Organochlorine pesticides (OCP) were monitored at the Weissfluhjoch site (Switzerland) and the top of Mt. Sonnblick (Austria) with a low volume active air sampler and semipermeable membrane devices (SPMD). The air sampling rates (R (air)) of the SPMD for OCP were calculated. Statistical tests showed that there was no significant difference between R (air) at the two different sampling sites. Quantitative structure-property relationship (QSPR) models of the R (air) of the SPMD were developed for OCP using partial least square (PLS) regression. Quantum chemical descriptors computed by the semi-empirical PM6 method were used as predictor variables. The cumulative variance of the dependent variable explained by the PLS components and determined by cross-validation (Q (2) (cum)) was > 0.818 for each optimal model. This indicates that the model has good predictive ability and robustness. The R (air) of the SPMD for OCP is related to the total energy, the van der Waals area and the total dipole moment of the OCP molecules. The main factors governing R (air) values of OCP are intermolecular interactions and the energy required for cavity-formation in dissolution of OCP into triolein of the SPMD. The linear correlation coefficient between predicted and experimental values were all > 0.921.