Nineteenth century Parisian smoke variations inferred from Eiffel Tower atmospheric electrical observations

Atmospheric electrical measurements provide proxy data from which historic smoke pollution levels can be determined. This approach is applied to infer autumnal Parisian smoke levels in the 1890s, based on atmospheric electric potential measurements made at the surface and the summit of the Eiffel Tower (48.7degreesN, 2.4degreesE). A theoretical model of the development of the autumn convective boundary layer is used to determine when local pollution effects dominated the Eiffel Tower potential measurements. The diurnal variation of the Eiffel Tower potential showed a single oscillation, but it differs from the standard oceanic air potential gradient (PG) variations during the period 09-17 UT, when the model indicates that the Eiffel Tower summit should be within the boundary layer. Outside these hours, the potential changes closely follow the clean air PG variation: this finding is used to calibrate the Eiffel Tower measurements. The surface smoke pollution concentration found during the morning maximum was 60 +/- 30 mug m(-3), substantially lower than the values previously inferred for Kew in 1863; A vertical smoke profile was also derived using a combination of the atmospheric electrical data and boundary layer meteorology theory. Midday smoke concentration decreased with height from 60 mug m(-3) at the surface to 15 mug m(-3) at the top of the Eiffel Tower. The 19th century PG measurements in both polluted and clean Parisian air present a unique resource for European air pollution and atmospheric composition studies, and early evidence of the global atmospheric electrical circuit. (C) 2003 Elsevier Ltd. All rights reserved.