Using δ15N to screen for nitrogen fixation: Reference plant position and species

Abstract

The δ15N natural abundance technique is frequently used to quantify the proportion of legume nitrogen derived from atmospheric fixation (%Ndfa). This method compares the δ15N of a legume with a nonlegume reference plant to estimate the %Ndfa. For accurate estimations, it is recommended to pair each legume plot with a nonlegume reference plant to account for spatial variation. However, this pairing scheme is not feasible in large screening trials. Here, we tested the feasibility of eliminating reference plants to screen a faba bean (Vicia faba L.) population for %Ndfa. The first experiment screened 63 faba bean genotypes for %Ndfa by comparing estimates derived from an adjacent wheat (Triticum aestivum L.) reference plant in a nearest neighbor scheme (NN) with an arbitrary δ15N reference value (AR) of 7‰. Average and standard deviation of the genotypes’ %Ndfa were 58 ± 13% in 2019 and 72 ± 8% in 2020 using the NN model and 61 ± 11% in 2019 and 75 ± 7% in 2020 using the AR model. The AR model was able to identify a majority of the high and low %Ndfa genotypes in both years. In a second study, wheat and a dicot weed within the field were used as reference plants to estimate the %Ndfa. The weed reference plant resulted in 3 to 13% higher %Ndfa values compared with the wheat in three out of four testing environments, but the variation did not alter ranking of the genotypes for %Ndfa.