Diurnal and seasonal variation of carbonaceous aerosols at a remote MBL site of La Reunion island

In this study, we characterise the variability of black; carbon (BC) obtained at Sainte-Rose, a tropical marine site of La Reunion island (21.5 degreesS, 55 degreesE) from one and a half-year (1998-1999) semi-continuous ground-based measurements. BC diurnal variation shows low concentrations during night-time (mean similar to 10-20 ng/m(3), following the season) and moderate levels at daytime (mean similar to 35 ng/m(3)), while BC peaks are observed at the beginning and end of the day. Morning (mean similar to 50 ng/m(3) at 0700 h) and evening (mean similar to 70 ng/m(3) at 1800 h) BC peaks are thought to be due to local pollution induced by anthropogenic activities, most probably car exhausts from nearby roads. Night-time BC exhibits a seasonal variation too, with maximum levels observed during autumn and winter(similar to 20 ng/m(3)) and lowest values measured during spring and summer (similar to 10 ng/m(3)). BC seasonality suggests a local-to-regional anthropogenic origin of carbonaceous aerosols following the season. These suggestions are confirmed by wind sector and BC concentrations analysis at Sainte-Rose, which show that daytime BC aerosols originate mainly from inland while night-time BC concentrations originate from northeasterly (regional inhabited areas) to southeasterly (marine) regions, following the season. Wind sector analysis also points out important seasonal shift in origin of night-time air samples. These assumptions are confirmed by trajectory analysis indicating a spatial shift in the origin of air samples following the season. Finally, it is shown that during austral spring and summer, low carbonaceous aerosol concentrations encountered at this site are weakly subjected to scavenging processes. Results point out the potential of ground-based measurements to assess the origin and causes of carbonaceous aerosol variability at remote locations under favourable conditions (minimum local anthropogenic inputs and meteorology). (C) 2001 Elsevier Science B.V. All rights reserved.