The total concentrations of particulate trace metals (Al. Cd, Co, Cu, Fe. Mn, Na. Ni. Pb, Zn) with analytical quality assurance, have been determined in atmospheric aerosols at two coastal sites, and during cruises in the Celtic Sea. Sampling at a site on the western English Channel covered 19 months and represents one of the most comprehensive time series of trace metals in atmospheric aerosols. Aerosol concentrations of Cd. Cu, Pb and Zn for the English Channel were lower than previously reported and Al. Co and Mn concentrations were similar to literature values. The elements were grouped according to behaviour such that Group 1 elements (Cd, Fe. Mn, Pb and Zn) displayed enhanced concentrations in autumn/winter 1994 and 1995. whereas Group 2 elements (Al, Co, Cu, Na and Ni) had enhanced concentrations during winter 1995 only. This was ascribed to source functions being dependant on wind direction with Group 1 elements being carried mainly by north easterly air masses, whereas Group 2 elements originated mainly with air masses from the south west. Dissolved trace metal (Al, Co, Cu, Ni, Pb) analyses were carried out on rain waters collected at the English Channel site. The rain water analyses showed that the soluble trace metal fraction was in the order Co = Cu > Ni >Pb > Na » Al and wet and dry depositional fluxes accounted for the differential behaviour and the solubility of aerosol trace metals on contact with sea water. Aerosol concentrations are also reported for the north western Mediterranean and the Celtic Sea. In the former location, the aerosol trace metal concentrations were lower than those reported in literature and in the Celtic Sea there was a gradient in the trace metal composition of the aerosol from land to open sea. The solid state speciation of Cd and Pb gave an increasingly matrix-bound fraction in the order north western Mediterranean < western English Channel < Celtic Sea and the Celtic Sea aerosols had the greatest sea water solubilities of all elements, except Ni. Concentrations of Cu, Ni and Pb in English Channel aerosols, rain waters and the fraction partitioning from aerosols on contact with sea water, together with sea water concentrations from the literature, were used to devise a trace metal flux model for the English Channel. The model showed that of the total trace metal fluxes into the English Channel, the atmospheric fluxes were in the order of importance Pb > Cu > Ni. The overall budgets revealed discrepancies in the mass balances, which were identified as sediment-water exchange for Ni and Pb and adsorption onto particles for Cu. These budget calculations provide a basis for the development of more advanced modelling concepts involving coupled atmosphere-ocean interactions.

Trace element concentrations vary substantially during snowmelt runoff, with changes in the dissolved versus particulate fractions potentially impacting their movement at the catchment scale. To investigate trace element behavior in a snowmelt dominated stream, we measured concentrations in different size fractions in the Provo River (northern Utah, USA). We sampled the river at three locations during water years 2016-2018 and 2021-2023 for trace metal and major ion concentrations. During the final year, we collected three fractions (unfiltered, <0.45 µm filtered, and <0.22 µm filtered) for trace metal chemistry and calculated the particulate concentration as the difference between the unfiltered and 0.45 µm fraction. Stream measurements included pH, discharge, turbidity, and fluorescent dissolved organic matter (fDOM). We also sampled water sources (soil water, ephemeral streams, and snowpack) to compare trace metal concentrations across the watershed with the stream. Trace metal (Al, Be, Fe, Pb) and rare earth element (REE +Y) concentrations had the highest particulate and dissolved concentrations during snowmelt runoff. In contrast, major cations were primarily found in the dissolved fraction with lowest concentrations during snowmelt runoff. Major cation and particulate metal concentrations increased from upstream to downstream. The increased trace metal and REE + Y concentrations during snowmelt relative to baseflow may be explained by increased discharge and associated increased turbidity and fDOM. A comparison with water sources suggests that dissolved and particulate trace metals in the stream are sourced from flushed soil water, as trace metal transport through the watershed is facilitated by suspended sediment and dissolved organic matter. For most of the trace metals and REE +Y, concentrations were similar in the <0.45 µm and <0.22 µm fractions. Yet Al tended to have higher concentrations in the <0.45 µm fraction, suggesting a colloidal form of Al between 0.22 µm and 0.45 µm may exist. Differences from upstream to downstream may be related to changes in pH, from ~7 at the upper site to ~8 at the lower sites, which would change sorption capacity, saturation indices, or speciation. Our study demonstrates that trace metal concentrations are variable during snowmelt as the metals interact with suspended particles or dissolved organic matter and are influenced by changing water chemistry, with implications for understanding water quality impairments in snowmelt-dominated streams.