Predicting Nitrogen Use Traits for Non-Destructive Selection in Winter Wheat Using Proximal Sensing.

Reducing nitrogen pollution by breeding nitrogen use efficient wheat varieties is an urgent need, but methods for evaluating nitrogen use (NU) traits are expensive, time and labor consuming, and destructive. A faster, non-destructive, and efficient phenotyping method is needed to allow breeding programs to select winter wheat varieties with good nitrogen use characteristics. Therefore we used non-destructive and high throughput proximal sensing during the grain fill period to calculate vegetation indices (VI) that estimate canopy chlorophyll content. We collected VI phenotypes on a 299 genotype hard winter wheat association mapping panel planted in Ithaca, NE in 2012 and 2013.  We then investigated the relationship between VI and nitrogen use traits such as nitrogen harvest index, post anthesis nitrogen uptake, and grain nitrogen yield. We found two VI, Boochs2 and Maccioni, that performed consistently across environment and were significantly related to NU traits. These VI were also more heritable than NU traits suggesting that NU traits could be improved using indirect selection based on VI values. This method is more time and cost efficient than directly selecting for NU which requires destructive biomass harvest from the plots. The results of this study indicate that VIs, in particular the Maccioni and Boochs2 indices, could be used in wheat breeding programs to non-destructively phenotype for NU traits. Therefore, high throughput proximal sensing can be used to reduce phenotyping labor and time costs and allow wheat breeding programs to efficiently breed for NU traits.