Is it feasible to reduce tillage and N use while improving maize yield in irrigated Mediterranean agroecosystems?

Mediterranean rainfed areas are transformed into irrigation to stabilize or increase crop yields. The gradual occupation of irrigation leads to an increase in nitrogen use and intensity of tillage. The aim of this work was to evaluate the combined impact of tillage systems and mineral N fertilization rates on maize grain yield, water and nitrogen use efficiencies (WUE and NUE) under Mediterranean irrigated conditions. The study was carried out in NE Spain during three maize growing seasons (i.e. years 2015, 2016 and 2017). A long-term (LTE) tillage and N rate field experiment established in 1996 under rainfed conditions was transformed into irrigation with maize (Zea mays L.) monoculture as cropping system in 2015. Three types of tillage (conventional tillage, CT; reduced tillage, RT; no-tillage, NT) and three mineral N fertilization rates (0, 200, 400 kg N ha-1) were compared in a randomized block design with three replications. In 2015, an adjacent experiment (short-term experiment, STE) with the same layout was set up in an area previously managed under long-term rainfed NT for the last 21 years. Soil water (SWC) and nitrate (SNC) content were quantified. Maize above ground biomass and N uptake, grain yield and yield components, grain N were measured at harvest. The WUE for above ground biomass and yield (WUEB and WUEY, respectively) and NUE, as well as other N-related indexes (nitrogen harvest index, NHI; apparent nitrogen recovery efficiency, NAR) were calculated. In the long-term tillage and N fertilization combination (LTE), the reduction of tillage (NT and RT) led to greater grain yield when applying 200 and 400 kg N ha-1 compared to the use of the same rates under CT. Differently, in the sort-term experiment with preceding NT (STE), tillage systems did not influence grain yields, while N application led to greater yields than the control (0 kg N ha-1). In both situations (LTE and STE), NT and RT enhanced SWC before planting leading to greater crop growth compared to CT. The lack of available water under CT caused lower maize above-ground biomass, yield, and yield components in LTE and, therefore, lower WUEB and WUEY. In LTE, the use of long-term CT led to a significant accumulation of nitrate compared to NT. Differently, in the STE, SNC did not show differences between tillage systems. In the LTE, water and N were used more efficiently to produce above-ground biomass and grain yield in RT and NT. Our study shows that in Mediterranean agroecosystems transformed into irrigation the use of NT and RT with medium rates of N leads to greater maize yield, WUE and NUE than the traditional management based on CT with high rates of mineral N. In rainfed areas with long-term history of no-till, this soil management system can be successfully maintained if transformed into irrigation.

We would like to thank the field and laboratory technicians Javier Bareche, Carlos Cortés, Barbara Jelcic and Silvia Martí. This research work was financially supported by the Ministerio de Economía y Competitividad of Spain (project AGL2013-49062-C4-1-R; PhD fellowship BES‐2014‐070039). DPB received a Juan de la Cierva postdoctoral grant from the Ministerio de Economía y Competitividad of Spain (IJCI-2016-27784).