1 Introduction

Air quality is a critical environmental factor that profoundly impacts public health, ecosystems, and overall quality of life over some time [1]. Nigeria's air quality landscape has witnessed dynamic shifts due to factors such as urbanization, industrialization, and increased vehicular traffic, as these changes have resulted in the alteration of pollutant emissions and the composition of the atmosphere [2, 3]. The World Health Organization [4] stated that “poor air quality has increased to over 5 million premature deaths annually in 2019 from the increased burden of stroke, heart diseases, lung cancer, neurological damage, chronic to acute respiratory diseases such as asthma”, as limited monitoring and evaluations are conducted to mitigate how air emissions in certain locations using sustainable green practices [5]. The burden of diseases associated with poor air quality is increasing in Nigeria, a rapidly developing nation that has experienced significant changes in its atmospheric composition over the years such as acid rain, increasing temperature, desertification, particulates and black carbon soot [6,7,8,9].

Air pollution has been defined by numerous entities as the release of harmful gases and/or particulate matter into the atmosphere from natural and anthropogenic sources that tend to interact biochemically to form byproducts that can cause harm to the organisms and the environment [10,11,12]. As defined by the U.S. Environmental Protection Agency [3] and the World Health Organization [2] several air pollutants have been categorized such as nitrogen dioxide (NO2), Sulphur dioxide (SO2), Carbon monoxide (CO), Ozone (O3), Particulate matter (PM) and Lead which are known to cause health and environmental issues over a time [11, 13, 14]

The measurement of air pollutants has been used for quantitative and qualitative modes using autosamplers (analyzers, drones, satellites, and airplanes) and offline laboratory-based analysis, which has immensely improved remote monitoring over time [15, 16]. The use of sophisticated software, computational algorithms and models has made it easy for the chemical quantification of these air pollutants and as such provides crucial information about pollution source, transportation, distribution, and impact over a period of time through remote sensing and aids in improving effective air quality management and mitigation strategies for government and policymakers [17,18,19,20,21]. Currently in Nigeria, there is increasing use of remote sensing for air quality monitoring such as SO2 [22, 23]; AI [24,25,26]; NO2 [27]; temperature [28]; relative humidity [29]; land use changes and desertification [30, 31] and particulate matter [32], which has focused on a short term focus, providing spatial resolution of satellite temporal data and limited focus on health effect and large area of coverage in Nigeria.

Therefore, the study seeks to delve into long-term trends of these specific atmospheric constituents such as O3, NO2, HCHO, SO2, CO, AI at 340/380 nm and 354/388 nm), temperature and their implications for air quality in Nigeria from April 2018 till December 2022 using Sentinel 5 Precursor and 3A/B data via EO browser. We will explore the atmospheric data resources, statistical and implication, related to these pollutants. By examining Nigeria's evolving air quality scenario, we can gain valuable insights into the challenges and opportunities for sustainable development and environmental preservation in the region.

2 Materials and methods

2.1 Study area and climate

Nigeria is situated in West Africa and is bordered by the Gulf of Guinea to the south, Benin to the west, Niger to the north, and Chad and Cameroon to the east (See Fig. 1). The country comprises Northern Nigeria and Southern Nigeria, which includes 36 states, including the capital city, Abuja. Nigeria, with a population exceeding 211 million people in 2022, ranks as the seventh most populous country globally [33]. The nation is divided into six geopolitical zones, with the North-West having the highest population and the South-East having the lowest. Nigeria experiences rapid population growth, with an annual growth rate of 2.6%, and it is projected to reach 733 million by 2100 [34, 35], as this growth presents challenges such as increased demand for resources and pressure on social services. Southern Nigeria is primarily driven by oil and gas production, petrochemical, transportation, and commercial activities as compared to Northern Nigeria, which is driven by agriculture and transportation, although this trend is changing with the discovery of oil and gas in Northeastern states of Borno, Bauchi, Gombe) in 2022 [35, 36]. Despite its strengths, the region faces challenges, including poverty, infrastructure deficiencies, and occasional ethnic and religious conflicts. The climate of Nigeria is characterized by wet and dry seasons as the wet season spans from April to October, while the dry season occurs from November to March. The climatic condition is generally warm and humid, with average temperatures ranging from 25 to 35 °C depending on location, as rainfall varies across the country, with the southern regions receiving more rainfall than the north.

Fig. 1
figure 1

Map of Nigeria Topography with the population density distribution

Full size image

2.2 Data collection

Sentinel-5P (Sentinel-5 Precursor) was launched on 13 October 2017 by the European Copernicus mission satellite service to monitor the Earth's atmosphere [37]. It carries an advanced multi-spectral imaging spectroscopic instrument called TROPOMI (TROPOspheric Monitoring Instrument), which is a nadir-viewing 108° field-of-view covering a wavelength for ultraviolet–visible (UV–Vis at 270 nm to 495 nm), near-infrared (NIR at 675 to 775 nm) and Short-wave infrared (SWIR at 2305 to 2385 nm) spectrometers; capable of measuring a wide range of atmospheric gases and pollutants. TROPOMI maps the earth’s atmosphere daily with a resolution of 7 km × 3.5 km which is more sophisticated than other monitoring instruments such as OMI (Ozone Monitoring Instrument) at 24 km × 13 km; GOME2 (Global Ozone Monitoring Experiment-2) at 80 km × 40 km and SCIAMACHY (SCanning Imaging Absorption spectroMeter for Atmospheric CartograpHY) at 200 km × 30 km for detection of gases such as NO2; SO2; O3; HCHO; CH4; CO and aerosol optical depth at high spatial, temporal and spectral resolutions [38,39,40,41], which can be viewed via three data mode: near real-time stream (NRTS) available within 3 h after data acquisition while, offline stream (OFS) and reprocessed stream (RPS) are available within a few days after data processing respectively [42,43,44,45].

Sentinel 3A/B was launched on 16 February 2016 and 25 April 2018 taken into orbit from the Plesetsk Cosmodrome in Northern Russia by the European Space Agency (ESA) [46]. The satellite comprises multiple sensing instruments for SLSTR (Sea and Land Surface Temperature Radiometer), OLCI (Ocean and Land colour instrument), SRAL (SAR Altimeter), DORIS (Doppler Orbitography and Radio positioning Integrated by Satellite) and MWR (Microwave Radiometer) [46]. These optical instruments orbit at a repeat cycle of 27 days with inclination at 98.65°, orbit height and type at 814.5 km and sun-synchronous. The Sea and Land Surface Temperature (SLSTR) instrument on board Sentinel-3 measures the global and regional sea and land surface temperature that covers the spatial resolution of 500 m × 500 m for visible, near- and short-wave infrared (0.555, 0.659, 0.865, 1.375, 1.61 and 2.25 µm), and 1 km × 1 km for thermal infrared wavelengths (3.74, 10.85 and 12 μm) of the electromagnetic spectrum. Land Surface Temperature (LST) data was accessed from Level 2 data of SLSTR in the EO Browser platform with a revisit time of > 23 h on the same area of interest [47].

2.3 Quality assurance

For this research, the Sentinel 3A/B and Sentinel 5P TROPOMI offline stream (OFS and reprocessed stream (RPS) data set were processed and visualized using EO Browser (https://apps.sentinel-hub.com/eo-browser/) and Copernicus Open Access Hub (https://scihub.copernicus.eu/dhus/#/home) for April 2018 to December 2022 for Nigeria. The Sentinel 5P and 3A/B offline data (Table 1) were subjected to quality assurance (qa_value) by analyzing each grid cell dependent on the area of interest based on latitude/longitude WGS84 projection, cloud cover ≤ 50, surface interference and geometry and physical assessment (screen poor quality pixel) [48,49,50]. The physical data variables were determined from mean pixel data of all locations and errors and problematic retrievals were filtered out following Copernicus specifications at qa_value > 0.75 (NO2) and qa_value > 50 (other product for a continuous range of 0 (poor data) to 1 (good data), as we used 75% for NO2 and 50% for other products [51].

Table 1 Dataset used in the study
Full size table

2.4 Data analysis

Several mapping, statistical and graphical analyses were conducted using ArcGIS Pro 2.5; Microsoft Excel v2019 and Paleontological Data v4.12b software from the number of datasets available [54,55,56], which are:

1. Mean (Average): The mean is calculated by summing up all the values in a dataset and dividing by the number of values (n).

$$ Mean \left( {\overline{x}} \right) = \frac{{\left( {\Sigma x} \right)}}{n} $$
(1)

where:

x̄ represents the mean.

Σx represents the sum of all values.

n is the number of values in the dataset.

2. Minimum (Min): The minimum is the smallest value in a dataset.

3. Maximum (Max): The maximum is the largest value in a dataset.

4. Standard Deviation: The standard deviation measures the dispersion or spread of data points in a dataset. It quantifies how much individual data points deviate from the mean:

$$ Standard Deviation \left( {SD} \right) = \sqrt {\left( {\frac{{\Sigma \left( {x - x^{ - } } \right)) }}{{\left( {n - 1} \right)}}} \right)} $$
(2)

where:

SD represents the standard deviation.

Σ(x-x−)2 is the sum of the squared differences between each data point (x) and the mean \(\left( {x ^{ - } } \right)\).

n is the number of values in the dataset.

5. Pearson Correlation: Pearson correlation measures the strength and direction of the linear relationship between two continuous variables [55]. It quantifies how well the variation in one variable can be predicted from the variation in another variable. The Pearson correlation coefficient ® ranges from -1 to 1, where -1 indicates a perfect negative linear relationship; 0 indicates no linear relationship; + 1 indicates a perfect positive linear relationship.

6. Factor analysis (FA): is a useful statistical tool for analysing shared variance among all the variables that typically begins with data collection, where information on several variables are interconnected into factor groups with a correlation matrix computed to understand the relationships between pairs of variables [56]. These factor groups could have different rotation mechanism (varimax or non-varimax rotation (promax) used to reposition the factors into clearer separation of variables with high loading for easier interpretation. FA has the ability to unveil patterns in a complex datasets.

7. Graphical representation: Data sets were analyzed using Excel and PAST Software such as line with markers, dendrogram clustering and factor analysis scatter plots.

3 Result

3.1 Atmospheric parameters assessment

Table 2 and Fig. 2 show various atmospheric parameters (NO2, SO2, O3, HCHO, CH4, CO, AI at 340-380 nm, AI at 354–388 nm) for different periods from April 2018 to December 2022 assessed using Sentinel 5P TROPOMI offline dataset, which was analyzed for mean (average) values and their corresponding ranges in tandem with a standard deviation that indicates the degree of variability or dispersion around the mean value offering insights into the data consistency. The mean NO2 (nitrogen dioxide) levels remain relatively stable over the years, with a slight decrease from 2018 to 2022, with small variations having 1.57E-05 and 1.48E-05 mol/m2. The mean SO2 (sulphur dioxide) concentration exhibits minimal changes over 5.71E-05–1.11E-05 mol/m2, with a low standard deviation from 2018 and 2022. The Ozone (O3) shows slight fluctuations with a mean value with minor fluctuations to 0.12 mol/m2, while ranges are consistent with and stable. Formaldehyde (HCHO) concentration shows modest variations with consistent measurements having mean range concentration at 1.15E-04 mol/m2 and 1.80E-05 mol/m2 for 2020 and 2018 with 2019 and 2022 having 1.20E-04 mol/m2 respectively. The CH4 mean concentration levels, measured in parts per billion (ppb) steadily increased from 1860.61 ppb to 1914.84 ppb from 2018 to 2022. The mean CO concentrations exhibit a gradual increase from 2018 (3.75E-02 mol/m2) that peaked in 2020 (4.39E-02 mol/m2), thereafter decreasing to 2022 (4.19E-02 mol/m2), while minimum and maximum values reflect similar downward trend. The aerosol optical indices (AI) at 340–380 nm and 354–388 nm (having unitless measurement) showed minor decreasing fluctuations, with relatively consistent measurements over the years with minimum and maximum values of 0.84 (year 2022)–1.12 (year 2018 and 2020) and 0.91 (year 2022)–1.18 (year 2020), as AI at 340–380 nm and 354–388 nm showed strong correlation between each other at 1.0 indicating consistent and close relationship regarding aerosol presence and satellite spectrometric detection at close nm ranges. The mean daytime temperature for these five years shows some variability but generally hovers around the high 30 °C, with the highest mean in 2021 at 39.48 °C and the lowest in 2020 at 38.45 °C, while the minimum and maximum range from 28.04 °C in 2020 to 30.96 °C in 2021 and 46.55 °C in 2018 to 47.85 °C in 2021. Nighttime temperatures are generally cooler than daytime temperatures, as expected, the mean nighttime temperature remains relatively stable over these years, ranging from 21.63 °C in 2022 to 22.24 °C in 2021, while the minimum and maximum temperatures ranged from 16.67 °C in 2018 to 18.85 °C in 2021 and 25.24 °C in 2022 to 25.76 °C in 2020 respectively. The entire atmospheric parameter showed that NO2, SO2, O3, and HCHO show relatively stable levels over the years with minimal fluctuations, whereas CH4 and CO exhibit slight fluctuations over time, which could be due to emission reductions or photochemical changes in atmospheric chemistry to CO2 as a stable component. Aerosol optical indices remain relatively consistent, indicating consistent aerosol presence influenced by local emissions, weather conditions and atmospheric processes, as daytime temperature during the annual regime was consistently high compared to nighttime temperature, which was mildly cooler and stable.

Table 2 Descriptive Statistics of Atmospheric Parameters for Nigeria from 2018 till 2022–
Full size table
Fig. 2
figure 2

Changes from year of NO2; SO2; O3; HCHO; CH4; CO; Aerosol Index at 340–380 nm and Aerosol Index at 354–388 nm

Full size image

Figure 3 shows the monthly trend of atmospheric parameters (NO2, SO2, O3, HCHO, CH4, CO, AI@340-380 nm, AI@354-388 nm) from 2018 to 2022, which produced several trends, variations, and possible atmospheric interpretations over the period analyzed. NO2 level showed variations with higher values observed during the cooler months (November to February) and lower values in the warmer months (June to September) as mean concentration were 1.57E-05 mol/m2; 1.55E-05 mol/m2; 1.52E-05 mol/m2; 1.53E-05 mol/m2; 1.48E-05 mol/m2 via 2018 (April-December to 2019–2022 (January–December) respectively. SO2 levels also exhibited occasional fluctuations with mean concentrations from 2018 to 2022 as 5.71E-05 mol/m2; 4.74E-06 mol/m2; 7.11E-06 mol/m2; 9.86E-06 mol/m2 and 1.11E-05 mol/m2, which were relatively stable wherein there was no clear seasonal pattern, although the various occasional spikes could be due to wind pattern, industrial to transboundary atmospheric releases. O3 levels were relatively stable throughout the year (2018–2022) averaging via 0.122 mol/m2; 0.118 mol/m2; 0.121 mol/m2; 0.121 mol/m2 and 0.122 mol/m2 with O3 levels higher peak warm months (June–September) and lower for cooler months (October–May), which is a typical pattern due to the increased photochemical production of O3 in warm, sunny conditions. HCHO levels followed a seasonal pattern, with higher values during the cooler months (June–September) and lower values during the warmer months (December–April) influenced by temperature, atmospheric chemistry of formaldehyde and other anthropogenic release sources, as the cumulative mean concentration for 2018 to 2022 are 1.80E-04 mol/m2; 1.20E-04 mol/m2; 1.15E-04 mol/m2; 1.21E-04 mol/m2 and 1.20E-04 mol/m2. CH4 increased steadily as the year progressed from 2018 till 2022 was consistent with undulating monthly increase (February–May; September–November) and decrease (June–August; December-January) seen via the line chart in Fig. 3 due to locational variation, atmospheric dynamics and transboundary emissions, as average CH4 concentrations are 1860.61 ppb; 1871.36 ppb; 1884.81 ppb; 1902.13 ppb and 1914.84 ppb for 2018 to 2022. CO levels showed some seasonal variations as decreasing (April–September) and increasing (October–March) were consistent across all years (2018–2022) analyzed averaging at 3.75E-02 mol/m2; 4.19E-02 mol/m2; 4.39E-02 mol/m2; 4.31E-02 mol/m2 and 4.19E-02 mol/m2 that could be associated to wet and dry influences from emission to ionic conversion to CO2 in the atmosphere. The aerosol indices (340–380 nm and 354–388 nm) vary over time with similar seasonal patterns showing decreased monthly regime from May–August and increased monthly trend from September–April ranging between 1.01 and 1.23; 0.95 and 1.28; 0.84 to 1.24; 0.72 to 1.14 and 0.57 to 1.38 via 2018 to 2022. Daytime and nighttime temperatures in Nigeria can have significant impacts on human health and climate, as the daytime temperatures especially during April to October, are relatively high, reaching as high as 47.85 °C in April 2021 and vice versa in August 2019 having 29.32 °C. The nighttime temperatures were relatively cooler than daytime temperatures, as lower temperatures were seen during December and January of each year, but still can be warm especially during April to October.

Fig. 3
figure 3

Monthly trend of NO2; SO2; O3; HCHO; CH4; CO; Aerosol index at 340–380 nm and Aerosol Index at 354–388 nm

Full size image

3.2 Multivariate analysis of atmospheric parameters

3.2.1 Factor analysis

Factor analysis is a multivariate statistical technique that identifies common patterns or factors among a set of variables. Principal Components (PCs) are a linear combination of original variables that capture significant variance in a set of data. Table 3 shows the set of PCs computed from the monthly trend of NO2, SO2, O3, HCHO, CH4, CO, AI@340-380 nm, AI@354-388 nm from April 2018 to December 2022 that produced eight (8) PCs with PC1 having the highest eigenvalue of 3.15 and percentage variance of 39.43% among the parameters, suggesting it represents a significant underlying pattern. The loading of O3 (–0.73), NO2 (–0.68), CO (0.86); AI@340–380 nm (0.76) and AI@354–388 nm (0.79) on PC1 indicate a potential shared source and chemical relationship among these pollutants, as this suggests there might be common sources or chemical process from incomplete combustion and vehicular gaseous emissions contributing to photochemical smog from aerosols as indicated from Aerosol index from two specific wavelength ranges (340–380 nm; 354–388 nm) at 0.76 and 0.79 leading to increased temperatures and poor radiative balance from particulates. One possibility is photochemical reactions in urban areas, where NO2 and VOCs (like HCHO) can lead to ozone formation (photochemical smog). CO, a byproduct of incomplete combustion, is also commonly associated with urban pollution. PC2 captures a substantial portion of the eigenvalue of 2.01; variance at 25.18% represents another important factor for loading for NO2 (0.50); HCHO (0.73); AI@340-380 nm (0.59); AI@354-388 nm (0.54), which shows the influence of volatile organic compounds (VOCs) impact to contribute to photolysis with aerosols in replica to PC1 causing decreased visibility, cloudiness from photochemical smog formation from Nitrogen oxides (NOx), as free radicals in the form of nitrate aerosols (-NH4NO3–ammonium nitrates, NH4HSO4–ammonium hydrogen sulphate,(NH4)2SO4–ammonium sulphate) [10]. According to Atkinson [57], the oxidative reaction of NOx with VOCs (from HCHO) can form organic acid species such as aldehydes, ketones, and increased carbon monoxide concentration at atmospheric temperature. PC3 shows a decreasing eigenvalue of 0.98 and a variance of 12.27% which showed appreciable loading of SO2 and CH4 at 0.62 and 0.68 respectively. As SO2 undergoes oxidation and forms sulphate aerosols and its ionic interactions with methane, it increases CO–CO2 concentrations in the atmosphere over a continuous period, influencing air quality and climate. PC4 to PC8 (Eigenvalues < 1, Explained Variance < 10%) represent less significance with loadings that may not indicate strong relationships among the parameters having less dominant, sources or processes.

Table 3 Factor analysis of atmospheric constituents
Full size table

3.2.2 Pearson correlation coefficient

The Pearson correlation coefficients provided for the atmospheric parameters (NO2, SO2, O3, HCHO, CH4, CO, AI@340-380 nm, AI@354-388 nm and temperature measurements (DT, daytime and NT, night-time) indicate the strength and direction of linear relationships and potential sources of pollution between these parameters as shown in Fig. 4. NO2 exhibits a strong positive correlation with O3 (0.72) and HCHO (0.56) indicating that higher levels of NO2 are associated with higher levels of O3 and HCHO suggesting that NO2, HCHO and O3 as gaseous precursors forms nitrogen complexes, peroxides, and depletion of O3 through photochemical reactions, and lower nitrogen aerosols as seen in the Aerosol indices (-0.14 and -0.19). NO2 has negative correlations with SO2 (− 0.12) CH4 (− 0.37) and CO (-0.61) implying that higher levels of NO2 are associated with lower levels of SO2, CH4 and CO respectively. SO2 showed weak positive correlations with AI@340-380 nm (0.35) and AI@354-388 nm (0.36) suggesting that a considerable level of sulphate aerosols would be formed from thermo-ionic interactions through atmospheric oxidation predominantly during daytime temperature levels (0.34) (Moradi and Zeuss, 2023). O3 shows moderate positive correlations with HCHO (0.45) and nighttime temperature (0.62) implying that strong levels of NO2 (0.72) will form volatile organic carbons (VOCs) that undergo photochemical reactions. Also, O3 had a negative correlation with CO (-0.64) and daytime temperature (− 0.59) implying lower level of CO is likely since CO2 would be formed from the depletion of ozone in the atmosphere. HCHO shows a moderate positive correlation with NO2 (0.56) and a weak positive correlation with O3 (0.45) implying strong physiochemical reaction is possible during nighttime temperature (0.46); whereas CO (− 0.27) and daytime temperature (− 0.25) are weakly correlated indicating lower CO level is likely as HCHO increases. CH4 (Methane) showed a weak correlation with SO2 (0.08) and NO2 (-0.37) implying no strong relationship, whereas correlated with CO (0.27) signifying probable interactions from the depletion of ozone during night temperature (0.46), as they are both greenhouse gases that increase CO2 concentration, atmospheric temperature, and lead to climate action. CO is often released from incomplete combustion processes, which exhibits negative correlations with O3 (− 0.64) and NO2 (− 0.61), and positive correlations with SO2 (0.16), CH4 (0.27) and Aerosol Optical Indices (AI@340-380 nm and AI@354-388 nm) at 0.51 and 0.54; the negative correlations suggest that higher levels of CO are associated with lower levels of O3, NO2, and vice versa for Aerosol indices from higher temperature during daytime (0.76). Aerosol optical Indices (AI) were positively correlated with each other (1.00), indicating that they provide consistent information about aerosol presence at 340–380 nm and 354–388 nm analysed from the data matrices, with SO2 and CO having a high proportion of aerosol formations influenced by the presence of these complex particulate pollutants predominantly during daytime temperature (0.78; 0.81) and weakly during nighttime temperature (0.34; 0.29) periods. Temperature measurements for daytime and night-time were weakly correlated at 0.08 suggesting little relationship between each component, as daytime temperature correlated positively with SO2, CH4, CO and AI, whereas nighttime temperature correlated with NO2, O3, HCHO and AI respectively.

Fig. 4
figure 4

Pearson Correlation of Atmospheric Parameters

Full size image

3.2.3 Scatter plot

Figure 5 shows the scatter plot determined as a multivariate factor analysis of atmospheric constituents in Nigeria from the monthly trend of April 2018 till December 2022, which was integrated as a data visualization with linkage spinning tree for individual data of interest at significance, p ≥ 0.95 plotted on Component 1 (PC 1) and Component 2 (PC 2) respectively. As seen, the scatter plot showed more interactions between all atmospheric constituents analyzed with NO2, O3, HCHO, SO2, AI@340-380 nm and AI@354-388 nm above one (1) from PC 1, whereas CO and CH4 were below one(1); for PC 2, SO2, AI@340-380 nm, AI@354-388 nm, CO and CH4 were above one (1) and vice versa for O3, NO2 and HCHO implying dissimilar relationship and interactions from each monthly variations shown from the cluster groups. The monthly trend showed that April and May 2018 were above 95% significance ellipse while the linkage spanning tree continued from August, 2018 to August, 2020, 2022 showing predominance near and similar pollution sources. As seen also, there were branched linkages from June 2019; June 2020; July 2022; August 2022; December 2022; January 2022; March 2022 and February 2020 showing close grouped clustering and interaction from atmospheric components.

Fig. 5
figure 5

Scatter Plot incorporated with linkage spanning tree at significance, p ≥ 0.95

Full size image

3.2.4 Dendrogram clustering

Figure 6 shows the dendrogram clustering of atmospheric constituents and monthly trends analyzed from the initial data set, as it can be used to merge similar data points into clusters and grouped into tree-like structures. As seen from the monthly trends, April-June 2018 were linked to multiple clusters July 2018 till June 2020 signifying close similarity in merged cluster groups, whereas July 2020 to September 2020 and July 2022 to December 2022 were merged and separated into the general merged group showing lower similarity due to the presence of longer branches and finally October 2021 till June 2022 showed a combination of shorter and longer branches showing high similarity and vice versa. For the atmospheric components, showed that AI@354-388 nm and AI@340-380 nm were very similar in their relationship that was linked to CO and CH4 and dissimilarity takes place for HCHO and O3 linked to SO2 and clusters to NO2 implying that the heights of the branches indicate levels of dissimilarity or no form of relationship with each merged cluster showing strong or close relationship among each components.

Fig. 6
figure 6

Dendrogram Clustering with heat map for atmospheric components and monthly trends

Full size image

4 Discussions

The processing of Sentinel 5P and 3 data from the TROPOMI and SLSTR sensors, which was monitored via spatiotemporal assessment between April 2018–December 2022 showed processing changes across mean and annual concentrations. Nitrogen dioxide (NO2) is released from both biogeochemical and anthropogenic combustion, which contribute to the total nitrogen atmospheric composition, as Muniraj et al. [58] NO2 study in India showed 51.54% and 85.81% reduction in 2018 and 2019 as compared to 2020. Olusola et al. [27] assessment showed that NO2 concentration in December 2019 was higher as compared to April 2020 due to decreased activities from COVID-19 lockdown restrictions across Nigeria; they stated that government measures lead to a decrease in anthropogenic activities such as land, sea and air transportation in addition to commercial activities (oil and gas, industries, power generating) that lead to increased NOx transport emissions in urban environment [59]. Although this study (Fig. 7) showed a 32–33% NO2 reduction in 2020 and 2022, it is critical to know that activities in southern Nigeria contributed to 56% NO2 as compared to Northern Nigeria due to oil and gas exploration and combustion, transportation, open burning of waste and transboundary releases from the offshore shipping and platforms that are released into the atmosphere leading to depletion of oxygen and ozone leading to formation of acid rain (HNO3), free radicals (peroxides and oxyacid) and peroxyl alkyl nitrate (RO2, RO2NO2) and aerosols (NH4NO3; (NH4)HSO4; (NH4)2SO4) as the inhalation of these elevated NO2 can lead to nasal irritation, asthma and breathing problems. [10, 60].

Fig. 7
figure 7

Nitrogen dioxide (NO2) from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image

SO2 is released from natural (volcanic and sulfur-bearing vents) and anthropogenic (industrial, metal smelting plant, vehicles, power plants) sources, which is a toxic pollutant that reacts with oxygen to form sulphur oxides and aerosols thereby causing photochemical smog, acid rain (H2SO3, H2SO4), increase soil acidity and health effects such as skin irritation and blisters [61]. The present study showed a slight reduction from 2018 to 2022 for mean SO2 concentration, which is similar to Karim and Rappenglück [62] assessment in Pakistan from 2020 and 2021, as the occasional spikes in SO2 (Fig. 8) levels could be attributed to factors like wind patterns or industrial emissions. Muriraj et al. [58] noted a 45.57% and 71.61% reduction from 2018 and 2019 assessments in India as compared to 2020. O3 is formed from the photolytic reaction of oxygen, the photochemical reaction of nitrogen, oxides (NOx), carbon oxides (COx) and volatile organic carbons (VOCs), as the present study showed elevated ozone from April to September each year due to sunny period and long-distance wind movement, as annual O3 (Fig. 9) concentration was higher in 2018, 2020–2022 and less in 2019. Moradi and Zeuss et al. [63] saw a 6.8% and 5.67% decrease in O3 in Tehran and Arak in Iran, which was lower than this study which had 3.5% decrease in Nigeria in 2019 respectively. The breathing of elevated ozone at ground level causes asthma is probable in children who are exposed to outdoor activities and affects photosensitive vegetation and the ecosystem [59].

Fig. 8
figure 8

Sulphur dioxide (SO2) from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image
Fig. 9
figure 9

Ozone (O3) from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image

HCHO is formed from the oxidation of non-methane volatile organic compounds (VOCs) such as aliphatic and aromatic hydrocarbons, sulfur-containing organics that are predominantly involved in the photochemical production of tropospheric ozone [57, 64, 65]. The oxidative reaction of nitrogen oxide (NOx) leads to the thermal decomposition of ROx (R-alkyl group) to carbon dioxide and organic acids that cause the greenhouse effect during nighttime from the generation of infrared radiated and increased temperature, solar intensity, and atmospheric pressure (Fig. 10) [66, 67]. CH4 is a greenhouse gas that plays a critical role in atmospheric and climatic chemistry, it is released from the anaerobic decomposition of plants, landfills and wetlands, and industrial and fossil fuel production [68]. The combustion of CH4 in the presence of O2 or O3 leads to the formation of CO2 that increases temperature and potentially impacts metrological indices such as relative humidity, wind pattern, rain pattern, cloud–-gas–aerosol mass mixing potential (Fig. 11) [69, 70]. CO is formed from incomplete combustion of carbon materials such as fossil fuel, biomass burning and wildfires, as CO is a highly toxic gas that can cause serious health problems such as CO blood poisoning, death, vomiting, oxygen deficiency in the blood and haemoglobin damage. The present study (Fig. 12) showed that October–February each year showed an elevated increase in CO and as such there were not many changes although 2018 had the lowest CO concentration as compared to other months. Halder et al. [71] in their study in India showed a decrease in fossil fuel combustion during the covid lockdown in early 2020 led to a decrease in CO combustion, as mining and calcination increased CO–CO2 concentration.

Fig. 10
figure 10

Formaldehyde (HCHO) from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image
Fig. 11
figure 11

Methane (CH4) from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image
Fig. 12
figure 12

Carbon monoxide (CO) from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image

AI is the ultraviolet aerosol concentration determined at 340–380 nm and 354–388 nm for particulate matter via the aerosol column, as it is expressed between (−1)–(+ 1) implying lower aerosol index and vice versa for higher aerosol index between > 1–5, as this shows the presence of aerosol particles across different locations dependent on various concentrations from anthropogenic releases [72, 73]. Previous research by Gbodjinou et al. [74] in their aerosol index assessment in the Benin Republic showed that the mean range of 2019 was between 0.0438 and 0.9094 with February 2019 having 1.859, which was lower than Nigeria's data between 0.661 and 1.93 with April been the maximum aerosol index at 2.01. A careful review of this study (Figs. 13 and 14) in Nigeria showed that January–April of each year had significant aerosol particulate movement as compared to other months and that the Northern Sahara wind pattern blowing towards Southern Nigeria (Atlantic Ocean) led to increased atmospheric circulation of aerosols as sea salt, dust particles, urban haze, which impacts precipitation gradient, metrological factors and also seasonal variation from the atmosphere to ground level, which is in agreement to Gbodinou et al. [74]. The influence of industrial activities, transportation, and other anthropogenic activities can increase aerosol concentration across different particle sizes that when inhaled by humans, flora and fauna cause significant health effects such as skin inflammation and irritation, asthma, respiratory inflammation and tract infection, extreme coughing, chest tightness and death [71], and ecological influences such as discoloration and bleaching of flora leaf, stem deterioration and decreased fruit yield over time [9].

Fig. 13
figure 13

Aerosol Index at 340–380 nm from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image
Fig. 14
figure 14

Aerosol Index at 354–388 nm from Sentinel-5p in Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image

Land surface temperature (LST) measures how cold or hot the earth’s surface would feel to the touch via the mean radiative temperature of all objects measured by spaceborne remote instruments [47]. At infrared (IR) wavelengths using Sentinel 3, the surface temperature determines the amount of energy emitted by the earth's surface for the determination of heat and water fluxes in the atmosphere. The daytime temperature showed that March to April had the highest temperature influx (46–47 °C) as compared to August (28–30 °C that was least, while the nighttime temperature showed March to April had highest temperatures between 24–25 °C and lowest in December and January 18–19.5 °C due to northern Nigeria as compared to southern Nigeria. As seen from both daytime and nighttime temperature differing values (Fig. 15), it implies that climatic, and physiochemical reactions from anthropogenic activities can impact metrological conditions invariably impact both daytime and night-time temperatures from thermal radiation, conduction, convection, and advection leading to absorption of solar radiation, the release of heart from the earth surface and air mass movement at different temperatures [75].

Fig. 15
figure 15

Surface Temperature from Sentinel 3A/B for Nigeria derived from EO Browser (https://apps.sentinel-hub.com/eo-browser/) and map developed in ArcGIS Software (https://www.arcgis.com/index.html)

Full size image

The overall trend analysis showed the influence of industrialization and emission release that can impact air quality dynamics in Nigeria. The study underscores the importance of primary and secondary pollutants including NO2, SO2, O3, HCHO, CH4, CO, and AI, which can impact the atmospheric chemistry and interaction over a long period as climatic conditions changes. The comprehensive examination of various atmospheric constituents and their intricate relationships provides a foundation for informed decision-making. The study's insights into regional variations, pollutant sources, and their implications on health and the environment underscore the importance of tailored environmental management strategies. As Nigeria strives for sustainable development, continuous monitoring and research efforts are imperative to address emerging challenges and ensure a healthier, more resilient future for its populace.

5 Limitation and recommendation

While this study provides valuable insights, certain limitations should be acknowledged. The reliance on Sentinel-5 Precursor TROPOMI and Sentinel 3A/B SLSTR data, while advanced, may have inherent uncertainties or biases. The absence of ground-based measurements may limit the precision of the findings, and future research could benefit from integrating ground-based data for validation. The expansion of the scope to include a broader spectrum of pollutants and socio-economic factors would provide a more comprehensive understanding of the impact of air quality on both the environment and human health. Collaborative efforts between governmental bodies, research institutions, and industries are recommended to develop effective strategies for reducing anthropogenic contributions to air pollution.

6 Conclusions

The study presents a tropospheric analysis of sentinel-5P and 3 data obtained from TROPOMI and SLSTR sensors, focusing on the spatiotemporal assessment of air quality in Nigeria from April 2018 to December 2022. The investigation encompasses the mean and annual concentrations of various atmospheric constituents, including Nitrogen dioxide (NO2) attributed to 56% contribution from oil and gas exploration, Sulphur dioxide (SO2) indicates a slight reduction in annual mean concentrations from 2018 to 2022, Ozone (O3) showed that elevated sunshine influenced O3 increment and reduced by ionic reactions, Formaldehyde (HCHO) incremental impact due to nighttime greenhouse gas pollutants (NOx), Methane (CH4) and Carbon monoxide (CO) influence on temperature increment and greenhouse pollution of CO2, and Aerosol indices that impact health ecological and metrological factors over a time frame. The findings reveal intriguing patterns in pollutant concentrations, influenced by several factors such as anthropogenic activities, climatic conditions, and natural events. This underscores the importance of continuous monitoring and in-depth analyses for a holistic understanding of air quality dynamics in Nigeria. The identified patterns and influences provide crucial insights for policymakers, emphasizing the need for targeted interventions to mitigate environmental and health risks associated with specific atmospheric constituents.