Establishing a measure of reproducibility of ultrahigh-resolution mass spectra for complex mixtures of natural organic matter.

This study describes a method for evaluating the reproducibility of replicate mass spectra acquired for complex natural organic matter (NOM) samples analyzed by electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry, with regard to both peak detection and peak magnitude. Because studies of NOM characterization utilize not only the emergence and disappearance of peaks but also changes in relative peak magnitude, it is important to establish that the differences between samples are significantly larger than those between sample replicates. Here, a method is developed for correcting strict signal-to-noise thresholds, along with a new scheme for assessing the reproducibility of peak magnitudes. Furthermore, a systematic approach for discerning when the comparison of samples by the presence or absence of peaks is appropriate and when it is necessary to compare based on the relative magnitude of the peaks is presented. A variety of 10 different types of NOM samples are analyzed in duplicate or triplicate instrumental injections or experimental extractions. A framework for these procedures is provided, and acceptable reproducibility levels are recommended.