Respirable Dust Exposure in Western Australian Mining: Trends, Variability, and Implications for Occupational Health Authors: Renee N. Carey, Susan Peters, Nicholas de Klerk, Fraser Brims, Peter Franklin Published: 2026 ABSTRACT ABSTRACTBackgroundMining workers are exposed to a range of respiratory hazards, including respirable dust. While exposure to respirable crystalline silica in the mining industry has been found to be common, less is known about trends in measured levels of exposure to respirable dust overall.MethodsThis study analyzed 109,442 personal respirable dust measurements collected from Western Australian mines between 1986 and 2014. Data were sourced from an industry‐wide monitoring database and modeled using mixed‐effects regression, adjusting for sampling characteristics (year of measurement, duration, reason), mine type, and job title. Differences in trends over time between mine types were also examined.ResultsOverall geometric mean respirable dust concentrations declined from 0.393 mg/m³ in 1986–1990 to 0.181 mg/m³ in 2011–2014, representing an average annual decrease of 3.51% (95% confidence interval −3.63% to −3.39%). Modeled exposure levels varied by mine type, with the highest levels observed in iron ore mines (GM 0.489 mg/m3 in 1986 through to 0.197 mg/m3 in 2014) and exploration (GM 0.665 mg/m3 in 1994 through to 0.172 mg/m3 in 2014). Temporal trends varied by mine type, with a steeper decline in exposure levels seen in exploration activities in particular.ConclusionsExposure to respirable dust in Western Australian mines has decreased substantially over the past three decades, although higher levels persist in some jobs and activities. While this suggests that existing controls have been broadly effective, the presence of these higher exposures indicates that further measures may be required. Introduction Mining is known to be a high‐risk industry for occupational injury and disease, with workers exposed to a variety of occupational hazards [1]. In particular, mine workers are exposed to a range of respiratory hazards, including welding fume, diesel particulate matter, and other airborne dusts [2]. Accordingly, employment in the mining industry is a risk factor for respiratory diseases such as pneumoconiosis, chronic obstructive pulmonary disease (COPD), and respiratory cancers [3]. Respiratory symptoms in the absence of disease, such as cough, wheeze, and breathlessness, have also been found to be common among miners, with higher symptom rates among underground miners and those who currently smoke [4]. For example, a study of individuals working in Western Australian mines found that around 30% reported at least one respiratory symptom, with the most commonly reported symptoms being cough (reported by 15%) and wheeze (13%) [5]. The dusts created from mining activities vary in composition and size, depending on the geology, type of mining, and type of activity creating the dust [6]. Respirable dusts are defined as those with a particle size of less than 10 microns, meaning that they are able to be inhaled deep into the lungs and penetrate to the unciliated airways [7]. Recent attention has focused on respirable crystalline silica (RCS), exposure to which has been found to be particularly common in the mining industry, with 92% of Australian miners estimated to be exposed, and 62.5% at a high level [8]. Modeling of exposure measurements from Western Australian mines from 1986 to 2014 showed an overall decreasing trend in RCS exposures over time, although some operations were still associated with levels of exposure which may be associated with adverse health effects [9]. Less attention has been paid to respirable dust in general. The composition of respirable dust varies, but in mining and quarry environments may include coal, silica, asbestos, and metal dusts [2], alongside dusts not otherwise specified (NOS), sometimes characterized as “nuisance” dusts, which are by definition of inherently low toxicity and free from toxic impurities [10]. Accordingly, respirable dust collected in mining areas in Western Australia has been found to consist primarily of silica, iron, and aluminum, with smaller amounts of asbestos and other metals such as cadmium, chromium, and nickel [11]. There is growing concern about the respiratory health effects of these dusts, with international studies finding exposure to be associated with symptoms including reduced lung function and the development of chronic obstructive pulmonary disease (COPD) [12]. The composition of the inhaled particles may convey additional risks, with evidence that exposure to metal dusts such as lead and nickel may cause pulmonary fibrosis and shortness of breath [13, 14]. Likewise, there is some evidence that inhalation of iron particles may lead to reversible lung tissue damage and may exacerbate responses to other elements or to existing respiratory infection, while aluminum inhalation is known to be associated with lung diseases including asthma and pulmonary fibrosis [15]. The exposure standard for respirable dust in Western Australian mineral mines is currently 3 mg/m3 [16]. Exposure standards establish a legal maximum upper limit of airborne concentrations in a worker's breathing zone, and can be defined as a level at which adverse health effects should not occur in the majority of workers [2]. However, there is evidence that respiratory symptoms and impaired lung function can occur with exposures below this value over a working life, and it has been suggested that the respirable dust exposure standards should be lowered to 1 mg/m3 [10, 12]. In practice, measured levels of respirable dust concentrations tend to be below this exposure standard. International studies have found median concentrations of between 0.12 mg/m3 and 0.68 mg/m3, with an overall decreasing trend over time [1, 17]. Likewise, in Australia, a recent study of exposure to respirable dust at Western Australian lithium mines found an overall mean concentration of 0.29 mg/m3, with important differences by gender and occupation [18]. However, while mean exposures to respirable dust are generally below the exposure standard, maximum exposures for some occupations and tasks are well above this standard, with maximum exposures as high as 18.0 mg/m3 in some instances [1]. This study aimed to explore the trends in exposure to respirable dust in Western Australian mines using an industry‐wide database containing monitoring data from 1986 to 2014. Trends were investigated over time for different mine types. Materials and Methods Exposure Data In Western Australia, each mine is required to undertake a regular structured risk assessment of all occupational health hazards, including exposure to airborne contaminants. Based on this risk assessment and a workforce survey, sampling quotas for airborne contaminants are set for each mine site, with the aim of ensuring that samples are representative of the workforce and exposure conditions. Quota samples (93.4% of the measurements in this dataset) are therefore intended to represent typical exposure scenarios rather than high‐ (or low‐) risk tasks. Mines may also collect additional samples (6.6% of the measurements in this dataset), sometimes under the direction of the regulator, and these samples may represent high or atypical exposure scenarios. Personal exposure monitoring is required to be undertaken by a registered hygiene sampler and results reported to the Department of Energy, Mines, Industry Regulation and Safety (DEMIRS), where they are stored in an electronic database, previously known as CONTAM [19]. For each sampling result, information about the mine site and employee (ID, job title, shift length, and use of respirator) is collected, alongside details of the sample, including purpose of sampling (quota or other), date, duration, type of sampling (either time‐weighted average, short‐term exposure limit, or peak exposure), equipment used, and flow rate. All identifying details were removed before data were provided to the research team. For this study, respirable dust exposure measurements from June 1986 to December 2014 were available. All samples were collected in accordance with Australian Standard AS2985:2009, using a cyclone sampler with either a 25 mm or 37 mm filter. In total, 110,002 personal exposure measurement results were reported for respirable dust. In line with guidance from AS2985:2009 regarding sampling duration, we excluded samples taken for less than 240 min (n = 44) as well as those taken for greater than 840 min (n = 20), with extended shift lengths typical in Western Australian mines [20]. We also excluded samples collected with a noncompliant device (n = 43) and samples designed to measure peak or short‐term exposures (as opposed to a time‐weighted average) (n = 4). The upper limit of determination for respirable dust was set at 10 mg/m3 in accordance with NIOSH guidelines [21]; a total of 449 samples were above this limit and were excluded from analysis. This left 109,442 samples for analysis. Ethics approval for this study was obtained from the Western Australian Department of Health Human Research Ethics Committee (approval number RGS0000006906) and the University of Western Australia Human Research Ethics Committee (approval number 2024/ET000578). The analysis used deidentified regulatory exposure monitoring data provided under the Department of Local Government, Industry Regulation and Safety (formerly Department of Energy, Mines, Industry Regulation and Safety) governance framework. Occupational Grouping Each monitoring result in the database was coded with the job title of the monitored employee. These job codes followed a hierarchical structure with nine major divisions, namely: “Management and supervisory,” “Underground production and services,” “Mining production and services (surface),” “Ore treatment occupations,” “Railway operations occupations,” “Metal working processing trades,” “Electrical or electronic trades,” “Miscellaneous trades or utilities,” and “Material handling—store or warehouse occupations.” To account for the lack of measurements for some job titles, these were collapsed into broader occupational groups aligned directly with these hierarchical divisions, ensuring that roles with similar tasks were combined. Occupational groups were therefore created on the basis of job tasks rather than exposure data. A total of 18 surface mining groups and five underground mining groups were created (23 occupational groups in total; as shown in Table 1). TABLE 1Summary statistics of respirable dust measurements (mg/m3) in Western Australian mines, 1986–2014.N%< LODAMGMGSDOverall109,44232.50.5230.2633.19Years1986–199013,74817.40.7390.3933.101991–199521,09823.00.6390.3433.131996–200018,61527.90.5530.2903.092001–200516,87734.00.4960.2493.152006–201020,99242.60.4040.2033.082011–201418,11246.60.3560.1812.98Mine typeGold48,14329.60.5500.2823.17Iron ore23,19932.90.5200.2583.20Industrial minerals13,27642.70.4170.2043.12Nickel11,68031.50.4920.2653.08Other minerals811135.60.5020.2453.22Base metals322329.00.5490.2903.10Exploration181027.60.8640.3483.85Excavation typeSurface51,57534.00.5280.2563.25Underground57,86731.20.5180.2693.13Occupation groupManagement/administration304245.30.3340.1832.90Engineers/professionals279749.80.3250.1732.87Blast hole drillers638829.30.5300.2753.11Blasting workers251029.40.4810.2712.96Exploration drillers319224.61.0150.4223.93Excavation equipment881838.80.3670.2062.87Mobile plant operators483933.00.4670.2513.04Drivers964141.60.3380.1932.82Service occupations180735.80.4220.2263.05Ore processing plant operators18,04029.30.5750.2903.21Ore mobile plant operators184350.80.3220.1642.95Ore product handlers176638.70.5400.2403.50Ore samplers880924.10.8120.3763.54Railway occupations27554.50.5230.1913.67Metal working trades960737.70.5050.2373.30Electrical/electronic trades228943.00.3970.1963.04Miscellaneous trades/utilities160346.20.4130.1853.17Material handling occupations60353.10.2540.1502.64Underground miners557118.80.6070.3642.88Underground drillers316123.80.4960.2992.80Underground loading/transport10,18123.50.5750.3273.01Underground service occupations222426.10.4950.2862.89Underground winding/hoisting43624.80.5770.3233.14Abbreviations: N, number of samples; %< LOD, percentage of samples below limit of detection (0.15 mg/m3); AM, arithmetic mean; GM, geometric mean; GSD, geometric standard deviation.John Wiley & Sons, Ltd. Summary statistics of respirable dust measurements (mg/m3) in Western Australian mines, 1986–2014. Abbreviations: N, number of samples; %< LOD, percentage of samples below limit of detection (0.15 mg/m3); AM, arithmetic mean; GM, geometric mean; GSD, geometric standard deviation. Statistical Analysis All analyses were undertaken in SAS v9.4 (SAS Institute, Cary, NC). Respirable dust concentrations were summarized as arithmetic mean (AM), geometric mean (GM), and geometric standard deviation (GSD). Concentrations were log‐normally distributed and analyses were performed on natural log‐transformed data. The limit of detection (LOD) was defined as 0.15 mg/m3 following NIOSH guidelines, whereby the estimated LOD for respirable particulates is 0.03 mg per sample, and assuming a sampling volume of 0.20m3 or 200 L [21]. Measurements below the LOD (32.5% of all samples) were imputed using maximum likelihood estimation following the method described by Lubin and colleagues [22]. This method assumes that values below the LOD follow the same log‐normal distribution as observed data, and as such a single log‐normal distribution was fit across all measurements. True values were assumed to lie between a background level of 0.01 mg/m3 (selected as a plausible lower bound consistent with typical ambient background concentrations of airborne particulate matter) and the LOD of 0.15 mg/m3. As previously described [9, 23], the PROC MIXED restricted maximum likelihood method was used to develop the exposure prediction model. Models were compared using the Akaike Information Criterion (AIC). Fixed effects included year of measurement (modeled as a continuous variable), sampling duration (in minutes), reason for sampling (quota or additional), and mine type (based on mineral mined), as well as the interaction between year of measurement and mine type. Mine type was categorized as gold, nickel, iron ore, base metals (lead, copper, or zinc), industrial minerals, other minerals or exploration. The other minerals group primarily comprised bauxite/alumina (45% of measurements in this sample) and tin/tantalum/lithium (26%). Random effect terms included job title and mine site identification number, assuming a variance‐components covariance structure and independent residual errors. As worker identification numbers were not assigned prior to 2000, these were not included in the model (due to the high level of missing data). Results Measurements of respirable dust (n = 109,442) were collected from individuals representing 246 different occupations. The majority of measurements came from gold (n = 48,143, 44.0%) and iron ore (n = 23,199, 21.2%) mines, with a relatively even split of measurements from underground (n = 57,867, 52.9%) and surface (n = 51,575, 47.1%) mines (Table 1). Geometric means (GMs) ranged from 0.393 mg/m3 in 1986–1990 to 0.181 mg/m3 in 2011–2014, with the highest levels observed in exploration operations (GM 0.348 mg/m3), followed by base metal (0.290 mg/m3) and gold mines (0.282 mg/m3) (Table 1). Jobs with the highest exposure to respirable dust included exploration drillers (0.422 mg/m3), ore samplers (0.376 mg/m3), and underground miners (0.364 mg/m3). Exposure to respirable dust has decreased over time by an average of 3.51% per year (95% confidence interval −3.63% to −3.39%). The interaction between year of measurement and mine type was significant (p < 0.001), indicating that temporal trends vary by mine type (Figure 1). Exploration activities showed a markedly different trend over time to the other mine types. The highest exposure levels were seen in iron ore mining (GM 0.489 mg/m3 in 1986 through to 0.197 mg/m3 in 2014) and exploration (GM 0.665 mg/m3 in 1994 through to 0.172 mg/m3 in 2014), while the lowest levels were seen in nickel mining (GM 0.369 mg/m3 in 1986 through to 0.135 mg/m3 in 2014) (Table 2). Respirable dust exposure levels (mg/m3) in the Western Australian mining industry over time (1986–2014) for selected mine types, based on modeled exposure data. TABLE 2Modeled respirable dust exposure levels (mg/m3) a in the Western Australian mining industry, by mine type and year.Mine type198620002014Gold0.4640.2790.167Iron ore0.4890.3100.197Industrial minerals0.4650.2940.185Nickel0.3690.2230.135Other minerals0.5560.3020.164Base metals0.5670.2890.147Exploration‐b 0.4430.172 a Model adjusted for sampling reason, occupation code, and mine site code b Measurements in exploration activities commenced in 1994John Wiley & Sons, Ltd. Modeled respirable dust exposure levels (mg/m3) a in the Western Australian mining industry, by mine type and year. Model adjusted for sampling reason, occupation code, and mine site code Measurements in exploration activities commenced in 1994 Table 3 shows the modeled respirable dust exposure levels for 12‐h shifts (as a typical shift length) [20]. for the 15 highest exposed jobs over time. As shown in Table 3, the highest exposure levels were estimated for spray painters (GM 1.497 mg/m3 in 1986 through to 0.539 mg/m3 in 2014), underground ground and roof support workers (GM 1.382 mg/m3 in 1986 through to 0.498 mg/m3 in 2014), and underground crushing operators (GM 1.070 mg/m3 in 1986 through to 0.385 mg/m3 in 2014). Modeled exposure levels for the full list of jobs with at least 10 measurements available (n = 191) are provided in Supporting Information S1: Table 1. TABLE 3Modeled respirable dust exposure levels (mg/m3) a in the Western Australian mining industry for the 15 highest exposed jobs b , by job title and year.Job title198620002014Spray painter (motor/engine trades)1.4970.8980.539Ground or roof support worker (underground)1.3820.8290.498Crushing operator (underground)1.0700.6420.385Hydraulic fill operator (underground)0.9790.5870.352Mechanical bogger driver (underground)0.9520.5710.343Roofbolter (underground)0.9420.5650.339Sand blaster (motor/engine trades)0.8980.5390.323Track laying and maintenance (rail operations)0.8810.5280.317Driller's assistant (surface)0.8480.5090.305Conveyor attendant or operator (underground)0.8410.5050.303Diesel loader operator (underground)0.8350.5010.301Boilermaker0.8230.4940.296Blasthole drill operator's assistant (surface)0.8230.4940.296Exploration driller (surface)0.8210.4930.296Shotfirer (underground)0.8080.4850.291 a Model adjusted for sampling reason and mine site code, standardized to gold mining as the reference category, and estimated for 12‐h shifts. b Jobs with at least 10 measurements available.John Wiley & Sons, Ltd. Modeled respirable dust exposure levels (mg/m3) a in the Western Australian mining industry for the 15 highest exposed jobs b , by job title and year. Model adjusted for sampling reason and mine site code, standardized to gold mining as the reference category, and estimated for 12‐h shifts. Jobs with at least 10 measurements available. Discussion An overall downward trend in levels of exposure to respirable dust was observed between 1986 and 2014 in the Western Australian mining industry, consistent with previously observed trends in exposure to RCS [9]. This trend was observed across all mine types and occupations, although the exact nature of the trend varied by mine type. Decreasing exposures to respirable dust have also been observed in other studies both within Australia and internationally. For example, a study of over 8500 dust samples from Western Australian mines found a general declining trend in respirable dust exposure from 2001 to 2012, with an overall geometric mean of 0.26 mg/m3 over this time period, in line with the current findings [4]. Likewise, a study of over 31,000 respirable dust measurements collected across Europe in the industrial mineral sector between 2000 and 2016 found an overall downward trend in respirable dust exposure of 9% per year on average [17]. This is a steeper decline than estimated in the current study (3.51% per year), potentially reflecting differences in geology, mining practices, and tasks completed, as well as the workers sampled. Several broader changes in the mining industry over this period may have contributed to the observed decline in exposures. For example, improved ventilation and dust suppression technologies, alongside the wider adoption of enclosed cabins and remote‐controlled equipment, may have reduced worker exposure to dust‐generating processes [24, 25]. In addition, increased mechanization and automation of mining processes, such as the use of automated shotcreting and rockbolting devices and intelligent drill rigs capable of more precise drilling [26], together with the concomitant changes in workforce composition from predominantly manual to more technical and remote operation roles [27] may also have contributed to lower exposure levels. Stronger regulations around exposure have also been introduced over time, with the earlier recommended limit of 5 mg/m3 for respirable dust exposure (8‐h time‐weighted average) being replaced by the lower limit of 3 mg/m3 in mineral mines in Western Australia in 1995 [28], following the American Conference of Government Industrial Hygienists (ACGIH) 1994 recommendation [29]. Although overall exposures declined, exposures were found to differ by job, with highest exposures seen among spray painters, underground support workers, and underground crushing operators. These were also among the jobs found to have the highest exposure to RCS in our previous analyses [9]. Crushing operators in particular have been previously described as having high levels of exposure to dust, with a study of lithium mine workers in Western Australia, for example, finding that crusher personnel recorded among the highest levels of respirable dust exposure with a mean exposure of 0.558 mg/m3 [18]. In addition, a recent scoping review identified the crushing process as one of the most consistent contributors to high levels of respirable dust exposure in the mining environment [30]. The wide range of jobs represented among those with the highest exposures indicates that these exposures likely arise from multiple sources of respirable dust, and as such the composition of the dust is likely to differ by job. Many of the jobs which showed the highest levels of exposure (eight of the top 15) were underground mining jobs, although there was no substantial difference in respirable dust exposure between surface and underground mines found. Our previous analyses showed that exposures to diesel particulate matter were generally higher in underground than surface mining jobs [23], although exposures to RCS were generally comparable between underground and surface mining operations [9]. Previous research in coal mines in the United States has shown that respirable dust exposure varies between underground and surface mines, with exposures approximately three times higher in underground mines [31, 32]. The overall geometric mean concentrations for over 650,000 respirable dust samples collected from underground mines over the period 1982 to 2017 was 0.55 mg/m3, with 5.5% of samples exceeding the permissible exposure limit at the time (2.0 mg/m3) [32], while the corresponding mean in surface mines was 0.17 mg/m3, with 1.6% exceeding the exposure limit [31]. In both mine types, the mean exposure declined over time [31, 32]. Exposures also varied by mine type, with the highest levels estimated in iron ore mining and exploration. Higher exposure levels during mining exploration, in particular, may be due to an increased frequency of drilling and blasting activities, possibly without the level of dust control that would be in place on an operational mine site [24]. A recent analysis of RCS exposure in Western Australian mines also found exploration activities to be associated with the highest levels of exposure, with exceedances of the exposure standard being four times more common during exploration activities compared with operational mining activities [24]. Exposures in exploration activities also showed a steeper decrease over time than those in other mine types. This may be related to changes in the nature of exploration activities, including a move towards brownfields exploration (around existing deposits) and the use of more sophisticated drilling and exploration methods, as well as changes in the minerals targeted for exploration [33, 34]. In addition to exposure levels, the mine type is likely to influence dust composition, which may have important consequences for the associated health impacts [6], with differences in the incidence of mine dust lung diseases seen between mines and mining regions [35]. Indeed, even within the same mine, the composition of respirable dust may vary considerably [35, 36]. The observed overall decline in respirable dust exposure levels, alongside the finding that most measurements were below the recommended limit of 3 mg/m3, suggests that existing controls have been broadly effective. However, the presence of higher exposures in specific mine types and occupations indicates that further measures may be required. In particular, jobs such as spray painters, underground support workers, and crushing operators, as well as those working in mine types with consistently elevated exposures, represent priority groups for targeted engineering controls and enhanced surveillance. This is particularly important given concerns that the current exposure standards may not be sufficiently protective, and that exposures below these standards may still pose a risk to worker health [12]. There were some limitations to the data used in the current study to model exposures, which may influence the generalizability of these results. Information regarding the use of RPE was limited to a dichotomous variable of “worn” or “not worn,” with the type of RPE or frequency of use not captured. In addition, it is not known whether the samples were taken inside or outside of the respiratory equipment, although in line with common measurement practices [37] it is likely that measurements were taken outside of the respirator and so may not accurately capture the exposure to the individual's breathing zone. This also means, however, that the decline in exposure levels over time is unlikely to be related to the increased use of RPE observed over the same time period, and accordingly, there was no relationship between exposure to respirable dust and use of RPE at the job level (data not shown). The representativeness of the measurements in the exposure database is unknown, as we have previously noted [9, 23]. While some overestimation of exposure may be possible in circumstances where exposure measurements were taken in areas or under circumstances where higher exposures were expected, there may also have been underestimation of exposure where measurements were taken under “best case” conditions. Only 0.8% of measurements (n = 904) were reported to have been taken under other than usual conditions, and the vast majority of measurements (n = 102,196; 93.4%) were taken to satisfy quota requirements which are intended to represent typical exposure scenarios. As quota samples represent the bulk of measurements in the dataset, the modeled exposure estimates are likely to reflect typical working conditions, although high or atypical exposure scenarios may be under‐represented. Additionally, we controlled for sampling reason in the analyses. Taken together, these factors suggest that any bias introduced by the reason for sampling is likely to be small, and the temporal trends observed unlikely to be meaningfully affected. Our results are also in general agreement with other findings from Western Australia and internationally [4, 17, 18]. The findings of this study show that exposure to respirable dust has decreased substantially over time, with most measurements considerably lower than the occupational exposure standard. However, some operations and occupations are still associated with high respirable dust exposures, indicating that workplace controls and continued monitoring and surveillance is still required, particularly as the health impacts of respirable dust remain largely unknown. Author Contributions Renee N. Carey, Susan Peters, Peter Franklin, and Fraser Brims were involved in the conception and design of this work. Analysis and interpretation of data were performed by Renee N Carey with assistance from Susan Peters and Nicholas de Klerk. Renee N. Carey drafted the work and all authors revised it critically. All authors have provided final approval of the version to be published and agree to be accountable for all aspects of the work. Ethics Statement The work in this manuscript was performed at the University of Western Australia. The Western Australian Department of Health and University of Western Australia Human Research Ethics Committees provided ethical approval. This project used pre‐existing or routinely collected data and so informed consent was not provided. Supporting information Supporting File