Improving rice nitrogen utilization efficiency (NUtE) is imperative to maximizing future food productivity while minimizing environmental threats, yet knowledge of its variation and the underlying regulatory factors is still lacking. Here, we integrated a dataset with 21,571 data compiled by available data from peer-reviewed literature and a large-scale field survey to address this knowledge gap. The overall results revealed great variations in rice NUtE, which were mainly associated with human activities, climate conditions and rice variety. Specifically, N supply rate, temperature and precipitation were the foremost determinants of rice NUtE, and NUtE responses to climatic change differed among rice varieties. Further prediction highlighted the improved rice NUtE with the increasing latitude or longitude. The indica and hybrid rice exhibited higher NUtE in low-latitude regions compared to japonica and inbred rice, respectively. Collectively, our results evaluated the primary drivers of rice NUtE variations and predicted the geographical responses of NUtE in different varieties. Linking the global variations in rice NUtE with environmental factors and geographical adaptability provides valuable agronomic and ecological insights into the regulation of rice NUtE.

Peer-reviewed literature published from 1981 to 2017 was searched using the Web of Science (http://apps.webofknowledge.com/), Google Scholar (http://scholar.google.com/), and China National Knowledge Infrastructure (http://www.cnki.net/) to establish a compiled database. The search terms were “(nitrogen content OR concentration) AND (biomass) AND (rice)”. Studies included in our dataset for further analysis must fulfill the following criteria: (1) the experiment was performed in Asia (with latitudes from 14.11 to 45.22 °N and longitudes from 73.09 to 140.87 °E), and only field rather than soil-pot or hydroponic studies were included; (2) the selected study should contain information on rice cultivars (indica or japonica, and inbred or hybrid rice), experimental sites and soil chemical parameters; (3) studies utilizing inorganic N fertilizer were included in the dataset, while those using compound fertilizer (containing other nutrients) and organic fertilizer were excluded due to their complex compositions; (4) rice shoot N content and shoot biomass were directly documented or could be derived from the results; (5) the application amount of N, phosphorus and potassium fertilizers could be obtained from the study; and (6) the most recent study was selected if more than one publication reported results from the same experiment. Herein, 56 published articles with 922 experiments were retained in the database (Table S1). Additionally, another dataset from 20,649 farm experiments across the major rice-growing regions of China (Data sources: a nationwide field survey performed by the National Agricultural Technology Extension and Service Center (Beijing, China) from 2005–2009), including rice growth parameters and the affiliated soil properties, fertilization management and environmental information, were also incorporated to form a complete compiled database. Finally, 21,571 observations distributed in 7 countries were comprised in our dataset for further analysis.

The data file can be opened with the EXCEL software.