AbstractTillage is an important cause of release of particulate matter (PM) from soils. Measurements in Europe indicate that emission of dust by tillage operations is many times higher than by natural wind erosion. In Europe there are two periods when most tillage operations are carried out – in spring and in late summer. These operations are carried out under different soil moisture conditions. In spring soil moisture is close to field capacity and the surface soil dries mainly by evaporation. At this surface a dry layer is established, which is only few centimetres thick and impedes further evaporation. A different situation occurs in late summer, when the crops have depleted the soil water in the root zone. In this case the soils are dried‐up to a depth of some decametres. Tillage operations differ in depth and intensity, ranging from few centimetres of mixing at seedbed preparations to a few decametres of turning by ploughing. Hence, tillage‐related emissions are the combined result of vertical soil moisture distribution and depth and intensity of tillage operations.The objective of our study is to determine the influence of soil moisture on the particulate matter emission of soils subjected to mechanical disturbance, particularly tillage. Because the emission is a result of many variables that cannot be controlled in field experiments, we followed a stepwise analysis of the main influencing factors. First, we used a wind tunnel as a cross‐flow gravitational separator to investigate the relation between soil type, soil moisture and fine dust emission. Twelve soils of different texture were investigated with regard to their water content within a range from 0 to 40 per cent mass. The results show that soils can emit dust over a certain range of moisture, but even a small increase in soil moisture causes a distinct reduction of dust emission. The threshold water content for fine dust emission of soils was between 2 and 5 per cent mass for sandy soils, 5–10 per cent mass for silty soils, about 30 per cent mass for the clayey soils and 25–45 per cent mass for organic soils.The wind‐tunnel results were used to calculate the PM10 emission potential of a sandy soil in spring and late summer for a tillage depth of 20 cm by taking the measured vertical soil moisture profiles at these times into consideration. In spring only the upper 2·5 cm was dry enough to emit PM10, whereas in summer the soil was desiccated to the entire tillage depth. The calculated PM10 emission potential resulted in 13·4 g m−2 for the soil moisture conditions in spring and 76·8 g m−2 in summer. Finally, field measurements of the PM10 emission were made when ploughing the soil on both occasions. Modelling the measured data with the Lagrangian dispersion model GRAL resulted in PM10 emissions of 0·12 g m−2 for ploughing under the conditions in spring and 1·05 g m−2 in summer.Our results show the importance of the vertical soil moisture profile on the PM emission of soils. Therefore, the emission factors resulting from field operations should preferably be related to the affected amount/volume of a soil that is dry enough to emit PM rather than to the affected area. This is confirmed by the good correlation between the proportions of affected soil to the measured emissions. Our findings show that the differences between results obtained from laboratory methods and field measurements need further clarification. Copyright © 2008 John Wiley & Sons, Ltd.