Analysis and Characterization of the Behavior of Air Pollutants and Their Relationship with Climate Variability in the Main Industrial Zones of Hidalgo State, México

Abstract

The concentration of air pollutants could be affected by climate change in industrial park zones in Hidalgo state, Mexico (IPHSs). The goals of this work were: (a) to describe the aerosols’ behavior (PM₁₀ and PM_2.5) and air pollutants (SO₂, NO₂, O₃, and CO) in the IPHSs and (b) determine the climate variable behavior regarding the presence in IPHSs. The methodology consisted of structuring the time series of climate variables and air pollutants in six analysis regions. Afterwards, an annual average calculation, a count of days exceeding the allowed limits set by the official Mexican norms, an analysis of annual behavior by season, the Sen slope calculation, and correlation among variables were performed. Results demonstrated that Zone 2 is the most polluted, exceeding the allowed limits in the annual average (PM₁₀ > 36 μg/m³, PM_2.5 > 10 μg/m³, and NO₂ > 0.021 ppm) and having more than 1000, 96, and 11 days where the daily limit was exceeded in PM₁₀, PM_2.5, and SO₂, respectively. The minimum concentrations of the pollutants were observed during the summer–autumn seasons, coinciding with the highest precipitation. Regarding the correlations, the pollutants are negatively and statistically significantly correlated with precipitation (ranging from −0.81 to −0.43); meanwhile, the maximum temperature (ranging from +0.41 to +0.51) and evaporation (ranging from +0.39 to +0.54) are positively and statistically significantly correlated. In conclusion, the results could suggest that the presence of pollutants in the atmosphere may be influenced by the behavior of nearby regional climatic conditions in the IPHSs.

1. Introduction

The impacts of human activities on a region’s climate are becoming increasingly evident. In this context, some organizations, such as the United Nations Framework Convention on Climate Change (UNFCCC), develop international policies for climate change research. In the Paris Agreement (2015), the relevance of keeping the world’s average temperature increase below 2° relative to the pre-industrial levels, as well as the search for low anthropogenic emissions development to reduce the risks and effects of climate change, was established [1]. Moreover, this year, the Sustainable Development Goals were established, incorporating Goal 13, known as Climate Action. This goal stipulates that the emissions contributing to climate change must be reduced by almost half by 2030 [2].

According to the Intergovernmental Panel on Climate Change (IPCC), an anthropogenic emission is composed of greenhouse gases (GHGs), atmospheric pollutants (APs), and aerosols (AEs) generated by human activities, such as fossil fuel burning, deforestation, changes in land use, livestock production, fertilization, waste management, and industrial processes, which generate considerable pollution in the atmosphere [3]. GHGs, APs, and AEs absorb and emit radiation at specific spectral wavelengths emitted by the Earth’s surface, the atmosphere itself, and the clouds. Primary GHGs and APs include water vapor (H₂O), carbon dioxide (CO₂), methane (CH₄), and ozone (O₃), and nitrogen dioxide (NO₂). On the other hand, AEs are particulate matter (PM), such as sea salt, organic carbon, black carbon, mineral species (mainly dust), sulfate, nitrate, and ammonium [3,4,5].

Worldwide, the effects caused by GHGs, APs, and AEs in the atmosphere have been studied because they are considered air pollutants. These effects are mostly related to the health sector [6,7,8,9,10,11,12,13], the environment [14,15,16] and the climate [17,18,19,20,21,22]. Recent changes in local and regional climates have been identified as potentially degrading air quality through altered wind patterns (ventilation and dispersion), temperature, and precipitation [17,18,20]. Moreover, the occurrence of extreme events, potentiated by regional climate change, such as more frequent and prolonged drought and increased temperatures, can directly affect air pollutant concentrations [23]. Similarly, the combination of changing weather patterns and altered atmospheric chemistry can promote the formation and accumulation of atmospheric air pollutants, resulting in a phenomenon known as the climate penalty [24,25,26].

The World Health Organization (WHO) establishes the permissible levels for each GHG, AP, and AE across several time scales to ensure the health of the population is not endangered [27]. In particular, in Mexico, there are various official standards (Norma Oficial Mexicana, NOM) related to the monitoring of these pollutants, that set permitted emissions levels in the country. These NOMs are NOM-025-SSA1-2021, NOM-020-SSA1-2021, NOM-022-SSA1-2019, NOM-023-SSA1-2021, and NOM-021-SSA1-2021, used for monitoring the PM₁₀, PM_2.5, O₃, SO₂, NO₂ and CO, respectively [28,29,30,31,32].

On the other hand, in Mexico, there is the Sistema Nacional de Información de la Calidad del Aire (SINAICA), which belongs to the Instituto Nacional de Ecología y Cambio Climático (INECC). SINAICA is a series of computational programs for recording, transmitting, and publishing atmospheric conditions reported by monitoring stations in several states that have adequate infrastructure for these measurements [33]. The pollutants monitored are particulate matter less than or equal to 10 μg (PM₁₀) and less than or equal to 2.5 μg (PM_2.5), sulfur dioxide (SO₂), NO₂, O₃ and carbon monoxide (CO). Moreover, the Servicio Meteorológico Nacional (SMN) is the government institution responsible for monitoring weather conditions nationwide. SMN operates a meteorological monitoring network of stations across the country, reporting daily on temperature, precipitation, and evaporation, among other parameters [34].

From this, the IPCC has developed several future scenarios related to human activities, considering economic, population, and GHG emissions growth, called the Shared Socioeconomic Pathways (SSPs). In the five possible scenarios (SSP1–SSP5), a worldwide increase in temperature of 1.5 °C (SSP1, zero greenhouse gas concentrations scenario) to 5 °C (SSP5, very high emissions scenario and fossil fuel-based development) is expected. In addition, rainfall patterns are also expected to be modified, with increases and decreases of up to 40% concerning their annual average. In Mexico, temperatures will rise from 2 °C to 5.5 °C across all scenarios. Precipitation is expected to decrease by up to 20% of its annual average [4].

Hidalgo state is one of Mexico’s 32 states, located in the central-eastern part of the country. The industry is the state’s primary activity, accounting for approximately 24% of the state’s Gross Domestic Product and concentrated in the southern part of the state [35,36,37]. There are approximately 12 industrial parks in Hidalgo state (IPHSs) connected to the country’s main economic sectors, located in the states of Mexico City, Queretaro, Mexico state, Puebla, and Tlaxcala. Regarding its public administration, the state seeks to develop more industrial parks in its territory [38]. Although several studies examine atmospheric pollutants in this state [35,39,40,41], none examine the behavior of climate variables in the presence of these pollutants in such locations.

Consequently, it is necessary to study climate change behavior and its relationship to air pollutant levels by analyzing spatiotemporal trends in pollutants and meteorological variables within the primary IPHS. Therefore, the goals of this research are: (1) to describe the AE (PM₁₀ and PM_2.5), AP (NO₂) and GHG (SO₂, O₃, and CO) behavior in the main IPHS and (2) to determine the behavior of the temperature, precipitation, and evaporation variables regarding the presence of the pollutants above.

2. Materials and Methods

2.1. Study Area

The study area is located between the geographic coordinates 19.8° and 20.5° N and 98.2° and 99.8° W, corresponding to the south, southeast, and southwest regions of the state of Hidalgo, Mexico (see Figure 1). This area was selected because it concentrates the 12 IPHSs grouped in six small zones. Furthermore, the study region’s topography ranges from 2000 to 2800 m above sea level (masl), providing a range of climatic conditions. Regarding its geographic location, the study area is delimited by the states of Mexico City, Queretaro, Mexico state, Puebla, and Tlaxcala, which represent one of the most relevant regions of the country due to their high economic value [38,42].

Figure 1. Study area. Colors indicate the six small regions where the 12 IPHSs are distributed. Green circles represent the location of the SINAICA monitoring stations [33]. Black triangles indicate the location of the SMN meteorological stations [34].The color red represents the state of Hidalgo, Mexico.

2.2. Data

2.2.1. Air Pollutant Information

The monitoring information of the six air pollutants (PM₁₀, PM_2.5, SO₂, NO₂, O₃, and CO) was provided by nine distributed stations in the south of Hidalgo state, Mexico (see Table A1). These stations belong to the INECC’s monitoring network of air quality (see Figure 1). The analysis includes the years from 2016 to 2023. The information used for PM₁₀, PM_2.5, and SO₂ was the 24 h average; meanwhile, the one-hour maximum daily was used for NO₂ and O₃. Finally, the maximum daily eight-hour moving averages were employed for CO [33]. These criteria were selected due to time-scale constraints of the time series and the limits stipulated by the different NOMs for monitoring each pollutant. Table 1 presents the time scale description and the allowed limits set by the NOMs. Moreover, pollutant stations with at least 80% of recorded data were considered.

Table 1. Description of the different time series scales and the allowed limits for the six pollutants regarding several NOMs.

Pollutant	NOM	Stipulated Limit
PM10	NOM-025-SSA1-2021	24 h: 70 μg/m3 Annual: 36 μg/m3
PM2.5		24 h: 41 μg/m3 Annual: 10 μg/m3
SO2	NOM-022-SSA1-2019	1 h: 0.075 ppm 24 h: 0.04 ppm
NO2	NOM-023-SSA1-2021	1 h: 0.106 ppm Annual: 0.021 ppm
O3	NOM-020-SSA1-2021	1 h: 0.090 ppm 8 h: 0.065 ppm
CO	NOM-021-SSA1-2021	1 h: 26 ppm 8 h: 9 ppm

2.2.2. Climate Information

Monthly values of maximum, average, and minimum temperatures, and evaporation from the SMN monitoring network were employed [34]. This decision was made because these variables are considered as reference by the IPCC to determine the presence of pollutants in the atmosphere [3,4]. The period selected for this study spans from 1990 to 2023, as a minimum of 30 years is required for describing a place’s climate [43]. Moreover, meteorological stations with the following conditions were considered:

(a)

The distance between the meteorological station and the pollutant monitoring station must be less than 10 km.

(b)

The meteorological station supplied at least 80% of the data for the monthly averages of the selected variables. The World Meteorological Organization (WMO) suggests this percentage value [43].

(c)

The variables’ time series must pass a homogeneous test and quality control [44].

A total of 16 meteorological stations with the requirements mentioned above were obtained. The distribution of each station is illustrated in Figure 1. For more details about the stations, see Table A2.

2.3. Zone Distribution and Structuration

Six small regions were structured in the study area. At least one meteorological station and one pollutant monitoring station must be located in a region to be considered. Moreover, the zones must include at least one IPHS or be strategic and economically relevant regions for the Hidalgo state. In Table 2, the zone distribution, the contamination monitoring stations, and the meteorological stations are described.

Table 2. Description of the integration of the six analysis zones regarding pollutant and meteorological stations.

Zone	Pollutant Station	Meteorological Station
1	Hospital, Huichapan (HUI)	Tlaxcalilla Presa Madero
2	Universidad Tecnológica de Tula-Tepeji, Tula (TUL) Primaria Melchor Ocampo, Tepeji (TPJ) Centro de Salud, Atitalaquia (ATI) Primaria Revolución, Atotonilco (ATO)	Ajacuba (DGE) El Banco Presa Endho Presa Requena
3	Biblioteca, Tizayuca (TIZ)	El manantial
4	Museo del Rehilete, Pachuca (PAC)	Pachuca (OBS) El Cerezo Presa El Girón Real Del Monte
5	Estación de Bomberos de Cd. Sahagún, Tepeapulco (TEP)	San Jerónimo Emiliano Zapata
6	Tulancingo, Tulancingo (TLN)	Santiago Tulantepec Presa La Esperanza Tulancingo (OBS)

2.4. Time Series Configuration

Once the pollutant and meteorological stations were identified in each analysis zone, the time series was structured. For the pollutant stations, daily data were obtained, so monthly averages were calculated for each pollutant. In regions with multiple stations, the average of all stations is calculated to create the time series representing the zone. Moreover, the pollutants were grouped by the season in which they were recorded. This grouping was performed because the data showed a clear trend. For example, the summer months have higher precipitation and greater temperature increases, while the winter months have lower precipitation and greater temperature decreases [45]. Therefore, the monthly average for spring comprises March, April, and May. For summer, June, July, and August were considered. For the autumn case, the months of September, October, and November were used. Finally, for winter, December, January, and February were employed. At the meteorological stations, the data were already available daily, and the same structure for the pollutants data was used.

A normality test was performed on the time series to analyze its behavior. The test results indicate that the time series does not follow a normal distribution at the 5% significance level. The Shapiro–Wilk test, conducted using SigmaStat version 3.5, was used to assess normality.

2.5. Analysis of the Pollutant and Climate Variables

To describe the behavior of the pollutants across multiple time scales, analyses were performed using annual and aggregated data by climate station, as well as the number of days exceeding the NOM-allowed limits [28,29,30,31,32]. On the other hand, to describe the behavior of climate variables, climographs for each region were constructed according to the WMO [43].

2.6. Trend and Correlation Analysis

For the trend analysis, the nonparametric Mann–Kendall test was applied to all time series, without assuming a data distribution. This test compares segment pairs from both time series using the S statistic, calculated by Equation (1) [46], where T_i and T_j are the two time series, and n is the number of time series in the sample.

$$\begin{array}{c}S=\sum _{i=0}^n\sum _{j=0}^{i-1}sign\left(T_i-T_j\right)\\ sign\left(D\right)=\left\{\begin{array}{c}1, D>0\\ 0,D=0\\ -1, D<0\end{array}\right.\end{array},$$

(1)

Positive values of S indicate an upward trend in the time series. On the contrary, negative values indicate a trend toward low values. Meanwhile, values equal to zero indicate a lack of trend.

For n > 10, the Mann–Kendall test is calculated through an approximation using the Gaussian statistic Z [47]. Equation (2) shows the statistic calculation, where δ² is the S statistic variance.

$$Z=\left\{\begin{array}{c}{\displaystyle \frac{S-1}{\sqrt{{\delta }^2\left(S\right)}}}, S>0\\ \hspace{1em}\hspace{1em} 0, S=0\\ {\displaystyle \frac{S+1}{\sqrt{{\delta }^2\left(S\right)}}}, S<0\end{array}\right.$$

(2)

The value of δ² depends on the presence of tied groups. A tied group is one in which all S statistic values are equal. Equation (3) is used to calculate δ² when there are no ties for T_i. On the contrary, Equation (4) is employed, where τr is the frequency of rank r.

$${\delta }^2\left(S\right)={\displaystyle \frac{n\left(n-1\right)\left(2n+5\right)}{18}}$$

(3)

$${\delta }^2\left(S\right)={\displaystyle \frac{n\left(n-1\right)\left(2n+1\right)-\sum {\tau }_r({\tau }_r-1)(2{\tau }_r+5)}{18}},$$

(4)

The null hypothesis of this test is that there is no trend in the data, while the alternative hypothesis is that there is a monotonic trend.

Subsequently, the Sen Slope method is employed to quantify the trend magnitude. This method is used to identify trends in hydrometeorological or non-normal distribution data and is calculated as the change in measurements over time [48]. Equation (5) illustrates the slope calculation, where Δ is the y_i and y_j point slope in the time segment delimited by t_i and t_j, respectively.

$$\mathsf{\Delta }={\displaystyle \frac{y_i-y_j}{t_i-t_j}}$$

(5)

Finally, the Sen Slope estimation is determined by Equation (6), which is the median of the obtained slopes, where m is the number of computed slopes.

$$\beta =f\left(x\right)=\left\{\begin{array}{c}\hspace{1em}\hspace{1em}\hspace{1em}{∆}_{{\displaystyle \frac{m+1}{2}}}, m is odd\\ {\displaystyle \frac{1}{2}}\left({\mathsf{\Delta }}_{{\displaystyle \frac{m}{2}}}+{\mathsf{\Delta }}_{{\displaystyle \frac{m+2}{2}}}\right), m is even,\end{array}\right.$$

(6)

Positive values of β indicate a trend increase; on the contrary, negative values indicate a trend decrease in the time series [48].

In this analysis, the Python library pyMannKendal (Version v1.4.3) was used to compute the Mann–Kendall statistic and the Sen Slope [49]. These tests were performed at a 95% confidence level.

The correlation analysis was performed using Spearman’s rank correlation coefficient at the 95% confidence level to evaluate the relationship between the pollutant and climate variables, given non-normal data distributions. This coefficient was computed using SigmaStat version 3.5. Finally, the period employed for the trend and correlation analysis was from 2016 to 2023, as this timeframe coincided with all the time series.

3. Results

3.1. Pollutant Behavior

Figure 2 shows the annual average concentrations of pollutants in several regions. For PM₁₀ (Figure 2a), Zones 2 and 3 show the highest values across all time series, with levels above 60 μg/m³ in 2022 and 2023, respectively. Zone 5 showed a significant increase in 2020, recording the highest PM₁₀ levels (54 μg/m³). On the other hand, the lowest average concentrations in 2022 were in Zones 1 and 4, while Zone 6 showed an upward trend between 2017 and 2022. It is important to mention that Zones 2, 3, and 5 recorded values exceeding the allowed limit stipulated by the NOM, reaching 62 μg/m³. Moreover, all zones exceed the WHO-defined limit, and the average values of all the time series exceed the NOM limit.

Figure 2. Behavior of the annual average concentration of six air pollutants in several analysis zones from 2016 to 2023. (a) PM₁₀, (b) PM_2.5, (c) SO₂, (d) NO₂, (e) O₃, and (f) CO. The purple dotted line shows the average behavior of all the time series. The gray dotted line indicates the allowed annual limits by the NOM [29,30]. The green dotted line denotes the WHO-allowed limits [27].

Regarding PM_2.5 (Figure 2b), the average values across all time series exceeded the NOM and WHO limits. Zones 2 and 4 reached the highest values, although the latter presents a considerable decrease of almost 50% from 2020. Meanwhile, for SO₂ (Figure 2c), Zone 2 had the highest values (>0.006 ppm), whereas Zone 4 had the lowest values in more than 60% of the years.

In the case of NO₂ (Figure 2d), all the zones exceeded the WHO limit, recording values up to 200% above the stated limit. Additionally, Zones 2, 3, 4, and 5 exceeded the NOM limit, although Zone 5 showed a decrease in concentration from 2020. It is worth noting that Zone 1 had values close to the NOM limit, but it was never exceeded (<0.021 ppm).

For O₃ (Figure 2e), Zones 2, 3, and 4 present increases from 2020, while Zone 5 showed a considerable reduction of almost 50% respect to its highest value. This behavior is similar to NO₂, with the lowest concentrations recorded in Zone 5. Finally, regarding CO (Figure 2f), Zone 3 recorded the highest annual average (>1.0 ppm); in contrast, Zone 4 recorded the lowest annual average (<0.65 ppm), with a downward trend. In summary, Zones 2 and 3 exhibit the highest contamination levels, while Zones 1 and 5 show the lowest.

Figure 3 describes the number of days during which the pollutant concentrations exceeded the allowed limits for several NOMs. PM₁₀ is the pollutant with most days in which NOM-allowed limits were exceeded across all zones. In particular, Zone 2 had the most exceeded days during the analysis period (1000 days), followed by Zone 3 with approximately 450 days. Zones 4 and 5 exceeded between 180 and 210 days, respectively.

Figure 3. Distribution of the number of exceeded days regarding the NOM allowed limits for each pollutant in (a) Zone 1, (b) Zone 2, (c) Zone 3, (d) Zone 4, (e) Zone 5, and (f) Zone 6.

For PM_2.5, Zone 2 recorded the highest number of exceeded days, totaling approximately 96, followed by Zones 6 and 4, with 84 and 47 exceeded days, respectively. Notably, O₃ had the highest number of exceedance days in Zone 4, totaling 158 days, exceeding both PM₁₀ and PM_2.5.

Regarding SO₂, Zone 2 had the highest number of days exceeding the NOM limits (11 days) while Zone 5 reported only 1. For this component, these zones were the only ones where the limit was exceeded. In contrast, Zone 2 had the highest number of days with NO₂ levels exceeding the daily NOM, with 2 days exceeding it. Zones 1 and 6 only exceeded the NOM on one day each. Finally, CO did not report days exceeding the stipulated limits in any analysis zone.

Consequently, the findings suggest that PM₁₀, PM_2.5, and O₃ are the pollutants with the highest number of exceedance days relative to the NOM-stated limits. Zone 2 recorded 1300 days exceeding the limit, more than any other zone.

Furthermore, Figure 4 describes the behavior of data grouped by seasons. The minimum concentrations of pollutants are observed in summer and autumn, while the maximum values are observed in spring and winter. Zone 1 showed different behavior for PM_2.5 (Figure 4b), with the minimum occurring in winter. On the other hand, Zone 2 recorded the highest levels of PM₁₀, PM_2.5, SO₂, and NO₂. Likewise, Zone 3 shows the same behavior as Zone 2, except in CO (Figure 4f), where a constant change is observed, with the highest values. Nonetheless, Zone 6 recorded the lowest levels of PM₁₀, PM_2.5, and O₃, while Zone 4 recorded the lowest levels of SO₂ and CO. Moreover, Zone 1 recorded the lowest NO₂ levels. Finally, Zone 5 showed the lowest PM_2.5 level in autumn across all time series, around 10 μg/m³.

Figure 4. Average behavior per season in six air pollutants for several study zones. (a) PM₁₀, (b) PM_2.5, (c) SO₂, (d) NO₂, (e) O₃, and (f) CO. The analysis period comprises 2016 to 2023.

Table 3 presents the trend in pollutant concentrations across the study zones. For PM₁₀, an upward trend is observed in Zones 2, 3, 5, and 6, with the most significant increase in Zone 5 (+1.01 μg/m³ per year). In contrast, Zones 1 and 4 showed negative trends, with Zone 1 showing the lowest (−1.32 μg/m³). The statistically significant differences are presented in Zones 1 and 6. Regarding PM_2.5, increases are observed in Zones 3 and 6, with Zone 3 showing the largest increase and a statistically significant difference (1.74 μg/m³). Moreover, significant reductions were identified in Zones 2 and 4, with the lowest value (−0.99 μg/m³) in Zone 4. SO₂ shows a negative trend across all zones, with statistical significance in all but Zone 3. The lowest reduction is presented in Zone 2 (−12.3 × 10⁻⁴ ppm per year), and the highest reduction is presented in Zone 3 (−1.71 × 10⁻⁴ ppm per year). In For NO₂, the trend is downward across all zones, with Zones 5 and 6 showing the largest statistically significant reductions (−8.77 × 10⁻⁴ ppm and −7.83 × 10⁻⁴ ppm, respectively). On the other hand, O₃ shows reductions across all zones, except Zone 2, where an increase is observed (+12.7 × 10⁻⁴ ppm). The statistical test shows a significant difference in Zones 1 and 5 for this pollutant. Finally, CO shows positive trends in Zones 1 and 6, with Zone 1 showing the largest significant increase (0.02 ppm per year), while statistically significant reductions are observed in Zones 2 and 3.

Table 3. Behavior of the annual trend of the pollutant concentrations in the study zones.

Pollutants	Zone 1		Zone 2		Zone 3		Zone 4		Zone 5		Zone 6
	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope
PM10	−33.34 *	−1.32	2.33	0.24	5.47	0.87	−3.69	−0.46	5.30	1.01	11.97 *	0.81
PM2.5	−6.66	−0.51	−8.96 *	−0.38	9.57 *	1.74	−25.72 *	−0.99	−1.63	−0.05	2.75	0.18
SO2	−18.58 *	−6.2	−15.26 *	−12.3	−4.70	−1.71	−14.42 *	−6.04	−9.43 *	−6.12	−10.42 *	−2.47
NO2	−5.03	−1.61	−1.28	−1.37	−7.24	−2.83	−3.88	−3.72	−12.90 *	−8.77	−19.30 *	−7.83
O3	−10.58 *	−15.2	0.00	12.7	−1.41	−1.65	−3.83	−5.74	−16.07 *	−36.6	−6.96	−8.56
CO	8.07 *	0.02	−41.26 *	−0.03	−17.25 *	−0.04	−21.94 *	−0.03	−0.34	$-1.73\times {10}^{-3}$	1.04	$5.04\times {10}^{-3}$

The values of SO₂, NO₂, and O₃ are multiplied by ${10}^{-4}$. The asterisk indicates a statistically significant trend based on the Mann–Kendall test at the 95% confidence level.

3.2. Climate Variables’ Behavior

Figure 5 shows the climograph behavior in each study variable in the analysis zones. In the case of precipitation, the maximum values are presented in summer (around 100 mm). In contrast, the minimum values are presented in all winters in all analysis zones (values above 20 mm). Regarding evaporation, the maximum values are presented in spring (over 150 mm) and the minimum values in autumn (around 100 mm). It is important to mention that Zone 3 recorded values for all variables except evaporation. For maximum temperature, the highest values are reached in spring (above 25 °C), and the lowest values occur between autumn and winter (around 20 °C). On the other hand, for the average temperature, the maximum values oscillate between spring and summer, surpassing 15 °C, while the minimum values are presented around 10 °C in winter. Finally, the minimum temperature shows a behavior similar to the average temperature, with the highest values in summer and the lowest in winter.

Figure 5. Climate variables climographs for each study zone. (a) Zone 1, (b) Zone 2, (c) Zone 3, (d) Zone 4, (e) Zone 5, and (f) Zone 6. The analysis period comprises 1990 to 2023. In Zone 3, no information on evaporation was provided, but data on the other variables were included.

Table 4 shows trends in the climate variables. For precipitation, all zones exhibit a negative trend, suggesting future reductions. The most statistically significant negative trends were observed in Zones 4, 5, and 6, with values fluctuating around −10 mm per year. In the evaporation case, trend increases are observed in Zones 1, 2, and 4, with values around +0.63 and +3.28 mm per year, while Zones 5 and 6 show negative, statistically significant trends of −1.43 mm and −2.22 mm, respectively. Regarding maximum temperature, an upward trend is observed in all zones except Zone 5, with values ranging from +0.1 °C to +0.57 °C per year. Statistically significant differences were observed in Zones 2 and 3. Likewise, the average temperature followed a pattern similar to the maximum temperature, except in Zone 4, where a negative trend was observed, close to zero. Finally, the minimum temperature recorded shows an increasing trend in Zones 1 and 2, whereas in the other zones the values were negative (between −0.01 and −0.39 °C per year). Statistically significant differences were observed in Zones 3 and 5.

Table 4. Trend behavior of the climatic variables in the different analysis areas.

Pollutants	Zone 1		Zone 2		Zone 3		Zone 4		Zone 5		Zone 6
	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope	Z Value	Sen Slope
Precipitation	−4.43	−2.93	−4.70	−5.35	−0.01	−7.00	−11.31 *	−10.72	−7.90 *	−9.85	−10.21 *	−10.13
Evaporation	0.43	0.63	4.29	1.94	NA	NA	5.38	3.28	−9.61 *	−1.43	−9.08 *	−2.22
Maximum Temperature	2.95	0.13	11.22 *	0.49	12.52 *	0.57	5.89	0.15	−0.71	−0.02	2.10	0.10
Average Temperature	2.27	0.08	13.76 *	0.30	9.81 *	0.26	−0.70	−0.04	−4.98	−0.19	0.87	0.04
Minimum Temperature	1.20	0.02	1.36	0.05	−12.73 *	−0.28	−0.15	−0.01	−12.00 *	−0.39	−0.99	−0.02

The asterisk indicates a statistically significant trend based on the Mann–Kendall test at the 95% confidence level. NA indicates that no information is available for the variable.

The observed behavior may suggest that the maximum and minimum temperature values are diverging, leading to increasingly extreme temperatures. Conversely, the potential reductions in precipitation across all zones are exhibited.

3.3. Correlations’ Behavior

Figure 6 illustrates the correlation coefficient behavior between the analysis variables and the study zones. Regarding precipitation, negative values were observed for all pollutants and zones, except in Zone 1, where PM_2.5 and SO₂ showed positive values close to zero (+0.09 and +0.03, respectively). Furthermore, PM₁₀, NO₂, and CO exhibited significant inverse correlations across all zones, with correlation values ranging from −0.43 to −0.81. PM_2.5 showed statistically significant inverse correlations in four zones, ranging from −0.70 to −0.46. O₃ showed similar correlations in Zones 4 and 6. SO₂ had the fewest correlation values, with no statistically significant differences. For maximum temperature, positive correlations were observed in most zones, with the highest and statistically significant values in Zones 2, 4, 5, and 6 (ranging from +0.41 to +0.51). Moreover, the maximum temperatures are positively correlated with PM₁₀, PM_2.5, and O₃, and negatively correlated with CO and NO₂, achieving negative, statistically significant correlation values in Zones 2, 3, and 6 (ranging from −0.45 to −0.61). On the other hand, the average temperature showed positive correlations with PM₁₀, PM_2.5, and O₃ in Zones 1, 4, and 6, although these differences were not statistically significant. Conversely, SO₂, NO₂, and CO showed negative, statistically significant correlations with the average temperature, ranging from −0.73 to −0.49. Regarding minimum temperature, negative values were observed in all analysis zones and across all pollutants, but NO₂ and CO showed statistically significant values in all zones. It is important to mention that, in Zones 2, 3, 4, and 6, the correlation values were negative. Finally, evaporation showed positive values in Zones 1, 4, 5, and 6 for most pollutants, and statistically significant differences in PM₁₀, PM_2.5, and O₃, with values ranging from +0.39 to +0.54. The only significant negative value was found in Zone 2 for CO (−0.46). This behavior is also evident in maximum temperature, indicating that these variables exhibit similar patterns.

Figure 6. Behavior of the rho correlation coefficient between climate and pollutant variables in the study regions. The black-dotted square indicates a statistically significant difference at the 95% confidence level. (a) Zone 1, (b) Zone 2, (c) Zone 3, (d) Zone 4, (e) Zone 5, and (f) Zone 6. Blue indicates negative correlation, and red indicates positive correlation.

4. Discussion

Results demonstrate that pollutant behavior is constantly changing. Zones 2 and 3 showed values exceeding the annual limits established by the NOM and WHO (PM₁₀, PM_2.5, and NO₂) and reported higher values for pollutants for which no annual reference is available (SO₂, O₃, and CO) (see Figure 2). Moreover, these zones present a higher number of days surpassing the daily values allowed by the NOM for all pollutants (see Figure 3). This behavior is due to the presence of industrial parks for the railway, automotive, chemical, energy, and metal–mechanical sectors. Meanwhile, for Zone 3, industrial parks for logistics, paper, food, and beverage are identified [38]. Moreover, the thermal power station “Francisco Pérez Ríos” of the Comisión Federal de Electricidad (CFE) and the refinery “Miguel Hidalgo” of Petróleos Mexicanos (PEMEX) are located in Zone 2. The thermal power station has been operating since 1975 and has a capacity of 2095 GW. Furthermore, it is the fifth-largest thermal power station in Mexico and the leading electricity supplier in central Mexico. This thermal power station produces electricity using conventional steam and utilizes both natural gas and fuel oil. The refinery, which opened in 1976, is the second largest in the country, with a capacity to process approximately 612,000 barrels of fuel per day [35,50,51,52]. Both industries can generate the highest pollutant concentrations, accounting for approximately 90% of the total atmospheric emissions in the Hidalgo state [53]. These emissions concentrations are primarily due to the use of fossil fuels with high sulfur content, which generate significant sources of PM₁₀, PM_2.5, and various oxide emissions in the country [51,54]. Figure 2 and Figure 3 illustrate this behavior, showing that PM₁₀, PM_2.5, SO₂, NO₂, and CO reached the highest annual values, and the highest number of days exceeding the NOM limits, compared to other zones.

Regarding O₃, a significant decrease has been observed in Zone 5 since 2020. This trend is observed in the same zone with NO₂ (Figure 2d,e). The reason is that tropospheric ozone (O₃) concentrations depend on a photochemical process involving reactions with certain precursor pollutants, such as nitrogen oxides (NO₂ and NO), meaning that reductions in NO₂ lead to reductions in O₃ [55].According to the Gaceta Oficial de la Ciudad de México, a government initiative to regulate air quality in Mexico City, called Programa para Prevenir y Responder a Contingencias Ambientales, was created in 2019 [56]. This initiative establishes that the thermal power station would reduce fuel oil consumption by 30%, a goal achieved by the CFE in 2021, when it reduced this fuel by 80% [57]. These actions are observed in SO₂ and CO, with a downward trend. Nonetheless, this initiative does not decrease the concentrations of the other four air pollutants, which have significantly increased since 2020.

Another factor contributing to air contamination in the study zones is mobile sources (internal combustion vehicles), which use fossil fuels as their primary energy source [16,58,59,60,61,62]. Furthermore, sandstorms and strong wind currents are known to increase PM_2.5 and PM₁₀ concentrations by transporting dust and other particulate matter [35,63]. On the other hand, future annual trends indicate a decrease in the concentrations of most pollutants across all zones (see Table 3). This situation suggests that the companies and industrial parks may be complying with the limits set by various NOMs [28,29,30,31,32] or international standards [1,2].

Regarding air pollutants, some studies have demonstrated that exposure to PM₁₀ above 35 μg/m³ can lead to health effects such as hospitalizations [64], cardiopulmonary disorders [65,66], and respiratory mortality [67]. Moreover, exposure to PM_2.5, above or similar to 10 ppm (10 μg/m³) can produce diseases such as diabetes [68], lung cancer [69], cardiovascular diseases [70] and chronic hospitalizations [64]. On the other hand, prolonged exposure to low concentrations of SO₂ < 0.00136 ppm (<1.36 μg/m³) is associated with a higher risk of mental illness, depressive disorders, and anxiety disorders [71]. Also, SO₂ values around 0.0038 ppm (10 μg/m³) increase the cardiovascular and respiratory mortality [66]. In the case of NO₂, exposure to approximately 0.0053 ppm (10 μg/m³) is associated with chronic obstructive pulmonary disease (COPD) and asthma [72,73]. In conclusion, pollutant concentrations could cause health effects in the population, with Zone 2 posing the highest risk.

For climate variables, results suggest a strong correlation with the pollutant concentrations. Precipitation is negatively correlated across all study zones, indicating an inverse relationship with pollutants, with statistically significant differences in 72% of the time series (Figure 6). This behavior is observed in Figure 4 and Figure 5, where the maximum pollutant concentrations occur in spring and winter, and the minimum concentrations occur in summer and autumn. These behaviors have been described in the literature, where authors affirm that the maximum pollutant concentration occurs in winter [74,75,76], because precipitation is important for the dissipation, transport, and deposition of pollutants in the atmosphere [13,77,78]. In tropical regions, such as Mexico, most precipitation occurs in summer and autumn due to increased humidity from the two ocean basins, cyclogenesis zones, convective instability, low pressures, and other factors that help create climate phenomena such as tropical cyclones [45,79,80]. Higher humidity is associated with more intense precipitation events, which reduce atmospheric particle concentrations through aerosol scavenging. As a result, there is an inverse relationship between precipitation and particulate matter in ambient air [75]. However, results suggest a negative trend of this variable in several study zones, indicating a decrease in precipitation in the future. This behavior has been highlighted by authors, who point out that rainfall across the country will decrease due to various anthropogenic activities, intensifying drought [81,82,83,84,85] and hindering the dissipation of pollutant concentrations across all zones.

For maximum temperature and evaporation, a pattern similar to its maximum and minimum was observed, driven by higher tropospheric water vapor availability and higher temperatures, which increase sensible and latent heat flux at the surface, leading to greater evaporation of available moisture [86,87,88,89]. Regarding pollutant concentrations, the maximum and minimum temperatures show statistically significant differences with O₃, PM₁₀, and PM_2.5. Several authors report a direct, positive relationship between tropospheric O₃ and maximum temperature, because the ozone layer forms through photochemical reactions involving solar radiation and high temperatures, which may lead to higher ozone concentrations [90,91,92,93]. In the case of PM₁₀ and PM_2.5, maximum temperature and humidity are related because warm weather conditions also result in stagnant air masses, which reduce dispersion and dilution of pollutants, thereby exacerbating their concentrations [94]. In contrast, the minimum temperature shows statistically significant negative correlations with several pollutants (PM₁₀, PM_2.5, NO₂, and CO), consistent with [13], who found lower temperatures are associated with higher pollutant concentrations in ambient air, highlighting a significant relationship between these factors. Finally, temperature trends are positive, indicating that they could increase in the future. This behavior has been reported in other studies, indicating that in Mexico, temperature increases due to anthropogenic activities are expected [50,95,96], which could increase concentrations of various pollutants. The main limitations when studying air pollutants and climate variability using monitoring stations are limited spatial coverage, limited data availability, and difficulty in capturing complex processes due to the spatial and temporal heterogeneity of the data, which could limit the accuracy of methodologies for assessing the relationships between pollutants and climate variables.

5. Conclusions

Air pollutants can exacerbate climate change in specific areas. In this work, we concluded that air pollutants are closely related to the behavior of climatic variables in several IPHSs. Zones 3 and 4 reported the highest annual average values compared to the allowed limits established by the NOMs (PM₁₀ > 36 μg/m³, PM_2.5 > 10 μg/m³, and NO₂ > 0.021 ppm). In terms of the number of days exceeding these limits, Zone 3 and 4 recorded over 1000 and 500 days for PM₁₀, more than 90 and 8 days for PM_2.5, 11 days and 1 day for SO₂, and over 140 and 100 days for O₃, respectively. Regarding the annual distribution, pollutant concentrations follow a well-defined pattern, with the maximum values in spring and winter and the minimum values in summer and autumn. This behavior indicates an inverse relationship with precipitation, with 72% of the pollutants showing statistically significant differences (ranging from −0.81 to −0.43) across several zones.

On the other hand, statistically significant, positive relations were reported for maximum temperature (+0.41 to +0.51) and evaporation (+0.39 to +0.54). In the context of future trends, the Sen Slope method indicates a possible decrease in pollutant concentrations across several study regions, suggesting that decision-makers are taking action to control these pollutants. However, industrial zones are not the only human activities that affect pollution; vehicle presence and population growth also significantly contribute. For future trends in climate variables, a decrease in precipitation (−10 mm per year) is projected, leading to lower pollutant dispersion and higher concentrations. In contrast, temperatures show substantial increases (ranging from +0.1 °C to +0.57 °C), which could lead to higher pollutant concentrations due to atmospheric stability. As a result, climate conditions can alter the presence of several pollutants, suggesting that potential climate change could affect pollutant concentrations or dissipation in regions with industrial areas in the state of Hidalgo.

As future research, we suggest using multispectral images to monitor air pollutant concentrations, thereby reducing limitations in the distribution for the existing monitoring network in Mexico, missing data, and the periodicity of the information.