C:N ratios of bulk soils and particle-size fractions: Global trends and major drivers

Main drivers and affecting factors of global soil C:N ratios. Blue arrows indicate positive correlations, red arrow indicate negative correlations, and arrow thickness indicate the strength of correlations. • Globally, climate exerts strong control on C and N contents in soils. • Particle-size distribution is the main driver of global soil C:N ratios. • C:N ratios in the sand fraction are the highest and have the highest variability. • C:N ratios in the clay are the lowest (∼10) and mostly affected by precipitation. • Soils with <15% clay have N-saturated clays, affecting N cycling and availability. C:N ratios are the most ancillary and available proxy for soil organic matter (SOM) chemical composition and quality in the literature, but there is surprising little information on how it is affected by environmental factors. To address this knowledge gap, we searched the global literature and gathered information from 74 studies conducted between 1980 and 2019, covering 29 countries and a wide range of variation environmental settings. We compiled data on soil organic carbon (SOC), total N (TN) and C:N ratios in bulk soil and, for the first time, within soil particle-size fractions – sand, silt, and clay. Correlation and random forest statistical analyses showed that SOC and TN contents were mostly controlled by macroscale, climatic drivers, namely temperature and precipitation, whereas soil C:N ratios were more responsive to microscale drivers, e.g. soil particle-size distribution. Specifically, C:N ratios in bulk soil were most strongly correlated with sand contents ( r = 0.52; p < 0.001), consistent with the high C:N of sand-sized SOM (median 16.7). Otherwise, clay-sized SOM showed much less variation in C:N ratios (5.0–17.6, median 9.6), regardless of sampling depth and land use. Clay C:N ratios were only consistently lower under arid climate types. Silt-sized SOM showed intermediate C:N ratios (median 13.1) relative to sand and clay-sized fractions. Hence, our results suggest a preferential immobilization of N-enriched SOM onto clay-sized particles coupled to a segregation of N-depleted SOM in sand-sized fractions. Although bulk soil C:N increased with sand contents, the accumulation of N-depleted SOM within sand-sized fractions increased with clay content. Overall, by identifying the complementarity of macro- and microscale factors underpinning soil C and N pools and their C:N ratios, our results contribute to reduce the inherent complexity to model SOM cycling across global to local scales.