Agronomic measures improve crop yield and water and nitrogen use efficiency under brackish water irrigation: A global meta-analysis

CONTEXT Brackish groundwater irrigation presents a feasible solution to alleviate water scarcity; however, it also induces salt stress in crops. The effects of agronomic measures on crop yield, water use efficiency (WUE), irrigation water use efficiency (IWUE) and nitrogen use efficiency (NUE) under brackish water irrigation are still unclear. OBJECTIVE The objectives of this study were to: (i) systematically analyze the effects of brackish water irrigation on yield, WUE, IWUE and NUE of cotton and grain crops across different climates and soil types; (ii) investigate how agronomic measures such as irrigation amounts, fertilizer application , and biochar addition affect crop yield, WUE, and NUE under brackish water irrigation; (iii) explore the effects of meteorological conditions, soil properties, and agronomic measures on yield, WUE, IWUE and NUE using random forest model and correlation analyses. METHODS This study conducted a comprehensive meta–analysis of 636 comparisons from 81 global studies on brackish water irrigation. In addition, machine learning was used to assess the effects of environmental conditions and agronomic measures on crop yield, WUE, IWUE and NUE under brackish water irrigation, as well as to determine the relative importance of different influencing factors. RESULTS AND CONCLUSIONS Brackish water irrigation significantly reduced crop yield, WUE, IWUE, and NUE compared to freshwater irrigation. Cotton showed the strongest adaptability to brackish water, followed by wheat, while maize exhibited the poorest adaptability. Acidic soil with low bulk density and loamy texture were most conducive to brackish water irrigation, particularly in semi–arid and semi–humid regions. Increasing irrigation amount by 25–50 % can both increase crop yields and maintain high WUE. Although increased nitrogen application enhanced grain yields and WUE, it was also associated with the risk of significant decline in NUE. Random forest model analysis suggested that prioritizing nitrogen application is essential for cotton under brackish water irrigation, while increasing irrigation water is more critical for wheat and maize. SIGNIFICANCE This study provided valuable insights into the response of crop production indicators under brackish water irrigation and offered recommendations for its rational application.