Particulate aluminium in boreal streams: Towards a better understanding of its sources and influence on dissolved aluminium speciation

The adverse impacts of the inorganic labile monomeric Al (Ali) fraction on aquatic organisms have meant that Al (Altot) determination and even speciation has become a routine part of environmental monitoring and assessment. However, if samples are not filtered prior to analysis then particulate Al (Altot(p)) could influence the determination of Altot, and therefore the determination of the more toxicologically important (Ali), both when it is measured analytically or modelled from Altot. This paper shows that the Al/DOC ratio in unfiltered samples can identify the Altot(p) fraction, and thus improve the speciation of Ali. These findings are based on data from a study in a 67 km2 catchment in northern Sweden during the snowmelt-driven spring flood of two consecutive years. Filtered and unfiltered samples were studied to determine the spatial and temporal patterns in Altot(p). The concentrations of Altot(p) were greatest in larger downstream sites where significant silt deposits are located. The sites with no silt in their drainage area showed a mean difference between filtered (Altot(f)) and unfiltered (Altot(uf)) samples of 6%, while sites with silt deposits had a mean difference of 65%. The difference between filtered and unfiltered samples was greatest at peak flow. Spikes in Altot(p) did not behave consistently during fractionation with a cation exchange column, resulting in increases in either measured Ali(f) or non-labile monomeric Al (Alo(f)). Altot(p) spikes were associated with sharp increases in the Al:DOC ratio. The baseflow Al:DOC ratio could be used to model filtered Altot from DOC with a Spearman rho of 0.75.