Use of 15 N‐Labeled Soil in Measuring Nitrogen Fertilizer Recovery Efficiency in Transplanted Rice

In transplanted flooded rice ( Oryza sativa L.) culture in Asia, N fertilizer recovery efficiency (RE) estimates are usually greater with the difference method than with the 15 N method. These reported differences are greater and more common for transplanted irrigated rice than for upland crops and might be associated with basal N applications when plant demand is low. Added N interactions (ANI), or greater soil N accumulation in fertilized plants than in unfertilized plants, are often cited as the reason for these differences. The causes of ANI can be stimulation of mineralization of soil organic matter by fertilizer or greater root exploration in fertilized plants. The objectives of this pot study were (i) to compare the 15 N dilution method using enriched 15 N‐urea (5.0 atom %), and 14 N‐urea (0.366 atom %) plus 15 N‐labeled (0.447 atom %, Maahas clay [isohyperthermic Andaqueptic Haplaquoll]) soil, with the difference method of measuring RE of N, as affected by timing of application, and (ii) to use 15 N‐labeled (0.447 atom %) soil to directly measure ANI as 15 N accumulation in 14 N fertilized pots minus 15 N accumulation in unfertilized pots. Recovery efficiency measures at physiological maturity were higher with the difference method (54%) than with the two isotope dilution methods (44%), which were in turn similar. The real ANI measured was negligible, as the root biomass in this clay soil was not greatly affected by fertilization. Therefore, the ANI was apparent and due to isotope substitution. The RE estimate by difference (54.4%) was therefore more accurate than the RE measure by 15 N methods (44.4%). The 33.2% of added 15 N not accounted for in the plant–soil system was a measure of N fertilizer losses (NH 3 volatilization from floodwater, denitrification, NH 3 loss through the plant during grain fill) that were not affected by isotope substitution. Recovery efficiency estimated with 15 N‐enriched soil plus ordinary urea was identical to the RE estimate using enriched 15 N‐urea, although both were underestimations because of isotope substitution.