Air pollution is one of the most serious environmental threats to both human health and the world’s economies (Sánchez-Triana et al., 2014). It has also been regarded as one of the growing environmental problems in South Asian developing countries, such as Pakistan. Since 1990, massive land use and land cover change (LULCC), urbanization, rapid industrialization, and economic development have improved in Pakistan (Ali et al., 2022). Pakistan is the second fastest-growing country showing rapid urbanization and a steep record of growth in industries such as the energy and transportation sectors. It is the 33rd largest country in the world, with a population of more than 207.8 million.

Urban areas are the densest for anthropogenic activities, and maximum air pollutant concentrations accumulate in the atmosphere. During the last two decades, transportation and industrial emissions have been the major air pollution contributors in Pakistan (Shahid et al., 2015); however, a new significant contributor has emerged—fossil fuel consumption in thermal power plants (Alkon et al., 2019). Similarly, in major cities of South Asia, air pollutant concentrations are significantly deteriorating the ambient air quality and human health in the metropolitan cities of Pakistan (Pilarczyk et al., 2019; Fatmi et al., 2020). The population in the urban centers is highly exposed to the effects of air pollution. Among all the air pollutants, particulate matter (PM) with an aerodynamic diameter of approximately 2.5 µm (PM_2.5) has been regarded as extremely harmful to human health. Long-term exposure to high concentrations of PM_2.5 dramatically increases health problems such as cardiovascular diseases, lung cancer, and respiratory problems. The health burden of exposure to PM_2.5 concentrations on an annual basis is higher than 10 μg/m³. However, this threshold value was set by the WHO. According to the WHO, PM and nitrogen dioxide (NO₂) concentrations were higher than the threshold value (Colbeck et al., 2010), and a higher mortality rate due to air pollution was noted, with 200 mortalities for a population of 100,000 in Pakistan. The concentrations of sulfur dioxide (SO₂) increased by 78% from 2005 to 2016 in Pakistan (Jabeen and Khokhar, 2019).

The solid and liquid particles suspended in the air are very complex and are known as atmospheric aerosols (Leonardi et al., 2020). The main sources of air pollution in Pakistan are power plants, transportation, petroleum factories, refineries, and increasing industries. The northeast region of Pakistan has severe air pollutant concentrations due to the industrial zone (Tabinda et al., 2020; Malhi et al., 2022), and likewise, India has stagnant pollution over Chandigarh, Bathinda, Amritsar, and Nangal cities (Raja et al., 2010). Different studies explored the air pollution situation in different areas of Pakistan with limited objectives. Lahore is the second most polluted city in Pakistan. A study by Lodhi et al. (2009) explored the PM_2.5 concentrations across Lahore from 2005 to 2006. It showed that road dust (18%), primary industrial emissions (26%), and secondary aerosols (51%) contributed to severe PM_2.5 in Lahore (Lodhi et al., 2009). Alam et al. (2014) revealed that PM₁₀ concentrations varied between 254 and 555 during March 2010 in Lahore (Alam et al., 2014). Urban air pollution mainly consists of SO₂, NO₂, ozone (O₃), and a mixture of carbon monoxide (CO), volatile organic matter, and suspended particulate matter (SPM) (Kaliyaperumal and Sharma, 2021; Joshi, 2023). These air pollutants are emitted from the combustion of fuel in industries and vehicles. In urban areas, photochemical oxidants including NO₂ and O₃ are the major concern among air pollutants because they are capable of deteriorating the environment and human health, respectively (Maring et al., 2023; Norazrin et al., 2023). NO₂ plays the role of primary pollutants in atmospheric chemistry in the formation of secondary pollutants such as nitrate aerosols, peroxyacetyl nitrate (PAN), and tropospheric O₃ (Sun et al., 2018; Feng et al., 2020). These pollutants reduce visibility in the atmosphere by absorbing radiation and have the potential to enforce global warming by unbalancing the radiation budget.

The climate of Earth is undergoing substantial change, mainly due to anthropogenic activities. Climate change is a worldwide phenomenon that has become more obvious in the last three decades (Destek and Sarkodie, 2019; Tang, 2019). There has been increasing interest in examining the impacts of climate change, which is a global phenomenon. As reported by the Fifth Assessment Report from the Intergovernmental Panel on Climate Change (IPCC), global land and ocean temperatures increased by 0.85 C between 1880 and 2012 (Ira, 2018). Air pollution is a prime concern across the globe due to human activities and adverse meteorological conditions (Kumari and Toshniwal, 2020; Agarwal et al., 2021). Climate change and air pollution are intrinsically linked. Radiative forcing (RF) induced the present-day atmospheric and climate system perturbation due to changes in different atmospheric pollutants (Christensen et al., 2022). In the context of climate change, air pollutants are perturbing the amount of incoming sunlight that is absorbed or reflected by the atmosphere. They have significant impacts on climate change. Anthropogenic emissions are contributing to adverse meteorological conditions (Fu et al., 2021).

Moreover, air pollutants alter the surface climate including the precipitation and temperature patterns (Falloon and Betts, 2010). Ramanathan et al. (2005) and Sillmann et al. (2017) studied the impact of aerosols on precipitation pattern changes and concluded that air pollutants significantly alter the precipitation pattern, including the monsoon cycle, across the world (Sillmann et al., 2017; Gautam et al., 2022).

Anthropogenic activities such as the release of greenhouse gases (GHGs), aerosols, and changes in land use have altered the Earth’s radiative balance and albedo. LULCC has the power to influence external forces on the atmosphere as well as natural and anthropogenic emissions of air particles. In addition, it affects the formation, migration, and movement of primary and secondary air pollutants (Zhou et al., 2016). Previously, the relationship between air pollution variation and land cover changes was investigated, revealing that urban expansion and cultivated land are increasing PM_2.5, while forests serve as a contributing factor in reducing PM_2.5 (Superczynski and Christopher, 2011; Zou et al., 2016; Lu et al., 2020; Xu et al., 2021; Ding et al., 2022). Generally, LULCC has shown that an urbanized environment, such as high-density buildings, contributes to increasing atmospheric pollutant concentrations, but low-density buildings are taking part in significantly reducing the atmospheric pollutant concentrations (Jiang et al., 2023). The northern region of Pakistan exhibited a long-term trend in the intensification of regional climate in tropical regions because of extreme deforestation in the area, which indicates the importance of vegetation cover in regulating aerosols and climate interactions from 1980 to 2018 (Chakraborty and Lee, 2019).

Pakistan has the highest urban population in South Asia, at 36.4%, surpassing that of Bangladesh, at 35.1% (Pasha, 2018). The uncontrolled expansion and densification of urban areas can disrupt the balance between impervious surfaces and green spaces, leading to the emergence of the surface urban heat island (SUHI) effect. This phenomenon is frequently observed in modern cities worldwide, contributing to the broader issue of global urban heat islands (UHIs) (Sadik-Zada and Gatto, 2022). The meteorological characteristics of an area undergo degradation due to urbanization, resulting in increasing air temperatures, reducing rainfall, increasing relative humidity, and increasing energy consumption (Arshad et al., 2019; Dilawar et al., 2021a; Dilawar et al., 2021b; Ali et al., 2020; Ali et al. 2021; Ali et al. 2023; Dilawar et al., 2021c). Riaz et al. (2014) revealed that Lahore’s urban land area increased from 33.28% (58,977 ha) in 1972 to 55.97% (99,173 ha) in 2009, accompanied by a proportional decrease in green cover during the same timeframe (Riaz et al., 2014). Similar LULCC patterns were also observed in studies by Bhatti et al. (2015) and Akhtar et al. (2020). These substantial transitions in Lahore’s urban landscape led to a significant increase in the average surface temperature in summer by 0.7 C and an annual temperature rise of 0.021 C from 1981 to 2013 (Arshad et al., 2022). Humbal et al. (2023) investigated anthropogenic influences on urban climate by emphasizing the impact of urbanization on the land surface temperature (LST) pattern in Pakistan (Humbal et al., 2023). Dilawar et al. (2021a) extended the study to evaluate the influence of urbanization on the SUHI in six major cities of the Punjab province in Pakistan. The findings consistently revealed a decline in vegetation cover at the expense of urbanization in all cities, contributing to a noticeable increase in SUHI over time (Dilawar et al., 2021a).

The occurrence of stagnant air pollution episodes in Pakistan shows that there is still a gap in understanding the driving factors of air pollution. In the current study, we focused on two prominent drivers: LULCC and climatic factors. To know how differently they affect air pollution changes, we aim to gain insights into their distinct contributions. There is a significant need to explore the impact of LULCC and climatic factors on air pollution dynamics, particularly in Pakistan. However, there is still a lack of a comprehensive methodology framework to investigate the association of air pollution with LULCC and regional climate variables. Developing such a framework is essential for effective air pollution control and management in Pakistan. Therefore, this study aims to bridge this gap and provide important insights into the complex relationships between air pollution and driving factors such as LULCC and climate change. It would be helpful to understand the human contribution toward the increase in air pollution and develop sustainable cities and the environment. To fully understand the air pollution control mechanisms and apply the policies adequately, it would be helpful to clarify the challenges associated with it. The current study comprehensively estimates the potential impacts of LULCC and regional climate change on air pollution variations during 2004–2021 in Pakistan by covering the following objectives:

1. To identify the long-term variability of air pollutants and evaluate the comprehensive changes in air pollution severity

In the current study, we have analyzed the LULCC, climate variability, and air pollution variation from 2004 to 2021 using satellite remote sensing and ground station data over Pakistan. The most prominent increasing trend for the PM was 0.68 per year. The entire population of Pakistan has been highly exposed to PM concentrations for many years (Bilal et al., 2021). It is caused by anthropogenic emissions associated with economic growth, including traffic flow and urbanization (Luo et al., 2021). PM concentrations decreased in the north and central southeast regions of Pakistan, with the highest decreasing trend at 0.55. In the case of NO₂ and SO₂, the increasing trend was 0.89 × 10 ¹⁵ molecules/cm² per year and 0.54 DU/year during 2004–2021. Most parts of Pakistan are facing an increasing trend of NO₂ emissions, but the highest increasing trend was found in the central, southcentral, and southeast regions of Pakistan (Rana et al., 2019). Transportation, industries, and urban areas are contributing to the increase in NO₂ emissions in Pakistan (Pervaiz et al., 2020). NOx emission reduction from power plants and vehicle emissions has been focused on because both on- and off-road vehicles significantly contribute to increasing air pollution, and multiple studies revealed a significant correlation between increasing air pollution and vehicle traffic (Huang et al., 2018; Jamali et al., 2020).

The SO₂ spatial variation showed a mixed trend across the country but showed a clear increasing trend over Lahore city, Punjab, and west–south region of Pakistan. Bilal et al. (2021) revealed that tropospheric NO₂ and SO₂ concentrations are high over Lahore, Punjab, Pakistan (Bilal et al., 2021). Previously, it was observed that column densities of SO₂ are higher in major cities like Lahore, Pakistan (Jabeen and Khokhar, 2019). In the case of O₃ trend concentrations, visible increasing and decreasing patterns were observed over Pakistan. In addition, the north region and west–south region experienced the highest increase (Zeb et al., 2019) and the highest decreasing trend per year, respectively.

LCM, MGWR, and proposed mathematical models have been used to determine the degree of contribution of LULCC and climatic factors to air pollution dynamics. Recently, a study used a simple GWR model based on their coefficients to describe the relationship between the two parameters in Bangladesh (Rahman et al., 2022). The complex topography conditions, intensive LULCC, and variant meteorological conditions during the past 20 years have affected the relationship with significant heterogeneity in Pakistan. We used the land change model to comprehensively study the LULCC during 2004–2021 and computed the spatial relationship on geographic variant location by considering the non-uniform effect of the surrounding element through MGWR. Second, the effect of climatology on air pollution was calculated by integrating the MGWR-based spatial variability relation with the differential equation-based contribution analysis during 2004–2021 over Pakistan.

In all LULCC types, the mean air pollutant concentrations significantly increased from 2004 to 2021, and the same results were found in Alam et al. (2015). All land cover types were increased except for barren lands. In addition, shrublands and grasslands had the largest increasing rates (Fan et al., 2020). Urban lands had the lowest increasing rate among all lands. The increasing trends indicate that the air pollution situation worsened during the study period. The increasing SDs mean that the spatial difference in air pollutant concentrations has increased. Indeed, the whole study area is facing a serious air pollution situation.

As shown in Table 2, in 2004, the mainland type in Pakistan was shrublands, whose area was 8.12 × 10⁴ mile², accounting for 32% of the total area, followed by grasslands and croplands, accounting for 24.02% and 38.32%, respectively. Urban lands covered 3.5 × 10⁴ mile², accounting for 1.3% of the total area. In 2021, although shrublands covered the second-largest area, their area increased to 8.74 × 10⁴ mile², accounting for 34.49% of the total area. The area of grasslands decreased by 5.9 × 10⁴ mile², accounting for 23.48% of the total area, and became the third-largest land type. The area of croplands and urban lands increased, accounting for 40% and 1.4%, respectively, and their proportion increased to 9.56% and 9%, respectively. The main conversion sources of urban lands were shrublands, grasslands, croplands, and barren lands, and the total conversion area was 0.98 × 10⁴ mile², which was caused by rapid urbanization. PM concentrations increased to their highest values in the PM trend. In addition, increasing trends were observed when grasslands were converted to urban land, wetlands, and shrublands. Urban development changed the grasslands, and forestlands contributed to a dramatic increase in PM_2.5 (Zhou et al., 2022). Moreover, some land cover transitions showed no trend, such as shrublands to barren lands and shrublands to grasslands. For NO₂ concentrations, land transitions from croplands and barren lands showed the highest increasing and lowest increasing trends, respectively. In the case of SO₂ concentrations, conversion of croplands to grasslands and that of barren lands to urban lands showed the second-highest increasing and lowest decreasing trends, respectively. The largest increasing trend was for grasslands converted to shrublands (Stavi, 2019), while the highest decreasing trend was for barren lands converted to croplands with the trend values of 1.5 DU and 0.7 DU for O₃. Tian et al. (2012) studied the responses of the ecosystem and revealed that O₃ concentrations are increased by increasing the shrublands (Tian et al., 2012).

In general, the trend was increasing for all parameters across the study region.

In the context of the climatic factor’s contribution, Precp showed a prominent increasing trend in the central and southern parts of Pakistan, with a maximum value of 1 mm per year. A study by Ahmad et al. (2017) concluded that the annual mean precipitation in Pakistan is increasing (Ahmed et al., 2018). The higher wind speed transported the large amounts of moisture that caused increased rainfall across the country. SP showed the maximum increasing trend of 0.01 per year while decreasing by 0.0038 Kpa per year, respectively. The WS trend was also increased but prominent in the north, central-south, southeast, and southwest parts of Pakistan (Ullah et al., 2022), where its increase was 0.04 m/s per year during 2004–2021. In the case of Tmax, the trend increased and decreased across the region, with the highest increase being 0.28°C/year and highest decrease being 0.08°C/year. The maximum decreasing and increasing trends for Tmin were 0.16°C/year and 0.19°C/year, respectively. The findings of the trend analysis showed geographical variation because of the impact of monsoon climate changes (Safdar et al., 2021).

The positive relationship means Precp contributed to the increase in PM, reaching its highest value of 10 by changing the per-unit Precp. The precipitation increases the PM because it forms rain droplets that act as PM (Lou et al., 2017). Both SP and WS showed a decreasing contribution of PM pollution in the northern region, with the highest values of −0.33 and −0.04, respectively. Hossain et al. (2021) found that maximum wind speed causes a decrease in the average concentrations of air pollutants (Hossain et al., 2021). Low wind speed causes stagnant pollution episodes, while the WS transports the pollutants. Tmax and Tmin positively contributed to the increase in the PM concentrations, particularly in the central region of Pakistan (Syed et al., 2021). Moreover, Tmin is most affected by PM concentrations than by Tmax across Pakistan.

NO₂ concentrations were increased by 0.19 × 10 ¹⁵ molecules/cm² per year for every unit change in Precp while reducing the NO₂ concentrations to 0.08 × 10 ¹⁵ molecules/cm² per year across the country. SP had significant positive and negative contributions, ranging from −0.01 to 0.007, while WS had positive contributions in most parts of the country, except in a small portion of the central region of Pakistan. The highest decreasing contribution in NO₂ concentrations was 0.007 × 10 ¹⁵ molecules/cm² per year by the change in unit Tmax. The relationship between Tmin and NO₂ concentrations had an increasing and decreasing trend in major regions of Pakistan. In general, Precp and WS contributed to increasing the NO₂ concentration over Pakistan. Meteorological conditions are significantly worsening the air pollutant concentrations, and it was concluded that wind speed played a role in increasing the NO₂ concentrations (Syed et al., 2021). Overall, we concluded that Precp and WS were the most influential climatic parameters for the SO₂ concentrations across Pakistan, with the highest contribution values of 0.004 DU and 0.003 DU, respectively. In general, a comparison of the contribution between O₃ and meteorological factors revealed that Precp was the most influential variable, followed by Tmax with the maximum contribution of 0.21 DU and 0.18 DU, respectively.

There are many contributors to the air pollution dynamics in the study region. Among them, the two most likely factors are LULCC and climate factors. The potential impact evaluation of LULCC and climate on air pollution variation provides effective information about air pollution driving forces and mechanisms. The current study evaluates Pakistan’s land use changes, regional climate variations, and air pollution changes from 2004 to 2021. The MK test and MGWR model were integrated with the differential equation model to assess the contribution of each parameter and how they are taking part in changing the pollution in Pakistan. It was assessed that the most prominent increasing trend for the PM was 0.68 per year. For NO₂ and SO₂, the increasing trend was 0.89 × 10 ¹⁵ molecules/cm² per year and 0.54 DU/year during 2004–2021. O₃ concentrations showed visible increasing and decreasing patterns over Pakistan: the north and west–south regions were experiencing the highest increase. In 2004, the shrublands, with an area of 8.12 × 10⁴ mile², accounted for 32% of the total area; although the area increased to 8.74 × 10⁴ mile², it accounted for 34.49% of the total area. The main conversion sources of urban lands were shrublands, grasslands, croplands, and barren lands, and the total conversion area was 0.98 × 10⁴ mile², which was caused by rapid urbanization.

During 2021, increasing trends indicated that the air pollution situation worsened during the study period. Increasing trends for PM were seen when grasslands were converted to urban lands, wetlands, and shrublands. Moreover, some land change transitions showed no trend, such as shrublands to barren lands and shrublands to grasslands. SO₂ concentrations in conversion of croplands to grasslands show the highest increasing trend. The largest increase was for grasslands converted to shrublands, with a trend value of 1.5 DU for O_3. The regional climate played a very important role in making air pollution stagnant across the country. The positive relationship means Precp contributed to the increase in PM, reaching its highest value of 10 when changing the per-unit Precp. In general, Precp and WS contributed highly to increasing NO₂ concentrations in Pakistan. Precp in SO₂ concentrations across Pakistan had the highest contribution values (0.004 DU). For O₃, the most influential climate factor was Precp.

Based on the results, we concluded that the dynamics of air pollution in Pakistan are significantly and inextricably linked with changes in land cover and meteorological variables. The type of pollutants and geographical region can explain the effects of land surface conditions and weather parameters on air pollution. These results, on a long-term temporal scale, comprehensively evaluate how climate variability and LULCC are altering ambient air quality in Pakistan. The current study can aid in determining the relevant key stakeholders and directing attention toward the best strategy for environmental protection and management. These useful insights may be beneficial for policymakers to develop suitable guidelines and regulations for preventing air pollution and keeping an eye on Pakistan’s air quality. It is encouraged to carry out more research on the interconnections between pollutants and climate variability by combining finer spatial and temporal resolution data with the intensity of anthropogenic activities and environmental governance activities. Furthermore, Pakistan has different climate zones, so there is a need to consider them to analyze the impact assessment of climatic parameters on air pollution across the zones. Second, the potential land use scenarios should be considered, including the effects of rapid urbanization and industrialization on air pollution episodes. Understanding the effects of land use management on variability in air pollution in Pakistan is still inadequate. It will help achieve the SDGs as recommended by the United Nations.