Filter-based optical measurements of light absorbing particulate matter at awavelength of about 550 nm, referred to as soot, in air have been performed during theperiod from 1st June 2005 to 31th May 2009 at Godavari in Nepal, Sinhagad in India andHanimaadhoo in the Maldives. In order to reduce systematic errors due to the lightscattering of non-absorbing particles co-deposited on the filter, such as inorganic saltsand mineral dust, an additional sensor recording backscattered light was implemented.Two protocols of corrections (optical and chemical) were applied to the samplescollected at the observatories. The Indian monsoon circulation with its two annualphases in combination with the location of the combustion sources and their contributionrelative other non-anthropogenic sources dominated the observed patterns of soot at theobservatories in India and Maldives. The observatory in Nepal was however mainlyinfluenced by combustion sources all year around concealing possible variability relatedto the monsoon circulation. At the receptor observatory in the Maldives, peak values inthe soot absorption coefficient occurred during the winter season (December to April)when air was transported from the polluted Indian subcontinent out over the IndianOcean. A close to two orders of magnitude lower values were recorded in air that hadspent more than 10-days over the Indian Ocean during the monsoon season (July toSeptember), suggested to be dominated by particulate matter from remote marinebiogenic sources.

Of the many identified and potential effects ofatmospheric aerosol particles on climate, those of soot particlesare the most uncertain, in that analytical techniques concerningsoot are far from satisfactory. One concern whenapplying filter-based optical measurements of soot is thatthey suffer from systematic errors due to the light scatteringof non-absorbing particles co-deposited on the filter, suchas inorganic salts and mineral dust. In addition to an opticalcorrection of the non-absorbing material this study providesa protocol for correction of light scattering based onthe chemical quantification of the material, which is a novelty.A newly designed Particle Soot Absorption Photometerwas constructed to measure light transmission on particleaccumulating filters, which includes an additional sensorrecording backscattered light. The choice of polycarbonatemembrane filters avoided high chemical blank values and reducederrors associated with length of the light path throughthe filter.Two protocols of corrections were applied to aerosol samplescollected at the Maldives Climate Observatory Hanimaadhooduring episodes with either continentally influencedair from the Indian/Arabian subcontinents (wintermonsoon) or pristine air from the Southern Indian Ocean(summer monsoon). The two ways of correction (optical andchemical) lowered the particle light absorption of soot by 63 to 61 %, respectively, for data from the Arabian Sea sourcedgroup, resulting in median soot absorption coefficients of 4.2 and 3.5 Mm^-1. Corresponding values for the South IndianOcean data were 69 and 97 % (0.38 and 0.02 Mm^-1). A comparison with other studies in the area indicated anoverestimation of their soot levels, by up to two orders ofmagnitude. This raises the necessity for chemical correctionprotocols on optical filter-based determinations of soot, before even the sign on the radiative forcing based on their effectscan be assessed.

Simultaneous measurements of soot (absorbing material at 528 nm) and inorganic ions in aerosol and precipitation at the Maldives Climate Observatory Hanimaadhoo during the period May 2005 to February 2007 have made it possible to calculate the washout ratio (WR) of these components as a measure of how efficiently they are scavenged by precipitation. Based on air trajectories the data have been separated into days with polluted air arriving from the Indian subcontinent in a north-easterly sector during winter and clean monsoon days with southerly flow from the Indian Ocean. The average soot concentration was a factor of ten higher in the former situations.

Despite considerable scatter for individual days a systematic pattern emerged when the WR for the different components were compared with each other. During the monsoon season the WR for soot was similar to that of sulphate and other fine mode aerosol components, indicating that soot containing particles in these situations were efficient as cloud condensation nuclei. The origin of the light absorbing material during the monsoon season is unclear. We speculate that light absorbing material from the tropical ocean surface could contribute to the concentration of "soot" during the monsoon season.

During the polluted winter days, on the other hand, the WR for soot was 3 times smaller than that of sulphate. This indicates that, even after a travel time of several days, the soot containing particles from India have retained much of their hydrophobic property and that the soot must be mainly externally mixed. The low WR and the infrequent rain during this season probably contribute to extending the atmospheric lifetime of soot well beyond several days.

Surprisingly high concentrations of non sea salt calcium were measured during the monsoon season, substantially higher than during the winter season. The origin of these high values might be long-range transport from the Australian or African continents. Another possibility might be exopolymer gels derived from the ocean surface micro-layer.