Air Quality Predictions through Mathematical Modeling for Iron Ore Mine Project

Abstract

Mathematical modeling was deployed to predict air quality during the construction and operation phases of an iron ore mine project in Maharashtra, India. A survey of different models revealed that the ISCST3 model was the most applicable one to predict the air quality parameters, particularly the suspended particulate matter (SPM) and coarse particulate matter (PM₁₀). Baseline air quality data, emission rates, local meteorology, and terrain information were used to simulate the ground-level concentrations. The simulation predicted SPM and PM₁₀ peaks of 172 µg/m³ and 44 µg/m³, respectively. The prediction was within the prescribed limits of the national standards of 200 µg/m³ and 100 µg/m³, respectively, near the source, with minor exceedances in total SPM in two nearby villages and an impact on air quality due to proposed mining. Accordingly, mitigation strategies towards such villages were recommended and implemented. Later, the monitoring in the operation phase revealed that particulate matter could be controlled effectively with mitigation strategies and ensured compliance with air quality standards. The analysis also revealed strong correlations between the particulate matter and the distance of the localities and SPM and PM₁₀. Continuous monitoring and adaptive mitigation based on real-time data were thus emphasized for long-term sustainability and responsible mining practices.

1. Introduction

Mining activities are the cornerstone of modern civilization, as these provide essential resources for infrastructure, industry, and technology [1]. However, such benefits come at a significant environmental cost, as mining activities have significant environmental impacts, including the creation of solid, liquid, and gaseous wastes, leading to pollution and ecosystem degradation [2]. Among these environmental impacts, air quality stands as a critical issue due to its direct and demonstrably adverse effects on human health and ecological well-being [3,4].

The release of particulate matter (PM) from mining operation areas is a significant source of dust emissions, primarily arising from activities such as blasting, material handling, processing, and transportation. These activities disturb settled particles, leading to their suspension in the air, and, once airborne, they can translate into environmental hazards and lead to respiratory risks in human populations [2,5]. Additionally, wind erosion of waste disposal sites, such as waste rocks and tailings, also contributes to dust generation into the surrounding environment [6].

In recent years, the demand for low-grade iron ore has surged across domestic and export markets [7]. To meet this growing demand, iron ore mines often increase production, leading to a proportional increase in dust generation and associated air pollution [8]. This necessitates a comprehensive assessment of the potential air quality impact arising from such production expansions.

Environmental impacts associated with mining activities are broadly categorized into primary and secondary impacts [9]. Primary impacts directly result from the mining operations, while secondary impacts include broader societal and economic changes inflicted by the mining project. This study focused on the primary air quality impacts anticipated due to the planned production capacity at a specific iron ore mine (IOM) in the western part of India. Environmental Impact Assessments (EIAs) are crucial for evaluating the potential environmental consequences associated with various projects, including mining operations [10]. The air quality assessment is a key component of EIAs to ensure proper evaluation and implementation of control measures [11,12].

Air pollution modeling plays a crucial role in understanding, assessing, and regulating air quality and the distribution of toxic pollutants. Mathematical modeling serves as a valuable tool for anticipating these impacts, filling the gaps in the traditional assessment methods and enhancing decision-making processes in mining activities [11,13]. Established methodologies incorporating recent studies on iron ore-mine emissions and modeling have been used to model the air quality around such mining areas [14]. Researchers have primarily used models like AERMOD, CALPUFF, and HYSPLIT, which are effective for long-term and complex terrain assessments but less suitable for short-term predictions in flat terrains typical of many mining regions [14,15]. This study is an attempt to bridge the gap by employing the ISCST3 model for accurate short-term predictions of particulate-matter dispersion in a flat-terrain setting.

Various models, including Gaussian Dispersion Models and the Industrial Source Complex Dispersion Model, estimate pollutant concentrations from different sources during mining operations [15,16]. These models help predict the horizontal and vertical dispersion of pollutants under different emission scenarios and meteorological conditions, aiding in the establishment of air quality-monitoring stations for effective management [17]. The ISCST3 model is employed to conduct a comprehensive assessment of the potential air quality ramifications arising from the proposed mine [17,18]. In situations where mining is proposed, an opportunity exists to analyze the air quality impacts into two distinct phases, i.e., the construction phase and the operation phase [19,20,21]. The existing background-pollution levels are assumed to remain consistent, and long-term climatic variations are excluded.

Considering the short-term prediction, flat terrain, and proposed mining activities, an iron ore mine (IOM) was selected for the study. As mentioned, the construction and operation phases of suspended particulate matter (SPM) and coarse particulate matter (PM₁₀) were considered for analysis in the two phases. The study focused on the impacts of mining on primary air quality. The objectives of the study were as follows:

• Survey of the various air quality prediction models (AQMs) and selection of the most suitable model for predicting air quality impacts for an iron ore mine project for short-term area sources and a flat terrain, while accounting for the actual field conditions.
• To quantify SPM and PM₁₀ in the dry season during the construction phase and model the same with the established model and continue monitoring during operation phases.
• To propose mitigation measures based on simulations to minimize air quality impacts based on AQM predictions and their implementation for compliance with environmental standards during operation phase.
• To evaluate air quality trends over distance and the variability of SPM and PM₁₀ using exploratory analysis, while evaluating the effectiveness of implemented mitigation strategies.

2. Study Area

The study was conducted at an operational iron ore mine (IOM) located in the Sindhudurg District of Maharashtra, India (Figure 1a). The mine site, which is fully mechanized, spans geographic coordinates from Latitude N 15°44′19″ to 15°44′56″ and Longitude E 73°39′47″ to 73°43′22″. This region experiences a tropical climate with high humidity levels and significant rainfall during the monsoon season (June to September), while the rest of the year is marked by dry conditions with moderate-to-high temperatures.

The topography of the area is relatively flat with slight undulations in some sections. The area is subject to varying wind speeds and directions, with predominant wind directions from the northwest during the pre-monsoon season, shifting to the southwest during the monsoon season. These meteorological conditions are known to significantly influence the pollutant dispersion and deposition patterns.

To assess the air quality, air-monitoring stations (Figure 1b) were established in and around the lease area within a 10 km radius. These stations (Figure 1c) allowed for the collection of essential data for the baseline air quality assessment and subsequent modeling. The distinct wet and dry seasons influence the dispersion and settling of particulate matter, with the dry season exacerbating dust generation and dispersion due to reduced humidity and minimal vegetation cover.

Geologically, the area falls in the Redi Iron Formation of the Banda Group, which is characterized by the frequent occurrence of iron ore bodies with significant mineral content. The general succession of lithology from top to bottom observed in the study area is compiled in

Table 1. The general succession of lithology of the mine area.

Lithology	Thickness (m)
Laterite	10 to 20
Phyllitic clay	20 to 25
Lumpy iron ore	5 to 7
Blue dust	20 to 40
Limonitic/manganiferous clay	25 to 55
Siliceous clays/quartzite	30 to 60
Schists	Unknown (base)

.

The general strike of the ore body is NW-SE (around N270° to N290°), but in some places, it has taken a swing, and the direction changes to E-W. The general dip of the ore body is 42° to 55° towards NE. The mine uses a mechanized open-pit approach for mineral extraction, deploying heavy machinery such as excavators, dumpers, and dozers, replacing traditional drilling and blasting methods. The mine benches and broad view of the pit area are illustrated in Figure 2a,b.

3. Methodology

A comprehensive methodology was designed and deployed to achieve the objectives of the study. The method consisted of the selection of a mine for the study, collection of the baseline data before the ensuing mining activities, collection of data after the mining activity started, devising mitigation measures, a baseline exploratory analysis, advanced air quality modeling and simulation using an established method, the implementation of the mitigation measures, and a post-operation analysis of the data (Figure 3). Each step is designed to ensure a comprehensive assessment of air quality impacts due to mining activities in the study area.

3.1. Data Acquisition and Compilation

Data for the study were acquired in two distinct phases: the construction and operation phases. The collection was carried out using the Respirable Dust Sampler (Envirotech APM-460, Envirotech Instruments Pvt. Ltd., New Delhi, India) and the Particulate Sampler (Envirotech APM-550 mini, Envirotech Instruments Pvt. Ltd.) at different locations in and around the mine (Figure 1b). The selected locations included Core Zone (CoZ, average distance of around 60 m from active mining area), Tiroda Village (TiV, distance 440 m from the active area of the mine), Nanos Village (NaV, distance 630 m), Shiroda Village (SiV, distance 1520 m), and Redi Village (ReV, distance 3170 m). In the operational phase, the mine areas, including the material loading point (MLP), waste dump point (WDP), haulage and transportation road (HTR), and mobile screen plant (MSP), were also selected for the sampling. The emission data during excavation and loading, along with the meteorological data, were used for the analysis in AQM. The data collection focused on the SPM and PM₁₀ levels during the dry summer season, ensuring no rainfall interference. Meteorological data, including wind speed, direction, temperature, and humidity, were also collected to aid in the modeling process. The data were grouped into pre-mining (PreMin), mining operation (PostMin), and along with the group of locations considered as the source of the particulate matter (OOPM) for comparative analysis.

3.2. Survey of the Air Quality Models

A survey of various air quality-modeling techniques was conducted to identify the most suitable model for predicting air quality impacts in a flat-terrain setting typical of the study area.

Table 2. Survey of air quality prediction models for the study of IOM.

Sl. No.	Model Name	Advantages	Disadvantages	Citation
1	AERMOD	Regulatory applications, applicable to various source types, complex terrain	Complex, less suitable for short-term assessments	[14,15]
2	CALPUFF	Multi-layered model, long-term assessments, considers complex terrain, land-use changes	Not suitable for short-term assessments	[22]
3	HYSPLIT	Trajectory model, regional transport patterns, local dispersion modeling	Too complex, not suitable for short-term assessments	[23,24]
4	CMAQ	Comprehensive regional model, multiple pollutants, and complex atmospheric processes	Not suitable for short-term assessments	[25]
5	ISCST3	The gaussian plume model approach, needs less input data, less complicated, versatility for short-term area source modeling, suitable for flat terrains	Flat-terrain assumption may not be fully representative.	[8]

provides a comparative assessment of popular models.

It has been observed that models at Sl. Nos. 1 to 4 (Table 2) are either too complex or are applicable on a regional scale or complex terrain. So, the Industrial Source Complex Short Term (ISCST3) model was selected for air pollution modeling, specifically for assessing total SPM and PM₁₀ emissions from iron ore-mining operations, as also supported by various other authors, e.g., [26]. ISCST3 requires input parameters such as emission rates, meteorological data, and geographical information to simulate the dispersion of pollutants in the air [27]. The model is well-regarded for its applicability to short-term area source modeling in flat terrains, making it suitable for the study area [28,29].

The Gaussian plume approach in the ISCST3 allows for simultaneous consideration of various source types, capturing the dispersion of pollutants from mining activities comprehensively [27,30]. The ISCST3 is suitable for short-term area source modeling, which aligns with the need of this study, to assess impacts during both construction and operation phases, matching the emissions characteristics of mining activities. Studies have shown that ISCST3 is a refined dispersion modeling technique that provides accurate predictions of pollutants like nitrogen oxides (NO_X) and total suspended particles [31]. The model uses rural dispersion and regulatory defaults, in conformity with the Central Pollution Control Board (CPCB, New Delhi, India) guidelines, thus simplifying the implementation and ensuring timely results [32]. One of the advantages of the ISCST3 model is that it assumes a flat terrain and, hence, is suitable for the topography of the study area, ensuring appropriate predictions in the area.

3.3. Air Quality Assessment

The study was conducted in two phases: the construction and the operational phase, each with distinct characteristics affecting suspended particulate matter. These phases are described further to understand their difference in the context of the air quality.

3.3.1. Impact on Air Quality during Construction Phase

During the construction phase of an iron ore mine (IOM) project, emissions originate from equipment movement and site-specific dust-generation activities like grading, and earth and foundation works. The vehicles and equipment deployed produce exhaust emissions, potentially increasing pollutants like SO₂, NO_x, SPM, CO, and hydrocarbons. Concerns arise about air quality during dry months due to dust emissions, for reasons mentioned earlier, and explain the reason for selecting data for such a period for the analysis.

The construction activities cause local variations in SPM levels, but impacts are temporary, marginal, and reversible, with predominantly inorganic dust that settles quickly at the site, posing minimal toxicity [33]. Although effects are confined to project boundaries, proactive measures like vehicle maintenance, wetting roads, and construction sites and introducing vegetation significantly reduce these impacts. These measures demonstrate a proactive approach to addressing environmental consequences during construction. However, the construction phase, while causing temporary localized air quality impacts, offers an opportunity for effective mitigation of the particulate matter through conscientious practices and interventions.

3.3.2. Impact on Air Quality during Operation Phase

During the operational phase, emissions originated from the different mining activities, i.e., excavation, loading, transportation of ore and overburden, movement of heavy earth-moving machinery, and dumping of the ore or overburden. The particulate matter from such sources influences the air quality and needs assessment against baseline concentrations.

3.3.3. Air Quality Modeling

The ISCST3 model was deployed to assess the air quality impacts of mining operations. This model employs a Gaussian plume approach (Equation (1)), allowing for simultaneous consideration of various source types. The formula used for pollutant-concentration calculation is as follows:

$$C_{(x,y,0,H)}=\frac{Q}{\pi u{\sigma }_y{\sigma }_z}\exp \left(-\frac{1}{2}{\left(\frac{y}{{\sigma }_y}\right)}^2\right)\mathrm{e}\mathrm{x}\left(-\frac{1}{2}{\left(\frac{H}{{\sigma }_z}\right)}^2\right)$$

(1)

where C_(x,y,0,H) is the downwind concentration at ground level (g/m³), Q is the emission rate (g/s), σ_y and σ_z are the lateral dispersion and vertical dispersion parameters (m), u is the wind speed (m/s), y is the crosswind distance (m), and H is the effective stack height (m) (actual height + plume rise).

This model employs a short-term area-source approach with numerical integration across upwind and crosswind directions of the Gaussian plume formula, allowing for the inclusion of point, area, line, or volume sources. This flexibility enables the model to account for diverse sources and their contributions to overall air quality by considering source characteristics, meteorological conditions, and plume dispersion patterns.

Meteorological data of wind speed, direction, temperature rainfall, and relative humidity, recorded at one-hour intervals during the monitoring period by a continuous weather monitoring station, served as the model’s input.

Table 3. Stability classes for wind direction and dispersion parameters influencing the predicted pollutant concentrations for rural conditions and frequency of occurrence (%) in the pre-monsoon season in the study area.

Stability Class	Standard Deviation of Wind Direction Fluctuation $\left({\mathit{\sigma }}_{\mathit{\alpha }}\right)$ in Degrees	Lateral Dispersion $\left({\mathit{\sigma }}_{\mathit{y}}\right)$	Vertical Dispersion $\left({\mathit{\sigma }}_{\mathit{z}}\right)$	Frequency of Occurrence (%)
A	>22.5	$0.22x(1+0.0001x{)}^{-0.5}$	$0.20x$	8.33
B	22.4–17.5	$0.16x(1+0.0001x{)}^{-0.5}$	$0.12x$	8.33
C	17.4–12.5	$0.11x(1+0.0001x{)}^{-0.5}$	$0.08x(1+0.0002x{)}^{0.5}$	12.5
D	12.4–7.5	$0.08x(1+0.0001x{)}^{-0.5}$	$0.06x(1+0.0015x{)}^{0.5}$	25
E	7.4–3.5	$0.06x(1+0.0001x{)}^{-0.5}$	$0.03x(1+0.0003x{)}^{-1}$	16.67
F	<3.5	$0.04x(1+0.0001x{)}^{-0.5}$	$0.16x(1+0.0003x{)}^{-1}$	29.17

summarizes the distribution of stability classes observed during the pre-monsoon season in the study area. For extrapolating wind speed, the power law relationship was used to calculate wind speed at the stack level, ensuring an accurate estimation of wind conditions at the relevant height for pollutant-dispersion modeling [25]. The mixing-heights criterion from the Central Pollution Control Board (CPCB), India, was adopted in the model due to the absence of site-specific data [34,35,36]. This approach facilitates the simulation of a worst-case scenario based on typical pre-monsoon season variations in atmospheric mixing heights (see supplementary data). The complete modeling process of ISCST3 is explained in Figure 4.

3.3.4. Stability Classification and Dispersion Parameters

The stability classification, determined using the wind direction-fluctuation method recommended by the CPCB, India [37,38], allowed for the quantification of the atmospheric stability through the standard deviation of wind-direction fluctuation (σ_a), as shown in Equation (2).

$${\sigma }_a=\frac{W_{dr}}{6}$$

(2)

where σ_a is the standard deviation of wind-direction fluctuation, and W_dr is the overall wind-direction fluctuation or width of the wind direction in degrees.

The classification significantly influenced dispersion parameters and predicted pollutant concentrations, with stability classes ranging from A (very unstable) to F (very stable). Unstable conditions (A, B, and C) lead to greater mixing and dispersion, while stable conditions (E and F) restrict dispersion, leading to higher concentrations near the source. Different stability classes have significant implications for dispersion parameters and ultimately influence the predicted pollutant concentrations.

The project’s rural setting with flat terrain uses open country-dispersion parameters recommended by the CPCB, India [39,40]. These parameters, crucial for simulating pollutant plume behavior, vary with downwind distance and depend on the assigned stability class. Table 3 outlines these coefficients specifically for rural conditions, where σ_y represents lateral dispersion, and σ_z represents vertical dispersion. These parameters are used for modeling atmospheric dispersion in flat terrain and rural areas.

The lateral dispersion (σ_y) and vertical dispersion (σ_z) parameters vary with downwind distance (x) and are determined by different empirical formulas depending on the stability class [41,42], thus ensuring accurate modeling of pollutant dispersion under different atmospheric stability conditions, as is crucial for Environmental Impact Assessments and air quality management in rural settings.

Based on the model assumption, the mixing height of particulate matter during the day (Figure 5a), standard deviation of the wind-direction fluctuations along with frequency of occurrence by stability classes A to F (Figure 5b), and coefficients of the lateral and vertical dispersion in different stability classes (Figure 5c) were evaluated and recorded.

3.3.5. Modeling Procedure

The Key model settings for evaluating AQ at Ground-Level Concentrations (GLCs) for the study area are as follows:

• Plume rise estimation: Briggs formulae, limited to the mixing layer.
• Buoyancy induced dispersion: Captures plume dispersion during ascent.
• Processing routine: Calms by default.
• Wind profile exponents: Default ‘Irwin’.
• Terrain: Flat terrain computation
• Transformation and removal: No physiochemical transformations or pollutant removal assumed.
• Washout by rain: Not considered.

These model settings ensured the model accurately reflected site-specific conditions. The emission data for SPM stemming from various phases of the mining operation, including excavation, and loading activities, were also documented (

Table 4. Iron ore-mine emissions calculated for the proposed IOM.

Parameter	Excavation	Loading of the Ore—OB	Loading of the Ore—Core Zone
Ore production (million tons/year)	0.5	0.5	0.5
Total working days	220	220	220
Operational hours	10	10	10
Total operational hours	2200	2200	2200
Activity rate (t/h)	227.27	227.27	227.27
Uncontrolled emission rate (gm/t/h)	0.0236	0.0236	0.0236
Emission factor (gm/t)	0.0236	0.0236	0.0236
Influence area (m2)	70,300	70,300	70,300
Source emission rate in area (g/s/m2)	9 × 10−8	2 × 10−8	9 × 10−8
Control efficiency (%)	90%	90%	90%
Controlled emissions (g/s/m2)	9.3 × 10−9	2.1 × 10−8	2.1 × 10−8

).

This table outlines both the operational parameters and the resulting emissions calculations for the proposed iron ore mine (IOM). Parameters such as ‘ore production’, ‘total working days’, ‘operational hours’, and ‘activity rate’ are inputs used to estimate emissions. ‘Uncontrolled emission rate’ and ‘emission factor’ are used to derive ‘source emission rate in area’ and ‘controlled emissions’, which are the calculated emissions after applying a control efficiency of 90% thereby incorporating potential environmental impacts under typical operational conditions and implemented control measures.

4. Modeling, Analysis, and Results

4.1. Model Simulations and Analysis

To provide a detailed representation of particulate matter-concentration variations across the site and surrounding area, simulations were thus conducted using the high-resolution triple-joint frequency data available for every hour. These simulations covered a 20,000 m radius encompassing 16 directions emanating from the site, allowing for a thorough assessment of potential AQ impact.

The simulations estimated incremental concentration values over the entire monitoring period, identifying the highest concentrations observed at any receptor point within a 24-hour timescale. The results were visualized using isopleths, which illustrated the distribution of suspended particulate matter (SPM) concentrations during mining activities (Figure 6).

The isopleth simulation (Figure 6) depicted the spatial distribution of incremental SPM concentrations from the proposed mining operation, showing concentration contours of 15, 30, and 45 µg/m³. The concentric isopleths indicated a Gaussian dispersion pattern, with the highest pollutant levels near the emission source, gradually decreasing with distance. The peak incremental concentrations were observed within a 1000 m radius from the site, with values above 45 µg/m³. Beyond this radius, concentrations decreased, with values ranging from 30 to 45 µg/m³ up to 5000 m and further reducing to below 30 µg/m³ beyond 10,000 m. The model simulates a higher spread of SPM (15 µg/m³) incremental concentrations in SW-SE directions and a lower spread in the other directions, reflecting the influence of prevailing wind patterns.

The peak incremental Ground-Level Concentrations (GLCs) predicted by the model, the existing baseline levels of SPM and PM₁₀ measured during the pre-monsoon season, and the distances of monitoring locations are summarized in

Table 5. Resultant concentrations due to incremental GLCs (mining only).

Receptor Location	Cardinal Direction	Distance (m)	Baseline SPM (μg/m3)	Predicted Incremental SPM (μg/m3)	Estimated Total SPM (μg/m3)	Baseline PM10 + Predicted Incremental PM10 (μg/m3)
Core Zone	-	~60	117	49	166	44.0
Tiroda Village	N	440	152	20	172	41.9
Nanos Village	SE	630	132	16	148	34.8
Redi Village	SW	3170	101	34	135	41.6
Shiroda Village	NW	1520	78	13	91	26.9

Note: Permissible limits, as per NAAQ Standard, for SPM and PM₁₀ are 200 μg/m³ and 100 μg/m³, respectively.

. The baseline SPM and the estimated SPM are also presented in Figure 7.

Table 5 presents concentrations calculated under controlled emissions using the ISCST3 model, incorporating hourly wind data to simulate pollutant dispersion. The concentrations at receptor locations in surrounding villages exceed those in the Core Zone (single coordinate representing all the mining activities) due to the alignment with peak wind trajectories (as shown in Figure 6), which efficiently disperse pollutants from the mine. The Core Zone, positioned away from peak winds, showed lower levels. Without control measures, incremental SPM could rise significantly in the Core Zone, causing substantial increases at other locations, highlighting the importance of dust control measures in protecting air quality and understanding pollutant dispersion dynamics.

The model simulated the ground-level concentrations (GLCs) of SPM and PM₁₀. Table 5 reveals that the estimated total SPM (ETSPM) varies across the receptor locations. The minimum ETSPM of 91 μg/m³ is observed at Shiroda Village, located 1.52 km away, in the northwest direction. The maximum ETSPM of 172 μg/m³ is observed at 0.44 km away, in the north direction (Tiroda Village). Nanos Village, located 0.63 km away, in the southeast direction, shows an ETSPM of 148 μg/m³. The impact of predicted concentrations on the surrounding population is minimal, as the maximum incremental increase is only 34 μg/m³ at Redi Village.

The predicted concentrations for the Core Zone and surrounding villages remained within permissible limits, except for slight exceedances in Tiroda and Nanos Villages. However, to have better control over the AQ, mitigation measures (

Table 6. Details of the mitigation measures recommended and implemented in the mine in operation phase.

Sl. No.	Mitigation Measure	Details	Advantages
1	Water-spray systems	Dedicated haul road tankers (8 KL/day) for continuous dust suppression. Expansion of permanent sprinklers to critical areas (including dump slopes) for dust control and plantation irrigation. Strategic deployment of water-mist technology (WMT) for enhanced dust suppression.	Reduces airborne dust significantly and improves air quality for work persons and nearby communities. It also supports the plant growth and sustainability on-site
2	Stockpile management	Tarpaulin covers to minimize wind erosion across all stockpiles. Regular water-sprinkling programs to further control dust generation.	Prevents the dust from being airborne, reduces overall environmental adverse impact, and enhances site cleanliness.
3	Vegetation buffer expansion	Planting trees and shrubs around the mine perimeter and on dumps to create natural dust-dispersion barriers.	The vegetation buffer acts as a natural air filter, which also enhances the ecological aesthetics of the mining area and provides a habitat for local wildlife.
4	Administrative enhancements	Stricter HEMM maintenance schedules for minimized emissions. Expanded dust-control training for personnel. Upgraded worker PPE with potentially integrated air quality monitoring	Ensures compliance with environmental regulations and increases awareness and real-time air quality management

) were recommended and implemented in the field before the commencement of the operational phase (see supplementary data also for relevant details and figures).

The objective of the mitigation measures was to minimize dust generation at the source through a combination of engineering, administrative, and technological practices. The measures were implemented, and the operational-phase air quality data were analyzed in light of the construction-phase data.

4.2. Exploratory Data Analysis

The exploratory data analysis was performed by grouping the data into three classes: pre-mining data, data from the different places in the Core Zone with higher concentrations of the SPM (acting as a source during particulate matter dispersion), and post-mining initiation data. The data variability for the three classes is presented in Figure 8 (where PreMin, OOPM, and PostMin represent the above three categories of data, respectively. (The SPM is in µg/m³, designated as mcg in the figures).

Figure 8 shows higher variability in pre-mining data compared to post-mining, with significant variation observed in areas of mining activity. The concentration of SPM at various sites in the mines and surrounding villages shows a notable contrast in values.

Tiroda Village (TiV) exhibited the highest variability, followed by Shiroda Village (SiV) and Redi Village (ReV). Although Nanos Village (NaV) had higher concentrations, it showed less variability during the monitoring period (Figure 9a). Figure 9b illustrates a significant drop in SPM levels beyond 630 m from the source. The summarized data on air quality, along with the locations and other derived factors, are provided in

Table 7. Air quality along with the locations and other derived factors.

Mining Phase	Location	Distance (m)	SPM to (SPM Core Zone) Ratio	PM10 to (PM10 Core Zone) Ratio	Change in SPM Mining Phase (%)	Remarks
Construction	Core Zone	~60	1	1
Construction	Tiroda Village	440	1.3	1.15
Construction	Nanos Village	630	0.86	0.79
Construction	Redi Village	3170	0.67	0.67
Construction	Shiroda Village	1520	0.67	0.81
Operation	Core Zone	41	1	1	212.82
Operation	Tiroda Village	440	0.23	0.2	−43.42	Mitigation measure used
Operation	Nanos Village	630	0.24	0.24	−11.88	Mitigation measure used
Operation	Redi Village	3170	0.22	0.21	5.13	Increase 4 µg/m3
Operation	Shiroda Village	1520	0.23	0.18	10.26	Increase by 8 µg/m3

.

The concentrations of SPM and PM₁₀ along with their variation in the pre-mining phase and mining phase at different locations are given in Figure 10.

Figure 9 and Figure 10 indicate that while SPM and PM10 concentrations increased in the Core Zone during the mining phase, they showed a reduction or only minor increases in surrounding villages. It is important to note that other mines in the vicinity contributed to higher pre-mining-phase pollutant concentrations, particularly in Tiroda Village. However, the implementation of mitigation measures, such as green belt barriers, water sprinklers on haul roads, wet loading of ore and overburden, and geotextile application on stabilized dumps, resulted in decreased concentrations during the mining phase and only minor increases in Redi and Shiroda Villages.

The relationship of SPM and PM₁₀ monitored in the two phases shows a significant decrease in these pollutants with distance (Figure 11a,b) and assumes a power function.

Additionally, a significant relationship was found between the SPM and PM10 ratios relative to the Core Zone, as depicted in Figure 12. The logarithmic trend with an R² value of 0.98 between the ratio of PM₁₀ to ‘PM₁₀ Core Zone’ and the ratio of SPM to ‘SPM Core Zone’ implies a high degree of correlation between these parameters.

5. Conclusions

The ISCST3 model effectively simulated air quality impacts from an iron ore mine during the construction phase, proving to be the best-suited model for flat terrain and short-term applications. The model predicted that calculated SPM and PM₁₀ concentrations remained within national standards, although localized exceedances in Tiroda and Nanos Villages further highlighted the need for targeted mitigation.

During the operational phase, the implementation of green belt barriers, water sprinklers on haul roads and dumps, wet loading of ore and overburden, and Water Mist Technology (WMT) was effective for dustsuppression. Comparing emissions during construction and operation phases with the distance of the villages revealed that, despite increasing particulate matter in the Core Zone during mining, only a minor increase was observed in surrounding villages. The highest incremental SPM concentration, 172 µg/m³, was observed in Tiroda Village, 0.44 km north of the mine, establishing the effectiveness of mitigation measures in minimizing particulate matter generation and dispersion.

Furthermore, the strong logarithmic correlation between SPM and PM10 indicated that continuous and adaptive mitigation measures, along with continuous air quality monitoring, are crucial for maintaining air quality standards throughout the project’s life cycle and ensuring sustainability. The findings can be useful for evolving sustainable mining practices, balancing economic gains with the environment and safeguarding public health.

Future research can focus on more comprehensive modeling techniques, considering area sources and barriers, and conducting long-term health impact studies. Advanced air quality modeling with IoT-based sensing devices, integrating machine learning methods to such databases with enhanced predictive capabilities.