Total column ozone (O 3 ) and nitrogen dioxide (NO 2 ) via satellite and their relationship with the burned area and climatic data in biomes of Central-West Brazil

doi:10.21203/rs.3.rs-1878892/v1

The total column ozone (O₃) and nitrogen dioxide (NO₂) levels based on the satellite remote sensing for a period from 2005 to 2020 along with air temperature, rainfall and burned area in three biomes (Cerrado, Pantanal, and Atlantic Forest) from Mato Grosso do Sul - Brazil was evaluated. The annual variations of O₃ ranged from 260 DU for the Pantanal to 347 DU for the Cerrado. Whereas the NO₂ concentrations ranged from 2.95×10¹⁵ molecules for the Cerrado to 3.01×10¹⁵ molecules for the Atlantic Forest. The differences between NO₂ and O₃ concentrations on monthly and seasonal time scales, with higher values during the dry period (between July and October). The NO₂ and O₃ concentrations positively correlated with the burn areas in Pantanal and Cerrado, while the rainfall negatively correlated with these gases’ concentrations in these biomes The first principal component in all biomes is a comparison between rainfall and NO₂, O₃, a burned area, and air temperature with higher values of eigenvalues for the burned area in Pantanal, followed by Cerrado and Atlantic Forest, indicating more fires in Pantanal. Rainfall showed the highest value in the first principal components (PC). The second component in the Atlantic Forest was a balance between rainfall and air temperature with NO₂, O₃ and burned area. In the Cerrado and Pantanal, a comparison is made between burned area and NO₂, with rainfall, air temperature and O₃. There are differences in the behavior of NO₂ and O₃ concentrations in biomes, driven by different environmental and anthropic variables.

Urban pollutants

Satellite remote sensing

ecological risk

air quality

environmental degradation

Air pollution/air quality, climate change, deforestation, and, environmental degradation are the main challenges in the world. The ozone (O₃) and nitrogen dioxide (NO₂) are the main urban air pollutants (Souza and Ozonur, 2019; Silveira et al., 2020). The O₃ is one of the main greenhouse gases (GHG) and plays an important role in our climate system (Silveira et al., 2020). It absorbs the solar ultraviolet (UV) radiation and thus heats the stratosphere and troposphere. By photolysis of molecular oxygen at shorter UV wavelengths, the stratospheric ozone is produced and destroyed by catalytic cycles involving nitrogen (N₂), hydrogen (H₂), and halogen radicals (Miles et al. 2015; Souza et al. 2020a). Furthermore, the O₃ has adverse effects on human health and ecosystem productivity (Wang et al. 2019).

The changing O₃ concentration can have an impact on the climate, and these changes can affect the O₃ in a number of different ways. Therefore, it is a fact that the future distribution of O₃ will depend upon the emission and impact of other GHGs. The combustion of air leads to the formation of nitrogen oxides (NOx = [NO₂ + NO]) (Aragão et al. 2014), which is toxic to biosphere species. It converts to nitric acid (HNO₃) and contributes to the acid rain, which is harmful to the entire ecosystem (terrestrial + aquatic) (Gauss et al. 2007). The NO₂ is an important catalyst in the production of Tropospheric ozone (Zoran et al. 2020). The Stratospheric ozone has been depleted in recent decades, mainly due to the impact of chlorofluorocarbons (CFCs) on the O₃ layer (Aragão et al. 2014). Further, human activities also change the atmospheric gas concentrations such as tropospheric ozone, hydroxyl (OH), nitrous oxide (NO), chlorine oxide (ClO), and bromine oxide (BrO) - (Brasseur and Hitchman, 1988). The O₃ can be transported over long distances from its sources (Castell-Balaguer et al. 2012) and its remote effects are intensified in the occurrence of forest fires and also in urban environments with worsening air quality (Chen et al. 2017).

Many studies have suggested that the formation of O₃ and its transport depend on the meteorological factors (Thompson et al. 2001; Watson et al. 2016). Biomass burning is playing a dominant role in influencing the global climate system and biogeochemical cycles. Recently, some studies found a significant correlation between NO₂ and fire counts followed by the influence of land use and land cover and local industrialization (José et al. 2017; Wang et al. 2019). Therefore, the investigation of the dynamics between pollutant concentrations, climatic variables, and fire foci occurrences must be intensified around the world to understand the local functioning and its remote effects. However, studies relating the concentrations of O₃, NO₂, fire occurrences, and climatic conditions are scarce in underdeveloped countries.

In Brazil, for example, problems associated with the degradation of air quality via the analysis of O₃ and/or NO₂ concentrations have been investigated in large urban centers such as São Paulo (Carvalho et al. 2020), Rio de Janeiro (Siciliano et al. 2020; Silveira et al. 2020). In the Center-West of Brazil, agricultural and industrial activities and urbanization are growing, which in turn change the patterns of O₃ and NO₂ in the region. In addition, fire activities to clean areas are common in the region (Oliveira-Júnior et al. 2020). Thus, such information is relevant to the dynamics of O₃ and NO₂ and, therefore, the study to evaluates O₃ and NO₂ and their relationship with fire foci occurrence and climatic data (rainfall and air temperature) in three biomes in the state of Mato Grosso do Sul (MS) - Midwest of Brazil, followed by verifying differences between the concentrations of O₃ and NO₂ in the biomes and pointing out which are the main variables that determine the concentrations of O₃ and NO₂.

2.1. Study area

The state of MS is located in the Midwest of Brazil (Fig. 1a), with an area of approximately 358,159 km². The annual average air temperatures oscillate between 20 and 26°C and the annual total rainfall varies from 1,000 to 1,900 mm (Abreu et al. 2020). Based on the Köppen climate classification, the four climate types are defined (Fig. 1b): “Aw” (tropical zone with winter), mainly in the southeast (SE) and north (N); “Am” (tropical monsoon) in the central region; “Af” (tropical zone with no dry season) mainly in the southwest (SW); and “Cfa” (humid subtropical zone with hot summer) in the south (S) - (Alvares et al. 2013). The MS has two well-defined seasons, a rainy season from November to March and a dry season from April to September (Abreu et al. 2020).

Figure 1

2.2. The biomes

Three biomes cover the state (Fig. 1c) the Cerrado (Brazilian Cerrado − 61%), the Pantanal (wetlands − 25%) and the Atlantic Forest (14%) (Oliveira Junior et al. 2020; Abreu et al. 2020). The area of these biomes is being degraded by fires and natural areas are being converted by the use of fire (Oliveira Junior et al. 2020; Silva Junior et al. 2020), reinforcing the need for information on the frequency of fire foci.

2.2.1. Cerrado

The Cerrado is a biodiversity hotspot (Myers et al. 2000) and the largest national producer of soybeans, accounting for 52% of Brazilian productivity (Magalhães et al. 2020). The expansion of agricultural land has shaped the landscapes of the Brazilian Cerrado in recent decades (Grecchi et al. 2014), where implementation and management include fire as a cleaning tool (Magalhães et al. 2020; Oliveira Junior et al. 2020; Silva Junior et al. 2020). The environmental consequences of these transformations are still poorly evaluated. The Cerrado biome covered almost 2 million km² 60 years ago (Durigan and Ratter, 2015). Half of this area has been converted to pasture, in addition to annual crops, forestry, sugarcane, and intensive land use; the challenge is to reconcile conservation with agricultural production (Magalhães et al. 2020). The cerrado vegetation (sensu lato) comprises a physiognomic mosaic that varies from open pastures (campo puro, campo sujo, and campo cerrado - physiognomies with natural pastures and shrub trees), with an intermediate physiognomy of savannah (cerrado stricto sensu) predominant and the forest (cerradão or forest savannah) - (Ribeiro and Walter, 1998). Due to the flammability of grasses, fire tends to be more frequent in more open physiognomies (Newberry et al. 2020). Fire is capable of killing young arboreal individuals by opening their faces (Miranda et al. 2009).

2.2.2. Pantanal

The Pantanal is the largest tropical wetland and is recognized as a World Natural Heritage and Biosphere Reserve by the Scientific and Cultural Organization (UNESCO). In the Pantanal, fire, and flooding synergistically shape the vegetation structure (Manrique-Pineda et al. 2021). Flood-intolerant species are eliminated from the plant community (Manrique-Pineda et al. 2021), and fire eliminates intolerant trees, most of them with thin bark (Hoffmann et al. 2012). The flood pulse can promote drastic changes in the ecosystem and can alter the frequency and intensity of the fire, as suggested by Manrique-Pineda et al. (2021). The increase in anthropogenic fires can harm the recovery of vegetation and impoverish biodiversity (Oliveira Junior et al. 2020). Recently, the Brazilian government allowed the planting of sugarcane in the Pantanal (Lima et al. 2020). In addition, exotic grasses introduced for pasture can alter fire regimes (Schulz et al. 2019), increasing the incidence of fires in the region. The risk of fire focis can be high, significantly high, or extreme (Mota et al. 2019) according to forest fire risk zoning and the intensity of fire foci in the neighboring state (Mato Grosso - MT).

2.2.3. Atlantic forest

The Atlantic Forest is also a biodiversity hotspot (Myers et al. 2000), and the vegetation is not fire-prone. The structure of the Atlantic Forest in the state of MS presents physiognomies of semideciduous and deciduous forests (Damasceno-Junior et al. 2018). Semideciduous forests can lose 20–50% of leaves during the dry season, while deciduous forests can lose more than 50% (IBGE, 2022). The amount of dry leaf litter and canopy opening in combination can create conditions for fire propagation under the canopies (Newberry et al. 2020). In MS, the main reason for the reduction of the Atlantic Forest is related to the expansion of the agricultural area (Cunha et al. 2021), which puts this biome on alert for an increase in fires.

2.3. Data

TCO (Total Column Ozone) and NO₂ data for the period of 2005 to 2020 were obtained from the AURA/OMI (Ozone Monitoring Instrument) on the EOS (Earth Observation System) platform on the AURA Satellite. AURA is a heliosynchronous orbit satellite, which allows daily recoveries of species such as NO₂, O₃, BrO, SO₂, and HCHO. The OMI instrument is a contribution of the Dutch Agency for Aerospace Programs in collaboration with the Meteorological Institute of Finland for the EOS AURA mission and was launched on 15 July 2004. OMI-based O₃ measurement is performed using a nadir scanner that detects solar backscattered radiation in visible (350–500 nm) and UV (270–314 nm and 306–380 nm UV-UV) wavelength channels I and II. The OMI product is a version 3 and level 3 global gridded data with 0.25° x 0.25° spatial resolution based on the total O₃ mapping spectrometer algorithm. OMNO₂ is a daily data product produced from NO₂ measurements taken by OMI on the EOS-AURA spacecraft.

The data is populated into a grid with a horizontal resolution of 0.25° x 0.25° in latitude and longitude. It provides a vertical column density of tropospheric daily NO₂ at spatial resolutions and is useful for regional analysis. The IMO makes use of a radiative transfer model and a georeferencing scheme to determine trace gas column abundances for standard analysis of Earth spectral radiance measurements (Bucsela et al. 2006). Rainfall and air temperature data were obtained via CHIRPS (Climate Hazards Group InfraRed Precipitation with Station) and MOD11C3 v061 products, followed by land use and land cover (LULC) and burned data obtained from the MapBiomas platform (Souza et al. 2020b), respectively.

To assist in the discussions, data from LULC, population density, and road density in each biome, to verify the possible influences on NO₂ and O₃ concentrations. The LULC data were extracted from Map Biomass (Souza et al. 2020b) and the population density and road density data were from the Brazilian Institute of Geography and Statistics (IBGE, 2022).

2.4. Statistical analysis

The descriptive analysis consisted of obtaining position and dispersion statistics and graphical analysis of trend lines and notched boxplots. Monthly averages and monthly and seasonal standard deviations of NO₂, O₃ concentrations, amount of fire foci, rainfall, and air temperature were obtained on an annual, seasonal, and monthly scale. Notched box plots apply a "notch" of the box around the median. The notches are useful in providing a rough guide to the meaning of the median difference; this will provide evidence of a statistically significant difference between them (Abreu et al. 2020).

The NO₂ and O₃ concentration and burned area occurrences were compared between the biomes using the ANOVA two-way test, arranged in a factorial design. Factor 1 represents the temporal effects considering the 12 months of the year and the seasonal period (dry season and rainy season). Factor 2 represents the spatial effects considering the three biomes in the MS state. The replicate was considered as the mean value of NO₂ and O₃ concentration and number of fire foci occurrences for each month/seasonal period and in each biome. The normality using the Shapiro-Wilk (SW) test was tested before the ANOVA, at 5% probability. The natural logarithmic transformation (ln (x)) was applied throughout the non-normal dataset, as recommended by Carvalho et al. (2020). Scott-Knott’s test at 5% probability was used to compare the levels of the factor.

Principal Component Analysis (PCA) is a mathematical procedure that uses an orthogonal transformation (orthogonalization of vectors) to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. Through the eigenvalues and respective eigenvectors of a variance and covariance matrix or a correlation matrix between variables, it is possible to verify which variables are most important in explaining the variance of the data. In the present study, we adopted the correlation matrix to extract the eigenvalues, due to the unequal measurement units and the variance presenting a large difference between the variables (Souza and Santos, 2018).

In addition, the KMO statistic is observed, which assesses the adequacy of the sample in terms of the degree of partial correlations between the variables. Values vary between 0 and 1, and the closer to 1, the more appropriate is the use of the technique. Bartlett's sphericity test is used to evaluate the hypothesis that the correlation matrix can be the identity matrix with a determinant equal to 1. If the correlation matrix is an identity matrix, it means that the variables are not correlated, being inappropriate for the use of factor analysis. Otherwise, there will be indications that the correlations between the variables are significant (Fraga et al. 2020). The entire statistical procedure was performed in an R environment (R Team Core, 2018).

The annual variations of O₃, NO₂^,and fire foci for the three MS biomes for the period of 16 years from 2005 to 2020 are shown in Table 1 and Figure 2. The ANOVA and Scott–Knott test produced significant results, showing a two-way interaction between the factors biome and months for NO₂ and fire foci, but not for O₃. The seasonality was marked in the concentrations of O₃ and NO₂ and the occurrence of fires during the dry period. According to the Scott–Knott test (α = 0.05) for the interactions in NO₂, the occurrence of maximum NO₂ concentration in the Pantanal has lower concentrations between January and April and higher concentrations between August and September. In the Cerrado and Atlantic Forest, the highest concentrations of NO₂ occur from July to December. During the dry period, Pantanal had a higher concentration of NO₂ than Cerrado and Atlantic Forest. During the rainy period, the opposite was observed with a higher concentration of NO₂ in Cerrado and Atlantic Forest than in Pantanal. Regardless of the time of year, a higher O₃ was observed in Cerrado. The months with higher concentration were September and October. A lower concentration of O₃ was observed in April and May.

Figure 2

Table 1

The peak of burned area occurrence is in the dry period, in August and September in Cerrado and Atlantic Forest, while in Pantanal is from August to October. The biomass with the higher burned area is the Pantanal, despite being the smallest biome in Mato Grosso do Sul, followed by Cerrado and Atlantic Forest. There is an inverse relationship with a relative delay between the burned area and precipitation. The increase in burned areas occurs months before the rainy season.

The variability of atmospheric pollutant concentrations is dependent on specific emissions and general weather conditions (Seinfeld and Pandis, 2006; Aragão et al. 2014). NOx is a primary pollutant and O₃ is a secondary pollutant that originates in the atmosphere through a set of complex reactions (Seinfeld and Pandis, 2006). The monthly analysis of variability in O₃ and NO₂ concentrations showed the fire foci, rainfall, and temperature shown in Figure 3. The air temperature reached the highest values between September and November, except for Atlantic Forest, followed by intermediate values between December and November. March and the lowest temperatures were observed between May and July. Rain seasonality is evident in the region during the rainy months between November and March and dry months between April and September (Abreu et al. 2020; Oliveira-Júnior et al. 2021; Abreu et al. 2022). In the period of lower rainfall and higher temperatures, the occurrence of fire foci increases significantly, as well as the concentrations of O₃ and NO₂ in the three biomes.

Figure 3

Figure 4 shows the notched boxplot analysis for the rainy (October to March) and dry (April to September) periods, where the differences in O₃ and NO₂ concentrations between the periods were marked in the Pantanal, with a lower concentration of NO₂ in the rainy season. The burned area occurrences were higher in the dry period in the three MS biomes, a period of lower temperatures and rainfall.

Figure 4

In the evaluation of data suitability to the PCA (Table 2), the existence of significant correlations between the variables for the biomess (p-value < 0.05) was verified with Bartlett’s sphericity test. Regarding the KMO test, the values found were higher than 0.64, demonstrating the correlation of the variables and the PCA as a suitable statistical approach. The first two principal components (PC) were able to explain more than 70% of the data variance in Cerrado ad Atlantic Forest and 60% in Atlantic Forest (Table 3).

Table 2

Table 3

The PCA results using Spearman's correlation matrix are shown in Figure 5. Most of the correlations between the NO₂ and the other variables were significant, except with burned area in Atlantic Forest, with a positive correlation with temperature, a burned area, and O₃, and negative with rainfall. The O₃ was positively correlated with temperature in all biomes. The burned area was significantly and positively correlated with NO₂ and O₃ in Pantanal and Cerrado, but not in Atlantic Forest. In addition, the burned area is negatively correlated with rainfall, except in Atlantic Forest.

Figure 5

The behavior of the PC expresses the behavior of the variables in biomes (Figure 5 and Table 4). The first PC represents the relationship between the rainfall against the concentrations of O₃, NO₂, a burned area, and temperature. The weights of the eigenvalues, of each variable were well distributed in all biomes, with rainfall with a positive eigenvalue and the other variables with negative eigenvalue in Atlantic Forest and negative eigenvalue for rainfall in Pantanal and Cerrado and positive eigenvalue for the other variables. The rainfall variable has a greater weight in the Atlantic Forest, while in the Pantanal the weight of the burned area was higher than in the other biomes. The highest scores in this component in Cerrado and Pantanal represent higher temperatures, higher concentrations of O₃ and NO₂ and higher burn area, and lower rainfall. In Atlantic Forest, it indicates more rainfall and lower concentration of O₃ and NO₂, lower temperature, and burned area.

Table 4

The second component indicates a relationship between rainfall and temperature, against with O₃, NO₂and burned area, in Atlantic Forest. The higher scores indicate more rainfall and higher temperature and lower NO₂, O_3, and burned area. For Pantanal and Cerrado is a comparison between burned area and NO₂, with O₃, temperature, and rainfall. In this component, higher scores indicate the higher occurrence of burn area, higher NO₂ concentrations, lower rainfall, lower temperatures, and lower O₃ concentrations.

The highest population density and concentration of highways in MS State is observed in the Cerrado. The Pantanal has the lowest density with only one city with more than 35 inhabitants.km^-2 (Corumbá) and the lowest density of highways (Figure 6). Table 5 shows the LULC in biomes and the percentual of increase or decrease in each LULC from 2005 to 2020. The biomes in MS state had an increase in Agricultural (between 25 and 74%), a mosaic of agricultural and pasture (between 22 and 34%), and grassland (between 24 and 37%) areas. The mining area increases from 2005 to 2020 in Pantanal and Cerrado. In addition, there was a decrease in wetland areas in all biomes from 2005 to 2020, especially in Pantanal, where the reduction was around 170% of the area. The wetland area in Pantanal is dependent on rainfall, as described by Marengo et al. (2021). The proportion of each LULC in the state biomes shows the anthropization picture of the biomes. The largest proportion of areas in the Cerrado and Atlantic Forest were pasture, with 31 and 50% of the area in biomes. In the Cerrado, the increase in agriculture caused the proportion of agricultural areas to jump from 8% in 2005 to 29% in 2020. In the Pantanal, the LULC class with the highest proportion of area was grassland (44% in 2020). Wetlands in Pantanal corresponded to 15% of the area in 2005 and 2020 only 5% of the area of the biome.

Figure 6

Table 5

The elevated NO₂ concentrations were observed throughout the study period. Higher NO₂ concentrations can be attributed to greater vehicular influence/fires, characterized by a “heavy” fleet. The notched boxplot analysis showed differences between the biomes about NO₂ and O₃ concentrations, with exceptions in some periods of the year. The highest concentration of NO₂ is found in the Pantanal in the dry season, in contrast, the Atlantic Forest and Cerrado biomes, had higher concentrations than in Pantanal in the rainy season. There was variability for the O₃ concentrations with higher concentrations in Cerrado and greater variability in the dry months in all biomes. There are differences in the amount of burned area in the three biomes, where, regardless of the time of year, there is a higher occurrence in the Pantanal, followed by the Cerrado and Atlantic Forest (Silva-Junior et al. 2020; Oliveira-Júnior et al. 2020). The amount of rainfall and temperature showed different patterns in the dry and rainy seasons. The highest total rainfall records occur in the Atlantic Forest in the dry season. In the rainy season, there is less difference between the biomes. The temperature tends to be lower in the Atlantic Forest and similar in the Cerrado and Pantanal (Fig. 4).

The rainfall and air temperature have been reported as the agents that regulate the pollutants due to the dispersion and photolysis of polluting agents (Zhang, Liu and Zhao, 2018; Borge et al. 2019; Souza et al. 2022). The increase in air temperature results in an increase in the O₃ concentration, on the contrary, the humidity which adversely affects the O₃ concentration. Notably, the average annual burned area in the Pantanal is higher than in the Cerrado, which, in turn, is higher than in the Atlantic Forest (Table 1). Studies have reported a significant increase in fire foci in the Cerrado, associated with the dry season and reduced rainfall in the rainy season, followed by changes in land use and land cover in the region (Silva Junior et al. 2021; Oliveira-Júnior et al. 2020). We found the most fire in the Pantanal (wetlands), biome with lower population and highway density. The industrial emissions and combustion of coal in thermoelectric plants, fossil fuels used in transport and fire have been reported as contributors to air pollution in MS state (Silva-Júnior et al. 2021; Souza et al. 2022) polluting agents. Another important factor is the traffic load and population density in the state of MS, which is higher in Cerrado, reflect higher O₃ in this biome.

In general, the levels of NO₂ and O₃ concentrations in the cold-dry seasons (from April to September – dry period) for the three biomes had more variability and higher values. The dry period in MS State is characterized by cold, dry air temperatures due low levels of solar radiation that was supposed to reduce the rate of photochemical processes and at the end, reduce the dispersion rates and transportation of pollutants. Some studies have been reported that NO₂ has a longer life expectancy during winter compared to spring seasons (Duncan et al. 2010; Matandirotya and Burger, 2021). In addition, the high precipitation received during the rainy period means that the life expectancy of tropospheric NO₂ is reduced as it quickly reacts with water to form nitric acid (Matandirotya and Burger, 2021). During monsoon periods, minimal levels of O₃ were observed due to insufficient solar radiation (clouds interaction with radiation) and pollutant washout, as well as O₃ consumption by HO_x radicals. The sharp rate of increase in ozone concentration was from minimum (from April to June) to maximum (August-September) and therefore one can better understand the seasonal difference between temperatures and rainfall for the duration of sunshine duration (Ali et al. 2012) and pollutant washout. The air begins to cool in winter and the ozone-laden air becomes denser and descends to lower altitudes, resulting in a lower ozone peak (Ganguly et al. 2010). The maximum tropospheric NO₂ level was observed in the winter and the minimum in the monsoon months over the biomes. The maximum observation found in winter is due to local thermal activities and population density/fires (Kalita et al. 2011).

Some authors have observed an inverse relationship between O₃ and NO_x (Pancholi et al. 2018), since the formation of O₃ is initiated by the photolysis of NO₂, as NOx reacts quickly with O₃, thus generating NO₂ (Pancholi et al. 2018; Souza and Ozonur, 2019). In MS State there is a delay in the response of higher concentrations of O₃ and NO₂ in relation to mild temperatures, which does not occur with fire foci. On colder days, the photochemical formation of O₃ is inhibited by the lower availability of solar radiation (Souza and Ozonur, 2019; Souza et al. 2020a).

The level of suspended particulate matter (PM) is high, leading to regular fog and haze of smoke that causes pollution related problems such as lung cancer; traffic disturbance on roads, as well as reduced visibility in winter (Gunaseelan et al. 2014). For the seasonal variation in the biomes, the maximum rainfall occurs in the summer with the minimum in the winter in the three biomes. Planetary boundary layer (PBL) height may be the reason for low rainfall during winter (Ganguly et al. 2010). The maximum temperature observed occurs in the summer and in the monsoon season. In these stations, due to the intense solar radiation, the photochemical reaction of O₃ is high. The maximum concentration of NO₂ observed in winter and after the monsoon season is due to excessive consumption of heating fuels. In the monsoon season, the wind brings dry, cold and dense air masses, resulting in the scattering and dilution effects that produce NO₂ through NO oxidation and lead to a vertical transport of O₃. The maximum rainfall was observed in the monsoon season, when atmospheric stability is high, which delays the photochemical process and the reduction of the ozone layer by the deposition of water droplets. Thus, the O₃ concentration is strongly dependent on humidity (Sharma et al. 2013). The high values of NO₂ were observed in September with the contribution to emissions from the burning of NO_x from the soil (Yienger and Levy, 1995) due to fires to clean areas and the increase of the temperature in MS State. The transport of cleaner air from the north (Lower Chaco) results in the loss of NO₂ through reaction with OH radicals. Due to the higher moisture content, the wet deposition rate of NO₂ may also be faster in June-September (Sheel et al. 2010).

The difference in total rainfall in each biome, especially in the dry season, may have contributed to the behavior of NO₂ and O₃, since rain tends to reduce the concentrations of these gases, as described above. The seasonality of rainfall and temperature influence the planting schedule and the conversion of LULC into MS through the use of fire, which contributes to the increase in the concentration of these gases. The Cerrado is the biome with the high population density, highest highway density, and a large proportion of the biome composed of pasture. These indicators are effective, together with temperature, they are effective for the highest concentration of polluting gases, especially O₃, as the PCA demonstrated. In the Cerrado the concentrations of O₃ were noticeably higher. The area burned in the Cerrado was intermediate, between the Pantanal and the Atlantic Forest, and this also contributes to atmospheric pollution.

In the Pantanal, the concentration of NO₂ was higher than in the other biomes, despite the low population density and few roads. The Pantanal is the largest wetland in the world and has greenhouse gas emissions and PM with a diameter of less than 2.5µm (PM_2.5ìm) to the atmosphere (Cerri et al. 2009). However, the use of fire for the conversion of agricultural areas, due to the magnitude of the burned areas, effectively contributes to the higher concentration of NO₂ in the dry period, when the highest amount of fire occurs (Oliveira-Júnior et al. 2020).

The categories of land use and land cover (Table 5) associated with the occurrence of fire in the state, indicate these differences and determine the need for more detailed analysis in each biome. In Atlantic Forest and Cerrado, the change in land use and land cover indicates a reduction in pasture area and an increase in agriculture. In the Cerrado, there is still an increase in the area of a mosaic of agricultural and pasture. In the Pantanal, grassland areas increased significantly between 2005 and 2020, and a reduction in the Wetland area, which is closely linked to agricultural activities and fire in the biome (Leal Filho et al. 2021; Marengo et al. 2021; Abreu et al. 2022). In addition, the increase in urban infrastructure was effective in all biomes, especially in the Pantanal (Abreu et al. 2022). The growth of commercial activities is associated with air pollution (Chen and Kan, 2008).

Surprisingly, the positive correlations between NO₂ and O₃ corroborate the visual analysis (Fig. 2 and Fig. 3) in which the concentrations of these pollutants have simultaneous or close peaks, contrary to what was observed in Delhi, India (Taiwari et al. 2015), Milan, Italy (Zoran et al. 2020) and Hong Kong, China (Hossain et al. 2021), in which an increase in O₃ concentration induces a reduction in NO₂ concentration. The concentration of ozone depends on the absorption of solar radiation, the absolute concentrations of NO_X, VOCS, and the ratio of NO_X to VOCS (Sheel et al, 2010). During the day, the O₃ concentration increased due to the reactions of photolysis of NO₂ and photo-oxidation of VOCs, CO, hydrocarbons, and other O₃ precursors. Many studies have shown that the concentration of ozone increases with the increasing intensity of solar radiation and temperature on clear days.

The main components confirm the opposite relationship between O₃ and NO₂ concentrations since NO₂ is the precursor of O₃ (Tiwari et al. 2015). However, the concentrations in the MS state appear to be high enough that the highest concentrations of these pollutants are close. The components also indicate an opposite arrangement between rainfall and pollutants and a direct arrangement between fire foci and pollutants. Silva-Junior et al. (2020) found large carbon emissions, especially in the Cerrado and Amazonia, associated with the occurrence of fire. The burning of fossil fuels and biological materials does not exclusively generate carbon compounds as a product. The formation and reduction of NO₂ in the combustion of oxy-fuel involving lignite (NO + HO₂ ↔ NO₂ + OH), with the HO₂ mechanism being the main NO₂ former (Ndibe et al. 2013). According to Távora Maia and Bozelli (2022), the state of MS is a major contributor to GHG emissions, especially in areas dominated by the Pantanal and Cerrado. Also, emissions related to traffic activities are responsible for the largest share of atmospheric pollutants concentration in urban areas and, in Brazil, road transport is responsible for most of the movement of cargo and the fleet grows 12% per year (Vasques and Hoinaski, 2021).

In the formation, movement and dispersion of ozone, local climatic conditions play an important role. Variations in local climatic conditions such as solar radiation, wind speed, direction, precipitation, and relative humidity can influence the level of ozone in the atmosphere. Humidity is the precursor to the occurrence of rainfalls. In moderate rainfalls, the concentration of NO₂ and ozone increases, and in heavy rainfalls it can be reduced, and its concentrations are influenced by NO_X precursors and weather conditions (temperature and solar radiation), it is an important absorber of infrared and UV radiation and primary source of the most important oxidant in the atmosphere, the hydroxyl radical OH, which is highly reactive with organic and inorganic compounds, ozone is associated with lung and respiratory diseases, as well as mortality and its negative effect on vegetation, ecosystems (Fuhrer and Booker, 2003; Schaub et al. 2005; Ito et al. 2005; Ebi and McGregor, 2008).

The association of meteorological variables, such as solar radiation, temperature, cloudiness, precipitation, wind speed, horizontal transport, and PBL height, with peculiar factors of the topography and circulation of a given region, can considerably influence the concentrations and dispersion of atmospheric pollutants, through the determination of photochemical reaction rates (Oliveira-Júnior et al. 2020).

Among the meteorological phenomena, radiation and cloud cover are the ones that most influence the vertical distribution of O₃. The climate of the biome region is a combination of several factors, the most important of which is the availability of solar energy. Its climate is predominantly hot and humid, with average annual temperatures ranging between 24ºC and 26ºC, that is, the average thermal amplitude is around 2ºC, which characterizes a spatial and seasonal homogeneity of the local temperature (Reis et al, 2022; Souza et al. 2022, 2021). According to Souza et al. (2021) and Pobocikova et al. (2021), the seasonality of the rainfall regime in the region is defined as follows: the dry period occurs from May to September, and the month of October is considered a month of transition to the rainy season that extends from November to March, and April is the transition month to the dry season.

It is worth remembering that in the Region, the largest wetland in the world is located, the Pantanal, which is crossed by the Paraguay River and its tributaries, representing a large river network. Because it is located in the tropics, energy exchanges between the continental surface and the atmosphere are quite intense. Another meteorological forcing that modulates the climate in the Region is convection. These convective systems lead to an intense spatial and temporal variability not only in the hydrological cycle of biomes, which indicates the interaction of meteorological conditions with the vertical distribution of O₃.

The high intensity of UV radiation, combined with high humidity in the tropical atmosphere, results in an increase in the amount of OH radicals from the photolysis of O₃. At the same time, the wetland, as a wetland, emits large amounts of Non-Methane Hydrocarbons (NMHC), with large-scale biogenic activity, and CO from the burning of biomass, which, in turn, can be oxidized to produce O₃ with great efficiency (Kirchhoff and Rasmussen, 1990).

In these environments, the O₃ concentration tends to follow the intensity of solar radiation, resulting in a high concentration of O₃ during the day. In these cycles, the increase in the level of ozone concentration during the daytime period is attributed to the combined effect of the photochemical production of ozone in the mixed layer and the transport coming from the upper layers, which is favored at noon by the convective activity and, consequently, associated subsidence movements, both mechanisms being activated by solar radiation.

The emission and atmospheric transport of particles due to fires is a growing public health problem that mainly affects vulnerable communities and the most sensitive people, such as children, people with lung and/or heart diseases, and workers prone to occupational diseases and populations socially vulnerable (Souza et al. 2012; Souza et al. 2013). A significant and positive relationship was found between ozone concentrations during the fire period and hospitalizations for asthma (Souza et al. 2017) in areas close to a forest fire.

The study carried out shows the annual, seasonal, and monthly variations of O₃ and NO₂, and their relationships with the burned area and meteorological parameters in the three biomes in Mato Grosso do Sul state, Brazil. There are differences in the annual, seasonal, and monthly variations between the biomass with the influence of rainfall and air temperature conditions. The concentrations of O₃ and NO₂, mainly when there is a decrease in the rainfall and the occurrence of the burned area is higher. The monthly variations in the level of O₃ and NO₂ were higher in the months of August, September, and October in the three biomes, which corresponds to the dry period and transition to the rainy season in the state of MS-Brazil, on the contrary, the temperature that in this period is high in the state.

The notched boxplot analysis points out marked differences between the biomes in terms of NO₂ concentrations, while the O₃ concentrations are similar between both biomes. Highlight for the higher concentration of NO₂ in the Pantanal in the dry season, while in the rainy season the highest concentrations are in the Atlantic Forest and Cerrado. For the O₃ concentration, there is a similarity between both biomes, except for the Atlantic Forest with greater variability in the dry months. There is a correlation between the climatic variables, fire foci, and the concentrations of O₃ and NO₂. Fire foci and temperature were positively correlated with O₃ and NO₂ concentrations in the three biomes. Rainfall is inversely correlated with NO₂, unlike O₃, where there is no correlation.

The PC's indicate a significant influence of climatic variables on the concentrations of O₃ and NO₂ in the three biomes, this indicates an increase in the concentration of these pollutants when the temperature increases and lower rainfall records. Increasing exposure to emissions from forest fires is a clinical and public health problem, and investigations are important. Changes in weather patterns, including severe and prolonged droughts, significantly increase the risk of wildfires and comorbidities. Exposure to wildfire smoke in the three biomes impacts the health of populations most at risk, including those with chronic heart or lung disease, the elderly, and children not only locally but also remotely.

Consent to Participate

All authors declare their consent to participate in the article.

Authors Contributions

All authors contributed to the study conception and design; Material preparation, data collection and analysis; The first draft of the manuscript all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Conceptualization, A S, J F O J, M C A, G B L, M S F, U C D., methodology, A S, J F O J, M C A, G B L, M S F, U C D., validation, A S, J F O J, M C A, G B L, M S F, U C D formal analysis, A S, J F O J, M C A, G B L, M S F, U C D., writing - preparation of original draft, A S, J F O J, M C A, G B L, M S F, U C D writing - proofreading and editing, A S, J F O J, M C A, G B L, M S F, U C D visualization, A S, J F O J, M C A, G B L, M S F, U C D supervision, A S, J F O J, M C A, G B L, M S F, U C D

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Availability of data and materials

This research was supported by the Universities and by the UFMS Air Quality Laboratory.

Abreu MC, Lyra GB, Oliveira-Júnior JF, Souza A, Pobocikova I, Fraga MS, Abreu RCR (2022) Temporal and spatial patterns of fire activity in three biomes of Brazil. Sci Total Environ 844:157138. https://doi.org/10.1016/j.scitotenv.2022.157138
Abreu MC, Souza A, Lyra GB, Pobocikova I, Cecílio RA (2020) Analysis of monthly and annual rainfall variability using linear models in the state of Mato Grosso do Sul, Midwest of Brazil. Int J Climatol 41(S1):E2445–E2461. https://doi.org/10.1002/joc.6857
Ali K, Inamdar SR, Beig G, Ghude S, Peshin S (2012) Surface ozone scenario at Pune and Delhi during the decade of 1990s. J Earth Syst Sci 121(2):373–383. https://doi.org/10.1007/s12040-012-0170-1
Alvares CA, Stape JL, Sentelhas PC, Gonçalves JLM, Sparovek G (2013) Köppen’s climate classification map for Brazil. Meteorol Z 22(6):711–728. https://doi.org/10.1127/0941-2948/2013/0507
Aragão L, Perez Guerrero JS, Pimentel LCG (2014) Simplified Modeling of Tropospheric Ozone Formation Considering Alternative Fuels Using. Anuário do Instituto de Geociências UFRJ 37(2):151–160. https://doi.org/10.11137/2014_2_151_160
Borge R, Requia WJ, Yagüe C, Jhun I, Koutrakis P (2019) Impact of weather changes on air quality and related mortality in Spain over a 25 year period (1993–2017). 105272 Environment International, 133 (part B). https://doiorg/10.1016/j.envint.2019.105272
Brasseur G, Hitchman MH (1988) Stratospheric response to trace gas perturbations: Changes in ozone and temperature distributions. Science 240(4852):634–637. https://doi.org/10.1126/science.240.4852.634
Bucsela EJ, Celarier EA, Wenig MO, Gleason JF, Veefkind JP, Boersma KF, Brinksma EJ (2006) Algorithm for NO₂ vertical column retrieval from the ozone monitoring instrument. IEEE Trans Geoscience Remote Sens 44(5):1245–1258. 10.1109/TGRS.2005.863715
Carvalho VSB, Martins FB, Silveira WW, Campos B, Simões JB (2020) Variance analysis applied to ground-level ozone concentrations in the state of São Paulo, Brazil. Braz J Chem Eng 37(8):505–513. https://doi.org/10.1007/s43153-020-00045-7
Castell-Balaguer N, Tellez L, Mantilla E (2012) Daily, seasonal and monthly variations in ozone levels recorded at the Turia river basin in Valencia (Eastern Spain). Environ Sci Pollution Res 19(8):3461–3480. https://doi.org/10.1007/s11356-012-0881-5
Cerri CC, Maia SMF, Galdos MV, Cerri CEP, Feigl BJ, Bernoux M (2009) Brazilian Greenhouse Gas Emissions: The importance of agriculture and livestock. SciAgric 66(6):831–843. https://doi.org/10.1590/S0103-90162009000600017
Chen B, Kan H (2008) Air pollution and population health: a global challenge. Environ Health Prev Med 13(2):94–101. https://doi.org/10.1007/s12199-007-0018-5
Chen J, Li C, Ristovski Z, Milic A, Gu Y, Islam MS, Dumka UC (2017) A review of biomass burning: Emissions and impacts on air quality, health and climate in China. Sci Total Environ 579:1000–1034. https://doi.org/10.1016/j.scitotenv.2016.11.025
Cunha ER, Santos CAG, Silva RM, Bacani VM, Pott A (2021) Future scenarios based on a CA-Markov land use and land cover simulation model for a tropical humid basin in the Cerrado/Atlantic forest ecotone of Brazil. Land Use Policy 101105141. https://doi.org/10.1016/j.landusepol.2020.105141
Damasceno-Junior GA, Pott A, Neves DRM, Sciamarelli A, Fina BG (2018) Flora lenhosa de Florestas estacionais do estado de Mato Grosso do Sul: estado da arte Iheringia, Série Botânica 73 (supl.), 65–79. https://doi.org/10.21826/2446-8231201873s65
Duncan BN, Yoshida Y, Olson JR, Sillman S, Martin RV, Lamsal L, Hu Y, Pickering KE, Retscher C, Allen DJ, Crawford JH (2010) Application of OMI observations to a space-based indicator of NOx and VOC controls on surface ozone formation. Atmos Environ 44(18):2213–2223. https://doi.org/10.1016/j.atmosenv.2010.03.010
Durigan G, Ratter JA (2015) The need for a consistent fire policy for Cerrado conservation. J Appl Ecol 53(1):11–15. https://doi.org/10.1111/1365-2664.12559
Ebi K, McGregor G (2008) Climate change, tropospheric ozone and particulate matter, and health impacts. Environ Health Perspective 14(16):1449–1455. https://doi.org/10.1590/S1413-81232009000600037
Fraga MS, Reis GB, Silva DD, Guedes HAS, Elesbon AAA (2020) Use of multivariate statistical methods to analyze the monitoring of surface water quality in the Doce River basin, Minas Gerais, Brazil. Environ Sci Pollution Res 27(28):35303–35318. https://doi.org/10.1007/s11356-020-09783-0
Fuhrer J, Booker F (2003) Ecological issues related to ozone: Agricultural issues. Environ Interntational 29(2–3):141–154. https://doi.org/10.1016/S0160-4120(02)00157-5
Ganguly ND, Joel V (2010) Long term trend, diurnal and seasonal variations of atmospheric ozone at Indian Antarctic station Maitri. J Earth Sci India 3(III):174–180
Gauss M, Ellingsen K, Isaksen ISA, Dentener FJ, Stevenson DS, Amann M, Cofala J (2007) Changes in nitrogen dioxide and ozone over southeast and east Asia between year 2000 and 2030 with fixed meteorology. Terr Atmospheric Ocean Sciecnes 18(3):475–492. 10.3319/TAO.2007.18.3.475(EA)
Grecchi RC, Gwyn QHJ, Bénié GB, Formaggio AR, Fahl FC (2014) Land use and land cover changes in the Brazilian Cerrado: A multidisciplinary approach to assess the impacts of agricultural expansion. Appl Geogr 55:300–312. https://doi.org/10.1016/j.apgeog.2014.09.014
Gunaseelan I, Bhaskar BV, Muthuchelian K (2014) The effect of aerosol optical depth on rainfall with reference to meteorology over metro cities in India. Environ Sci Pollution Res 21(13):8188–8197. https://doi.org/10.1007/s11356-014-2711-4
Hoffmann WA, Geiger EL, Gotsch SG, Rossatto DR, Silva LCR, Lau OL, Haridasan M, Franco AC (2012) Ecological thresholds at the savanna-forest boundary: howplant traits, resources and fire govern the distribution oftropical biomes. Ecol Lett 15:759–768. https://doi.org/10.1111/j.1461-0248.2012.01789.x
Hossain MS, Frey HC, Louie PKK, Lau AKH (2021) Combined effects of increased O₃ and reduced NO₂ concentrations on short-term air pollution health risks in Hong Kong. Environ Pollut 270:116280. https://doi.org/10.1016/j.envpol.2020.116280
IBGE (2010) - Censo Demográfico 2010. Available at: https://censo2010.ibge.gov.br/resultados.html. Access: 23/03/2022
Ito K, De Leon S, Samantha F, Lippmann M (2005) Associations between ozone and daily mortality: Analysis and meta-analysis. Epidemiology 16:446–457. https://doi.org/10.1097/01.ede.0000165821.90114.7f
José RS, Pérez JL, González RM, Palacios M (2017) Improving air quality modelling systems by using on-line wild land fire forecasting tools coupled into WRF/Chem simulations over Europe. Urban Clim 22:2–18. https://doi.org/10.1016/j.uclim.2016.09.001
Kalita G, Bhuyan PK(2011) Spatial heterogeneity in tropospheric column ozone over the Indian subcontinent: Long-term climatology and possible association with natural and anthropogenic activities. Advances in Meteorology, 2011, 1–12. https://doi.org/10.1155/2011/924516
Kirchhoff VWJH (1996) Increasing concentrations of CO and O3 rising deforestation rates and increasing troposheric carbon monoxide and ozone in Amazonia. Environ Sci & Pollut Res 3:210–212. https://doi.org/10.1007/BF02986961
Leal Filho W, Azeiteiro UM, Salvia AL, Fritzen B, Libonati R (2021) Fire in Paradise: Why the Pantanal is burning. Environ Sci Policy 123:31–34. https://doi.org/10.1016/j.envsci.2021.05.005
Lima M, Silva Junior CA, Pelissari TD, Lourençoni T, Luz IMS, Lopes FJA (2020) Sugarcane: Brazilian public policies threaten the Amazon and Pantanal biomes. Perspect Ecol Conserv 19(3):210–212. https://doi.org/10.1016/j.pecon.2020.06.002
Magalhães IB, Pereira ASAP, Calijuri ML, Alves SC, Santos VJ, Lorentz JF (2020) Brazilian Cerrado and Soy moratorium: Effects on Brazilian Cerrado and Soy moratorium: Effects on biome preservation and consequences on grain production. Land Use Policy 99:105030. https://doi.org/10.1016/j.landusepol.2020.105030
Manrique-Pineda DA, Souza EB, Paranhos-Filho AC, Encina CCC, Damasceno-Junior GA (2021) Fire, flood and monodominance of Tabebuia aurea in Pantanal. For Ecol Manag 479:118599. https://doi.org/10.1016/j.foreco.2020.118599
Marengo JA, Cunha AP, Cuartas LA, Deusdará L, Broedel E, Seluchi ME, Michelin CM, Baião CFP, Angulo EE, Almeida EK, Kazmierczak ML, Mateus NPA, Silva RC, Bender F (2021) Extreme Drought in the Brazilian Pantanal in 2019–2020: Characterization, Causes, and Impacts. Front Water 3:639204. 10.3389/frwa.2021.639204
Matandirotya NR, Burger RP (2021) Spatiotemporal variability of tropospheric NO₂ over four megacities in Southern Africa: Implications for transboundary regional air pollution. Environ Challenges 5:100271. 10.1016/j.envc.2021.100271
Miles GM, Siddans R, Kerridge BJ, Latter BG, Richards NAD (2015) Tropospheric ozone and ozone profiles retrieved from GOME-2 and their validation. Atmos Meas Tech 8(1):385–398. https://doi.org/10.5194/amt-8-385-2015
Miranda HS, Sato MN, Neto WN, Aires FS (2009) Fires in the Cerrado, the Brazilian savanna. In: Cochrane MA (ed) Tropical Fire Ecology: Climate Change, Land Use and Ecosystem Dynamics. Springer-Praxis, Heidelberg, Germany, pp 427–450
Mota PHS, Rocha SJSS, Castro NLM, Marcatti GE, França LCJ, Schettini BLS, Villanova PH, Santos HT, Santos AR (2019) Forest fire hazard zoning in Mato Grosso State, Brazil. Land Use Policy 88:104206. https://doi.org/10.1016/j.landusepol.2019.104206
Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GA, Kent J (2000) Biodiversity hotspots for conservation priorities. Nature 403(6772):853–858. https://doi.org/10.1038/35002501
Ndibe C, Spörl R, Maier J, Scheffknecht G (2013) Experimental study of NO and NO₂ formation in a PF oxy-fuel firing system. Fuel 107:749–756. https://doi.org/10.1016/j.fuel.2013.01.055
Newberry BM, Power CR, Abreu R, Durigan G, Rossatto DR, Hoffmann WA (2020) Flammability thresholds or flammability gradients? Determinants of fire across savanna-forest transitions. New Phytol 228(3):910–921. https://doi.org/10.1111/nph.16742
Oliveira-Júnior JF, Teodoro PE, Silva-Junior CA, Baio FHR, Gava R, Capristo-Silva GF, Gois G, Correia Filho WLF, Lima M, Santiago DB, Freitas WK, Santos PJ, Costa MS (2020) Fire foci related to rainfall and biomes of the state of Mato Grosso do Sul, Brazil. Agric For Meteorol 282–283. https://doi.org/10.1016/j.agrformet.2019.107861
Oliveira-Júnior JF, Silva Junior CA, Teodoro PE, Rossi FS, Blanco CJC, Lima M, Gois G, Correia Filho WLF, Barros Santiago D, Vanderley MHGS (2021) Confronting CHIRPS dataset and in situ stations in the detection of wet and drought conditions in the Brazilian Midwest. Int J Climatol 41:4478–4493. https://doi.org/10.1002/joc.7080
Pancholi P, Kumar A, Bikundia DS, Chourasiya S (2018) An observation of seasonal and diurnal behavior of O₃–NO x relationships and local/regional oxidant (OX = O₃ + NO₂) levels at a semi-arid urban site of western India. Sustainable Environ Res 28(2):79–89. https://doi.org/10.1016/j.serj.2017.11.001
Pobocikova I, Souza A, Abreu MC, Oliveira-Júnior JF, Santos CM, Lins TMP, Aristone F, Ramos PL (2021) The impacts of climate change on rainfall modeling in the Pantanal of Mato Grosso Do Sul, vol V43. Acta Scientiarum-Technology, pp P1–19. https://doi.org/10.4025/actascitechnol.v43i1.55112
R Team Core (2018) R: A language and environment for statistical computing
Reis CJ, Souza A, Graf R, Kossowski TM, Abreu MC, Oliveira-Júnior JF, Fernandes WA (2022) Modeling of the air temperature using the Extreme Value Theory for selected biomes in Mato Grosso do Sul (Brazil). Stoch Environ Res Risk Assess 22:1. https://doi.org/10.1007/s00477-022-02206-1
Ribeiro JF, Walter BM (1998) Fitofisionomias do bioma CerradoIn. In: Almeida SM (ed) Cerrado: ambiente flora. Sano, pp 87–166
Schaub M, Skelly J, Zhang J, Ferdinand J, Savage J, Stevenson R, Davis D, Steiner K (2005) Physiological and foliar symptom response in the crowns of Prunus serotina, Fraxinus americana and Acer rubrum canopy trees to ambient ozone under forest conditions. Environ Pollution 133(3):553–567. https://doi.org/10.1016/j.envpol.2004.06.012
Schulz C, Whitney BS, Rossetto OC, Neves DM, Crabb L, Oliveira EC, Lima PLT, Afzal M, Laing AL, Fernandes LCS, Silva CA, Steinke VA, Saito CH (2019) Physical, ecological and human dimensions of environmental change in Brazil’s Pantanal wetland: Synthesis and research agenda. Sci Total Environ 678:1011–1027. https://doi.org/10.1016/j.scitotenv.2019.06.023
Seinfeld HJ, Pandis SN (2006) Atmospheric chemistry and physics from air pollution to climate change. John Wiley and Sons, Hoboken, NJ
Sharma RK, Thanu TC, Elampari K (2013) Surface ozone measurements in southernmost tip of peninsular India. Univers J Environ Res Technol 3(2):255–265
Sheel V, Lal S, Richter A, Burrows JP (2010) Comparison of satellite observed tropospheric NO₂ over India with model simulations. Atmos Environ 44(27):3314–3321. https://doi.org/10.1016/j.atmosenv.2010.05.043
Siciliano B, Dantas G, Silva CM, Arbilla G (2020) Increased ozone levels during the COVID-19 lockdown: Analysis for the city of Rio de Janeiro, Brazil. Sci Total Environ 139765. doi: 10.1016/j.scitotenv.2020.1397
Silva Junior CA, Teodoro PE, Delgado RC, Ribeiro LP, Lima M, Pantaleão AA, Baio FHR, Azevedo GB, Azevedo GTOS, Capristo-Silva GF, Arvor D, Facco CU(2020) Persistent fire foci in all biomes undermine the Paris Agreement in Brazil. Sci Rep 10, 16246 (2020). https://doi.org/10.1038/s41598-020-72571-w
Silveira VR, Oliveira Júnior JF, Silva MS, Silva C, Alves AR, Souza Pontes A, Pimentel LCG, Rotunno Filho OC (2020) Analysis of urban - industrial expansion and increasing level of ozone concentration as subsiding an environmental management plan for the east of Rio de Janeiro metropolitan area – Brazil. Land Use Policy 101:105148. https://doi.org/10.1016/j.landusepol.2020.105148
Souza A, Schujmann E, Fachel JMG, Fernandes WA(2013) Indicadores Ambientais e Doenças Respiratórias em CriançasMercator (FortalezaOnline), 12(27), 101–109
Souza JRCM, Shimbo JZ, Rosa MR, Parente LL, Alencar AA, Rudorff BF, Thasenack H, Matsumoto MG, Ferreira L, Souza-Filho PWM, de Oliveira SW, Rocha WF, Fonseca AV, Marques CB, Diniz CG, Costa D, Monteiro D, Rosa ER, Vélez-Martin E, Weber EJ, Lenti FEB, Paternost FF, Pareyn FG, Siqueira JV, Viera JL, Ferreira Neto LC, Saraiva MM, Sales MH, Salgado MPG, Vasconcelos R, Galano S, Mesquita VV, Azevedo T (2020b) Reconstructing Three Decades of Land Use and Land Cover Changes in Brazilian Biomes with Landsat Archive and Earth Engine. Remote Sens 12:2735. https://doi.org/10.3390/rs12172735
Souza A, Santos DAS (2018) Análise das componentes principais no processo de monitoramento ambiental. Nativa 6(6):639–647. https://doi.org/10.31413/nativa.v6i6.6453
Souza A, Ozonur D(2019) Statistical Behavior of O₃, O_X, NO, NO₂, and NO_x in Urban Environment. Ozone: Science & Engineering, 42(1), 1–13. https://doi.org/10.1080/01919512.2019.1602468
Souza A, Aristone F, Abreu MC, Oliveira-Júnior JF, Fernandes W, Pobocikova I (2022) Spatio-temporal variations of tropospheric nitrogen dioxide in South Mato Grosso based on remote sensing by satellite. Meteorol Atmos Phys 134:19. https://doi.org/10.1007/s00703-021-00855-5
Souza A, Aristone F, Fernandes WA, Oliveira APG, Olaofe Z, Abreu MC, Pobocikova I(2020a) Analysis of Ozone Concentrations Using Probability Distributions. Ozone: Science & Engineering, 42(6), 539–550. 10.1080/01919512.2020.1736987
Souza A, Fernandes WA, Pavao H, Lastoria G, Albrez EA (2012) Potential impacts of climate variability on respiratory morbidity in children, infants, and adults. Jornal Brasileiro De Pneumologia 38(6):708–715. https://doi.org/10.1590/S1806-37132012000600005
Souza A, Santos DAS, Kofanovski AZ, Kovač-Andrić I (2017) Condições climáticas e o risco de exacerbação da asma. Revista Saúde (Santa Maria) 43(1):244–253. https://Doi.Org/10.5902/2236583423462
Souza A, Abreu MC, Oliveira-Júnior JF, Aristone F, Fernandes WA, Aviv-Sharon E, Graf R (2022) Climate regionalization in Mato Grosso do Sul: a combination of hierarchical and non-hierarchical clustering analyses based on precipitation and temperature. Brazilian Archives of Biology and Technology 65:e22210331. https://doi.org/10.1590/1678-4324-2022210331
Souza A, dos Santos CM, Ihaddadene R, Cavazzana G, Abreu MC, Oliveira-Júnior JF, Pobocikova I, de Gois G, Lins TMP (2021) Analysis of extreme monthly and annual air temperatures variability using regression model in Mato Grosso do Sul, Brazil. Model Earth Syst Environmen 8:647–663. https://doi.org/10.1007/s40808-021-01096-6
Távora Maia RG, Bozelli H (2022) The importance of GHG emissions from land use change for biofuels in Brazil: An assessment for current and 2030 scenarios. Resour Conserv Recycl 179:106131. https://doi.org/10.1016/j.resconrec.2021.106131
Thompson ML, Reynolds J, Cox LH, Guttorp P, Sampson PD (2001) A review of statistical methods for the meteorological adjustment of tropospheric ozone. Atmos Environ 35(3):617–630. https://doi.org/10.1016/S1352-2310(00)00261-2
Tiwari S, Dahiya A, Kumar N (2015) Investigation into relationships among NO, NO₂, NO_x, O₃, and CO at an urban background site in Delhi, India. Atmos Res 157:119–126. https://doi.org/10.1016/j.atmosres.2015.01.00
Vasques TV, Hoinaski K(2021) Brazilian vehicular emission inventory software – BRAVES. Transportation Research Part D: Transport and Environment, 100, 103041. https://doi.org/10.1016/j.trd.2021.103041
Wang Z, Lv J, Tan Y, Guo M, Gu Y, Xu S, Zhou Y (2019) Temporospatial variations and Spearman correlation analysis of ozone concentrations to nitrogen dioxide, sulfur dioxide, particulate matters and carbon monoxide in ambient air, China. Atmospheric Pollution Research 10(4):1203–1210. https://doi.org/10.1016/j.apr.2019.02.003
Watson L, Lacressonnière G, Gauss M, Engardt M, Andersson C, Josse B, Vautard R (2016) Impact of emissions and + 2°C climate change upon future ozone and nitrogen dioxide over Europe. Atmos Environ 142:271–285. https://doi.org/10.1016/j.atmosenv.2016.07.05
Yienger JJ, Levy H (1995) Empirical model of global soil biogenic NOX emissions. J Geophys Research-Atmospheres 100(D6):11447–11464. https://doi.org/10.1029/95JD00370
Zhang L, Liu Y, Zhao F (2018) Singular value decomposition analysis of spatial relationships between monthly weather and air pollution index in China. Stoch Environ Res Risk Assess 32:733–748. https://doi.org/10.1007/s00477-017-1491-z
Zoran MA, Savastru RS, Savastru DM, Tautan MN (2020) Assessing the relationship between ground levels of ozone (O₃) and nitrogen dioxide (NO₂) with coronavirus (COVID-19) in Milan, Italy. Sci Total Environ 740:140005. https://doi.org/10.1016/j.scitotenv.2020.1400

Table 1 Comparison of means for the variables NO₂, O₃, and fire foci occurrence in month/seasonal and biomes od Mato Grosso do Sul state.

This means followed by the same capital letter in the column (biomes) and lowercase letter in line (month or season), for each variable, do not differ among themselves by Scott-Knott test (5% probability). The numbers and letters in red indicate no interaction between the two factors (biomes and months/season). The letters (A or B, a or b) were set up in descending order.

Months/period

NO₂ (molec.cm^-2)

O₃ (DU)

Burned area (ha)

Pantanal

Cerrado

Atlantic Forest

Average

Pantanal

Cerrado

Atlantic Forest

Average

Pantanal

Cerrado

Atlantic Forest

Average

Jan

2.63Bd

2.76Ae

2.88Ac

2.76e

258.10Bc

344.62Ac

258.65Bd

287.13e

9240.82cA

386.70dB

257.14cC

3294.89e

Feb

2.55Bd

2.76Ae

2.84Ac

2.72e

257.02Bc

343.05Ac

256.69Bd

285.59e

4056.12cA

950.83dA

522.15cB

1843.03e

Mar

2.52Bd

2.71Ae

2.73Ad

2.66e

252.25Bd

332.88Ac

254.96Bd

283.95f

2834.76cA

1657.44cA

143.39cB

1545.19e

Apr

2.44Bd

2.61Af

2.70Ad

2.58f

252.01Bd

335.58Ad

253.28Bd

280.32g

4682.13cA

2231.72cA

892.60bB

2602.15d

May

2.45Ad

2.56Af

2.57Ad

2.53f

252.25Bd

332.88Ad

254.96Bd

280.03g

7721.4cA

1937.99cA

275.37cB

3311.59d

Jun

2.58Ad

2.63Af

2.62Ad

2.61f

255.83Bd

337.06Ad

257.33Bd

283.41f

6470.19cA

6118.57bA

818.61bB

4469.12d

Jul

3.12Ab

2.99Ad

3.03Ac

3.04c

260.98Bc

345.05Ac

263.84Bc

289.96d

27944.97bA

9610.75aA

447.96aB

12667.89b

Aug

4.15Aa

3.42Cb

3.72Ba

3.77a

267.66Bb

353.90Ab

271.24Ba

297.60b

144224.95aA

20945.03aB

1019.66aC

55396.55a

Sep

4.14Aa

3.65Ba

3.56Ba

3.78a

273.38Ba

361.50Aa

274.78Ba

303.22a

223832.84aA

13961.02aB

1082.48aC

79625.45a

Oct

3.23Ab

3.26Ac

3.24Ab

3.24b

271.56Ba

360.60Aa

271.86Ba

301.34a

115057.53aA

7846.67bB

406.57bC

41103.59b

Nov

3.01Bc

3.14Bc

3.27Bb

3.14b

266.11Bb

354.41Ab

267.07Bb

295.87b

56989.02bA

1532.43cB

925.98bC

19815.81c

Dec

2.84Ac

2.94Ad

2.99Ac

2.92d

262.85Bb

350.22Ab

262.91Bc

291.99c

19933.44cA

816.57dB

415.75cC

7055.25e

Average

2.97A

2.95A

3.01A

261.08B

346.63A

262.39B

51915.68A

5666.31B

600.64C

Dry period

3.15aA

2.98aB

3.03aB

3.05a

260.36Ba

344.33Aa

263.57Ba

289.09a

59652.55a

8780.60a

750.54a

23061.23a

Rainy period

2.79bB

2.93aA

2.99aA

2.91b

261.81Ba

348.92Ab

262.20Ba

290.98a

29962.68b

1464.00b

371.88b

10599.52b

Table 2 Kaiser-Meyer-Olkin (KMO) and Bartlett's sphericity test to evaluate principal component analysis as a statistical approach to NO₂, O₃, fire, rainfall, and temperature analysis.

Biome	KMO	Bartlett test
Atlantic Forest	0.64	56.28*
Cerrado	0.68	60.71*
Pantanal	0.65	58.00*

*Significant in Bartlett's sphericity test 5% error probability.

Table 3 Proportion is explained by the main components in each biome.

Atlantic Forest	PC1	PC2	PC3	PC4
Standard deviation	1.522	1.148	0.795	0.676
Proportion of variance	0.464	0.264	0.126	0.091
Cumulative proportion	0.464	0.727	0.853	0.945
Cerrado	PC1	PC2	PC3	PC4
Standard deviation	1.613	1.163	0.655	0.571
Proportion of variance	0.520	0.271	0.086	0.065
Cumulative proportion	0.520	0.791	0.876	0.942
Pantanal	PC1	PC2	PC3	PC4
Standard deviation	1.605	1.151	0.736	0.559
Proportion of variance	0.515	0.265	0.108	0.062
Cumulative proportion	0.515	0.780	0.889	0.951

Table 4 The eigenvalue for principal components analysis in each biome.

Variables	Atlantic Forest		Cerrado		Pantanal
Variables	PC1	PC2	PC1	PC2	PC1	PC2
NO₂	-0.67	-0.01	0.58	-0.09	0.54	-0.21
O₃	-0.63	-0.03	0.56	0.14	0.49	0.15
Fire	-0.10	-0.28	0.32	-0.55	0.52	-0.08
Rainfall	0.22	0.70	-0.10	0.68	-0.19	0.78
Temperature	-0.32	0.66	0.49	0.45	0.41	0.57

Table 5 Land use and land cover areas (ha) in Atlantic Forest, Cerrado and Pantanal, in 2005 and 2020, and area variation (%) between 2005 and 2020, in Mato Grosso do Sul.

LULC/Biomes	Atlantic Forest area (ha)		Cerrado area (ha)		Pantanal area (ha)		Atlantic Forest	Cerrado	Pantanal
LULC/Biomes	2005	2020	2005	2020	2005	2020	Area variation (%) between 2005-2020
Agriculture	290394.37	1100092.35	1039701.32	2729805.75	7371.8	9887.21	73.60	61.91	25.44
Forest Plantation	304.09	4789.68	97748.56	801233.1	10.73	NA	93.65	87.80	NA
MAP	340300.38	440205.41	1255380.69	1876718.5	77.7	116.52	22.70	33.11	33.32
Pasture	2148529.77	1161700.35	12981843.28	11097263.22	1078191.16	1611523.85	-84.95	-16.98	33.09
Mining	NA	NA	98.37	163.08	8.97	15.23	NA	39.68	41.10
Other Non Vegetated Area	1409.03	4633.86	11881.56	17860.56	NA	NA	69.59	33.48	NA
Urban Infrastructure	13584.52	17427.6	47172.97	61051	3227.45	3456.44	22.05	22.73	6.63
Savanna Formation	75.6	52.93	2486056.61	1723630.87	1424108.41	1240863.19	-42.83	-44.23	-14.77
Forest Formation	364261.67	367041.58	3617131.65	3238164.64	1967538.55	1750830.11	0.76	-11.70	-12.38
Grassland	29.22	46.43	90692.93	120093.47	3100570.21	4303819.63	37.07	24.48	27.96
Wetland	413657.52	398208.3	530190.07	495472.03	1432090.86	528413.76	-3.88	-7.01	-171.02
Non Observed	0.16	NA	8.49	11.23	154.83	166.86	NA	24.40	NA

No competing interests reported.

Total column ozone (O 3 ) and nitrogen dioxide (NO 2 ) via satellite and their relationship with the burned area and climatic data in biomes of Central-West Brazil

Status:

Version 1

Abstract

Figures

1. Introduction

2. Material And Methods

2.1. Study area

2.2. The biomes

2.2.1. Cerrado

2.2.2. Pantanal

2.2.3. Atlantic forest

2.3. Data

2.4. Statistical analysis

3. Results

4. Discussion

5. Conclusions

Declarations

References

Tables

Additional Declarations

Status:

Version 1