Etschmann, B.Donner, E.Brugger, J.Howard, D.de Jonge, M.Paterson, D.Naidu, R.Scheckel, K.Ryan, C.Lombi, E.2014-10-122014-10-122014Environmental Chemistry, 2014; 11(3):341-3501448-25171449-8979http://hdl.handle.net/2440/86024Fast X-ray detectors with large solid angles and high dynamic ranges open the door to XANES imaging, in which millions of spectra are collected to image the speciation of metals at micrometre resolution, over areas up to several square centimetres. This paper explores how such multispectral datasets can be analysed in order to provide further insights into the distribution of Cu species in fresh and stockpiled biosolids. The approach demonstrated uses Principal Components Analysis to extract the ‘significant’ spectral information from the XANES maps, followed by cluster analysis to locate regions of contrasting spectral signatures. Following this model-free analysis, pixel-by-pixel linear combination fits are used to provide a direct link between bulk and imaging XANES spectroscopy. The results indicate that both the speciation and distribution of Cu species are significantly affected by ageing. The majority of heterogeneously distributed micrometre-sized Cu sulfide particles present in fresh biosolids disappear during the oxidative stockpiling process. In aged biosolids most of the Cu is homogeneously redistributed on organic matter suggesting that Cu mobility is temporarily increased during this redistribution process. This manuscript demonstrates how large XANES imaging datasets could be analysed and used to gain a deep understanding of metal speciation in environmental samples.enJournal compilation © CSIRO 2014Agriculture; biosolids; copperJournal article003000828210.1071/EN131890003402702000132-s2.0-8490331668373317