Ultrahigh resolution mass spectrometry (UHRMS) routinely detects and identifies thousands of molecular formulas (MFs) in natural organic matter (NOM). However, multiple assignments (MultiAs) occur when the several chemically plausible MFs are assigned to one single mass peak. MultiAs for a mass peak consist of one common core MF but different “formula residuals”, or replacement pairs, and increase as more heteroatoms and isotopes are being considered during the assignment process. This poses a major problem for the reliable interpretation of NOM composition in a biogeochemical context. A number of approaches have been proposed to rule out false assignments based on structural constraints or isotopologue detection and intensity ratios. But this becomes increasingly challenging for low abundance mass peaks or when stable isotope labeling (e.g. with 15N, 2H) is employed. Here, we present a new approach based on mass error distributions for the identification of true and false-assignments among MultiAs. An automatic workflow was developed for the detection and exclusion of false assignments in MultiAs based on their recurring replacement pairs and Kendrick mass defect values. The workflow can validate MFs for mass peaks that are close to detection limit or where naturally occurring isotopes are rare (e.g. 15N) or absent (e.g. P, F), substantially increasing the reliability of MFs assignments and broadening the applicability of UHRMS in characterization of NOM, e.g. for organic nitrogen and organic phosphorus in different environmental compartments, which are key components for global elemental cycles.