Search

Soybean seed cost has dramatically increased in recent decades which has led to producer interest in lowering input cost through reductions in seeding rate. Fifty-eight seeding rate trials of soybean were conducted at field-scale in Indiana from 2010 to 2021 to update recommendations of seeding rates and plant population. The objectives were to determine the agronomic optimal seeding rate (AOSR) and plant population (AOPP) based on planting equipment, tillage practices, and planting date. Economic optimal seeding rate (EOSR) was also determined based on these field scenarios. Harvest AOPP was not influenced by planting equipment (~212,000 plants ha-1) or tillage (~239,000 plants ha-1), but AOSR varied. Soybean seeded with a row-crop planter optimized grain yield with 352,600 seeds ha-1; whereas, the grain drill required 75,200 more seeds ha-1. Soybean seeded into conventional tillage maximized grain yield at 380,400 seeds ha-1; whereas, under no-till conditions 41,400 more seeds ha-1 were required. Timely planting required 417,300 seeds ha-1 to optimize grain yield, which resulted in harvest AOPP of 216,700 plants ha-1. Conversely, late plantings required 102,800 fewer seeds ha-1 but 36,200 more plants ha-1 than timely planting. Depending on seed cost and soybean market price, seeding rates could be reduced 13,700 to 92,800 seeds ha-1 below AOSR to maximize profit. Secondly, digital imagery with high spatial resolution was collected with an unmanned aerial vehicle (UAV) to develop a simple and practical method to segment soybean from non-plant pixels. The best vegetation indices were selected to segment young soybean plants (VC to V6). Two field-scale trials of soybean were planted in 2020 with the agronomic trial design of two varieties x five seeding rates with three replications. The imagery was collected during this period as it coincides with the time for determining whether a soybean stand should be replanted. Five relative vegetative indices based on the red, green, and blue (RGB) imagery were evaluated: excess greenness index (ExG), excess redness index (ExR), green leaf index (GLI), normalized green-red difference index (NGRDI) and visible atmospheric resistance index (VARI). Both GLI and ExG were superior in overall accuracy compared to all other vegetative indices with very small soybean plants (VC to V1 growth stages). VARI and NGRDI had relatively poor overall accuracy at VC and V1, but had similar overall accuracy to GLI as soybean plants grew larger (V2 to V6 growth stages). Across all growth stages and locations, ExR performed the poorest. Moreover, GLI had consistent performance across the range of growth stages, suggesting its suitability for early soybean stand assessment methods. Six field-scale trials were established in 2020 and 2021 in Indiana with two varieties seeded from 123,000 to 618,000 seeds ha-1. Canopy cover was calculated using GLI to create binary segmentation of plant pixels and non-plant pixels. UAV-derived canopy cover measurements were correlated with plant population of soybean from VC to V4 and growing degree days (GDD) after planting. Yield potential (75, 80, 85, 90, 95, 100%) was correlated with canopy cover from VC to V4 and GDD after planting. Canopy cover of 2.1, 5.0, 8.9 and 13.8% by 150, 250, 350, and 450 GDD°C after planting, respectively, would maximize yield. Canopy cover for 75% yield potential was one-fourth as much as the 100% yield potential. Recommended threshold for replant decisions should be based on canopy cover to attain 95% yield potential. Field observations below a canopy cover of 1.8, 4.2, 7.4, and 11.5% canopy cover by 150, 250, 350, and 450 GDD°C after planting respectively, would consider replanting.

Throughout the United States, large windfarms, containing anywhere from 5 to 150 turbines, have been installed due to the increased demand for alternative energy. Since 2008, 1,264 turbines have been installed in Indiana, with the majority of installations occurring on agricultural fields. Despite the large number of turbine installations, impacts of these installations on soil and crop health is unknown. Turbine installation requires the traffic of heavy construction equipment within agriculture fields which may compact the soil altering its physical properties and negatively impact crop growth. To better understand the impact of turbine installation on soil and crop health, we developed a remote-sensing based method to quantify the areal extent of soil and crop impairment due to turbine installation. The method compares the normalized difference vegetation index (NDVI) from satellite images from areas of potential impairment to areas with no known impairment to determine statistical differences in NDVI between impaired and unimpaired areas and then calculates the area of potential impairment. We tested this methodology on two windfarms in Indiana. Our results showed that in the year following turbine installation, turbine installation was associated with an average impairment of 1.8 hectares per turbine and the area of impairment decreased approximately 15-30 percent in each subsequent year. Our results also suggest that soil texture and drainage have an effect on the magnitude and recovery rate of impairment. Coarse textured and/or well drained soils experienced very little to no impairment while fine textured and/or poorly drained soils experienced significant impairment and had not returned to pre-installation levels of impairment after three years. Our findings will allow landowners the opportunity to review current points of negotiation with windfarm developers as well as provide information regarding the potential loss of productivity in crops at these sites.