Assessing the Contribution of Atmospheric Transport to the Total Pollution Load of the Mediterranean Sea: Facts and Models

Abstract

It is now recognized that a noticeable part of the pollution entering the ocean derives from sources located on land via atmospheric input (NAS, 1978; Waldichuck, 1982). Pollutants of major concern are heavy metals and metalloids such as Pb, Cd, Hg, As, and Sn, petroleum hydrocarbons, chlorinated hydrocarbons and pathogenic microorganisms. Such elements or substances are present in the air in the particulate form (in the aerosol-particle size range) and/or in the gaseous form. The atmospheric life-time of such materials is generally long enough (gt;1 day) to allow them to be transported far from their sources (gt; 1000 km). On the other hand, these life-times are often too short (lt; 1 month) to allow uniform horizontal and vertical mixing of pollutant material in the global atmosphere.

Over the last decade, significant progress has been obtained in understanding the atmospheric transport and removal processes of such materials. However, although the importance of this transport path can now be inferred, the data base available is too limited to allow quantitative estimates of atmospheric fluxes to the ocean. In this context, the major unresolved issue is to define the spatial and temporal variability of atmospheric concentrations and fluxes. How do these fluxes vary with weather, climate and human activity ? Understanding this variability is essential and implies a thorough understanding of the governing meteorological processes (wind systems, precipitation patterns) and to evaluate their effects on the chemical fluxes. On a global scale, scientific efforts have been conducted over the last years, especially over the Pacific Ocean during the SEAREX Programme (Atlas and Giam, 1981; Chesselet et al., 1981; Duce et al., 1983; Fitzgerald et al., 1983; Gagosian et al., 1981; Settle and Patterson, 1982). Some data are now available for pericontinental areas or regional seas such as the Baltic Sea, the North Sea and the Mediterranean Sea (Rodhe et al., 1980; Cambray et al., 1975; Arnold et al, 1982; Chester et al, 1981; Ho et al, 1982). They indicate that airbone inputs, especially for metals, are in the same range as riverine inputs. Such findings have led international concern about this topic. For example, in connection with the UNEP Regional Seas Programme, and especially in the context of the long-term programme for pollution monitoring and research in the Mediterranean Sea (MED-POL Phase II), a working group (GESAMP working group on the interchange of pollutants between the atmosphere and the oceans, led by WMO) has been entrusted to describe atmospheric transport processes and to assess pathways and fluxes, using the Mediterranean Sea as the first example.

A current thinking is that, ideally, every single pollutant of interest should be continuously monitored, which would imply that many sampling stations should be established in order to account for the expected geographical variability of atmospheric deposition patterns. However, such a strategy is unrealistic and would be extremely expensive because of operational, instrumental and analytical costs. Indeed, the concentrations of pollutants in marine air and rain are often extremely low (10⁻⁸ to 10⁻¹¹ g.m⁻³ of air; 10⁻⁵ to 10⁻⁹ g.1⁻¹ of rain) so that sampling and analysis cannot be considered as a routine exercise. Moreover, sample contamination during sampling and analysis is a very critical problem which can only be addressed by highly qualified personnel.

It is therefore more realistic, as a first step, to consider if our present knowledge of sources, atmospheric transport processes, chemical transformations in the atmosphere and deposition processes on the sea surface can help to provide accurate estimates of atmospheric inputs to the Mediterranean Sea. In recent years, transport models have been quite successfully applied to land areas, especially for sulfur and more recently for metals. We will evaluate here how such models could be applied to the Mediterranean Sea. It should be kept in mind that the behaviour of organic pollutants may be extremely difficult to assess. For example, whereas most metals are primarily attached to aerosol particles and undergo little or no chemical transformation during atmospheric transport, organic pollutants may be partitioned between gaseous and particulate forms of different life-times in the atmosphere. Our knowledge of such a partitioning and its spatial and temporal variability is almost non-existent. Moreover, we know very little about the chemical conversion rates of most of these compounds in the atmosphere.