric circulation patterns. Consequently, there is a mutual dependence among regional emissions, as pollutants released in one area can significantly affect air quality across borders. Thus, the results of the scientometric analysis provide additional arguments supporting the choice of the Central Asian region as the focus of this study.

3.2. Identification of the Central Asia Economy Sectors Features

The CACs are characterized by a heterogeneous economic structure. Thus, the structure of Kazakhstan’s GDP (Figure 2) is typical of countries with an above-average income level: the economy has a significant share of services (55.8%) and industry (27%), while agriculture, with huge land reserves, accounts for only 5.3%.

In contrast to Kazakhstan, an analysis of the GDP structure of Kyrgyzstan, Tajikistan and Uzbekistan reveals that their economies are characteristic of lower-middle-income countries, where agriculture plays a more significant role (15.3–26.1%). In Turkmenistan, industry dominates (49.5%), primarily due to the country’s specialization in mineral extraction (Figure 2). By comparison, in global production, the services and trade sector hold a stable leading position, accounting for approximately 64%, while agriculture contributes only 4.27%. This disparity suggests that much of the value created by humanity is driven by the pursuit of marginal income rather than the generation of real added value. Such trends are reshaping societal awareness of the need for environmental protection.

3.2.1. A Comparative Assessment of Key Fuel and Energy Resources

In Central Asia, organic fuel occupies a dominant position in the production of key fuel and energy resources. The region’s primary exports include oil, gas, coal, metals, and other resources from the extractive sector [11,36]. Raw material exports remain a significant factor in the regional economy (Figure 3). A comparative analysis of the trends in income from natural resources as a percentage of GDP over a 30-year period (1990–2022) shows that the share of natural resource income in GDP for CACs varies from 1.91% in Tajikistan to 38.55% in Turkmenistan. In Kazakhstan, this figure ranged from 3.40% to 33.25% during the same period. Among the main export products, fuel and energy resources accounted for 67.7% in 2016, with ferrous metals at 5.4%, copper at 4.4%, and ore at 2.4%. Furthermore, between 2017 and 2022, Kazakhstan dominated total commodity exports in the Central Asian region, accounting for 72%. The shares of the remaining CACs were distributed as follows: Uzbekistan (15.6%), Turkmenistan (8.4%), Kyrgyzstan (2.3%), and Tajikistan (1.7%). This uneven distribution of export revenues is largely attributable to variations in the reserves and production levels of minerals in demand on international markets [46,47].

Between 2000 and 2007, Kazakhstan’s economy experienced an average annual growth rate of 10%, while between 2010 and 2014, it grew by nearly 6%. These high growth rates were primarily driven by the exploitation of hydrocarbon fields in Western Kazakhstan, coupled with the rise in global raw material prices. In terms of oil reserves, Kazakhstan ranks 12th globally and is among the top twenty countries with the largest natural gas reserves. This strong correlation between Kazakhstan’s economy and oil prices underscores its resource dependence.

The region also holds immense potential in the water and energy sectors. Not only are some of the world’s richest oil and gas fields located here, but there is also significant, yet underutilized, potential for RES. For instance, hydropower resources are abundant due to the region’s mountainous terrain, which covers more than 17% of the total area. Additionally, solar energy potential is substantial, as two major deserts—the Karakum and Kyzylkum—account for over 16% of the region’s total land area [5].

Turkmenistan, Uzbekistan and Kazakhstan possess substantial natural gas reserves, with Kazakhstan also holding significant coal deposits. In contrast, Tajikistan and Kyrgyzstan are rich in hydropower resources, controlling approximately 60% of the region’s water reserves, despite having minimal accessible fossil fuel reserves [17,47]. Over the period from 1990 to 2020, the share of RES in Tajikistan and Kyrgyzstan averaged 32.37% and 25.48%, respectively [28]. In other countries within the region, which have significant potential for expanding solar and wind energy, the share of these energy sources remains low. The share of RES (excluding hydropower) in the region’s energy balance remains low: 3% in Kazakhstan, 1.2% in Uzbekistan, and less than 0.5% in Turkmenistan [47]. This is due to a number of factors, including outdated infrastructure, lack of investment and the absence of a clear government energy policy.

The data presented in Figure 4 indicate that approximately three-quarters of electricity in Central Asia is generated by power plants utilizing coal, oil, and natural gas. The power generation profiles of each of CACs highlight the regional imbalance in available energy sources. Kazakhstan, Turkmenistan and Uzbekistan predominantly rely on fossil fuels for electricity generation (approximately 90%), whereas Kyrgyzstan and Tajikistan generate over 90% of their electricity from hydropower [17,47].

The increase in energy consumption in the region can be attributed to several factors:

-

Population growth (in 2023, the population growth rate in the region ranged from 1.3% to 2.1%, compared to the global average of 0.9%);

-

Rising urbanization (urbanization rates in the region ranged from 38% to 58% in 2023, compared to the global average of 57%);

-

An increase in housing construction (for example, between 1990 and 2019, Kyrgyzstan’s housing stock increased by 30 million m²);

-

Infrastructure development (in 2023, the share of services in Kazakhstan’s GDP reached 55.97%, aligning closely with the global average of 54.52%) [34].

Notably, Kazakhstan exhibits a high level of energy consumption, with per capita energy use reaching 5890 kWh in 2023, significantly higher than the global average of 3200 kWh per capita (Figure 4). Furthermore, Figure 4 reveals that between 2014 and 2023, the CACs experienced a decline of 8–27% in GDP per unit of energy consumed, indicating a decrease in energy efficiency. In contrast, during the same period, global GDP per unit of energy consumed increased by 4%. This decline in energy efficiency in Central Asia is primarily driven by rising energy consumption and the expansion of energy-intensive industries within the regional economies. It is worth noting that the growing construction sector currently accounts for more than a fifth of global emissions in energy demand and emissions. The building sector is seeing a resurgence in expansion, which has a tremendous direct and indirect effect on the environment taken as one of the most waste-producing and waste-producing sectors of the economy [48]. The global construction sector emitted a total of 5.7 billion tons in 2009, accounting for 23% of the total CO₂ emissions generated by economic activities worldwide [48,49]. The residential sector is the largest consumer of electricity after the mining and industrial sectors in CACs. About 75% of buildings in the region were built in the 1950s–1990s and do not meet modern energy efficiency standards. For example, energy consumption in the residential sector of Kazakhstan is about 270 kWh/m², which is 2.7 times higher than in Europe (https://research.nu.edu.kz/en/publications/household-energy-consumption-and-energy-poverty-in-kazakhstan, accessed on 18 August 2025).

The link between economic structure and emissions levels is confirmed by econometric indicators: the carbon intensity of GDP in countries dependent on hydrocarbon exports (Kazakhstan, Turkmenistan), is 0.5–0.6 kg CO₂/$1 GDP, while in Kyrgyzstan and Tajikistan this figure is lower (0.2–0.3 kg CO₂/$1 GDP). Evidence of excessive energy consumption in CACs, particularly in Uzbekistan and Kazakhstan, is demonstrated by the energy intensity indicator, measured as energy use per $1000 of GDP for the period 1990–2023 (Figure 5). This indicator is significantly higher than that of developed countries, such as Norway (2–4 times higher) and China (1.5–2 times higher), which were selected for comparison [32]. High energy intensity is often associated with the predominant use of fossil fuels for energy generation, leading to elevated GHG emissions. The data presented herein suggest that Kazakhstan and Uzbekistan exhibit the highest levels of energy and carbon intensity within the Europe and Central Asia region.

The high energy demand per unit of GDP in Kazakhstan can be attributed to several factors, including the following:

• The historically established energy-intensive structure of the industrial economy with low added value;
• The overall technological backwardness of production processes;
• The severe continental climate, characterized by prolonged and harsh winters;
• The vast territory and extensive transport infrastructure, including oil and gas pipelines, electricity transmission lines and water supply systems.

This disparity between the relatively low share of industrial output and the disproportionately high level of energy consumption is, in our view, a critical factor exacerbating air pollution not only in Kazakhstan but also in other countries within the region.

3.2.2. The Relationship Between the Economic Structure of Countries and the Level of Pollution

An integral indicator for assessing anthropogenic impacts within the context of a resource-based economy is the total emissions of pollutants per unit of GDP. For comparative analysis, the dynamics of changes in CO₂ emissions per unit of GDP (Figure 6a) were examined. This metric serves as a universal indicator for evaluating the environmental impact of energy production, industrial activities, agriculture and municipal services. It reflects energy intensity, the efficiency of manufacturing technologies, and the extent of fossil fuel consumption. The analysis revealed that the highest values of this indicator are observed in Kazakhstan, Turkmenistan, and Uzbekistan.

Additionally, to assess the contribution of each Central Asian country to global emissions, the dynamics of specific CO₂ emissions per capita were analyzed. In Kyrgyzstan (Figure 6b), specific emissions have declined since 1990, with a slight increase in recent years, reaching just over 2 tons per person. This trend is likely due to the country’s growing emphasis on agriculture. For comparison, CO₂ emissions per capita in Kazakhstan reached 16.7 tons per person in 2011 [36]. Kazakhstan and Turkmenistan, as significant global exporters of oil and gas, have increased their per capita emissions by more than 70% from 1998 to 2014, nearing the levels recorded in 1990 [28,50].

High energy intensity and air pollution are interconnected elements in the broader issue of sustainable development and environmental protection. Against this background, the issue of using clean energy as an alternative solution to combat climate change also occupies an important place. Switching to RES and applying energy conservation technologies can help reduce both energy intensity and air pollution. The development of RES is hampered by limited access to financing, weak institutional support and a lack of regional coordination in environmental policy. This indicates the region’s high dependence on fossil fuels and a lack of investment in alternative energy.

Thus, in Central Asia despite the relatively high average yearly economic growth rate—6.2%, in the last 30 years, the main part of the commodity exports of Kazakhstan and Turkmenistan has been fossil natural resources and products of their primary processing (73.6–81.4% of all commodity exports in 2023). This is characterized by a low level of product diversification, accompanied by high energy consumption, and demonstrates an export-dependent economy vulnerable to external shocks, which is exacerbated by a lack of financing. CACs face vulnerabilities not only from the risks linked to climate change and global warming but also from external economic shocks, heavy dependence on resource rents and volatility of global oil prices. This means that natural resources start as a “resource blessing” but eventually turn into a “resource curse” [51].

3.3. Assessment of the Impact of Ecological-Climatic Changes on the Regional Air Quality

3.3.1. Comparative Analysis of National GHG Emission Reduction Strategies

A major contribution to current climate change is the substantial influx of GHGs into the atmosphere resulting from unregulated human economic activity. The reports from the IPCC, 2022, 2023 affirm with a strong level of certainty that there exists an almost linear correlation between total human-caused carbon emissions and the resulting global warming they induce [52]. Restricting global warming caused by human activities to a specific level necessitates controlling the total CO₂ emissions as well as substantially reducing emissions of other GHGs. In order to evaluate the effects of environmental and climatic changes on atmospheric air, an analysis of changes in the volume of GHGs in CACs from 1990 to 2021 (Figure 6) was conducted using available data sources.

In Kazakhstan, the primary source of GHG emissions (Figure 7) is the energy sector, which accounted for 83.02% in 1990 and 77% in 2023 of the total emissions, which is equivalent to 261.9 million tons of CO₂-equivalent (CO₂-eq.) [36]. Of this value, 75% of pollution originates from stationary sources, including power plants. In addition to traditional energy and heat production, the Energy Activities sector, according to the IPCC classification, includes the following categories: Manufacturing and Construction, Transport, Other Sectors [41]. An analysis of total national emissions showed that GHG emissions fell by only 1.55% between 2012 and 2021 [53]. In Kazakhstan, coal occupies the dominant portion of total primary energy use, making up 49.6% of the fuel and energy balance. The following biggest contributors to total primary energy use are natural gas at 26.7%, oil and oil products at 21.7%. More than 90% of all balance reserves of hydrocarbons are concentrated in the western region of Kazakhstan, including 13,466.9 million tons of geological and 4457.3 million tons of recoverable oil reserves, as well as 1322.6 billion m³ of free gas. At the same time, the Atyrau region accounts for 73% of oil reserves, and the West Kazakhstan region accounts for 62% of free gas reserves [35].

GHG emissions in industry rose by 26.1% from 2012 to 2021, according to the results of the national inventory. In 1990, the industry accounted for 19,292.85 thousand tons of CO₂-eq, which is 5.05%, and in 2020—24,960.81 thousand tons of CO₂-eq or 6.3%. The primary contributors of GHG emissions include the production of iron and steel, ferroalloys, aluminum, cement, use of limestone and dolomite. During the specified period, the volume of GHG emissions from waste increased by 58.17%, the share of which amounted to 2.03% in 2020 [36].

In Kyrgyzstan, the main volume of GHG emissions (Figure 7) came from the energy sector: from 73% in 1990, 54% in 2010, to 68.3% in 2021. The second largest source of GHGs in the republic is agriculture. Thus, the share of the “Agriculture” sector accounted for 33.5% in 2010, 31.05% in 2018. The share of Industrial Processes in the Energy sector in 1990 was 2.5%, and in 2021 it was 8.5%. Overall, GHG emissions from the Energy sector have declined, driven by increased use of hydropower [37].

An analysis of GHG emissions by sector in the Republic of Tajikistan (Figure 7) reveals that in 1990, 60% of the country’s total GHG emissions originated from the energy sector, followed by agriculture (35%) and industry (9%) [54]. Within the energy sector, the largest contributors to CO₂ emissions were sub sectors such as energy production, manufacturing and construction, transport, as well as cement and aluminum production [38]. By 2021, GHG emissions in CO₂-eq. from the energy sector accounted for 42.91% of the country’s total emissions, while agriculture contributed 38.54%, industry 13.4%, and waste 5.21% (Tajikistan’s NDC The updated NDC of the Republic of Tajikistan, https://www.climatewatchdata.org/ which uses https://unfccc.int/sites/default/files/resource/4NC_TJK_ru_0.pdf, accessed on 18 August 2025). Climate Watch is an online platform designed to empower policymakers, researchers, media and other stakeholders with the open climate data, visualizations and resources they need to gather insights on national and global progress on climate change.

Climate Watch brings together dozens of datasets for the first time to let users analyze and compare the Nationally Determined Contributions (NDCs) under the Paris Agreement, access historical emissions data, discover how countries can leverage their climate goals to achieve their sustainable development objectives and use models to map new pathways to a lower carbon, prosperous future.

This free platform enables users to create and share custom data visualizations and comparisons of national climate commitments. It contributes to the goals of the Paris Agreement by using open data to increase transparency and accountability and providing actionable analysis on how countries can enhance their efforts to combat climate change.

In Uzbekistan, GHG emissions increased by 42.4% between 1990 and 2020 (Figure 7). The energy sector remained the dominant contributor, accounting for 67.95% of total emissions in 2020. Agriculture ranked second, contributing 16.07% in 2020, with a significant increase of 172.7% over the 1990–2020 period. The industrial sector was the third-largest contributor, accounting for 14.56% of emissions in 2020, with a growth rate of 121% over the same period [39].

In 2021, Turkmenistan emitted 189 million metric tons of CO₂-eq., representing a 2.3% increase compared to 2020. The primary sources of GHG emissions in Turkmenistan include the energy, industrial, transport, agricultural, and residential and communal services sectors. The energy sector was the largest emitter, producing 174 million tons of GHG emissions in 2021, which accounted for 92.1% of the total (Turkmenistan Summary of updated First Nationally Determined Contribution (NDC) https://www.climatewatchdata.org/, accessed on 22 August 2025, which uses official inventories reported to the UNFCCC as a basis https://www.undp.org/sites/g/files/zskgke326/files/migration/tm/undp-tm-NCCS_ENG.pdf (accessed on 18 August 2025).

Agriculture and industrial processes were the second and third largest contributors, accounting for 6.3% and 0.79% of total emissions, respectively. Waste accounted for 1.5 million tons, or 0.79%, of total emissions in 2021 (Turkmenistan Summary of updated First Nationally Determined Contribution (NDC)). This picture is similar to the trends in Uzbekistan, Tajikistan and Kyrgyzstan, where agriculture produces the second largest volume of GHG emissions. The situation in Kazakhstan shows that by 2024 industry has already reached and exceeded the level of agricultural emissions. Despite the fact that data on Turkmenistan regarding greenhouse gas emissions are scattered and are not available in the form of systematic national reports in the public domain for research, substitute sources of information on GHG emissions were used in this work.

Accurate data on the dynamics of GHG emissions in Turkmenistan for the entire period from 1990 to 2021 remain unavailable due to contradictory and insufficient information. However, estimates indicate that the total volume of GHG emissions in the country more than doubled between 2000 and 2017. Emissions from fuel combustion increased by 62%, primarily driven by the combustion of flare gas. The highest emissions occur during fuel combustion, as well as during the extraction, transportation, and storage of oil and gas [40]. According to another source, the average annual growth rate of emissions since 1990 has been 0.85% (Turkmenistan Summary of updated First Nationally Determined Contribution (NDC)). Nevertheless, the energy sector remains the primary source of GHG emissions in Turkmenistan.

Thus, the primary contribution to GHG emissions in all CACs is made by the Energy sector, which is linked to the production and increasing utilization of organic fuel. The main source of emissions from burning fossil fuels increased by more than 60% between 2001 and 2021 (http://eco.rian.ru/, accessed on 10 October 2024). A comparative analysis of national emission reduction strategies has found that, despite international commitments, actual emission reductions remain small. Kazakhstan has committed to reducing CO₂ emissions by 15% by 2030, but as of 2024 their level remains higher than in 1990.Moreover, in terms of GHG emissions, the largest amount was emitted by Kazakhstan. Uzbekistan plans to cut by 10%, but dependence on natural gas prevents significant improvements. Over the same period, Kyrgyzstan showed a smaller increase in emissions—by 43.3%, and Tajikistan—by 29.7%, which is obviously related to the use of hydropower, which accounts for 90–98% of electricity generation. While Kyrgyzstan and Tajikistan are betting on hydropower, investment growth in green energy remains slow. The data underscores the high reliance of CACs on fossil energy, which presents obstacles to meeting climate objectives and shifting towards RES. Unlike the EU and China, which have strict emissions quotas and carbon trading systems, Central Asian countries lack an effective carbon regulation mechanism.

The average CO₂ emission factor (carbon factor), which is the ratio of CO₂ emissions to primary energy consumption for the CACs, is the highest globally, exceeding 2.0–2.5 tCO₂ per tons of oil equivalent (toe) in both 1990 and in 2023. For Kazakhstan it was 3.18 in 2023 [32] (Figure 8). Accordingly, Kazakhstan, Uzbekistan and Turkmenistan rank among the 100 countries globally with highest carbon footprints and significant CO₂ emissions [54,55,56].

3.3.2. Study of the Dynamics of Fossil CO₂, PM_2.5 and SO₂ Emissions in Central Asian Countries

This subsection examines the dynamics of fossil CO₂, PM2.5 and SO₂ emissions in CACs, as well as their geographical distribution and relationship with the economic structure of the region. Considering that the main atmospheric emissions in the CACs are determined by emissions from fuel combustion, it was considered appropriate in this paper to examine in more detail the dynamics of CO₂ emissions from fuel combustion (Figure 9).

Since CO₂ accounts for 99.7% of the volume when burning fuel, other greenhouse gases CH₄ and N₂O account for 0.3%, and all information in the international classification of countries in the ratings of the World Bank, UNECE, US Energy Information Administration, etc., is given for CO₂ [28,30,32,34]. The data obtained indicate that in CACs, CO₂ emissions associated with the combustion of fossil fuels have not decreased for many years. Interpretation of these results shows that the high CO₂ emissions in Kazakhstan and Uzbekistan are associated with the continued high share of coal-fired generation and industrial production, while the low emissions in Kyrgyzstan and Tajikistan are due to the high share of hydropower. However, this makes their energy systems vulnerable to seasonal climate changes. If we judge the dynamics of CO₂ emissions from the combustion of fossil fuels in Turkmenistan, the data presented in Figure 9 indicate that the general situation for the period under study remains practically at the same level. Moreover, in 2019, the level of emissions increased slightly—by 18% compared to 2018, but subsequently dropped to the 2011 level. The study of data on Turkmenistan draws the attention of researchers to the problem of the objectivity of data on this country. In general, the process of collecting data on this topic and these countries, checking this data for their applicability showed the presence of a significant problem with the formation of environmental data. This indicates that against the backdrop of growth in energy, industrial and agricultural production (Figure 2) and population growth in Turkmenistan (by 30%), certain results from decarbonization processes are having an effect. This indicates that against the backdrop of growth in energy, industrial and agricultural production (Figure 2) and population growth in Turkmenistan (by 30%), certain results from decarbonization processes are having an effect. For visualization, the map (Figure 10A) shows a comparison of CO₂ emissions from fuel combustion in CACs in 2020 (Kyrgyzstan—9089, Tajikistan—9329, Kazakhstan—211,890, Turkmenistan—63,665, Uzbekistan—115,589 tons).

The map (Figure 10B) shows CO₂ emissions per capita in metric tons, allowing a comparison of the environmental burden created by the average person in each country. In terms of emissions per capita, Kazakhstan is among the world’s largest emitters, ranking 12th in 2020 [47].

The map (Figure 10C) shows the ratio of specific CO₂ emissions to GDP (in kilograms per unit of GDP) (Tajikistan—0.21, Kyrgyzstan—0.26, Uzbekistan—0.43, Kazakhstan—0.44, Turkmenistan—0.61 kg CO₂ per unit of GDP), which allows us to estimate the carbon intensity of each country’s energy sector. As can be seen from (Figure 10C), in Kazakhstan CO₂ emissions per unit of GDP are 73%, higher than the global average (0.25). Kyrgyzstan and especially Tajikistan have significantly lower indicators, which is explained by the large share of clean energy, on average 25.48%, 52.37%, respectively [47]. Thus, a comparative analysis of the spatial distribution of CO₂ emissions revealed significant territorial differences in pollution levels. These data indicate the need for a territorially differentiated approach to environmental policy, taking into account the characteristics of natural conditions and industrial development.

Simultaneously, the combustion of various fuels releases several over pollutants, including carbon monoxide (CO), non-methane volatile organic compounds (NMVOC), sulfur dioxide (SO₂), nitrogen oxides (NO_X), and PM, all of which contribute to air pollution on local, regional, and even global scales [33]. PM_2.5 are dangerous for human health air pollutants, which includes black carbon and tropospheric ozone [26]. Below are data (Figure 11) showing PM_2,5 emissions in CACs since 1990. Almost the entire population of the region is exposed to their impact for a long time, which is associated with the climatic, geographical and meteorological characteristics of the region, which create conditions for the accumulation of polluting particles in the air, primarily in large cities and industrial centers, especially during the heating season (the concentration of PM_2.5 increases sharply). This problem is particularly relevant for some large cities in CA, where it can be aggravated by socio-economic factors and geographical features [47].

Typically, urban and industrialized areas with high population density have the highest levels of air pollution. Although the sources of PM_2.5 vary somewhat from country to country, they are definitely related to activities such as fuel combustion, agricultural waste, municipal solid waste, industrial activities, etc. An analysis of PM_2.5 emission sources obtained with a model using satellite measurements from a global low-resolution network dataset showed the following sources: wind-blown dust—33%, energy (CHP and heating)—40%, waste—7%, industry—4% and transport—3% PM levels indicate significant excess of World Health Organization (WHO) recommended values (5 µg/m³) [57]. The data obtained correlate with the findings of other studies on PM_2.5 emissions [9,44,58].

The map (Figure 12) displays the annual average levels of PM_2.5 air pollution in the region. In addition, the average monthly concentrations (January 2020) in five cities of Central Asia obtained in the work [9] are presented in the form of colored circles of different sizes. These data were grouped into ranges (27, 74, 99, 110 and 112 μg/m³) to reflect the degree of pollution in each city. Each color demonstrates a certain level of air pollution, indicated in the legend. It was shown that average monthly concentrations of PM_2.5 in these Central Asia cities exceeded WHO standards (5.0 μg/m³) by 4.3–12.6 times [9]. High concentrations of PM_2.5 in the cities of Almaty, Tashkent and Bishkek in winter exceed 100 μg/m³, which is significantly higher than WHO standards. Elevated PM2.5 concentrations in major cities indicate the impact of motor vehicle transport, coal heating and adverse weather conditions in winter. According to the 2021 study, the authors [9,57] associate higher concentrations of PM _2.5 in winter in the aforementioned Central Asia cities with heating, as well as a high frequency of ground-level inversions, stagnant weather conditions with lower planetary boundary layer heights and calmer wind conditions that impair PM_2.5 dispersion and dilution. It should be noted that in 2021, PM_2.5 levels in Kazakhstan exceeded WHO recommended limits by 2–15 times in 25 out of 27 cities [58]. According to IQAir, a global air quality monitoring platform, the concentration of PM_2.5 in the city of Uralsk in 2022 is 2.5 times higher than the WHO recommended values. Thus, in large industrial centers, concentrations of PM _2,5 reach critical levels, while in mountainous and sparsely populated areas the level of pollution remains low.

Over 80% of SO₂ emissions are anthropogenic [59]. Below are data (Figure 13) showing SO₂ emissions in CACs since 1990.

According to 2019 data, three of the world’s top 20 SO₂ emitters were in Central Asia: Kazakhstan (10th place), Uzbekistan (14th place) and Turkmenistan (17th place) [59]. Of course, the presented data are episodic in nature, since they are related to the activities of external experts, or within the framework of the implementation of grant projects. They need to be systematized and confirmed by ground measurements, but at the same time, they allow us to identify the main reasons for the decline in air quality in the region so that governments can make appropriate decisions.

Thus, the analysis of the results shows that the high level of emissions in the region is due to the structure of the economy based on fossil fuels, underdeveloped emission regulation mechanisms and weak integration into global initiatives to reduce air pollution. Moreover, the main contribution to GHG emissions is made by energy. In CACs, the proportion of CO₂ emissions from the combustion of organic fuels has not changed significantly over the past 30 years and ranges from 54 to 78%. As a result, carbon factor for CACs is the highest in the world—over 2.0–2.5 t CO₂/toe both in 1990 and in 2023. This is due to high energy and resource intensity of the region’s economy, there is inefficient use of natural resources and high energy consumption. This requires a comprehensive approach to reducing emissions, including the development of RES, the modernization of old coal-fired power plants, the introduction of stricter environmental standards and increased regional cooperation on air quality management.

3.4. Econometric Modeling in the R Program

To test the hypotheses put forward, econometric modeling was used to identify the level of dependencies between the dependent variable and its determinants. The collection and processing of data for five countries allowed us to form a panel dataset for the period 2012–2023, which included the following:

Variables
y1	CO2 emissions from fuel combustion, Mt	x6	GDP, billion USD
x1	coal consumption, Mt	x7	Energy consumption, Mtoe
x2	Oil product consumption, Mt	x8	SO2 emissons, kt
x3	natural gas consumption, Mt	x9	PM2.5 emission, kt
x4	global carbon factor, tCO2/toe	x10	share of RES, %
x5	energy intensity, koe/$15p	x11	CO2 emissions per capita, t/person

Correlation analysis of the data across countries revealed the following level of closeness of relationships, which allows the selection of variables for further modeling (Appendix B).

The graph in Figure 14 shows the correlation matrix for five countries for the period 2012–2023. The X and Y axes represent the variables, and the Z axis represents the correlation coefficient. The color scale shows the level of correlation: dark colors indicate negative relationships.

Strong positive correlations (yellow dots, closer to 1): CO₂ and GDP: high positive correlation. CO₂ and energy consumption: confirms the relationship between energy consumption and CO₂ emissions. SO₂ and coal: coal-fired generation increases sulfur emissions.

Correlation analysis of country-level data revealed the following strength of relationships, enabling variable selection for further modeling (Appendix B).

The plot in Figure 14 shows the correlation matrix for five countries over the period 2012–2023. The X and Y axes represent variables, while the Z axis indicates the correlation coefficient. The color scale reflects the correlation level, with darker colors indicating negative relationships.

Strong positive correlations (yellow dots, close to 1):

CO₂ and GDP: high positive correlation.

CO₂ and energy consumption: confirms the relationship between energy use and CO₂ emissions.

SO₂ and coal: coal-based power generation increases sulfur emissions.

Strong negative correlations (dark purple dots, closer to −1): RES_Share and Oil: renewable energy sources replace oil. RE_Share and Coal Consumption: renewable energy sources reduce coal consumption. Energy Efficiency and Oil Consumption: the higher the oil consumption, the lower the energy efficiency.

Medium strength of the relationship (green-blue dots): Carbon_Factor and GDP: moderate relationship via the energy intensity of the economy. SO₂ and PM_2.5: SO₂ emissions are associated with PM_2.5 pollution.

RES_Share is highly correlated with Oil, Coal and Energy_Intensity, which requires its exclusion from the modeling. The analysis of Energy_Intensity shows a high dependence on coal and oil, which also requires its exclusion. Since oil and coal consumption are highly correlated (−0.79), coal consumption is used for the modeling, as its impact on CO₂ is more significant. We build the model in R using fixed (FE) and random (RE) effects [60]. The analysis is based on panel data for five countries—Kazakhstan, Kyrgyzstan, Uzbekistan, Turkmenistan and Tajikistan—for the period 2012–2023. The dataset includes 12 variables, where the dependent variable is CO₂ emissions from fuel combustion, and the explanatory variables are energy consumption, GDP, coal and gas consumption, carbon factor, energy intensity, SO₂ and PM_2.5 emissions, the share of renewable energy and CO₂ emissions per capita (Appendix A). The data cover 12 years for each country, giving 60 observations (12 years × 5 countries). Information is obtained from the Enerdata Yearbook and World Bank databases (https://yearbook.enerdata.net/, accessed on 10 October 2024, https://data.worldbank.org/assets/images/placeholder.png accessed on 12 October 2024, https://data.worldbank.org/indicator/EN.GHG.CO2.WA.MT.CE.AR5, accessed on 12 October 2024).

To assess the factors affecting CO₂ levels, a fixed effects model was built, including five key factors that were tested for association and multicolinearity. As a result of the analysis, three variables were statistically significant: carbon intensity (Carbon_Factor), GDP (GDP), and energy consumption (Energy_Consumption). All of them showed a significant impact on CO₂ emissions, as confirmed by their p-levels. The model is based on Kaya Identity, which expresses CO₂ emissions as a product of factors: GDP per capita (G/P), energy intensity (E/G), and carbon intensity (F/E).

To estimate the impact of variables on CO₂ emissions, panel data models were used: fixed effects (FE) and random effects (RE) [60], (

Table 1. Initial conditions for modeling effects in R.

Type of Model	Indicator	Model Equation	Data
Fixed effects model (FE)	model = “within”	model_ <- plm(CO2~Energy_Consumption + GDP + Coal + Gas + Carbon_Factor	panel_data between period of 2012–2023г across 5 countries, 60 historical values
Random effects model (RE)	model = “random”

). Both models demonstrated high explanatory power, with R-square exceeding 0.99, indicating an excellent fit to the observed data.

Further testing of the data confirmed the possibility of using the fixed effects model for several reasons: (1) comparison of the results of the FE and RE models (

Table 2. Comparison of FE and RE models.

Variable	Coefficient (FE)	Std. Error (FE)	p-Value (FE)	Coefficient (RE)	Std. Error (RE)	p-Value (RE)
x1	0.222	0.231	0.343	0.306	0.196	0.118
x2	−0.309	0.278	0.271	−0.322	0.252	0.202
x3	9.926	3.108	0.002	11.016	2.957	0.00019
x4	0.121	0.045	0.009	0.124	0.045	0.005
x5	2.492	0.249	0.0	2.560	0.213	0.0

), (2) conclusions based on graphical visualization of the data, (3) the significance of the conclusions for further forecasting.

Based on the obtained results of the model coefficients, we visually compare the significance of the two equations and plot a graph (Figure 15).

The graph shows how the coefficient estimates differ between the fixed effects (FE) and random effects (RE) models. The FE model shows more stable and interpretable results, especially for the key variables (x3, x4, x5). The use of the fixed effects model allows for time-invariant characteristics of each country, such as geography, natural resources, and industrialization levels, which can affect CO₂ emissions.

In the course of the study, both approaches to estimating the random effects model were tested for comparison between fixed effects and random effects models—the classical Wallace–Hussain method, as well as the more robust Amemiya method [61], recommended for analyzing small panel samples (n = 5 countries).

As a result of the analysis, the random effects model was applied using the Amemiya method [61]—a more stable approach when dealing with a limited number of panel units (n = 5 countries). This method minimizes bias and instability in variance estimates, especially in small panels. A repeated Hausman test using Amemiya estimation yielded a result (p = 0.7493), which is explained by a reduced discrepancy between RE and FE estimates, typical for robust methods.

Despite differences in results, the fixed effects model was selected for interpretation and building the forecasting model in the study because:

-

it accounts for potential correlation between individual effects and explanatory variables;

-

it provides better interpretability with a small number of panel units.

As we can see, the null hypothesis was rejected in favor of the alternative, which confirms the use of the fixed effects model. As a result, the following data were obtained for significant variables (predictors) (

Table 3. Results of the fixed effects model for CO₂ (within model) with three significant predictors.

Variable	Evaluation of the Coefficient	Standard Error (FE)	t-Value	p-Value	Significance
x3- Carbon_Factor	9.8701	3.0907	3.1935	0.002388	**
x4- GDP	0.1133	0.0414	2.7369	0.008468	**
x5-Energy_Consumption	2.4903	0.2456	10.1394	6.43 × 10−14	***

).

Before interpreting the results of a panel regression model, it is important to ensure that the key assumptions of classical linear regression are met, especially homoscedasticity and the absence of autocorrelation. Violating these conditions can lead to biased standard errors, distorted p-values, and reduced reliability of the conclusions.

To test for homoscedasticity, the Breusch–Pagan test was used, which showed p-value = 0.09375, indicating the absence of heteroscedasticity. This confirms the correctness of using conventional standard errors. To test for autocorrelation, the Wooldridge and Baltega–Wu tests were used. The Wooldridge test (F = 6.60; p = 0.013) revealed first-order autocorrelation, which requires the use of robust standard errors. The Baltega–Wu test (χ² = 20.66; p = 0.056) showed borderline significance. To eliminate the influence of autocorrelation, robust standard errors were used according to the Arellano method [62], which ensures correct standard errors and stability of parameter estimates. The use of these methods increased the reliability of the results and allowed us to draw reasonable conclusions about the influence of factors on the level of CO₂ emissions in the region.

Retesting for autocorrelation after robust errors.

# Robust errors with autocorrelation and panel structure

coeftest(model_fe1, vcov = vcovHC(model_fe1, method = “arellano”, cluster = “group”))

result coeftest(model_fe1, vcov = vcovHC(model_fe1, method = “arellano”, cluster = “group”))

Estimate	Std. Error	t value	Pr(>|t|)
x3—9.8700997	4.8508973	2.0347	0.04699 *
x4—0.1133392	0.0075953	14.9223	<2.2 × 10−16 ***
x5—2.4902838	0.3897199	6.3899	4.64 × 10−8 ***

After diagnostics of the model correctness: heteroscedasticity and autocorrelation, an empirical analysis of the model was carried out. An increase in carbon intensity by 1 unit leads to an increase in CO₂ emissions by 9.8701 units, which confirms that the use of carbon-intensive energy sources contributes to an increase in emissions. This emphasizes the need to switch to less carbon-intensive sources to reduce the carbon footprint. An increase in GDP by 1 unit increases CO₂ emissions by 0.1133 units, which corresponds to the Kuznets curve hypothesis: at the initial stages of economic growth, emissions increase, but with further development of the economy they can decrease due to technological improvements. An increase in energy consumption by 1 unit leads to an increase in CO₂ emissions by 2.4903 units, which confirms that high energy consumption, especially from carbon-intensive sources, increases emissions. This sets the task of improving energy efficiency and switching to renewable energy sources as priority strategies in the fight against climate change.

3.5. CO₂ Emissions Forecast for 2024–2030 by Country

To forecast CO₂ emissions for (2024–2030) we use the calculated model

CO₂(i.t) = 9.8701·Carbon_Factor + 0.1133·GDP + 2.4903·Energy_Consumption

For the forecast for Kazakhstan, we use the following scenarios for changing factors, presented in

Table 4. Scenarios of change in factors for Kazakhstan.

Indicator	2024	2025	2026	2027	2028	2029	2030	Kazakhstan (2023)
Energy_Consumption	1.50%	1.20%	1.00%	0.80%	0.70%	0.50%	0.30%	88.0
GDP	3.00%	3.20%	3.50%	3.30%	3.10%	3.00%	2.80%	262.4
Coal Consumption	−1.00%	−1.50%	−2.00%	−2.50%	−3.00%	−3.50%	−4.00%	74.0
Carbon_Factor	−0.50%	−1.00%	−1.20%	−1.50%	−1.80%	−2.00%	−2.50%	3.23

.

The presented scenario is realistic, as it reflects current trends and plans of the Central Asian countries: historical data on the dynamics of economic growth, the level of energy efficiency and international climate commitments within the framework of the signed Paris Agreement.

The forecast for CO₂ emissions in Kazakhstan shows a slowdown in their growth, but the pace of this slowdown remains modest. (Appendix C). This underlines the conclusion that current measures are insufficient for a rapid transition to a low-carbon economy. Similar studies and assessments from organizations such as the UNFCCC, IMF and the World Bank confirm these findings. In particular, Climate Action Tracker (CAT) projects that CO₂ emissions in Kazakhstan will increase until 2035 under current policies, confirming the growth trend. Enerdata reports that CO₂ emissions from energy combustion will be 281 million tons in 2023, indicating stable emissions in recent years. The IMF also notes Kazakhstan’s high per capita CO₂ emissions, indicating that significant efforts are needed to reduce them.

The CO₂ emissions forecast for Kazakhstan for 2024–2030 (Figure 16A) shows that emissions will remain in the range of 290–299 million tons, above the 2012–2023 average of 274.6 million tons. This is due to rising energy consumption and economic development, which are increasing emissions, while reduced coal consumption and lower carbon intensity have little impact. The forecast model indicates that Kazakhstan is unlikely to meet its CO₂ emission reduction commitments, expecting a level of 298.6 million tons by 2030 instead of the stated 180–200 million tons. A more radical shift to clean energy sources and ambitious green economy measures will be needed to significantly reduce emissions.

Table 5. Forecast for Kazakhstan (manual algorithm for filling in the forecast table).

Year	Energy Consumption	GDP	Coal Consumption	Carbon Intensity	Calculation of Projected CO2 Emissions
Units	Mtoe	Billion USD	Mt	tCO2/toe	Mt
2024	88.0 × 1.015 = 89.32	262.4 × 1.03 = 270.272	74.0 × 0.99 = 73.267	3.23 × 0.995 = 3.214	284.48
2025	89.32 × 1.012 = 90.392	270.272 × 1.032 = 278.921	73.26 × 0.985 = 72.177	3.214 × 0.99 = 3.182	289.57
2026	90.392 × 1.01 = 91.296	278.921 × 1.035 = 288.683	72.17 × 0.98 = 70.737	3.182 × 0.988 = 3.144	290.84
2027	91.30 × 1.008 = 92.03	288.68 × 1.033 = 298.23	70.73 × 0.975 = 68.94	3.144 × 0.985 = 3.097	293.71
2028	92.03 × 1.007 = 92.68	298.23 × 1.031 = 307.48	68.94 × 0.97 = 66.87	3.097 × 0.982 = 3.048	295.42
2029	92.68 × 1.005 = 93.14	307.48 × 1.030 = 316.70	66.87 × 0.965 = 64.50	3.048 × 0.980 = 2.987	298.97
2030	93.14 × 1.003 = 93.42	316.70 × 1.028 = 325.59	64.50 × 0.96 = 61.92	2.987 × 0.975 = 2.913	298.59

Table 5. Forecast for Kazakhstan (manual algorithm for filling in the forecast table).

Year	Energy Consumption	GDP	Coal Consumption	Carbon Intensity	Calculation of Projected CO2 Emissions
Units	Mtoe	Billion USD	Mt	tCO2/toe	Mt
2024	88.0 × 1.015 = 89.32	262.4 × 1.03 = 270.272	74.0 × 0.99 = 73.267	3.23 × 0.995 = 3.214	284.48
2025	89.32 × 1.012 = 90.392	270.272 × 1.032 = 278.921	73.26 × 0.985 = 72.177	3.214 × 0.99 = 3.182	289.57
2026	90.392 × 1.01 = 91.296	278.921 × 1.035 = 288.683	72.17 × 0.98 = 70.737	3.182 × 0.988 = 3.144	290.84
2027	91.30 × 1.008 = 92.03	288.68 × 1.033 = 298.23	70.73 × 0.975 = 68.94	3.144 × 0.985 = 3.097	293.71
2028	92.03 × 1.007 = 92.68	298.23 × 1.031 = 307.48	68.94 × 0.97 = 66.87	3.097 × 0.982 = 3.048	295.42
2029	92.68 × 1.005 = 93.14	307.48 × 1.030 = 316.70	66.87 × 0.965 = 64.50	3.048 × 0.980 = 2.987	298.97
2030	93.14 × 1.003 = 93.42	316.70 × 1.028 = 325.59	64.50 × 0.96 = 61.92	2.987 × 0.975 = 2.913	298.59

The forecast for all other countries was made using the ARIMAX model, which takes into account exogenous variables such as economic growth and the introduction of green technologies. The ARIMAX model allows for not only the historical dynamics of emissions, but also the influence of external factors, which makes it especially useful in environmental studies. This is important because environmental processes are sensitive to changes in politics and economics. Using ARIMAX provides more accurate forecasts and helps in making informed decisions in the field of environmental protection. Thus, the ARIMAX model is a reliable tool for analyzing and forecasting CO₂ emissions, eliminating subjective adjustments and ensuring the objectivity of the results.

forecast_arimax <- forecast(model, xreg = as.matrix(future_xreg), h = 7) > forecast_arimax

As expected, the growth of emissions in Kyrgyzstan (

Table 6. Results of the CO₂ emission forecast for Kyrgyzstan.

Year	Point Forecast	Lo 80	Hi 80	Lo 95	Hi 95
2024	10.40497	10.02444	10.78549	9.823007	10.98693
2025	10.54485	10.16433	10.92538	9.962891	11.12681
2026	10.68474	10.30421	11.06526	10.102774	11.2667
2027	10.82462	10.44409	11.20514	10.242657	11.40658
2028	10.9645	10.58398	11.34503	10.382541	11.54646
2029	11.10439	10.72386	11.48491	10.522424	11.68635
2030	11.24427	10.86374	11.62479	10.662307	11.82623

) is possibly associated with the expected increase in GDP from 13.99 to 19.05 billion USD by 2030, which is one of the main factors driving the growth of emissions. According to Kaya Identity, economic growth leads to an increase in energy consumption and, as a result, CO₂ emissions. The expected growth of GDP from 13.99 billion USD (2023) to 19.05 billion USD (2030) increases energy consumption and CO₂ emissions. This pattern is confirmed by IEA reports and studies (IEA, 2023; Our World in Data). Kyrgyzstan’s energy sector still depends on fossil fuels—coal and natural gas. Renewable energy sources are developing slowly due to technological and institutional barriers (GIZ, UNECE, 2022). Therefore, an increase in energy consumption leads to an additional increase in CO₂ emissions. Based on the forecast data obtained, a forecast graph of CO₂ emissions for Kyrgyzstan was constructed.

The CO₂ forecast model for Kyrgyzstan demonstrates high reliability, excellent accuracy (MAPE ≈ 2.4%) and stable residuals. The forecast can be used for scientific, strategic and applied purposes. The forecast of carbon dioxide (CO₂) emissions in Uzbekistan (

Table 7. Results of the CO₂ emission forecast for Uzbekistan.

Year	CO2 Forecast	Lo 95	Hi 95
2024	107.48	102.66	112.3
2025	108.89	104.07	113.71
2026	109.07	104.25	113.89
2027	110.39	105.57	115.21
2028	110.57	105.75	115.39
2029	111.88	107.06	116.7
2030	112.06	107.24	116.88

), constructed using the ARIMAX model with exogenous variables (energy consumption, GDP, carbon intensity), is methodologically sound. The use of this model allows taking into account both economic activity and specific structural characteristics of the energy sector.

The CO₂ emission forecast for Uzbekistan shows that by 2030 the emission level will remain in the range of 107.0–112.1 million tons, which corresponds to the average value for 2012–2023 (102.5 million tons).

The results obtained demonstrate a moderate increase in CO₂ emissions by an average of 0.7% per year, which is consistent with current trends in the country’s economy and energy sector, as well as with the target indicators recorded in the updated Nationally Determined Contribution (NDC) of Uzbekistan—a reduction in greenhouse gas emissions by 35% by 2030 per unit of GDP (compared to the 2010 level) (UNFCCC, 2022 https://droughtclp.unccd.int/node/2353/printable/pdf, accessed on 10 October 2024). The scenario for the development of carbon emissions in Uzbekistan can be considered realistic and consistent with the strategic goals of sustainable development. The graph includes a forecast zone reflecting an optimistic and pessimistic forecast, since the model has errors due to the cyclical dynamics of historical data. However, the data indicate that Uzbekistan, like Kazakhstan and Kyrgyzstan, will not be able to achieve the declared environmental parameters by 2030.

The accuracy of the CO₂ emissions forecasting model for Uzbekistan was assessed using standard error metrics and residual autocorrelation analysis: RMSE (Root Mean Squared Error) = 2.01: a low value indicates a good approximation of the model to the actual values and the absence of large deviations. MAE (Mean Absolute Error) = 1.51: on average, the model deviates from the observed values by 1.5 CO₂ units, which is an acceptable level of accuracy. MAPE (Mean Absolute Percentage Error) = 1.48%: a forecast error of less than 2% demonstrates high predictive ability of the model and confirms its reliability. ACF1 (autocorrelation of the first-order residuals) = –0.05: a value close to zero indicates the absence of autocorrelation of the residuals, which confirms the correctness of the model specification. Thus, the ARIMAX model provides an adequate approximation of the CO₂ emissions dynamics and can be used for medium-term forecasting.

Metrics	Value (Training Set)	Metrics	Value (Training Set)
RMSE	2.007888	MAPE	1.475433
MAE	1.505546	MASE	0.3943097
MPE	−0.03831139	ACF1	−0.05035955

To estimate the dynamics of carbon dioxide (CO₂) emissions in Turkmenistan for 2024–2030, the ARIMAX model [63] was used, which combines time autodependence and the influence of external macroeconomic factors. The GDP growth forecast was +2.3% per year, according to estimates by the Asian Development Bank and the World Bank. The historical trend of energy consumption growth was set at +3.4% per year, and the scenario of reducing the carbon factor was set at 0.2% per year, which reflects the introduction of energy-efficient technologies. The model parameters (p, d, q) were determined automatically using the auto.arima function in R based on the minimum value of the Akaike criterion (AIC). The forecast values of exogenous variables were generated taking into account the official growth rates. The model was estimated with confidence intervals of 80% and 95%, which made it possible to take into account the uncertainty of the forecasts, the resulting data are presented in

Table 8. Results of CO₂ emission forecast for Turkmenistan.

Year	CO2 Forecast	80% Interval	95% Interval
2024	69.36	[67.40; 71.33]	[66.36; 72.37]
2025	70.38	[68.41; 72.35]	[67.37; 73.39]
2026	71.43	[69.46; 73.40]	[68.42; 74.44]
2027	72.52	[70.55; 74.49]	[69.51; 75.53]
2028	73.65	[71.68; 75.61]	[70.64; 76.65]
2029	74.81	[72.85; 76.78]	[71.80; 77.82]
2030	76.02	[74.05; 77.98]	[73.01; 79.03]

.

Visualization of the obtained data allowed us to construct the following forecast in R.

All trajectories considered cover the period 2024–2030 and reflect different approaches to forecasting carbon emissions (Figure 17). The ARIMAX model shows an increase in emissions from 69.4 million tons in 2024 to 76.0 million tons in 2030, which is consistent with the country’s economic and energy indicators. To assess the reliability of the constructed ARIMAX model, a comparison was made with the results of other authoritative sources and research initiatives.

Thus, the model using hybrid neural network algorithms [62] shows a similar growth rate with high accuracy (R² ≈ 0.99, AARD ≈ 3.7%). Historical data from Our World in Data and Worldometer (2022) [64] confirm the feasibility of the ARIMAX forecast at 69.9–70 million tons. The IEA/World Bank forecast (2013–2024) also points to an increase in emissions due to increased energy consumption and economic growth. At the same time, the alternative approach from GIZ (2024) assumes a reduction in emissions as part of green transition strategies. Overall, the ARIMAX model is consistent with consensus forecasts from international and regional sources, providing scientific validity for carbon modeling in Turkmenistan. In addition, the analysis of the errors of the ARIMAX model training sample showed its high predictive ability for key metrics. The mean absolute percentage error (MAPE) was 1.20%, and the mean absolute scaled error (MASE) was 0.575, indicating a significant superiority of the model over the naive forecast and high accuracy of predicting CO₂ emissions. The value of the first-order autocorrelation of the residuals (ACF1) is −0.38, which confirms the absence of significant inertia in the residuals and the correctness of accounting for the time structure of the series. The mean forecast error (ME) is +0.008, which is close to zero, indicating the absence of systematic overestimation or underestimation of emissions. Thus, the ARIMAX model provides an adequate approximation of the dynamics of CO₂ emissions in Turkmenistan and can be used for medium-term forecasting if current economic and energy trends are maintained.

The CO₂ emissions forecast for Tajikistan for the period 2024–2030 (

Table 9. Results of CO₂ emission forecast for Tajikistan.

Year	CO2 Forecast	80% Interval	95% Interval
2024	11.58	[10.93; 12.23]	[10.58; 12.57]
2025	12.27	[11.35; 13.19]	[10.86; 13.68]
2026	12.96	[11.83; 14.09]	[11.23; 14.68]
2027	13.64	[12.34; 14.95]	[11.65; 15.64]
2028	14.32	[12.86; 15.78]	[12.09; 16.55]
2029	15.00	[13.40; 16.59]	[12.56; 17.44]
2030	15.67	[13.95; 17.40]	[13.04; 18.31]

) is based on Kaya Identity, where emissions are considered as a function of economic activity, carbon intensity and energy consumption.

According to the data obtained, the influence of factors is as follows for Tajikistan. Carbon factor: it is expected to increase annually by 0.05 due to dependence on coal and fuel oil in the energy balance, especially in the cold season, despite the dominance of hydropower. Gross domestic product: average annual GDP growth is projected to be within 5.5–6.5%, which is equivalent to an increase of 0.8–0.9 billion US dollars per year. Energy consumption: after a temporary decline in 2020–2022, it will stabilize and reach 14.4 million tons of equivalent fuel by 2030. These parameters are used in the ARIMAX model to forecast emissions and assess the impact of individual factors in the medium term.

According to the data presented, CO₂ emissions from fuel combustion in Tajikistan show an upward trend: from 8.0 million tons in 2019 to a projected 15.7 million tons in 2030 (Figure 18). However, the long-term forecasts are influenced by many factors, including global economic conditions, domestic reforms and foreign economic relations Thus, although absolute CO₂ emissions in Tajikistan are growing, the country remains the leader in terms of low per capita emissions in the region. But, given the projected increase in emissions, it is important to continue to implement measures to reduce the carbon footprint and improve energy efficiency. It should be taken into account that long-term forecasts are influenced by many factors, including global economic conditions, domestic reforms and foreign economic relations.

All three trajectories show a similar moderate growth trend, which confirms the reliability of the model.

The more than threefold increase in CO₂ emissions over 22 years in Tajikistan (Figure 19) is associated with dependence on fossil fuels in winter, economic growth without green modernization, and the absence of political-environmental constraints. This context confirms the trend embedded in the ARIMAX model and strengthens its reliability when compared with historical data and cross-sector forecasts. An assessment of the predictive ability of the ARIMAX model for Tajikistan shows its high reliability. The mean error (ME) is close to zero, indicating the absence of bias. The root mean square error (RMSE = 0.388) and absolute error (MAE = 0.344) confirm the close match between the predicted and actual values. The MAPE = 5.14% indicates high reliability of the model, especially in macroeconomic time series, where errors of up to 10% are acceptable. The MASE is 0.51, which is more than twice lower than the naive forecast error, confirming the quality of ARIMAX [63]. The negative value of the first-order autocorrelation (ACF1 = –0.17) confirms the absence of systematic structure in the residuals, meeting the requirements of randomness and normality of the ARIMA model.

For Tajikistan, a slight dilution was applied according to the forecast schedule, since only in this country there is a steady increase in greenhouse gas emissions rather than their reduction. In this regard, achieving the goals of the Paris Agreement by 2030 is seen as the least possible.

Nevertheless, as the conducted study shows, confirmed by the results of econometric modeling, the Fixed effects model RE constructed taking into account historical data on CO₂ emissions for all five Central Asian countries demonstrated the validity of the proposed hypothesis H1. The closeness of the relationship between dependent and independent variables and their influence proved that the energy intensity of the structure of national economies of Central Asian countries is the main factor in the high level of emissions into the atmosphere. The study found that countries’ stated climate commitments under the Paris Agreement will fixed not be achieved by 2030. Some experts have already made similar assumptions about countries around the world in general [63]. However, the value of this study is that, by using various quantitative methods, it was econometrically proven that the measures currently being implemented by countries are insufficient. The question of whether external factors, such as geopolitical ones (e.g., the U.S. withdrawal from the Climate Agreement), population growth in some countries and demographic crises in others, will influence or adjust the results of modeling can be answered affirmatively. Moreover, against the backdrop of ongoing global warming processes and extreme weather conditions observed in many countries, the measures being implemented in Kazakhstan and Central Asian countries to reduce atmospheric emissions appear insufficient. Within the framework of possible scenarios for increasing the investment attractiveness of the Central Asian region for global business, agricultural development, and processing industries, without significant innovations in the implementation of alternative energy sources systems, a substantial reduction in greenhouse gas (GHG) emissions cannot be expected. This is because the modeling itself initially incorporates historical values that already generally characterized a non-breakthrough but gradual growth of all factors contributing to energy consumption and, consequently, GHG emissions. These include GDP per capita growth and a tilt towards energy-intensive sectors. However, this should not be considered as a limitation for forecasting. The sporadic dynamics of GHG emissions in Kazakhstan, Uzbekistan, Kyrgyzstan and Turkmenistan demonstrate that the measures being implemented in these countries for decarbonization and overall emission reduction do not correlate with national economic growth. Although, as the data show, Kyrgyzstan demonstrates the ability to achieve the stated commitments under the Paris Agreement more than other countries. Meanwhile, the dynamics of emissions in Tajikistan indicate that the transition to a green economy in the country is significantly lagging behind its core development. More precisely, as the modeling results show, a green economy in its fundamental essence has not become a developmental vector for Tajikistan. The main constraints for all states are financial resources needed for the development, acquisition, and subsequent implementation of eco-friendly technologies. Alongside this, there are issues with interstate relations concerning the joint use of natural resources (efficient and sustainable water consumption, wind energy utilization, etc.).

The conducted modeling and the subsequent forecasting of greenhouse gas emissions in Central Asian countries and Kazakhstan allowed us to obtain the most probable scenario. This is justified by the fact that the obtained forecast aligns with the main trends in the development of national economies. In this case, it can be recognized as an important macro-factor. On the other hand, the forecast was built on historical panel data over a 12-year period across five countries, without any consolidating or calculated adjustments to these values. Furthermore, the application of extrapolation confirms the correctness of acknowledging the probabilistic nature of the obtained forecasts. This allows us to recognize the validity of this data and the conclusions drawn from it.

The transition to a low-carbon economy is possible with a comprehensive approach that includes: the development of RES, increased energy efficiency, tighter environmental standards and deeper regional cooperation.

3.6. Strategies in Central Asia for Implementing Measures to Reduce Air Pollution

Central Asia strategies to implement measures for air pollution reduction

Analysis of the data presented in the study confirmed the assumption that Central Asia is one of the most energy-intensive and resource-intensive regional economies in the world. Experts believe that by 2030, CACs will consume 34% more energy resources due to population growth [64]. In our view, if this resource-intensive economic model continues, the expected rapid population and economic growth will lead to a significant increase in atmospheric emissions and will exert significant strain on essential natural resources and ecosystems.

Addressing the issue of air pollution necessitates a comprehensive approach and collaborative efforts of government entities, private sector and public organizations. At present, there are relatively few policy instruments in the CACs to regulate atmosphere pollution. According to an analysis of air quality legislation in 194 countries of the world, (with CACs), Global Assessment of Air Pollution Legislation (GAAPL) found that 34% of countries lacked any legislatively established standards for atmospheric air quality [37]. The establishment and enforcement of rigorous environmental regulations harmonized with the best world practices at the regional level will enhance collaboration in environmental protection efforts [64].

In the view of IEA, there is a potential for reducing atmospheric pollution, through enhancement of regional integration in the energy system and the revitalization of the unified energy system of the CACs. It is possible to develop the unique hydropower resources of the region. Thus, the annual hydropower potential of Tajikistan is about 527 TWh, but only 23 TWh has been developed [66]. However, one should not think that countries in the region such as Kazakhstan, Uzbekistan, and Turkmenistan, given the lack of investment, will be able to switch to renewable energy sources so quickly. For example, for Kazakhstan, Uzbekistan, and partially (in winter) for Tajikistan and Kyrgyzstan, the most acceptable and feasible option in practice at this stage may be the transition to low-emission and cleaner coal technologies. This requires upgrading existing coal-fired power plants, boiler houses and industrial plants with filters for PM, SO₂ and NO_x that meet modern emission standards, and installing equipment for continuous monitoring and control of atmospheric emissions. Research has shown that many of the air pollution control measures adopted under China’s 2 + 26 program have achieved notable declines in GHG emissions (mainly CO₂) within the existing energy structure. In addition, the “Action Plan for the Prevention and Control of Air Pollution” resulted in a reduction in SO₂, NO_x and PM_2.5 emissions by 36%, 31% and 30%, respectively [67,68]. Of course, simply increasing RES capacity is not enough to transition to a low-carbon energy system. This transition requires increased energy efficiency, which comes from investments in improved energy storage systems, carbon capture and sequestration systems, and smart energy distribution systems. [69,70].

The results of the analyses conducted and examination of reports of EDB, 2022; the UNECE, 2019; OECD, 2019; WMO, 2024, UNEP, 2021; along with data from the World Bank; TheGlobalEconomy.com; The U.S. Energy Information Administration.com and national reports of the CACs to the UNFCCC [17,20,23,27,29,31,33,34,36,37,38,39,40] as well as several additional sources that include environmental and economic indicators allow us to understand the connection between economic activity and the atmospheric emission levels [71,72,73,74,75,76,77,78,]. A matrix was constructed in the context of sustainable development of the region, (Table 10).

Table 10. Central Asia strategies to implement measures for air pollution reduction.

Factors/Conditions	Expected Results	Risks	Sustainable Development Goals: Indicators
Population and economic growth	Increase atmospheric emissions and place excessive pressure on key natural resources and ecosystems	Drying up of the Aral Sea, water shortages, droughts, extreme heat, unstable precipitation patterns, dust storms	SDG3: Good Health and Well Being; SDG2: Zero Hunger; SDG 8: Decent Work and Economic Growth; SDG10: Reduced Inequality; SDG 11: Sustainable Cities and Communities SDG 13: Climate Action
Decarbonization of the economy	National low-carbon development strategies, strengthening regional cooperation mechanisms to identify measures for adaptation to climate change, priorities for the most effective allocation of investments, development of a joint environmental protection program for sustainable development in Central Asia.	Financial burden on the budget of countries; social risks associated with job losses, difficulties in adaptation and implementation of new technologies	SDG 7: Affordable and Clean Energy; SDG5: Gender Equality; SDG 8: Decent Work and Economic Growth; SDG9: Industry, Innovation and Infrastructure; SDG11: Sustainable Cities and Communities; SDG12: Responsible Consumption and Production; SDG 13: Climate Change Adaptation.
Formation of a green strategy	Investments in low-carbon technologies, replacement of coal with gas fuel at thermal power plants, “green” projects, including RES, energy-efficient and resource-efficient technologies, modernization of the water and energy complex, land reclamation projects, combating desertification and reforestation, creating a modern transport and logistics system, “transition” to sustainable transport, efficient waste management	Significant investment in sustainable technologies and infrastructure will be required; social, technological and political risks	SDG 7: Affordable and Clean Energy; SDG 8: Decent Work and Economic Growth; SDG 11: Sustainable Cities and Communities; SDG12: Responsible Consumption and Production; SDG 13: Climate Change SDG 15: Life on Land
Implementation of international air quality standards, control systems and the best available technologies	Use of NetZero, water and waste management, Air Quality Control System, implementation of Air Quality Standards, energy strategy for Sustainability	Rising energy costs	SDG 3: Good Health and Well Being; SDG 9: Industry, Innovation and Infrastructure; SDG11: Sustainable Cityes and Communities; SDG12: Responsible Consumption and Production;
Environmental Aspects Registers (EnvAR) electronic registers of environmental aspects	Documentation and Analysis; Automatic tracking: waste, emissions, spills, energy consumption; Documentation and Analysis; Regulatory Compliance, Support for Environmental Management System; Transparency and Accountability	Additional costs	SDG3: Good Health and Well Being; SDG 8: Decent Work and Economic Growth; SDG11: Sustainable Cities and Communities; SDG12: Responsible Consumption and Production; SDG 13: Climate Change

Table 10. Central Asia strategies to implement measures for air pollution reduction.

Factors/Conditions	Expected Results	Risks	Sustainable Development Goals: Indicators
Population and economic growth	Increase atmospheric emissions and place excessive pressure on key natural resources and ecosystems	Drying up of the Aral Sea, water shortages, droughts, extreme heat, unstable precipitation patterns, dust storms	SDG3: Good Health and Well Being; SDG2: Zero Hunger; SDG 8: Decent Work and Economic Growth; SDG10: Reduced Inequality; SDG 11: Sustainable Cities and Communities SDG 13: Climate Action
Decarbonization of the economy	National low-carbon development strategies, strengthening regional cooperation mechanisms to identify measures for adaptation to climate change, priorities for the most effective allocation of investments, development of a joint environmental protection program for sustainable development in Central Asia.	Financial burden on the budget of countries; social risks associated with job losses, difficulties in adaptation and implementation of new technologies	SDG 7: Affordable and Clean Energy; SDG5: Gender Equality; SDG 8: Decent Work and Economic Growth; SDG9: Industry, Innovation and Infrastructure; SDG11: Sustainable Cities and Communities; SDG12: Responsible Consumption and Production; SDG 13: Climate Change Adaptation.
Formation of a green strategy	Investments in low-carbon technologies, replacement of coal with gas fuel at thermal power plants, “green” projects, including RES, energy-efficient and resource-efficient technologies, modernization of the water and energy complex, land reclamation projects, combating desertification and reforestation, creating a modern transport and logistics system, “transition” to sustainable transport, efficient waste management	Significant investment in sustainable technologies and infrastructure will be required; social, technological and political risks	SDG 7: Affordable and Clean Energy; SDG 8: Decent Work and Economic Growth; SDG 11: Sustainable Cities and Communities; SDG12: Responsible Consumption and Production; SDG 13: Climate Change SDG 15: Life on Land
Implementation of international air quality standards, control systems and the best available technologies	Use of NetZero, water and waste management, Air Quality Control System, implementation of Air Quality Standards, energy strategy for Sustainability	Rising energy costs	SDG 3: Good Health and Well Being; SDG 9: Industry, Innovation and Infrastructure; SDG11: Sustainable Cityes and Communities; SDG12: Responsible Consumption and Production;
Environmental Aspects Registers (EnvAR) electronic registers of environmental aspects	Documentation and Analysis; Automatic tracking: waste, emissions, spills, energy consumption; Documentation and Analysis; Regulatory Compliance, Support for Environmental Management System; Transparency and Accountability	Additional costs	SDG3: Good Health and Well Being; SDG 8: Decent Work and Economic Growth; SDG11: Sustainable Cities and Communities; SDG12: Responsible Consumption and Production; SDG 13: Climate Change

Recommendations for the establishment on effective air quality management system

The presented matrix systematizes factors, conditions, and possible risks in the implementation of air pollution tasks focusing on the establishment on effective air quality management system in the context of regional development. The following measures can be proposed to reduce air pollution:

1. 

Introducing United data metric

The creation of a unified air quality monitoring system will allow CACs to more effectively monitor the level of atmospheric pollution and promptly identify sources of pollution, which includes the following:

• Data standardization: development of common standards for collecting and analyzing air quality data, which will ensure comparability of data between countries. First of all, countries in the region need to move from declarative environmental regulation to strict, scientifically based standards governing emissions of CO₂, PM_2.5, SO₂ and other pollutants.
• Improved monitoring: the introduction of modern technologies for monitoring pollutants, which will allow for more accurate assessment of pollution levels and the adoption of measures to prevent them. This requires the creation of independent emission control agencies, the introduction of a mandatory air quality monitoring system based on modern technologies, including remote measurements and automated databases.
• The introduction of a digital reporting system and the integration of emission data into national and regional platforms will increase the transparency of environmental regulation and reduce corruption risks.

2. 

Changing the structure of national economies

Transition to less energy-intensive technologies and the introduction of RES are key steps to reduce air pollution, which include the following:

• Investment in RES: increasing the share of RES: hydropower, solar and wind energy, which will help reduce dependence on fossil fuels and reduce carbon emissions. Countries can work with international financial institutions, such as the World Bank and the Asian Development Bank, to obtain funding for RES projects. These organizations can offer both financial support and technical assistance. Also, China’s involvement in solar and wind energy development could be an important factor in increasing the share of RES in the region. Kazakhstan and Uzbekistan have set goals to significantly increase the share of RES [24,25,39]. Other CACs may also develop national strategies to increase the share of RES.
• Adopting of Best Available Technologies: stimulating the transition to technologies that require less energy to produce and operate.
• It is necessary to modernize the existing coal infrastructure by introducing carbon capture, utilization and storage (CCUS) technologies.
• Development of legislative initiatives: adoption of laws and regulations aimed at supporting clean technologies and reducing air pollution from stationary emission sources. For example, it is necessary to create and implement legislation that supports the use of RES, including tax incentives for investors, subsidies for solar and wind installations, and guarantees for the purchase of electricity from RES.

3. 

Strengthening regional and international cooperation

Given the transboundary nature of air pollution, it is important to develop comprehensive cooperation on the following issues:

• Mitigation of the effects of climate change and adaptation to it. Key aspects include the development of highly accurate models for predicting climate risks and assessing the impact of emissions on the environment and public health.
• Combating desertification.
• Environmental protection in general and, in particular, protection of atmospheric air.
• Rational use of water and energy resources. Implementation of joint energy projects. The creation of joint projects, including RES projects between Central Asian countries can facilitate the exchange of experience and technologies. For example, joint research and development in the field of RES can lead to more efficient solutions adapted to local conditions.
• A regional approach and the development of transboundary solutions are the only way to counteract the negative impacts and consequences of climate change in the long term. A combination of policy incentives and international cooperation is needed to significantly increase the share of RES in Central Asia. This includes the creation of a favorable legislative framework, targeted programs, and active interaction with international partners to attract investment and technology. High capital costs for projects, particularly in RES, put the region at a disadvantage compared to developed countries. The estimated climate finance need is USD 38 billion per year up to 2030, necessitating substantial efforts to bridge this gap.

Regional integration should be considered as a necessary condition for achieving the SDGs of ensuring water, energy and food security in the CACs, as well as the successful resolution of regional environmental problems. The collective Statement issued by the leaders of Central Asia noted the importance of further strengthening regional cooperation on climate change mitigation and adaptation, combating desertification, efficient consumption of water and energy resources, provision of clean drinking water, environmental protection, conservation of biodiversity and glaciers, and promotion of projects and programs in the specified areas [4,52]. However, as the analysis has shown, at present the main actions in the protection of atmospheric air are carried out to a greater extent at the national level with minimal regional integration. In terms of strengthening regional cooperation mechanisms, developing climate monitoring, information exchange and forecasting systems is essential.

To this end, it is necessary to increase the potential and capabilities for exchanging data on hydrometeorological indicators. To implement an effective system for managing the quality of atmospheric air in the region, it is necessary to create a current pollutant monitoring database employing GIS technologies that make it possible to predict the quality of atmospheric air taking into account the socio-economic development of the CACs. International cooperation is also essential to addressing air quality issues and can facilitate knowledge sharing, capacity building and technology transfer, allowing the region to benefit from global expertise and resources. This concerns, first of all, the protocol of the UNECE Convention on Long-Range Transboundary Air Pollution on the issue of financing the Joint Programme for Monitoring and Evaluation of the Spread of Atmospheric Pollutants.

Thus, the transition to a low-carbon economy in Central Asia is possible only if comprehensive measures are implemented, including strict environmental regulation, development of renewable energy sources, modernization of industry, transformation of transport infrastructure and strengthening of regional cooperation.

4. Discussion

In this work, an attempt was made to conduct a comprehensive analysis of both GHG and non-greenhouse emissions for the Central Asia region. The hypothesis formulated in the work that the energy-intensive structure of the national economies of the CACs determines the high level of air pollution in the region was fully confirmed during the study. At the same time, the opposite null hypothesis of the study was refuted. This study fills several important scientific gaps related to the analysis of pollutant emissions in CAs.

First, previous studies of emissions in the region have either focused on CO₂ emissions without taking into account PM and sulfur dioxide, or have looked at air quality without regard to economic structure. This study is the first to combine the analysis of CO₂, PM_2.5 and SO₂ emissions, allowing us not only to assess the level of pollution but also to identify its key sources.

Secondly, there is no systematic study in the scientific literature examining the impact of economic structure on emissions in Central Asia. Previous studies have limited themselves to analyzing energy consumption, but have not linked it to industrial sectors, energy efficiency levels, or structural changes in the economy. This study is the first to assess the carbon intensity of GDP and energy intensity of economies in the region, providing insight into which sectors contribute most to pollution and what changes could reduce emissions.

The third research gap is the lack of comparative analysis of national emission reduction strategies in Central Asia. While international organizations publish countries’ projections and targets under the Paris Agreement, there has been no detailed assessment of the effectiveness of these strategies compared to actual emissions trends. This study is the first to compare national goals of Central Asian countries with actual trends and to analyze barriers to meeting these commitments.

The fourth aspect is to identify regional differences in air pollution. Most previous studies presented average data across countries, without taking into account territorial heterogeneity. This study shows that both in industrially developed regions (Kazakhstan, Uzbekistan), and in countries with a high share of hydropower (Kyrgyzstan, Tajikistan) air pollution, although remaining at a relatively low level, emissions continue to grow over the forecast period. For the first time, a detailed analysis is being conducted of how geographic and climatic factors, including terrain features and air circulation, influence emissions dynamics in the region.

Furthermore, this study fills a gap in the area of applied recommendations for Central Asian countries. Most emission reduction proposals presented in international reports focus on developed economies or large developing countries such as China and India, and do not take into account regional specificities. Our study suggests tailored measures including upgrading coal power, promoting possible renewable energy and strengthening regional cooperation.

Thus, this study closes several scientific gaps at once:

-

The lack of a comprehensive analysis of CO₂, PM_2.5 and SO₂ emissions in relation to the economic structure of the region;

-

Insufficient study of the impact of the sectoral structure of the economy on the level of emissions;

-

The lack of a comparative analysis of the effectiveness of national strategies to reduce pollution;

-

Regional differences in the level of air pollution.

In addition, it provides science-based recommendations for reducing emissions that take into account the economic and institutional realities of Central Asia.

Along with this, it should be emphasized that the key regional atmospheric pollution are emissions from energy, industry, transport, agriculture and waste. Unsustainable agricultural practices and food security issues, poor road infrastructure, high wear and tear of power lines, heating networks—50.4%, water supply systems—60.5%, sewer systems—61.3%, minimal waste disposal (no more than 11.5%) lead to pollution of all components of the environment, and first of all, the atmosphere [10,11,17,23,79,80,81,82,83]. Based on the analysis of air pollution in Central Asia, the following can be noted:

• CACs are implementing measures to enhance air quality and decrease pollution. However, the imperfection of environmental taxation, limited financial resources, and the lack of effective systemic monitoring and control of air quality that meets modern international requirements and standards lead to the fact that there are no tangible changes in practice yet; the share of CO₂ emissions from the combustion of fossil fuels per capita and per unit of GDP is one of the highest in the world. Assessing the impact of these factors will help identify additional opportunities to address the raised issues; however, the range of these questions falls within the realm of other scientific knowledge and requires detailed research. A quantitative assessment of the impact of these factors will allow us to identify additional opportunities for solving the problems raised, but the range of issues relates to the sphere of other scientific knowledge and requires detailed research.
• It has been established that there is no coordinated water and energy policy, transport and logistics system that would allow for the optimization of resource management, provision of regulation and implementation of strategic planning in the management of atmospheric air quality both within a single country and at the regional level. The study of these problems could resolve issues of interstate and country management and regulation aimed at the sustainable development of CACs.
• To effectively reduce emissions, the region needs a comprehensive strategy that includes a transition to a low-carbon economy, large-scale development of RES, industrial modernization, the introduction of carbon regulation and environmental taxation mechanisms, and strengthening regional cooperation in the field of climate policy.
• Further research into pollutant emissions and their impact on the environment in Central Asia could cover several key areas. An in-depth analysis of the spatio-temporal dynamics of CO₂, PM_2.5 and SO₂ emissions is needed using high-precision modeling methods taking into account spatio-temporal changes and the influence of economic factors. It should be noted the role of NOₓ as a precursor to PM₂.₅, and propose the inclusion of NO_x in future studies using regional monitoring systems and atmospheric chemistry models. It is important to develop models that assess the effectiveness of pollution reduction mechanisms such as carbon taxation and incentives for renewable energy sources. A promising area is forecasting the consequences of climate change on air quality and the regional economy. Research into the impact of pollution on human health and ecosystems is also needed to help develop evidence-based strategies for reducing emissions and promoting sustainable development in Central Asia.

5. Conclusions

• The scientometric analysis of the study revealed that publication activity has increased several times since 2016, which is associated with the adoption of the SDGs (2015). At the same time, it has been established that since 2018, the focus of research has shifted from the processes of pollution itself to overcoming its consequences. Moreover, data on air pollution in Central Asia is quite limited and unsystematic. The information base is the main foundation for making management decisions in both political-social and undoubtedly economic directions. Achieving the SDGs for the sake of preserving our planet is impossible without adequate, timely, and systematic data on the state of various aspects related to the corresponding SDG indicators. There is a pressing need to establish unified mandatory markers for information on emissions and environmental pollution. The research conducted in the Central Asia region as part of this work revealed a significant lag in data both in terms of format and timing, content, and the approaches to their presentation.
• The results obtained in this study confirm the hypothesis that the high level of emissions in Central Asian countries is a consequence of the prevailing raw material and energy-intensive economic model based on the extraction and consumption of fossil fuels. A comprehensive analysis of the dynamics of CO₂, PM_2.5 and SO₂ emissions over the period 1990–2024 revealed that the most polluted regions remain countries with dominant coal and oil and gas industries, such as Kazakhstan, Uzbekistan and Turkmenistan, while Kyrgyzstan and Tajikistan, which rely on hydropower, show lower emissions. An examination of the trends in GHG and non-greenhouse emissions in the CACs have shown that the shares of greenhouse emissions from the combustion of organic fuels have not changed significantly over the past 30 years, ranging from 54% to 78%. The results obtained confirm the key role of the carbon intensity of GDP and the energy intensity of the economy in the formation of emissions, proving that a country’s economic model is a determining factor in its ecological footprint.
• The analysis of the countries ‘economic structure has shown that for more than 30 years, it relies heavily on the extraction and export of natural resources, which is linked to high energy consumption and carbon emissions. Thus, the main part of commodity exports of Kazakhstan and Turkmenistan consists of fossil natural resources and products of their primary processing (73.6–81.4%), which characterizes a low level of commodity diversification, accompanied by high energy intensity. This demonstrates an export-dependent economy that is vulnerable to external shocks, further exacerbated by a financing deficit.
• For the first time, a comprehensive comparative analysis of the spatial distribution of emissions was carried out and revealed significant territorial differences in pollution levels. It was found that the level of emissions in the region is not uniform and is determined not only by economic activity, but also by geographical and climatic factors. In large industrial centers, concentrations of pollutants reach critical levels, while in mountainous and sparsely populated areas the level of pollution remains low. These data indicate the need for a territorially differentiated approach to environmental policy, taking into account the characteristics of natural conditions and industrial development.
• An important scientific contribution of the study is the critical assessment of national emission reduction strategies of CACs. The analysis showed that despite the declarative commitments under the Paris Agreement and nationally determined contributions (NDCs), their implementation remains insufficiently effective. The analysis showed that despite the declarative commitments under the Paris Agreement and NDCs, their implementation remains insufficiently effective. The main obstacles are the lack of strict emission regulation mechanisms, limited funding for climate programs, insufficient institutional support and weak interstate cooperation. Unlike leading global economies such as the EU, China and the USA, CACs have not yet implemented effective carbon regulation instruments, including emissions trading systems, carbon taxes and strict energy efficiency standards, leaving the region vulnerable to global climate challenges.
• Econometric modeling was used to construct an integrated fixed effects model FE, expressing the dependence of CO₂ emissions on the level of energy consumption, coal consumption, GDP and carbon intensity of national economies for all five CACs. The closeness of the relationship between dependent and independent variables and their influence proved that the energy intensity of the structure of national economies of CACs is the main factor in the high level of atmospheric emissions. The study found that countries stated climate commitments under the Paris Agreement will not be achieved by 2030. The value of this study is that, using various quantitative methods, it was econometrically proven that the measures currently being implemented by countries are insufficient. The obtained results can be used to develop decarbonization targets for commitments extended to 2050.
• The key mechanisms for reducing atmospheric emissions are the transformation of the region’s economies in the context of regional development, interstate cooperation. It is a need to significantly increase investments for the modernization of the water and energy complex based on low-carbon technologies and projects, including RES, combating desertification and forest restoration, the formation of sustainable transport and logistics system. It is important to harmonize regional legislation of the CACs with the best world practices in matters of air protection, to implement the environmental norms, standards, to create the modern information database of pollutants using GIS technologies. Regional integration should be considered as a necessary condition for achieving the SDGs. It guarantees the security of water, energy and food of CACs, as well as successfully solving common environmental problems, including air pollution.
• The study not only achieved its goal—to comprehensively analyze the dynamics of atmospheric emissions, identify their key sources and assess possible ways of reduction—but also formed scientifically based recommendations focused on the realities of Central Asia. To effectively reduce emissions, the region needs a comprehensive strategy that includes a transition to a low-carbon economy, large-scale development of RES, industrial modernization, the introduction of carbon regulation and environmental taxation mechanisms, and strengthening regional cooperation in the field of climate policy. The recommendations developed during the study not only allow us to adapt advanced international practices to the conditions of Central Asia, but also form the basis for further scientific research aimed at finding the most effective tools for the environmental transformation of the region.