Investigating the Potential Climatic Effects of Atmospheric Pollution across China under the National Clean Air Action Plan

Abstract

To reduce air pollution, China adopted rigorous control mechanisms and announced the Air Pollution Prevention and Control Action Plan (APPCAP) in 2013. Here, using OMI satellite, the NASA Socioeconomic Data and Application Center (SEDAC), and Fifth ECMWF (ERA5) data at a 0.25° × 0.25° resolution, we explored changes in NO₂, PM, SO₂, and O₃ and climatology over China in response to the Action Plan between 2004 and 2021. This study attempts to investigate the long term trend analysis of air pollution and climatic variations during two scenarios before (2004–2013) and after (2013–2021) APPCAP. We investigated the climatic effects of air pollution in China before and after APPCAP adoption using geographically weighted regression (GWR) and differential models to assess the contribution of air pollution. The spatial representation analysis demonstrated how air pollution affected climatic factors before and after the APPCAP. Several important findings were derived: (1) the APPCAP significantly influenced air pollution reduction in China post-scenario (2013–2021); (2) the Mann Kendall test investigated that all pollutants showed an increasing trend pre-APPCAP, while they showed a decreasing trend, except for O₃, post-APPCAP; (3) for climatic factors, the MK test showed an increasing trend of precipitation and mean minimum air temperature t_min post-APPCAP; (4) innovative trend analysis (ITA) showed a reduction in NO₂, SO₂, and PM, although O₃ showed no trend post-APPCAP; and (5) pre-scenario, NO₂ contributed to an increase in the mean maximum air temperature (t_max) by 0.62 °C, PM contributed to raising t_min by 0.41 °C, while O₃ reduced the t_max(t_min) by 0.15 °C (0.05 °C). PM increased t_max and precipitation with a magnitude 0.38 °C (7.38 mm), and NO₂ contributed to increasing t_min by (0.35 °C), respectively, post-scenario. In particular, post-scenario led to an increase in t_min and precipitation across China. The results and discussion presented in this study can be beneficial for policymakers in China to establish long-term management plans for air pollution and climatological changes.

1. Introduction

Climatic change is a worldwide phenomenon that has become more obvious in the last three decades [1,2]. As a global phenomenon, there has been an increasing interest in examining the effects of climatic changes. According to the Fifth Assessment Report from the Intergovernmental Panel on Climate Change (IPCC), global land and ocean temperatures increased by 0.85 °C during 1880–2012 [3,4]. GHGs, aerosols, and change of land use patterns due to anthropogenic activities have altered the Earth’s radiative balance and albedo [5,6,7,8]. Alternations in the Earth’s radiative balance, through complex mechanisms, are responsible for changes in temperature, precipitation, and other climatic variables. Furthermore, significant climate crises can be attributed to the key drivers of climate variations, which include solar irradiance and volcanic eruptions [9,10]. Solar irradiance serves as the Earth’s primary energy source, directly impacting the climate system by influencing ozone concentrations, which induce substantial changes in atmospheric heating and circulation [11,12,13]. Volcanic eruptions are the most powerful natural phenomena of the industrial era, with minor eruptions occurring worldwide and emitting particles that remain confined in the troposphere for several weeks. On the other hand, explosive eruptions emit a significant amount of ash and SO₂ into the stratosphere, causing short-term climatic effects [14,15]. Urbanization has modified the local climate through the intensification of urban heat island (UHI) intensity effects, with UHI intensity continuously increasing, particularly in megacities [16,17,18,19].

Environmental pollution has affected the ecosystem and climate directly and indirectly, resulting in economic and social costs [20,21,22]. Air pollution is a prime global concern due to human activities and severe meteorological conditions [23,24,25]. Multiple studies have investigated the effects of air pollution impacts on human health, demonstrating reduced life expectancy and well-being, particularly in urban and industrialized areas [26,27,28,29]. The public awareness of environmental pollution and its challenges varies across the globe, with different stages of industrialization and urbanization in different countries. During the initial stages of industrialization, local economies were boosted, it raised prosperity, and people were less concerned with the negative effects of environmental pollution on the ecosystem [30,31]. China’s rapid economic expansion has led to significant pollution emissions and environmental devastation, with detrimental impacts on the local and regional environment [32,33]. Climatic change and air pollution are intrinsically linked, as air pollutants such as NO₂, PM, SO₂, and O₃ interact with solar and terrestrial radiation, contributing to climate perturbations [34,35]. SO₂, in particular, is a significant pollutant that causes severe health risks to individuals and environmental deterioration [36]. Changes in different atmospheric pollutants induce radiative forcing (RF), which perturbs the atmospheric and climate system [37].

In the context of climatic change, air pollutants disturb the amount of incoming sunlight that is absorbed or reflected by the atmosphere, resulting in significant impacts on the climate. Anthropogenic emissions cause adverse meteorological conditions [38,39]. Some studies have explored the systematic relationship between air pollution and climatic variables [40,41]. Moreover, air pollutants alter surface climates, including precipitation and temperature patterns [42]. Ramanathan (2005) and Sillmann (2017) investigated how aerosols could impact precipitation patterns and concluded that air pollution significantly alters precipitation patterns, including the monsoon cycle worldwide [43,44,45,46]. Over the last three decades, countries in Asia, such as China, India, Pakistan, and Nepal, have experienced severe air pollution because of industrialization and urbanization. The history of air pollution control shows that only increased awareness about the adverse impacts of pollution on health makes the governments take strict control measures [47,48]. Global industrialization, including in China today, has degraded ecosystems [49], depleted natural resources, and has had adverse effects on ozone levels that may harm plants and impair human health [30,50,51].

Due to the spatial scarcity of the consistent distribution of site stations, accurately assessing air pollution and the distribution of meteorological factors and their relationships becomes challenging. Under such conditions, satellite and reanalysis data provide a significant opportunity to assess these phenomena. Many studies have explored the long-term variation of air pollutants and their interaction with climatic factors using ozone monitoring instrument (OMI) satellite data [52,53]. OMIs have been used for a variety of studies, such as analyzing the relationship between air pollutants (NO₂, O₃, SO₂, CO) and meteorological factors over Bangladesh [54], analyzing the trends in NO₂ and SO₂ [55,56,57,58], evaluating the pollution emission inventory [59], as well as analyzing the long-term trends of NO₂ and SO₂ over East China [60]. Reanalysis datasets are potential alternatives for studying climate variability [61]. NASA SEDAC has provided high resolution PM data to analyze trends and impacts [62,63,64]. The ERA-interim and its new version, ERA5, are global reanalysis data made by the European Center for Medium Range Weather Forecasts (ECMWF). These reanalysis datasets include many climate state variables and provide ideal data for investigating climatic changes and improving water resource policies [65]. The hydrological regime of the Gilgit Basin in Pakistan was studied using ERA5 [66].

In response to heavy pollution and related concerns, the Chinese government has adopted certain measures to improve air quality and reduce environmental pollution [67]. China’s State Council has adopted three major action plans to control air, water, and soil pollution. In comparison to water and soil pollution, air pollution has gained more attention because it has significant spatial impacts and is highly visible [68,69]. To improve air quality in China, on 12 September 2013, the China State Council issued an action plan named APPCAP to guide the national efforts to prevent and reduce air pollution [70]. This action plan demanded a reduction in the mean annual concentrations of PM_2.5 between 15% and 25% across key regions. In this regard, the Chinese government has had the goal of reducing emissions of PM_2.5 and other pollutants since 2012 [71,72]. In 2013, the National Environmental Monitoring Center (NEMC) and the National Ministry of Environmental Protection (MEP) started monitoring public hourly and daily air quality data. As a consequence of the Chinese government’s enforcement of policies and measures, some types of pollution have been reduced over the last decade, such as smog, haze, and acid deposition [73,74,75]. By adopting the clean air action plan in 2013, China has achieved an average yearly reduction of 4.9%, 12.0%, and 5.3% for NO₂, SO₂, and CO, respectively, during 2014–2015. These reductions are considered to make APPCAP the most promising air pollution combat policy in China to date to improve air quality. To understand the promising efforts of China, the current research divided the study period into two different scenarios: before APPCAP (pre-scenario) and after APPCAP (post-scenario). The APPCAP is a milestone in air pollution control efforts in China and encourages researchers to evaluate the effect of air pollution management and its impacts at the national level in China. However, there is still lack of a methodology framework to investigate the air pollution association with the regional climatic variables under the context of air pollution control management in China. The current study analyzed the role of air pollution during the pre and post-scenarios based on APPCAP in regional meteorology changes (t_max, t_min, and precipitation) and associated impacts over China, with important implications for future air quality strategies in China. The climatic impacts of air pollution vary by location, and the air pollution-driven changes can be comparable to, or even more significant than, those caused by anthropogenic emissions [76]. Therefore, it is of great importance to identify the most influential air pollutant (NO₂, PM, SO₂, and O₃) for t_max, t_min, and precipitation changes over the whole of China since clean air actions were adopted.

This study aimed to gauge the effectiveness of the APPCAP policy in China and provide benchmarks and measurable implications for achieving air pollution mitigation strategies. To fully understand the air pollution phenomenon and apply the policy adequately, it would be helpful to clarify the aforementioned challenges, as shown in Figure 1. The current study comprehensively estimated the climatic impact of the APPCAP from 2004 to 2021 in China by following these objectives:

• To identify the long-term variability of air pollutants (NO₂, PM, SO₂, and O₃) under the pre-scenario and post-scenario scheme to evaluate the comprehensive goal of APPCAP;
• To determine the long-term trend of climatic factors (t_max, t_min, and precipitation) in compliance with the pollution control policy;
• To evaluate the spatial climatic contribution of air pollution control management over the past two decades at the national level in China.

2. Materials and Methods

2.1. Study Area

China is a vast and diverse country located in East Asia, covering an area of approximately 9.6 million square kilometers. It has a population of over 1.4 billion people, making it the most populous country in the world. China is divided into six major climate regions, including the tropical, subtropical, warm, middle temperate, cold, and plateau zones [77]. The subtropical zone covers the largest area of China, while the northern part of the country experiences year-round ice and snow. The eastern and southern regions of China are influenced by the Pacific Ocean and are characterized by monsoon climates such as hot, subtropical, and temperate monsoons [78]. These zones are further divided into sub regions based on Figure 2, which shows the location of these different climate zones, providing a useful reference for studying the climate and its impact on different aspects of life in China. Air pollution is a major issue in China, with many cities experiencing high levels of PM, SO₂, NO₂, and O₃. Rapid industrialization, urbanization, and the extensive use of fossil fuels have greatly contributed to China’s severe air pollution problem.

The northern parts of China, including Inner Mongolia and Xinjiang, experience a continental climate, with very cold winters and hot summers. The western region of China, such as Tibet and the Himalayas, has a plateau climate, characterized by low air pressure, thin air, and a large daily temperature range. The eastern and southern regions of China are influenced by the Pacific Ocean, with the monsoon climate dominating the area. The mountainous areas of China have a temperate climate, with cooler temperatures and lower humidity than other areas at the same altitude.

2.2. Data Description

Aura and Terra satellite-based OMI (https://giovanni.gsfc.nasa.gov/, accessed on 12 September 2022) measured the data of three air pollutants, including NO₂, SO₂, and O₃ [79,80,81]. PM data were collected from the NASA SEDAC, which provided global annual PM grids from MODIS, MISR, and SeaWiFS aerosol optical depth, v4.03 [62]. Aerosol optical depth retrievals from multiple satellite algorithms were calibrated based on Aerosol Robotic Network (AERONET) observations (https://sedac.ciesin.columbia.edu (10 September 2022)) [63,64]. Three meteorological parameters (temperature maximum, temperature minimum, and precipitation) were obtained from the Fifth generation (ERA5) (https://www.ecmwf.int/en, accessed on 28 September 2022) [82,83,84] as in the description written for

Table 1. Detailed information on data used in the current study. Note: ** Data were averaged into annual values.

Data Source	Parameter	Units	Spatial Resolution	Temporal Resolution
Ozone Monitoring Instrument (OMI) + NASA SEDAC	Nitrogen dioxide (NO2)	Molecules/cm2	0.25° × 0.25°	Daily **
	Ozone (O3)	Dobson Units (DU)	0.25° × 0.25°	Daily **
	Sulfur dioxide (SO2)	Dobson Units (DU)	0.25° × 0.25°	Daily **
	Particulate Matter (PM)	µgm−3	0.25° × 0.25°	Annual
Fifth-generation ECMWF (ERA5)	Temperature (max)	°C	0.25° × 0.25°	Monthly **
	Temperature (min)	°C	0.25° × 0.25°	Monthly **
	Precipitation	mm	0.25° × 0.25°	Monthly **

. Ground measurements were used to compare the satellite-estimated and NASA SEDAC data at some pollution monitoring stations as different locations as indicated in Figure 1. The ground measurements for air pollutants were accessed from a website (http://www.aqistudy.cn/, accessed on 20 February 2023) under the control of the Ministry of Ecology and Environment, China available from 2013.

Only NO₂, PM, and SO₂ ground concentrations were used to compare the satellite measurements and NASA SEDAC PM, and a high correlation was found between them, as shown in Figure 3.

2.3. Trend Analysis

In the present study, we used two non-parametric statistical approaches to determine whether the meteorological and air pollutants time series were increasing, decreasing, or showing no trend. ITA and MK tests were employed to identify trends in these variables. In particular, the ITA test was used to signify efforts to control air pollution in China. Traditional monotonic trend approaches detect trends using statistical calculations and cannot depict trends in different subcategories of data values. The ITA is an intuitive method for identifying trends in different time series subcategories, regardless of distribution assumptions [85,86,87]. Meanwhile, the MK test assumes that time series have no sequential correlation and is based on the assumption that all data are significant [88]. These techniques are widely used to analyze climatic parameters in particular [82,89].

2.3.1. ITA Trend Analysis

The ITA approach detects disparities in air pollution data time series using a non-parametric approach. Using ITA, the time series is divided into equal halves, and each subseries is sorted in ascending order separately. The first subseries is positioned (X_i; i = 1, 2, …, n/2) on the x-axis, and the second subseries is positioned (X_j; j = 1, 2, …, n/2) on the y-axis based on the Cartesian coordinate system. Both axes must have the same range to interpret the changed information. This approach provides a quick visual understanding of the nature of time series trends. The data values lie on a 1:1 (45°) straight line, indicating that the time series has no trend. The data points that lie above or below the area of triangulation show the increasing or decreasing trends of the time series, respectively [90,91]. The first half of the data points determines the trend change, and the trend indicator is determined by the difference between the two subseries divided by the mean of the first subseries [92].

The longitudinal trend is computed by taking the arithmetic average of the X_j and X_i values at every point in the time series. The vertical and horizontal distance from the linear line is used to calculate the absolute difference. The indicator of the trend [88,91] is computed using the following Equation (1):

$$S= \frac{2 \left(\overline{X_j}-\overline{X_i}\right)}{n}$$

(1)

where “S” shows the indicator of the trend, and n is the number of data points at $\overline{X_j} \mathrm{and} \overline{X_i},$ represented as the second and first subseries arithmetic averages, respectively. The positive and negative values of “S” show an increasing and decreasing trend. If the value of observation data is found odd in the original time series, the first observation might be discarded.

The significance of the slope was tested by the method proposed by Sen, and a description of the method was shown in [89].

Confidence limits (CLs) define the significance of the slope of the time series such as if the data points fall out of the CL, then it is statistically significant. To compute the CLs of the slope, it follows a Gaussian probability density function with zero mean and standard deviation as given below:

$$C{L}_{\left(1-\alpha \right)}=0\pm s_{cri}\times {\sigma }_s$$

(2)

where 𝛼 is the significance level at a 5% confidence that was used to evaluate the significance of the ITA trend, and $s\_cri$ is the standard normal PDF; the value of “$s\_cri$” can be calculated using the following equations:

$$s\_cri=\frac{Z-\alpha }{2}$$

(3)

where 𝛼 = 1 for confidence level. For example, if we have a 95% confidence level, then the corresponding level of alpha is as follows:

$$\mathsf{\alpha } =1-0.95=0.05$$

(4)

where ${\sigma }_s$ is defined as:

$${\sigma }_s =\frac{2\sqrt{2}}{n\sqrt{\mathrm{n}}}\sigma \sqrt{1-{\rho }_{\overline{xi},{\overline{x}}_j} }$$

(5)

where ${\rho }_{\overline{xi},{\overline{x}}_j}$ is the correlation coefficient between the two mean values.

$${\rho }_{\overline{xi},{\overline{x}}_j}=\frac{E \left({\overline{X}}_i{\overline{X}}_j\right)-E\left({\overline{X}}_i\right) E\left({\overline{X}}_j\right) }{{\sigma }_{x_i}{\sigma }_{x_j}}$$

(6)

2.3.2. Mann Kendall Trend Analysis

The aforementioned air pollution and meteorological data were evaluated to detect the spatial-temporal variation over the China region using two methods: a non-parametric MK test and Sen’s slope. The negative and positive magnitudes of trends were calculated to find the slope value and detect the trend changes in air pollution and meteorological factors in China. The MK test was used for the inhomogeneous time series data to illustrate the spatial and temporal patterns, and the ZMK test statistic was used [93,94]. The following mathematical equations were used to calculate the MK statistics V(S), S, and standardized test statistic Z:

$$\mathrm{S}={\displaystyle \sum }_{\mathrm{i}=1}^{\mathrm{n}-1}.{\displaystyle \sum }_{\mathrm{j}=\mathrm{i}+1}^{\mathrm{n}}\mathrm{sign} \left({\mathrm{X}}_{\mathrm{j}}-{\mathrm{X}}_{\mathrm{i}}\right),$$

$$\mathrm{sign} \left({\mathrm{X}}_{\mathrm{j}}-{\mathrm{X}}_{\mathrm{i}}\right)=\left\{\begin{array}{c}+1 \mathrm{if} \left({\mathrm{X}}_{\mathrm{j}}-{\mathrm{X}}_{\mathrm{i}}\right)>0\\ 0 \mathrm{if} \left({\mathrm{X}}_{\mathrm{j}}-{\mathrm{X}}_{\mathrm{i}}\right)=0\\ -1 \mathrm{if} \left({\mathrm{X}}_{\mathrm{j}}-{\mathrm{X}}_{\mathrm{i}}\right)<0,\end{array}\right.$$

$$\mathrm{V} \left(\mathrm{S}\right)=\left[\mathrm{n}\left(\mathrm{n}-1\right)\left(2\mathrm{n}+5\right)-{\displaystyle \sum }_{\mathrm{p}=1}^{\mathrm{q}}\mathrm{tp} \left(\mathrm{tp}-1\right)\left(2\mathrm{tp}+5\right)\right],$$

$$\mathrm{Z}=\left\{\begin{array}{c}\frac{\mathrm{S}-1}{\sqrt{\mathrm{VAR}\left(\mathrm{S}\right)}} \mathrm{if} \mathrm{S}>0\\ 0 \mathrm{if} \mathrm{S}=0\\ \frac{\mathrm{S}-1}{\sqrt{\mathrm{VAR}\left(\mathrm{S}\right)}} \mathrm{if} \mathrm{S}<0,\end{array}\right.$$

(7)

n is length of time, t_p is the tied values for the pth value, and q is the number of tied values, while X_i and X_j are the data values in chronological order. In the current study, the significance of the trend was tested at the Z-critical value > 1.96 with a significance level of 0.05. The null hypothesis of no trend must satisfy the condition if −1.96 > ZMK > 1.96.

2.4. Geographically Weighted Regression Model (GWR)

The GWR model was extensively used in this study to investigate the relationship between the climatic parameters and air pollutants at the national scale of China. GWR modeling is a type of regression modeling with geographically varying parameters (e.g., air pollutants), which differs from conventional regression modeling and was developed to achieve a higher performance in geographical analysis. In general, the GWR model can be expressed by the following Equations (8) and (9) [54,95]:

$$C_i={{\displaystyle \sum }}_k{\beta }_k\left(X_i{Y}_i\right)A{P}_{k,i}+{\epsilon }_i$$

(8)

where $C_i$ is the spatial dependent variable at the location 𝑖; $X_i{Y}_i$ is the x–y coordinate of the 𝑖_th location. ${\beta }_k\left(X_i{Y}_i\right)$ is the coefficient, a varying conditional at k_th, and the respective x-y location with the first coefficient by setting $A{P}_{0,i}$ = 1, that is ${\beta }_0\left(X_i{Y}_i\right)$ defined as the geographically varying intercept term. $A{P}_{k,i}$ is the independent variable at k_th and the respective x-y location. ${\epsilon }_i$ is the Gaussian error at the x-y coordinate of the 𝑖_th location:

$$C_i={\beta }_0\left(X_i,Y_i\right)+{\beta }_1\left(X_i,Y_i\right) N{O}_2+{\beta }_2\left(X_i,Y_i\right)\mathrm{PM}+{\beta }_3\left(X_i,Y_i\right)S{O}_2+{\beta }_4\left(X_i,Y_i\right)O_3+{\epsilon }_i$$

(9)

where $C_i$ stands for the dependent climatic factors (precipitation, temperature maximum, and temperature minimum). ${\beta }_0\left(X_i,Y_i\right)$ is the geographical intercept. ${\beta }_{1 \mathrm{to} 4}\left(X_i,Y_i\right)$ are the respective geographically varying coefficients, which are different from the coefficients of conventional regression. NO₂, PM, SO₂, and O₃ are the independent air pollutants at the x–y coordinate of the 𝑖_th location.

2.5. Impact Assessment of Air Pollution on Climate

The climatic impact of air pollution can be assessed using the following differential Equations (10) and (11):

$$\frac{\mathit{d}\mathit{C}\mathit{l}\mathit{i}{\mathit{m}}_{\mathit{i}}}{\mathit{d}\mathit{t}}=\frac{dCli{m}_i}{dPM}\times \frac{dPM}{dt}+\frac{dCli{m}_i}{dS{O}_2}\times \frac{dS{O}_2}{dt}+\frac{dCli{m}_i}{dN{O}_2}\times \frac{dN{O}_2}{dt}+\frac{dCli{m}_i}{d{O}_3}\times \frac{d{O}_3}{dt}$$

(10)

In the above Equation (10), $Cli{m}_i$ means different climatic parameters such as precipitation, temperature maximum, and temperature minimum. This equation is based on the chain rule of calculus, which states that the derivative of a composite function (in this case, “Clim_i” as a function of “PM, SO₂, NO₂, and O₃” and “PM, SO₂, NO₂, and O₃” as a function of time) is equal to the product of the derivatives of the individual functions. The equation suggests that the rate of change of a climate variable “Clim_i” can be influenced by changes in an air pollutants “PM, SO₂, NO₂, and O₃”. By understanding the relationship between these variables and their rates of change over time, we can gain insights into how changes in one variable may influence the other. This equation is a simple representation of the complex interactions between the atmosphere, land, and oceans that determine the Earth’s climate. It can be used to investigate the impacts of various human activities, such as emissions of air pollutants, on the climate system.

$$\Delta \mathit{C}\mathit{l}\mathit{i}{\mathit{m}}_{\mathit{i}}=a\times \Delta \mathrm{PM}+b\times \Delta S{O}_2+c\times \Delta N{O}_2+d\times \Delta O_3$$

(11)

Equation (11) explains in general that $\mathsf{\Delta }\mathrm{PM}, \mathsf{\Delta }S{O}_2, \mathsf{\Delta }N{O}_2, and \mathsf{\Delta }{O}_3$ are the inter-annual changes while a, b, c, and d mean coefficients.

3. Results

3.1. Trend Analysis of Air Pollution Variation under Pre- and Post-Scenarios

The current study presents the spatiotemporal characteristics of climatic variables and air pollutants from 2004 to 2021 using OMI, SEDAC, and ERA5 data. To analyze the spatial characteristics, we performed a trend analysis on the average annual concentrations of air pollutants from 2004 to 2021. We divided our study period into two scenarios based on APPCAP, namely pre- and post-scenarios, by accounting for the periods 2004–2013 and 2013–2021, respectively. We wanted to ensure the significance of the results of APPCAP in China. During the pre-scenario, it was observed that PM, SO₂, and NO₂ showed heterogeneous trends, whereas the trend for O₃ was homogenous across the study area. Figure 4 shows the most prominent increasing trend for NO₂, which was 1.21 × 10¹⁵ molecules/cm² per year. It is caused by anthropogenic emissions associated with economic growth, including traffic flow and urbanization [60].

In the case of PM and SO₂, the increasing trend was 4 µgm–3/year and 0.1 DU/year, respectively, during 2004–2013. In the pre-scenario group, the highest increasing trend for PM was recorded in northeast, east, and southeast China. For O₃, the trend over China was increasing, with the maximum value being 1 DU/year. On the other hand, the highest decreasing trend was recorded at 1.0 × 10¹⁵ molecules/cm² per year for NO₂ cross ponding to different hotspots in China.

The p values for PM and NO₂ were highly significant over most of the parts of the study region, as illustrated in Figure 5, whereas a mixed trend of p-value variation was observed for SO₂ and O₃.After adopting the APPCAP (post-scenario), it was observed that PM and NO₂ showed a significantly decreasing trend in most of the part of China, whereas the trend for O₃ was still increasing in China.

Figure 6 shows the most prominent decreasing trend, as the trend for NO₂ was 1.21 × 10¹⁵ molecules/cm² per year. In the northwest and north China, the trend was mixed for NO₂, while the northeast China PM trend increased after adopting the APPCAP. In the case of O₃, the trend decreased in the north part, while the rest of the area showed an increasing trend. The decreasing trend of O₃ was 2.0 DU/year during 2013–2021. In the post scenario, the trend for SO₂ was random, and a decreasing trend was found in central China, including the Shandong and Beijing areas. The highest decreasing trend for SO₂ was 0.10 DU/year, while the decreasing trend was recorded for PM in northeast, east, and southeast parts of China, with a peak value of 6 µgm⁻³/year. In contrast, O₃ showed an increasing trend with a maximum value of 1.2 DU/year. The highest increasing trends recorded were 0.05 DU/year and 1.21 × 10¹⁵ molecules/cm² per year for SO₂ and NO₂ in China. From Figure 7, the p values showed that the decreasing trend for PM and NO₂ was highly significant in the study region, whereas the increment in O₃ pollution was highly significant during the post-scenario period.

3.2. Trend Analysis of Climatic Factors Variation under Pre and Post Scenarios

Figure 8 shows the spatial variation in trends for climatic parameters, including t_max, t_min, and precipitation. In general, all parameters showed an increasing trend across the study region. During the pre-scenario, t_max and t_min showed a prominent increasing trend, while precipitation showed a less significant increasing trend. t_max showed the maximum increasing trend of 0.30 °C/year, while t_min represented 0.38 °C/year. The precipitation trend also increased, but it was prominent in the north and southeast regions of China, where it increased by value 253 mm/year. However, a prominent decreasing trend was found along the Shandong region to the south part. The precipitation trend decreased by 72 mm/year. In the case of t_max, the trend did not significantly increase in the northern region, whereas t_min showed a decreasing trend in the same region of China. The maximum decreasing trend for t_max and t_min was 0.35 °C/year and 0.42 °C/year, respectively. The trend variation in t_max, t_min, and precipitation was highly significant at p < 0.05 across the study area, as shown in Figure 9. After adopting the APPCAP, the trend variation was in contrast to the pre-scenario trend analysis results. Precipitation and t_max showed increasing trends with maximum trend values of 246 mm/year and 0.54 °C/year, respectively. The t_min trend seemed to slightly increase by 0.61 °C/year, especially in the north and southwest parts of China. In contrast, the precipitation trend decreased by 135 mm/year, particularly in the southeast part of China. Furthermore, t_max and t_min showed a decreasing trend over 2013–2021 with a magnitude of 0.26 °C/year and 0.37 °C/year, respectively.

3.3. ITA Analysis for Air Pollution Variation under Pre and Post Scenarios

The annual variations in air pollutants were studied to analyze the impact of APPCAP on air pollution reduction. The ITA analysis showed an increasing or decreasing trend over two decades, during pre-scenario and post-scenario periods. A combination of increasing and decreasing trends for data points above the 5% range from the 1:1 line exhibited a significant trend. The results showed that, overall, NO₂, PM, and SO₂ decreased after adopting APPCAP across China. Figure 10 suggests that air pollutants (NO₂, PM, SO₂, and O₃) significantly decreased. The results also showed that NO₂ data points fell within the −5% range, showing a decreasing trend during 2004 and 2021. The statistics for NO₂ exhibited significant negative values of ITA D. The O₃ data points were mostly on the diagonal line, indicating no trend/zero trends. The SO₂ data points fell above −5%, indicating that it decreased after adopting APPCAP. However, it was concluded that the decreasing trend of SO₂ was not significant over the period.

3.4. Impact Analysis of Air Pollutants on Climatic Factors before and after APPCAP

The impact analysis of air pollutants was investigated using the GWR model and differential Equations (8)–(11). We performed a spatial distribution analysis on the outputs from the differential equations to investigate the climatic impact of air pollution over two scenarios based on APPCAP in China.

3.4.1. Air Pollutant’s Contribution to Temperature Maximum Patterns

The spatial characteristics of the contribution of air pollutants to t_max patterns for the pre-scenario and post-scenario periods are represented in Figure 11. Higher positive values show a greater influence on increasing t_max over the study period. For the pre-scenario period, NO₂ had a higher value and increased t_max by 2.42 °C, while O₃ had a lower positive value of 1.10 °C over a decade. In the case of SO₂, the contribution pattern was random. In contrast, SO₂ had higher negative values and decreased t_max by 2.07 °C, while PM showed the lowest negative contribution with 0.55 °C over China. PM caused an increase in t_max in central east and southeast parts of China, as shown in Figure 11a. After adopting the APPCAP (post-scenario), PM showed the highest positive contribution to increasing t_max by 1.80 °C, while O₃ contributed the lowest positive value of 0.60 °C. SO₂ showed random spatial patterns of increasing or decreasing t_max over the whole study area. NO₂ contributed more to decreasing t_max by 2.27 °C, and spatial patterns showed that central east China was reducing t_max, as shown in Figure 11h.

O₃ reduced t_max in the southern part of China. The contribution pattern of O₃ indicated the lowest negative contribution to decreasing t_max by 0.47 °C. In general, a comparison of all air pollutant’s contributions indicated that, during the pre-scenario period, t_max positively changed by 2.42 °C, followed by NO₂, while in the post-scenario period, t_max positively changed by 1.80 °C, followed by PM. In contrast, the highest negative t_max change was by 2.07 °C, followed by SO₂, and 2.27 °C followed by NO₂, respectively. NO₂ and PM particles can contribute to global warming by trapping the heat in the atmosphere. They cause the warming effect on the atmosphere by absorbing the sunlight and re-emitting it as heat, leading to a rise in the temperature. The current effect is more in urbanized areas, where there are high traffic and industrial emissions. Further, NO₂ may also react with chemicals in the atmosphere to form secondary pollutants such as O₃, which can increase warming. Moreover, O₃ leads to an increase in temperature. SO₂ is primarily released from the burning process and volcanic eruption. It has harmful effects on human health as well as the environment. However, SO₂ emissions contribute to warming the atmosphere because it can react with other chemicals in the atmosphere to form sulfate aerosols and lead to an increase in temperature. This effect is more prominent in areas with high concentrations of SO₂, such as power and industrial plants. Overall, the pollutants significantly impact the temperature maximum, particularly in areas with high concentrations. The effect of the air pollutants on t_max depends on the concentration level, size of the particles, as well as other factors such as location and weather conditions.

3.4.2. Air Pollutant Contribution to Temperature Minimum Patterns

In this section, the spatial characteristics of the contribution to t_min changes for the pre-scenario and post-scenario periods are represented in Figure 12, which shows the spatially interpolated geographical air pollution contribution to climatic factors (t_min). The northern, western, and southeastern parts of China indicated a position association with air pollutants during the pre-scenario period (2004–2013). Figure 12a,d shows the higher positive influence on increasing t_min by 2.19 °C, followed by SO₂, while NO₂ had the second highest contribution to change in t_min by 1.96 °C during the pre-scenario period. The northeastern and southern parts indicated the rise in t_min due to the variation in PM. NO₂ caused higher t_min changes over the north, central, and southeast parts of China. In contrast, NO₂ contributed to a higher negative contribution to decrease t_min by 2.24 °C, while PM had a lower negative contribution with a value of 0.43 °C over a decade. Figure 12d indicates that NO₂ was mainly responsible for decreasing t_min mainly in the central part of China. O₃ and SO₂ had mixed patterns for changing t_min (Figure 12b,c).

In the post-scenario period, the west, north, and central parts of China showed an increasing contribution to t_min changes. NO₂ showed the highest positive contribution to increasing t_min by 2.24 °C, while O₃ had the lowest positive contribution with 0.60 °C. The second highest positive contribution to increasing t_min was with PM (1.69 °C). SO₂ and O₃ showed random spatial patterns to increase or decrease t_min over the whole study area. NO₂ had a positive contribution in central east China, as shown in Figure 12h. In contrast, NO₂ had the highest negative contribution for reducing t_min by 1.53 °C in the northeast and southeast parts of China.

Moreover, O₃ contribution patterns indicated the lowest negative contribution to decreasing the t_min by 0.47 °C. O₃ reduced t_min in the east-southeast part of China during the post-scenario period. Overall, a comparison of all air pollutants contributions indicated that, during the pre-scenario and post-scenario periods, the t_min changed the highest positively by 2.19 °C followed by the SO₂, while it was 2.44 °C followed by NO₂ during the pre-scenario period. In contrast, t_min has the highest negative change by 2.24 °C, while it was 1.53 °C followed by NO₂ in the case of the pre- and post-scenario periods, respectively. NO₂ has direct and indirect impacts on temperature minimum. It can have a warming effect. NO₂ absorbs sunlight and heat in the atmosphere that can lead to an increase in temperature, especially at night. It is important to note that impacts of SO₂ on t_min depend on the different factors such as season and altitude. The sulfate aerosols reflect solar radiation and reduce the energy on the Earth’s surface. This phenomena causes a cooling effect on temperature, including t_min. Moreover, NO₂ causes the urban heat island effect, where cities experience higher temperatures than surrounding rural areas due to human activity and the built environment. This effect can further increase the minimum temperature in urban areas, leading to a range of health and environment impacts.

This warming effect can have several consequences, including the melting of snow and ice, changes in precipitation patterns, and alternations in plant growth and ecosystem functioning.

3.4.3. Air Pollutant Contribution to Precipitation Patterns

Figure 13 shows spatial maps of the contribution of air pollutants in changing precipitation patterns over China in the pre- and post-scenario periods. Figure 13b,c shows the highest positive influence on precipitation, with an increase of 40 mm, followed by SO₂, while O₃ had the second highest contribution to changes in precipitation, with an increase of 39 mm during the pre-scenario period. The central-eastern part indicates the rise in precipitation due to variations in SO₂, while O₃ caused higher precipitation changes over the Southern part of China. In contrast, SO₂ had a negative contribution to decreasing precipitation by 22 mm, while NO₂ had lowest negative contribution, with a value of 0.01 mm over a decade. Figure 13b,d indicates the lower precipitation hotspots in the south of China caused by SO₂, while NO₂ showed hotspots in the east of China, respectively.

In the post-scenario period, the south, east, and central parts of China showed increasing contributions to precipitation alteration. PM made the highest positive contribution to increasing the precipitation by 27 mm, while NO₂ had the lowest positive contribution, with 0.007 mm. The second-highest positive contribution to raising precipitation was with SO₂ (24 mm). The PM contribution was positive in the central part of China, as shown in Figure 13e. In contrast, SO₂ had the highest negative contribution to reducing the precipitation magnitude by 19 mm in the central east (Shandong) part of China. Moreover, NO₂ contribution patterns indicated the lowest negative contribution to reducing precipitation by 0.01 mm. NO₂ reduced the precipitation in the north part of China during the post-scenario (2013–2021). Overall, a comparison of all air pollutant contributions indicated that during the pre-scenario period, the highest positive precipitation change was by 40 mm followed by SO₂, while it was 27 mm followed by PM in the case of the post-scenario period. In contrast, the highest negative precipitation change was by 22 mm, while it was 19 mm followed by SO₂ during the pre- and post-scenario periods, respectively. SO₂ and PM had significant impacts on the precipitation pattern and environment. SO₂ reacts with other chemicals to produce sulfate aerosols that may affect precipitation in different ways. SO₂ affects the precipitation pattern by altering the formation of clouds. Sulfate aerosols act as cloud condensation nuclei, which are important for the formation of clouds. PM particles promote the formation of clouds and change their properties through cloud condensation nuclei. However, these particles may affect the properties of clouds, including droplet size, their reflectivity, and the amount of precipitation they produce. As a result, the timing of precipitation as well as the pattern is affected. Additionally, SO₂ may impact precipitation patterns through its interactions with other atmospheric chemicals.

As an example, SO₂ reacts with nitrogen oxides to create fine PM particles that significantly affect precipitation patterns, in particular, the areas with high concentrations of both SO₂ and NO_x. PM emissions from urban and industrial areas can lead to forming urban heat islands, which alter local wind patterns and precipitation by changing the moisture content and temperature of the air.

4. Discussion

The State Council issued an action plan, APPCAP, to combat air pollution in 2013, which became the most ambitious policy to control air pollution concentrations [30,72]. APPCAP aimed to improve the following aspects: (1) control air pollution, (2) control pollution substances, (3) diversify air pollution control strategies, and (4) optimize pollution management mode [71,96,97]. The current study divided the entire study period into two separate scenarios to better comprehend the promising efforts of China: the pre-scenario (before APPCAP) and post-scenario (after APPCAP) periods. The APPCAP is a significant development in China’s efforts to curb air pollution, and motivates researchers to assess the implications of air pollution’s management at the national level. MK trend analysis was performed before and after APPCAP implementation on mean annual data. ITA is an intuitive method used to identify trends in different time series subcategories and can be applied irrespective of distribution assumptions [85,86,87], while the MK test assumes that time series have no sequence correlation and are least-based and all data are significant [88]. During the pre-scenario period, trend analysis results on PM, SO₂, and NO₂ showed heterogeneous trends, whereas the trend for O₃ was homogenous. The most prominent increase in NO₂ concentration was 1.21 × 10¹⁵ molecules/cm² per year [64,98,99]. In the case of PM and SO₂, the increasing trend was 4 µgm⁻³/year and 0.1 DU/year [88]. Northeastern, eastern, and southeastern China showed the highest increasing trend for PM in the pre-scenario period. For O₃, the trend increased across the region with a maximum value of 1 DU/year in the pre-scenario period.

In light of stringent clean air pollution actions, including APPCAP, adopted in China since 2013 (post-scenario), data analysis showed that PM and NO₂ had significantly decreasing trends in most of China, while the trend for O₃ did not reduce considerably and is still escalating [100,101]. In the post-scenario period, Beijing, Shandong, and other areas of central China showed a random and declining tendency for SO₂ [102], with the greatest rate of decline for SO₂ being 0.10 DU/year.

The spatial variation of the trend for the climatic parameters, including t_max, t_min, and precipitation, increased across the study region. During the pre-scenario period, t_max showed a decreasing mean annual trend of 0.025 °C/year, and t_min represented 0.02 °C/year, respectively, while precipitation had an increasing mean annual trend by a value of 90 mm/year. After adopting the APPCAP, trend variation was in contrast to the pre-scenario period for temperature trend analysis results. During the post scenario period, precipitation, t_max, and t_min showed increasing mean annual trends by 55 mm/year, 0.14 °C/year, and 0.12 °C/year, respectively, and were highly significant at (p < 0.05). Overall, t_max and t_min increased while the precipitation reduced.

A comparison of the effects of air pollution on climate between the pre- and post-scenario periods was analyzed for a two-decade period using ITA to determine whether there was an increasing or decreasing trend. The results showed that, when China adopted the APPCAP, overall concentrations of NO₂, SO₂, and PM declined. Figure 10 indicates that the levels of airborne pollutants (NO₂, SO₂, and PM) dramatically decreased. NO₂ data values above the −5% range showed a declining trend between 2013 and 2021, and NO₂ statistics exhibited significant negative values for ITA. However, despite the massive reduction in NOx emissions from power plants, vehicle emissions continued to increase. Both on- and off-road vehicles significantly contribute to air pollution, and multiple studies have revealed a strong correlation between rising vehicle traffic and air pollution in Chinese megacities [103,104,105,106]. Since the 1990s, the central government has implemented integrated emission control policies to reduce vehicle emissions, including emission standards for new vehicles, emission controls for existing vehicles, modifications in fuel quality, the promotion of advanced vehicles, sustainable transportation, and traffic management programs [107].

O₃ data points mostly fall on the (1:1) line, indicating no trend/zero trends. According to Liu [108], the sources of the ozone and its variation in concentrations in China, substantial declines in PM, and SO₂ emissions lead to a rise in urban O₃ due to the dynamic interaction between aerosols, radiation, and chemical reactions. O₃ concentrations also increase in the southwest due to climatic factors [109]. Additionally, several studies found that reducing the concentration of NOx increases the amount of O₃. During the COVID-19 lockdown, the transportation sector shutdown resulted in a reduction in NOx, which decreased the neutralization of NO into O₃, which raised the concentration of O₃ [110,111].

The SO₂ data points were above −5%, indicating that it started to decline after the APPCAP was implemented. Additionally, it was determined that the decreasing tendency of SO₂ over the period was significant and it was in line with findings of [97]. China’s environmental regulations resulted in significant reductions in SO₂ emissions, but overall, coal consumption has continued to rise. Since 1996, reducing SO₂ emissions from major industries, including coal-fired thermal power and iron/steel plants, has been top priority. The installation and strict management of online monitoring equipment in the main industries have become mandatory, and different policies and programs have been implemented during previous Five-Year Plans [107,112].

Our study has certain limitations. We examined the long-term changes in air pollutants and climatic factors, as well as their association based on different statistical models. The APPCAP adaptation resulted in an increase in t_min and precipitation. The rise in t_min may result in local warming in China for the most part, which requires more efforts to combat climatic change. To this end, there is a need to focus on the mechanisms underlying these regional climate changes, based on seasonal variation. The climatic effect of air pollution along seasonal variation remains unexplored, and coupling regional climate models to analyze the radiative effects of China’s emission reductions.

Climatic Impact of Air Pollutants for Pre and Post Scenarios

The current study describes the climatic changes due to variations in air pollutants in response to the adoption of the APPCAP. In the pre-scenario period, most of the air pollutants showed positive contributions to changing the climate. For example, O₃ played a role in reducing t_max (t_min) with a value of 0.15 °C (0.05 °C) and increasing precipitation with a value of 11mm, as shown in

Table 2. An average value of climatic variation in response to air pollutant changes during (pre- and post-) scenarios in the study area.

APPCAP	Parameter	Tmax °C	Tmin °C	Precp mm
Pre-Scenario	NO2	0.62	−0.14	0.0
	PM	0.47	0.41	0.31
	SO2	−0.06	0.41	9
	O3	−0.15	−0.05	11
Post-Scenario	NO2	−0.33	0.45	0.001
	PM	0.38	0.31	7.38
	SO2	−0.09	0.02	2.27
	O3	0.06	0.18	4.06

. PM showed a positive relationship with all climatic parameters with values of 0.47 °C (t_max), 0.41 °C (t_min), and 0.31 mm (precipitation). The analysis showed that NO₂ had positive and negative relationships with t_max (precipitation) 0.62 °C (0.1mm) and t_min (0.14 °C), respectively. SO₂ showed an increasing contribution with t_min (precipitation) 0.41 °C (9mm), while decreasing with t_max (0.06 °C).

During 2013–2021 (post scenario), most of the air pollutants (PM, SO₂) showed positive contributions to alter the climatic pattern including t_max, t_min, and precipitation. The analysis showed that PM had a positive relationship with t_max (t_min) 0.38 °C (0.31 °C), and precipitation (7.38 mm). NO₂ showed an increasing contribution with t_min (precipitation) 0.35 °C (0.001 mm), while decreasing with t_max (0.33 °C). Moreover, SO₂ indicated an increasing contribution to t_min (precipitation) 0.02 °C (2.2 mm), while decreasing with t_max (0.09 °C).

5. Conclusions

In summary, when China adopted the APPCAP, NO₂, SO₂, and PM concentrations declined dramatically. According to ITA, NO₂ data values above the –5% range showed significant declining trend between 2013 and 2021. Although O₃ concentrations have gradually increased, the air quality in most cities has immensely improved as a result of the stringent control measures by the State Council. The O₃ data points are mostly on the (1:1) line, which means that there is no trend/zero trends. The decreasing tendency of SO₂ over the period is not significant. Despite the tremendous progress made in reducing air pollution, PM and O₃ concentrations continue to be considerably above WHO guidelines. More efforts are required to focus on the research and development of new methods to address these ongoing challenges in air pollution emissions. The MK test investigated that all pollutants showed an increasing trend for pre-APPCAP, while showing a decreasing trend, except for O₃, during the post-APPCAP period. For climatic factors, the MK test showed an increasing trend of precipitation and t_min for the post-APPCAP period.

We used GWR and differential equation models to identify the climatic impact of air pollution before and after the implementation of APPCAP in 2013 to combat air pollution. Two scenarios based on APPCAP were made to account for the air pollution variation in China during 2004–2021. The contributions of the most influential air pollutants were calculated using the GWR and differential equation models. In the pre-scenario period, NO₂ contributed to increasing the mean maximum air temperature (t_max) by 0.62 °C, PM contributed to raising t_min by 0.41 °C, while O₃ reduced t_max(t_min) by 0.15 °C (0.05 °C). PM increased t_max and precipitation with magnitudes of 0.38 °C and 7.38mm, respectively, while NO₂ contributed to increasing t_min by 0.35 °C during the post-scenario period. In particular, the post-scenario period led to increases in t_min and precipitation across the country. The current study findings explain the significance of APPCAP and its role in changing the regional climate of China. Policymakers must implement measures to reduce emissions and maintain the climate warming level across the region. In the future, China needs to introduce more renewable energy to dominate the coal energy system, which may help with national and international climate mitigation measures.