Understanding dissolved organic matter (DOM) transformation is crucial for comprehending soil biogeochemical cycling. However, the extent that soil microbes mediate DOM transformation at the molecular level, and whether this is regulated by fertilization remain largely unknown. Here associate Prof. LIU Ming and his team investigated soil DOM transformations under long-term fertilization using Fourier-transform ion cyclotron resonance mass spectrometry, high-throughput sequencing, and machine learning. Fertilization greatly promoted transformation potential of DOM molecules. Organic fertilization increased the mean transformation number of DOM molecules by 260% compared to no-fertilization, while chemical fertilization increased it by 193%. Machine learning indicated that intrinsic DOM molecular characteristics could predict transformation potential, especially for medium- or low-transformation-potential molecules. However, high-transformation-potential DOM molecules were more influenced by soil microorganisms. Their study provides a parameter to characterize potential transformation capacity of DOM molecules, the effects of different fertilization treatments on this potential, and highlights microbial contributions to molecular transformation processes, identifying the key microbial groups.

This study was published in Communications Earth & Environment in February 2025.

Graphical abstract