A field study was conducted in two mines in order to determine the most suitable strategy for ambient exposure assessment in the framework of a European study aimed at validation of biological monitoring approaches for diesel exhaust (BIOMODEM). Exposure to dust and particle-associated 1-nitropyrene (1-NP) was studied in 20 miners of black coal by the long wall method (Czech Republic) and in 20 workers in oil shale mining by the room and pillar method (Estonia). The study in the oil shale mine was extended to include 100 workers in a second phase (main study). In each mine half of the study population worked underground as drivers of diesel-powered trains (black coal) and excavators (oil shale). The other half consisted of workers occupied in various non-diesel production assignments. Exposure to diesel exhaust was studied by measurement of inhalable and respirable dust at fixed locations and by personal air sampling of respirable dust. The ratio of geometric mean inhalable to respirable dust concentration was approximately two to one. The underground/surface ratio of respirable dust concentrations measured at fixed locations and in the breathing zones of the workers was 2-fold or greater. Respirable dust was 2- to 3-fold higher in the breathing zone than at fixed sampling locations. The 1-NP content in these dust fractions was determined by gas chromatography-mass spectrometry/mass spectrometry and ranged from 0.003 to 42.2 ng/m(3) in the breathing zones of the workers. In mine dust no 1-NP was detected. In both mines 1-NP was observed to be primarily associated with respirable particles. The 1-NP concentrations were also higher underground than on the surface (2- to 3-fold in the coal mine and 10-fold or more in the oil shale mine). Concentrations of 1-NP in the breathing zones were also higher than at fixed sites (2.5-fold in the coal mine and 10-fold in the oil shale mine). For individual exposure assessment personal air sampling is preferred over air sampling at fixed sites. This study also suggests that particle-associated 1-NP much better reflects the ambient exposure to diesel exhaust particles than dust concentrations. Therefore, measurement of particle-associated 1-NP is preferred over measurement of dust concentrations by gravimetry, when linking ambient exposure to biomonitoring outcomes such as protein and DNA adducts and excretion of urinary metabolites of genotoxic substances.