Fate of 15N-fertilizers in the soil-plant system of a forage rotation under conservation and plough tillage

For agriculture sustainability and biosphere conservation in the 21st century, to improve N use efficiency and to reduce the fertilization rates is fundamental. The N dynamics in the soil-plant system can be altered after the adoption of conservation practices, making necessary to evaluate the fertilizer efficiency and, if needed, to adapt N fertilization. Accordingly, during two years, we assess the effect of conventional plough tillage (PT) and conservation minimum tillage (MT) on the N dynamics in a 14-year-old ryegrass-maize forage rotation. We used a 15N-tracing approach to evaluate the recovery efficiency of N (REN) taking into account both the N immobilized in soil (0-5, 5-15, 15-30, 30-45 and 45-60 cm) and that exported at harvest. Adjacent PT (n = 9) and MT (n = 9) plots were randomly assigned in triplicate to three treatments to which 15NH415NO3 (10 atom% 15N) was applied in one of the three first fertilizations (15NOctober-, 15NMarch- and 15NMay-fertilizer), the others being done with unlabelled N. Plant N concentration (% N) was affected (p < 0.001; n = 18) by the crop [80% of variance explained: ryegrass-1 (2.6 ± 0.9%) > ryegrass-2 (1.9 ± 0.4%) > maize-2 (1.4 ± 0.1%) > maize-1 (1.1 ± 0.2%)] and the crop-tillage interaction (22% of variance explained). Jointly considering all data, more 15N-fertilizer was recovered in the MT (25 ± 4%) than in the PT soil profile (19 ± 6%) at the end of the experiment whereas the N exported with the crops was unaffected by the tillage system and varied from 5-6% (15NOctober-fertilizer) to 45-49% (15NMarch-fertilizer) and 52-53% (15NMay-fertilizer; despite only three instead of four subsequent crops were studied).The 15N unaccounted for in the case of 15NOctober-fertilizer (72 ± 5%) was more than twice that in 15NMarch- (34 ± 7%) and 15NMay-fertilizer (25 ± 14%). Considering soil, site and weather conditions, denitrification and nitrate leaching during the ryegrass-1 crop were the most likely processes explaining the high losses of the 15NOctober-fertilizer. Results suggested a higher initial immobilization of the applied 15N in the soil organic matter (SOM) of MT, that reduces 15N availability to the first crop, followed by an increase of the residual availability of the fertilizer 15N to the subsequent 2-3 crops.