Auxiliary material for Paper 2008GL033294 The continental margin is a key source of iron to the HNLC North Pacific Ocean Phoebe J. Lam Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA James K. B. Bishop Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA Department of Earth and Planetary Science, University of California, Berkeley, California, USA Lam, P. J., and J. K. B. Bishop (2008), The continental margin is a key source of iron to the HNLC North Pacific Ocean, Geophys. Res. Lett., 35, LXXXXX, doi:10.1029/2008GL033294. Introduction This auxiliary section contains a figure and a table related to the determination of the oxidation state of marine particulates, and a table of data. Figures showing an example of full XANES spectrum and the isolation of the pre-edge feature are given in Figure S1. The results to the determination of the pre-edge centroid are given in Table S1. Mn and Fe concentrations plotted in Figures 1a and 1b are reported in Table S2. 1. 2008gl033294-fs01.tif Normalized XANES spectrum of particulate sample from cast 7, 465 m. The pre-edge background was first removed with an elastic fit with fluorescence of 6650eV, then the XANES spectra were normalized for atomic absorption based on the average absorption coefficient between 7300-7400eV. The rectangular box encases the pre-edge feature along with the beginning of the main edge jump. b) Pre-edge feature before and after extraction. The pre-edge feature boxed in FigS1a is plotted (solid line). The feature was extracted from the edge jump contribution by removing a 3rd order polynomial fit to the background (dotted line) using data several eV before and after the feature (~7108-7110eV and ~7119-7122eV). The centroid position (*) was determined by integrating the background-subtracted peak area (dashed line) between 7108-7120 eV. 2. 2008gl033294-ts01.txt Pre-edge centroid positions and calculated redox ratios. Centroid positions were shifted by -1.045eV to line up with the Fe foil calibrations of Wilke et al. (2001) prior to redox ratio calculation. Sample "2002153" is a blank. Total Fe in samples "2001021" and "2002163" (60m) were very low, with 50-60% of the XANES signal attributable to the filter blank (2002153). The redox ratios for these samples were calculated with and without subtracting the contribution from the filter blank and are presented in the table, but were not plotted in Fig. 4. XANES for sample "2002164" were unusable because of a glitch near the pre-edge feature. 3.1 Column "SampleID", sample identification. 3.2 Column "Cast", MULVFS cast number. 3.3 Column "Depth", m. 3.4 Column "Centroid", eV, position of the centroid of the pre-edge feature after shifting by -1.045eV. 3.4 Column "Redox-ratio_percent", percentage, the percentage of total Fe that is in Fe(III) form (Fe(III)/Sum(Fe)*100). 3. 2008gl033294-ts02.txt Mn and Fe concentrations plotted in Figures 1a and 1b. All concentrations are blank corrected. "0" indicates that value was below detection. "null" indicates sample was not analyzed. 4.1 Column "SampleID", sample identification. 4.2 Column "Cast", MULVFS cast number. 4.3 Column "Depth_m", m. 4.4 Column "labileMn_pM", acid-leachable labile particulate Mn concentration determined by ICP-MS, pmol/L. 4.5 Column "totalMn_pM", total particulate Mn concentration determined by XRF, pmol/L. 4.6 Column "labileFe_pM", acid-leachable labile particulate Fe concentration determined by ICP-MS, pmol/L. 4.7 Column "totalFe_pM", total particulate Fe concentration determined by XRF, pmol/L. 4.8 Column "totalFe_sd_pM", standard deviation of total particulate Fe concentration determined by XRF (n=2), pmol/L.