Nitrogen (N) is a critical nutrient for crop growth, but excessive fertilizer use leads to low N recovery efficiency (NUEr) and environmental damage. Global food production needs to increase substantially by 2050, necessitating strategies to enhance NUEr, currently at 48% globally. Optimized agricultural management involves a combination of nutrient, crop, and soil management practices. Nutrient management focuses on strategies to synchronize crop demand and nutrient availability, using appropriate fertilizer types, rates, timing, and placement. Enhanced efficiency fertilizers and combinations of inorganic and organic fertilizers are examples of improved nutrient management practices. Crop management strategies aim to increase NUEr by using crop sequences with differing N uptake efficiencies. This involves crop rotation, cover crops, and recycling crop residues. Soil management often focuses on reducing soil carbon decomposition through reduced tillage, methods to enhance crop yields, soil biodiversity, and structural stability, and the management of organic residues to improve soil nutrient levels. High-income countries generally have higher NUEr than middle-income countries due to better fertilizer application and mechanization. However, there's regional variation in NUEr due to differences in N management and local conditions, highlighting the need for a global-scale assessment considering the interplay of management practices and site conditions. Previous meta-analyses often focused on single agronomic measures. This study addresses the gap by assessing the combined impact of nutrient, crop, and soil management practices on NUEr while accounting for the significant influence of site-specific conditions.

The research paper extensively reviews the existing literature on nitrogen use efficiency (NUEr) and the effects of various management practices. It cites numerous studies focusing on the impact of different fertilizer strategies (enhanced efficiency fertilizers, combined inorganic and organic fertilizers), crop management techniques (crop rotation, cover crops, residue retention), and soil management practices (tillage). The review highlights the disparities in NUEr between high-income and middle-income countries, attributing the differences to factors such as fertilizer regulations, subsidy policies, and access to technology. It acknowledges the lack of comprehensive global-scale analyses incorporating the interactive effects of management practices and site conditions. The authors point out the limitations of previous meta-analyses, which often concentrated on individual agronomic measures, failing to capture the complex interactions influencing NUEr. This sets the stage for the current study's novel approach, which integrates multiple meta-analyses and primary data to create a more comprehensive and nuanced understanding of the factors driving NUEr.

This study employed a combined meta-analytical and meta-regression approach. First, a meta-model was developed by integrating data from 29 existing meta-analyses on the effects of agronomic practices on NUEr. This meta-model predicted changes in NUEr in response to various management practices and its dependency on site conditions (mean annual temperature, precipitation, soil organic carbon, clay content, and pH). The study then used a single meta-regression model built on 2436 paired observations from 407 primary studies. This model used three different effect size measures: the log-transformed ratio of means (ROM), the raw mean difference (MD), and the standardized mean difference (SMD). The comparison aimed to account for spatial variability in site properties and quantify their interactions with NUEr. The meta-regression incorporated various management practices (nutrient, crop, and soil management) and site conditions as independent variables to predict changes in NUEr. The data was collected from various sources, including the Web of Science database for meta-analyses and direct extraction of data from primary studies. Site-specific climate and soil data was sourced from the CRU and SoilGrids databases. The statistical analysis included weighted mean calculations for the meta-model, and mixed-effects models with interaction terms for the meta-regression to account for the complex interplay between management practices and site conditions. Different response variables (ROM, MD, and SMD) were compared to assess the robustness of the results and minimize biases related to the type of effect size measure used. Akaike’s Information Criteria (AIC) and ANOVA were used to evaluate the contribution of each variable to NUEr and the significance of their interactions.

The analysis revealed that most agricultural management practices led to increased NUEr. Based on the meta-analysis of existing studies, practices like right fertilizer placement and crop rotation showed significant increases in relative NUEr (up to 38%). The meta-regression analysis of the primary data showed that efficient fertilizers, optimized fertilizer placement, rate, and timing, residue retention, and cover cropping significantly increased relative NUEr. However, organic fertilizers and reduced tillage had little to no impact on NUEr. When analyzing the change in absolute NUEr, most management practices resulted in a consistent increase (up to 10%). Enhanced efficiency fertilizers and combined fertilizers showed the most substantial positive effects. The meta-regression analysis showed that site conditions strongly influenced the effects of management practices on NUEr. Soil organic carbon (SOC), soil pH, and mean annual precipitation (MAP) had positive effects, while N application rate and clay content had negative effects. The model explained 57-65% of the variation in management-induced NUEr changes, highlighting the significant role of site conditions. Using the MD model, which showed the most robust explanation of NUEr variation, the study predicted the global potential of combined optimal management practices. It estimated an average increase of 30% in absolute NUEr (from 48% to 78%), primarily driven by nutrient (27%), crop (6.6%), and soil (0.6%) management improvements. The impact of optimal practices varied regionally, with higher increases expected in regions with currently lower NUEr values (e.g., parts of Africa, Central Asia, and South America).

The findings demonstrate the considerable potential for enhancing global NUEr through improved agronomic practices. The strong positive effects of the 4R nutrient management strategy (right fertilizer type, rate, timing, and placement) confirm its effectiveness in optimizing N utilization and minimizing losses. The positive impacts of enhanced efficiency fertilizers and integrated use of organic and inorganic fertilizers align with findings from long-term experiments. Similarly, the importance of crop residue management (retention, cover cropping) and crop diversification (rotation) in improving NUEr supports previous research. The limited effect of reduced tillage, however, suggests that its implementation might need to be coupled with other practices to prevent negative effects on NUEr. The significant regional variations in the impact of management practices emphasize the importance of site-specific approaches. Higher N application rates negatively affect NUEr, as expected, since excess N leads to losses to the environment. The regional differences also highlight the importance of addressing the specific challenges and opportunities presented in each region. The predicted increase in NUEr is substantial, suggesting a pathway toward sustainable intensification of agriculture.

This study provides a comprehensive global assessment of the potential to increase nitrogen recovery efficiency in croplands by optimizing nutrient, crop, and soil management practices. The results show a significant potential for increasing global mean NUEr by 30%, primarily through improvements in nutrient and crop management. This can contribute to both increased food production and reduced environmental impact from nitrogen losses. Future research should focus on developing and implementing site-specific recommendations for optimized management practices, taking into account the complex interactions between management practices and regional conditions. Further investigation into the long-term effects of these practices, combined with economic and social considerations, is needed for successful large-scale implementation.

The study's meta-regression model, while comprehensive, relies on existing data from primary studies. The quality and consistency of data across these studies might vary, potentially influencing the accuracy of the predictions. The model's predictive power is based on the included variables, and unmeasured factors could influence NUEr. Additionally, the study mainly focuses on major cereal crops and the extrapolation to other crops might not be directly applicable without further investigation. The study also notes that the uncertainty estimates in the model account for some variability but cannot entirely capture the full range of potential effects.