Background. Volumetric pollen traps are commonly used to assess pollen exposure. These traps are well suited for estimating the regional mean airborne pollen concentration but are likely not to provide an accurate index of personal exposure. In this study, we tested the hypothesis that hair sampling may provide different pollen counts from those from pollen traps, especially when the pollen exposure is diverse. Methods. We compared pollen counts in hair washes to counts provided by stationary volumetric and gravimetric pollen traps in 2 different settings: urban with volunteers living in short distance from one another and from the static trap and suburban in which volunteers live in a scattered environment, quite far from the static trap. Results. Pollen counts in hair washes are in full agreement with trap counts for uniform pollen exposure. In contrast, for diverse pollen exposure, .individual pollen counts in hair washes vary strongly in quantity and taxa composition between individuals and dates. These results demonstrate that the pollen counts method (hair washes vs. stationary pollen traps) may lead to different absolute and relative contributions of taxa to the total pollen count. Conclusions. In a geographic area with a high diversity of environmental exposure to pollen, static pollen traps, in contrast to hair washes, do not provide a reliable estimate of this higher diversity.