Time variations of gaseous and reactive mercury in the industrial area of Puertollano (south-central Spain). Temporal cycles with marked variations

Abstract

Puertollano (48,086 inhabitants) is the largest industrial city in the Castilla-La Mancha region (South-Central Spain). The city is located some 250 km South of Madrid; it was an important coal mining site during the last century and today it is the location of one of the most important Spanish oil refineries and the only refinery located away from the coast. Nowadays the area (which mainly includes the Ojailen valley) has a large open pit coal mine (Encasur), two power plants (Eon and Elcogas) and a petrochemical complex (Repsol) located S and SE from the town. These industries give rise to a complex scenario in terms of mercury emissions to the atmosphere: Repsol, Elcogas and Eon act as discrete sources, while coal mine and dumps acts as a general, diffuse source. The mercury contents in Puertollano town and the related industrial area were characterized during 2010 and 2011 by acquiring stationary data of Gaseous Elemental Mercury (GEM), Reactive Gaseous Mercury (RGM), meteorological parameters and other atmospheric contaminants (NO, NO2, SO2, benzene, toluene, xylene, ozone and PM10). In addition, several Total Gaseous Mercury (TGM) mobile surveys were carried out covering the Ojailen valley.Total Gaseous Mercury (TGM) in the whole valley was in the range 0-24 ng m-3 in all surveys, while higher levels were found near to the coal mine and in the vicinity of a coal power plant that employs clean technology (Elcogas).Tekran data showed low GEM levels during 2010-2011 (1.81 ng m-3 on average), while lower GEM levels were measured during autumn and summer, and maximum levels in spring (7.32 ng m-3 on average). RGM measurements were 0.0088 ng m-3, i.e., significantly lower than background levels in the USA and Europe (0.04 ng m-3). Concentrations of these mercury species' were higher during summer (0.0117 ng m-3).Multiple regression analysis was carried out and good relationships between GEM levels, meteorological parameters and other pollutants were identified. The best GEM predictors were temperature, relative humidity and NO2, whereas the best predictors for RGM were GEM, temperature and ozone. RGM variations seem to be explained predominantly by photoxidation processes, with GEM availability and transport processes of secondary importance.