Soybean [Glycine max (L.) Merr.] seed cost has increased dramatically with the introduction and adoption of herbicide-resistant cultivars, generating interest from growers to reduce seeding rates to the lowest possible level that does not affect yield. Research indicates that greater seeding rates are needed to maximize yield under low-yielding environments and less seed is needed in high-yielding environments, but this has not been confirmed with recent research in Virginia. The objectives of this research was to 1) determine the yield response of soybean cultivars with differing growth habits and maturities grown in full-season and double-crop systems to seeding rate under different yield environments; and 2) compare two seeding rates in large on-farm strip-plots to determine if the growth environment within the field affects the yield response to seeding rate. For objective 1, small-plot research was conducted on Piedmont and Coastal Plain sites across Virginia from 2017 thru 2020. Maturity group (MG) 4 or 5 cultivars were planted in 46-cm rows at the following seeding rates: full-season soybean – 74,130, 148,260, 222,390, 296,520, 370,650, and 444,780 seed ha-1; and double-crop soybean – 197,680, 296,520, 395,360, 494,200, 543,620, and 593,040 seed ha-1. One cultivar per MG was used in 2017 and 2018, but the experiments were expanded to include two cultivars, differing in canopy structure, within each MG in 2019 and 2020. On-farm research compared a high and low seeding rate with a 100,000 seed ha-1 difference based upon grower current practices. To determine growth influence on the yield response, normal difference vegetative index (NDVI) was measured at 2-week intervals from late-vegetative to late-reproductive stages in small-plot and on-farm experiments. Double-crop soybean required an average of 205,000 more seed ha-1 than full-season soybean. Although yield response varied with site and year, MG 4 cultivars usually yielded more than MG 5 at higher seeding rates, but less at lower seeding rate. No differences between cultivar canopy structure were present in full-season systems; differences were revealed in double-crop systems but were not consistent over sites. To obtain 95% of maximum yield, 170,000 to 390,000 seed ha-1 were required in full-season soybean and 470,000 to 550,000 seed ha-1 were required in double-crop soybean. While the NDVI response to seeding rate generally reflected the yield response at most site-years, relationship between yield and NDVI was weak. In on-farm experiments, higher seeding rates yielded more at 3 of 6 sites, but differences varied within the field. The yield-NDVI relationship was stronger due to greater variability within the field, but these differences due to seeding rate could not be discerned. Growing environment, primarily amount and distribution of rainfall, greatly influenced these results; therefore, more exact site-specific seed rate recommendations will be difficult in Virginia's environment. / Master of Science / Soybean is one of the world's most important seed legumes and contributes a major portion of global protein concentrate for livestock feeding and oil for human consumption therefore, it is essential to continually and sustainably increase yield while maintaining profitability for growers. Through adoption of precision agriculture technology by Southeastern farmers, site-specific management using variable rate application and/or automatic section control of inputs are able to be utilized when field variability is assessed. Additionally, on-farm experiments validate small plot research with larger field-scale evaluations with proper technology and design. In this study various seeding rates on soybean yield in both small-plot and on-farm experiments, as influenced by variable growth and yield differences within the field, was determined. On-farm research allowed better observation of yield response to seeding rate for each field, allowing the evaluation of production practices under realistic growing conditions. Ultimately, it is expected that more early-season growth in areas of the field with greater productivity will result in greater yields using less seed per hectare. The utilization of vegetative indices as well as remote sensing technology contributed largely to the analysis of yield and varying seeding rates. Accordingly, measurements were correlated with one another to determine if remote sensing techniques can be substituted for ground measurements in predicting yields.
| Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/104053 |
| Date | 28 June 2021 |
| Creators | Bowers, Lindsey Carolle |
| Contributors | Crop and Soil Environmental Sciences, Holshouser, David L., Thomason, Wade E., Galbraith, John M., Oakes, Joseph C. |
| Publisher | Virginia Tech |
| Source Sets | Virginia Tech Theses and Dissertation |
| Detected Language | English |
| Type | Thesis |
| Format | ETD, application/pdf, application/pdf |
| Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
Page generated in 0.002 seconds