Back to Contents6.2.1 Variable-Rate Seeding and Nitrogen Application
Investigators: Richard Barnhisel, Morris Bitzer, Scott Shearer, Lloyd Murdock and Paula Howe
It is well known that corn responds to nitrogen. Undoubtedly there have been hundreds if not thousands of studies on nitrogen response by corn conducted by agricultural experiment stations over the years. Unpublished work for a one-year study conducted at three locations, Marion, Shelby, and Woodford Counties, has suggested that varying the seeding as well as the nitrogen rates depending on the landscape position (topsoil depth) could have an economic benefit to the grower. Published research, in which both seeding rate and nitrogen were varied based on either topsoil thickness or landscape position, has not been identified in a literature search.
It is proposed that sites in Hardin, Hopkins, Nelson-Marion, Shelby, and Woodford Counties be selected for this investigation. Two test strips will be established in each of these counties. At the Hardin and the Woodford County Research Farm, sites will be on rolling topographies whereas at the Hopkins and Nelson-Marion County sites, the additional sites will be located in creek bottoms. At the Shelby county site, a whole field approach will be used. All locations will be soil sampled on a 30-meter grid spacing after spring planting. All samples will be analyzed for available nutrients, texture, organic matter, and water holding capacity. A rigorous topographic survey will be conducted using a survey grade GPS instrument at each of the investigation sites. Soil cores will be collected at deeper levels below the topsoil layer as well as taking cone penetrometer measurements on the Woodford County Research Farm as the time and cost of sampling will not allow this at other sites.
The seeding rate experiment will be duplicated at each site, one to receive a constant N treatment, the second in which the N rate is also varied. For those sites on rolling topographies, two constant seeding rates of 59 K and 74 K seeds per hectare (24 K and 30 K seeds per acre), for two hybrid seed corn varieties as control test strips on each side of the one variable-rate strip when located on rolling topographies. The variable seeding rates will consist of 30 K, 67 K and 82 K seeds per hectare (21 K, 27 K, and 33 K seeds per acre). Test strips will consist of four rows spaced 76 cm (30 inches) apart. One corn hybrid will be used at all locations for comparison purposes. This hybrid will be one that has wide adaptation characteristics. The second hybrid will be chosen in concert with individual cooperators to be consistent and representative of hybrids typically used in their operation. Because it is extremely difficult to find large enough fields, the above experiment will not be duplicated in the sense of classical statistics, but multiple observations will be taken along the test strips which are at least 275 m (900 ft.) in length. Fields are available at each location so that at least three regions occur for each landscape position.
On the rolling topographies, the N rate will be varied on one of the two variable seeding rate experimental strips at each topsoil depth and/or landscape position. A higher N rate will be used for the bottom or foot-slope positions, and a lower N rate on the shallow topsoil depths, which usually occur on the side-slopes or eroded regions. We plan to reduce the N by 22 kg/ha (20 lbs/acre) on the thinner soils and increase it by the same amount on the thicker soils. Assuming the proportion of each soil type is the same, then the overall N rate will be constant for each general test strip, only the distribution over the landscape. For the creek bottom sites in Hopkins and Nelson Counties, six strips varying in population, 59 K, 67 K, 74 K, and 82K seeds per hectare (24K, 27K, 30K, 33K and 36K seeds per acre) will be used. At these locations, the total N will be varied along each strip. The criterion for varying the N rate will be based on micro relief. A lower N rate will be applied on the higher relief positions and a higher N rate will be applied in the lower or depressional areas along the strip where de-nitrification usually occurs. It is expected that we can develop a protocol that will maximize economic production while reducing ground and surface water contamination.
For the Shelby county site, the cooperating farmer will plant the entire field with his 24-row corn drill, varying the seeding rate in a manner similar to that proposed for the strip tests as presented above. In addition to varying the seeding rate, N will be varied the first year and perhaps other nutrients will be variably applied in his fields the two subsequent years. He will harvest this field with his yield monitor equipped 12-row combine. The yield maps will be analyzed at similar landscape positions within the field where the population and N were the same to establish a mean yield for each landscape position. The net return from each position as well as the whole field results will be compared to an adjacent field in which seeding rates and N were held constant. These data should give insight on the economic benefit of precision agriculture. The change in total nutrients applied to a particular field will be a function of the relative proportion of each landscape position. We anticipate that the growers can meet their yield goals and maximize net return at the same time. In addition, considerations will also be made for environmental impacts of this experiment.
The N recommendations found in AGR-1 will be used as a guide. For no-till, the recommended rates are 140 to 168 kg/ha (125 to 150 lbs./acre) for well-drained soils and 196 to 224 kg/ha (175 to 200 lbs./acre) for moderately well drained soils. We plan to apply 28 kg/ha (25 lbs./acre) at planting to all sites and landscape positions. We then will apply as a side dressing, 140 kg/ha (125 lbs./acre) on the eroded sites, 168 kg/ha (150 lbs./acre) on the ridge tops where usually the intermediate soil depth occurs and 196 kg/ha (175 lbs./acre) on the bottom positions, which are also moderately well drained. For the creek bottom sites, we plan to side dress each the five populations with three rates of nitrogen, with higher N rates being applied for the higher seeding rates. Rates between 122 and 224 kg/ha (100 and 200 lbs./acre) will be used depending on the seeding rate. Twenty-eight percent nitrogen will be used for this project since all but one of the cooperating growers uses this form of N. Decisions regarding soil fertility (P and K) and weed control will remain the responsibility of the cooperator.
Using a plot combine, two of the four rows in each planter trip will be harvested, from measured areas where stand counts have been made. The remaining two rows will be harvested using a plot combine equipped with GPS and yield monitoring capabilities. For producers who have yield monitors, their yield data from adjacent planter trips will be compared with yield data from the test strips. Data sets will be merged into one map to illustrate the effect of seeding rates on a corn yield. In some cases the cooperating producer plans to vary their seeding rates as well, whereas the remaining fields will be planted to one of the rates we have used.
It is expected that we will generate criteria, which will give growers information to make test or generate recommendations for the feasibility of variable rate seed and fertilizer strategies on their individual fields or farms. This will include a range in costs of inputs and fixed costs so that they can adjust this method to reflect variations in input costs as well as expected sales of grain. The ultimate goal is to prepare a profit map to evaluate regions within each field where variable inputs were made as well as the overall net return from the field.