Spatial variation in actual and required nitrogen use efficiency and the potential to close the gap by management practices.

To boost crop production, China uses almost a third of the world's nitrogen (N) fertilizer. However, N losses due to enhanced application of N fertilizers has led to surface water and groundwater pollution. A reduction in N losses without reducing crop yields is possible by increasing nitrogen use efficiency (NUE), which is important for the effective management of local crop production and water quality. This study used two representative agricultural counties in China (Quzhou and Qiyang) to assess if it is possible to achieve N loss thresholds in surface and groundwater by optimizing N management measures while maintaining actual crop production. We used a spatially explicit N balance model to assess the spatial variation in actual N inputs to soil and N losses to water, and in critical N losses and associated agricultural N inputs. We also used this model to calculate the spatial variation in actual NUEs and the required NUE to align actual crop production with N thresholds. We then assessed the feasibility of achieving the necessary NUE changes through optimizing agricultural N management strategies. It was found that actual N input exceeded critical N input in 95 and 83 % of the agricultural area in Quzhou and Qiyang, respectively. To meet actual crop production without exceeding N loss thresholds, the NUE needs to increase with 11 to 15 % whereas the total N input needs to be reduced by 37 %. NUE gaps can be closed by reducing N rates, enhancing organic manure recycling, and using efficiency-enhancing fertilizers, with optimal combinations being dependent on site conditions.

Global mean nitrogen recovery efficiency in croplands can be enhanced by optimal nutrient, crop and soil management practices.

An increase in nitrogen (N) recovery efficiency, also denoted as N use efficiency (NUEr), is crucial to reconcile food production and environmental health. This study assessed the effects of nutrient, crop and soil management on NUEr accounting for its dependency on site conditions, including mean annual temperature and precipitation, soil organic carbon, clay and pH, by meta-regression models using 2436 pairs of observations from 407 primary studies. Nutrient management increased NUEr by 3.6-11%, crop management by 4.4-8%, while reduction in tillage had no significant impact. Site conditions strongly affected management induced changes in NUEr, highlighting their relevance for site-specific practices. Data driven models showed that the global mean NUEr can increase by 30%, from the current average of 48% to 78%, using optimal combinations of nutrient (27%), crop (6.6%) and soil (0.6%) management. This increase will in most cases allow to reconcile crop production with acceptable N losses to water. The predicted increase in NUEr was below average in most high-income regions but above average in middle-income regions.

Conversion of grassland to cropland altered soil nitrogen-related microbial communities at large scales.

Land-use changes may dramatically disturb underground microbial biodiversity, especially in the fragile ecological areas. However, the impact of conversion of grassland to cropland on soil nitrogen (N)-related microbial communities is not fully understood in the farming-pastoral ecotone of northern China. Therefore, 24 paired grassland and cropland soil samples were collected in this region to investigate the community structure and assembly processes of soil N-related microorganisms via amplicon sequencing of nifH, archaeal and bacterial amoA, and nxrB genes. The results showed higher ammonia-oxidizing bacteria (AOB) alpha diversity but a lower nitrite-oxidizing bacteria (NOB) diversity in cropland soil compared to grassland soil. Non-metric multidimensional scaling ordinations revealed that diazotroph, AOB and NOB communities differed considerably between grassland and cropland soil. Soil microbial co-occurrence networks showed that conversion of grassland to cropland significantly lowered the average degree, average clustering coefficient, total nodes and links, resulting in less complex microbial networks in cropland soil. Land-use change altered AOB community assembly processes, resulting from a stochasticity-dominated process in grassland soil to a determinism-dominated process in cropland soil. In contrast, deterministic processes were dominant in diazotroph community assembly, whereas stochastic processes were dominant in constructing ammonia-oxidizing archaea and NOB communities in both grassland and cropland soil. These results provide novel evidence that the conversion of grassland to cropland altered the diversity and assembly processes of soil microbial communities involved in soil N-cycling processes, which has important implications for the potential changes in soil functions under land-use changes.