A global synthesis reveals additive density design drives intercropping effects on soil N-cycling variables

Intercropping controls, a variety of agroecosystem processes that are crucial for effective crop production by increasing crop diversity. Prior studies have frequently concentrated on crop nitrogen uptake and apparent nitrogen balance, ignoring comprehensive evaluations of microbial genes involved in N-cycling, nitrogen pools, and nitrogen fluxes such as nitrification, denitrification, and nitrogen fixation processes. Furthermore, the impact of field configurations on these variables is still not well understood. To investigate the effects of intercropping on these soil N-cycling variables, we integrated data from 79 articles and retrieved 538 observations. Notably, intercropping increased amoA-AOA (−0.29%–45.59%, p = 0.053) and nifH gene abundance (9.81%–71.92%, p = 0.005), but had little impact on the others. Limited variations in soil N-cycling variables were explained by individual plant traits, including the photosynthesis assimilation pathway, crop stature, and crop species. This might be attributed to the mismatched responses of the crop's aboveground and belowground parts, as well as soil microbial activity, to intercropping. On the other hand, plant-population traits, such as spatial arrangement and density design, were key moderators of changes in N2O emission, amoA-AOA and nosZ gene abundances. Specifically, strip intercropping significantly reduced N2O emissions, increased nitrogen mineralization as well as amoA-AOA, narG, and nosZ gene abundances. While increasing microbial biomass nitrogen, amoA-AOA, amoA-AOB, and nifH gene abundances, additive intercropping significantly lowered N2O emission, nirK, and nosZ gene abundances. Additionally, compared to replacement intercropping, it creates closer covariation connections between microbial genes involved in N-cycling, nitrogen pools and fluxes. Moreover, changes in the mean annual precipitation, N fertilization rate, and initial pH were significantly correlated with amoA-AOA gene abundance, although their effects on N2O emissions were inversely related. Our findings indicate that population-level density design predominates intercropping effects on soil N-cycling variables. Strip and additive configurations may be key to enhancing soil nitrogen immobilization and reducing N2O emissions to promote the sustainable development of intercropping.