Evaluation of Multi-Line Seeding for Corn and Soybeans – Year 3 (Low Seeding Rate)

INTRODUCTION

This report very briefly reviews our third season of trials looking at variable-line seeding of corn and soybeans using a multi-hybrid planter. Where previously we had used a smaller 6-row prototype unit developed in collaboration with Raven Industries and Pioneer Hi-Bred; in 2015, Kinze Manufacturing kindly provided a 16-row planter which they have made commercially available. In the first season (2013) at the Tripp and Beresford sites we found on average a 5 bushel per acre yield gain with variable line planting in corn and a 3 bushel per acre yield gain in soybeans. In the second year of the study, we again found a 6 bushel yield advantage with corn with the right pairing of lines, but no advantage with corn or soybeans if the lines didn’t fit well. In this third year of the study we conducted we had 5 sets of plots for corn (all on-farm), and three sets for soybeans (two on-farm, and one at the research station).

The basic logic behind this approach is that given our rainfall distribution (which peaks in May and June) versus the water requirements of corn and soybean crops (which peak in August) there is a good chance that in the same field in the same season the lowland parts of the field may be yield limited by excess moisture early in the season, while the upland positions on the landscape will be yield limited by drought stress in late July and August. It seems logical that gains in productivity within a field might be achieved by using lines with a more horizontal root profile and tolerance to wet conditions in lowland portions of the landscape, and switching to lines with a more vertical root profile and resistance to drought conditions in the upland portions of the landscape. The primary objective of this project is to make an initial evaluation of improvements in grain yield for corn and soybeans grown with a variablegenotype planting system versus planting a single line across the landscape.

METHODS

The project looked at two pairs of soybean lines with only one given pair being tested at each site (Table 1). At each site treatments were upland line, lowland line, variable-line seeding according to landscape position, and variable-rate with variable line seeding. For soybeans, the standard and variable seed rates were 150,000 and 120/180,000, respectively, at all locations. All individual treatments were planted in field length plots, a minimum of 40’ wide. The number of replications at each site is given in Table 1. Yield data were subjected to analysis of variance with SAS statistical software using Proc GLM with all factors considered as fixed effects for each site. There were no significant site by treatment interaction for sites that shared the same lines, so data was pooled across sites where the lines were the same and there was more than one replication per site.

RESULTS AND DISCUSSION

We did not observe an impact of variable-line seeding with soybeans at Lennox or Beresford in the on-farm studies with ‘92Y51’ and ‘92Y70’ as the upland and lowland lines, respectively (Table 2). The trial at the research farm which used ‘P22T69’ as an upland line also did not show a yield response to variable line seeding (Table 3). Therefore, looking across seasons we have some mixed results as some of these same two lines (‘92Y51’ and ‘92Y70’) showed a significant 3 bu/ac benefit from variable-line seeding in the 2013 season, but did not show an advantage in 2015. The environment during seed-filling was remarkably good in our area in 2015, with adequate moisture and mild temperatures. Most years late-July and August are marked by more days with higher maximum temperatures and also some period of drought stress. The relatively ideal conditions in our area may have equally benefited all the lines, lessening the differences between them across the field.

Modified from:

https://www.sdstate.edu/sites/default/files/2015%20SERF%20AR_Final%20Web.pdf

by Peter Sexton∗, Douglas Prairie, Barry Anderson, Doug Johnson, Brandon Goette, and Dustin Theis

ACKNOWLEDGEMENTS

The authors would like to recognize Mr. Les Mehlhaff, Mr. Lee Brockmueller, Mr. Gordon Andersen, Mr. Jason Hausman, Mr. Jamie Tieszen, and Mr. Matt Loewe for being willing to put trials on their operations and for their work in implementing these trials. The efforts of the crew at the SDSU Southeast Research Farm, particularly Mr. Doug Johnson, were critical for the successful completion of this project. The authors wish to gratefully acknowledge the funding support from South Dakota Soybean Research and Promotion Council; as well as Pioneer Hi-Bred, Raven Industries, Kinze Manufacturing, and SDSU, CHS, and Country Pride Cooperative field personnel for supporting this project.