| dc.description.abstract | Seeding rate for maximum grain yield can differ for diverse hard red spring wheat
(HRSW) (Triticum aestivum L.) cultivars and can be derived from a seeding rate response curve.
Six groups of HRSW cultivars with combinations of Rht-B, Rht-D, and Ppd-D with two cultivars
per group were planted in 2013-2015 at five seeding rates in 23 trials throughout Minnesota
(MN) and eastern North Dakota (ND), USA. Seeding rates ranged from 1.59 – 5.55 million seeds
ha-1. Planting dates represented optimum and delayed seeding dates. Agronomic measurements
for plant height, lodging, stems per plant, protein, and yield were obtained. Stand loss
measurements, defined as the amount of viable seeds that did not become established plants,
ranged from 11-19% across seeding rates most commonly planted in the region. There was a
seeding rate by cultivar interaction for plant height, protein, lodging, stems plant-1, and yield. As
seeding rate increased stems per plant consistently decreased and there were large differences in
tillering capacity between cultivars. Increased seeding rate caused increased lodging for those
cultivars with a capacity to lodge. Seeding rate for maximum yield of the cultivars differed.
Combined over all cultivars, the seeding rate for maximum yield increased as the average yield
of an environment decreased. An analysis of covariance (ANCOVA) predictive model was built
for yield and tillering. The model for yield across all environments was not predictive with a
validation R2 of 0.01. However, when only the bottom six yielding environments out of the total
21 environments were used to build a yield model the predictions were more accurate with a
validation R2 of 0.44. The model built and validated for tillering was predictive for the validation
environments with an R2 of 0.71 for validation environments. Seeding rate trials continue to be
useful for producers making seeding rate decisions for a range of agronomic reasons.
Additionally, using regression predictions and separate training and validation datasets to predict yield and tillering with HRSW, genetic and geographic predictors show promise for
recommending seeding rates for future environments. | en_US |
