6.6   Evaluation of Topography Attributes on Corn Yield

Principal Investigators

Richard I. Barnhisel, Professor, Agronomy
Ronald Fleming, Assistant Professor, Agricultural Economics

Cooperators

Tom Luck, Luck Farms, Hanson KY
Peterson Dairy Farms, Bernard Peterson, Loretto, KY
Mr. George Kelly, Agr. Ext., Hopkins Co., Madisonville KY
Mr. Ronald Bowman, Agr. Ext., Nelson Co., Bardstown KY

Introduction and Literature Review

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.  We have demonstrated in Hopkins, Marion, Shelby, and Woodford Counties in a current study that varying the seeding and nitrogen rates based on landscape position (essentially this is the depth of the topsoil) has an economic benefit to the grower (Barnhisel et al., 1999).  However, in our literature search, we did not find any published research in which both seeding rate and nitrogen were varied based on either topsoil thickness or landscape position.

Nolan et al., presented data in 1998 on the effect of landscape classes on yield response to nitrogen.  Two poster papers were presented at the recent ASA meetings at Salt Lake City, Utah, on varying N in a precision Ag setting on this topic.  The reliability of corn yield potential as a guide to variable-rate applications of nitrogen was presented by White and Blackmer, 1999.  Various rates of N were applied in replicated strips that crossed several soil-mapping units.  Yields were related to the yield potentials of the soil types, thus suggesting that a precision Ag approach may likely increase grain yields in fields with multiple soil mapping units.  In another poster paper, the use of variable N rates was used as a tool for reducing N losses without yield reductions (Davis et al., 1999).  Both landscape position and soil properties were important determiners of N losses, and clearly site factors interacted with climate to influence losses and uptake of N.

Numerous economic studies have found that accounting for site factors can increase profitability (Anderson et. al., 1999; Feder and Slade, 1984; Lin, 1991, Rahm and Huffman 1984; Zepeda, 1994).  However, some recent studies suggest that precision farming may not always by profitable especially on small or otherwise resource limited farms (Debertin, 1997).  But missing from these and other studies is the impact of in-field variation that arises from tile drainage, edge effects, and micro-depressions that occur along creek or rivers.  Edge effects include losses in yield due to soil compaction from increased equipment travel, losses in yield to moisture and light competition with vegetation at the field’s edge (trees), and losses in yield due to crop consumption by wildlife (raccoons, deer, turkey, and other birds).  Furthermore, this study will evaluate actual in-field variation across complete fields over a number of years.  Most previous economic studies of the return to precision agriculture have had to rely on simulated farm data.

Because test plots and field strips are, generally, not adequate for capturing most forms of in-field variation, the impact on N use, yield, and net return of tile drainage, edge effects, and micro-depressions is best measured at the whole field level.  Specifically, drained field bottoms may be highly productive suggesting higher N and seeding rates with corresponding higher net returns.  Field edges, on the other hand, because of crop damage by wildlife are likely to exhibit lower rates of return even if the soil is classified as being highly productive.  Here the analysis becomes more complex.  Conceptually, because yields, hence economic returns are lower, it makes sense to reduce N and seed application rates.  The relevant question is how much lower should these application rates be?  Furthermore, if the first 12 or so rows are sufficiently damages as to result in economic losses, would it not be better for the grower to plant the edges to grass (create filter strips)?  This later solution is particularly appealing if the farmer is also able to receive conservation cost share money or other program funds for doing so.  It is also possible that eliminating the first 12 rows, for example, only shifts the edge effect interior on to more productive ground.

The specific objectives of the proposed work are:

1)               To determine the yield response, on a whole field basis, to variable seeding and nitrogen rates based on landscape or topographic position;

2)               To determine the effect of tiling micro depressions on corn yield;

3)               To determine the impact of border effects due to trees and wildlife on corn yield; and

4)               To evaluate the economic benefits of the above precision agricultural approaches and evaluate the economic impact of boarder effects due to trees and wildlife.

Background

            In 1998 and 1999, seeding and nitrogen rates have been studied at seven locations in Kentucky in response to topographic position.  Sites were chosen in these fields, which represented the maximum expected benefits due to variable seeding and nitrogen.  In 1996 through 1998, in an earlier study sponsored by the Ky Corn Growers Association, only seeding rates were varied in several additional locations.  In that study, some of the plots were located on creek or river bottom positions.  The practice of varying the seeding rates on these alluvial soils did not prove to be beneficial, except to substantiate that a seeding rate 27K to 30K was the proper seeding rate for these soils types most years.  At two of these creek or river bottom locations, and adjacent to the strips where the seeding rate study was conducted, micro depressions affected corn yield as seen in the yield maps provided by the cooperating grower.

During the past two years, yield data were collected from the strips where the seeding and nitrogen rates were varied for comparisons with yields where the seeding rate or nitrogen rates were constant for all landscape positions.  Varying the seeding and nitrogen rate increased the net return both years, by as much as $20 per acre in 1998 and $13 per acre in 1999, over that from adjacent strips with constant rates. 

Approach

            This is a three-year project.  Because the cooperating growers follow a crop rotation system, the same fields cannot be used all three years.  Where possible, the same fields will be evaluated the third year.  Both growers have AgLeaderâ yield monitors on their combines and have collected yield data from all fields at least two years.  These yield data have been made available for the tentative selection of the fields for the project proposal.

Whole fields will be selected in two regions of the state, Nelson and Hopkins Counties.  Several fields have been tentatively chosen in each area based on past yield map data collected in cooperation with the growers.  Two types of settings will be used in both regions, one having a rolling upland topography with forest or a “blue line stream” on at least one side of the field, with second setting being a creek or river bottom region.  There will be two fields with alluvial settings, one in which the field has been tiled, the other non-tiled.  In all cases, it is expected the fields will contain at least 30 acres.  For the tiled creek bottom fields, water samples will be taken to obtain a qualitative measure of N losses.  Water flow from these tiles will be estimated at the time of collection by measuring the flow with a bucket or some other device.

At one of the rolling upland fields in Nelson County, both the seeding and nitrogen rates will be varied by the cooperating grower.  In one each of the other two fields, either nitrogen or seeding rate will be varied.  At the Hopkins county site, seeding and nitrogen rates will be varied only in the test strips in 2000.  The fertilizer dealer, who applies nitrogen for the Luck Farms, plans to purchase a variable-rate nitrogen applicator.  If this actually takes place in time, it will be used to apply the nitrogen in 2000, but seeding rates will remain constant over the entire field since this grower does not have a variable seeding capability.

All fields will be intensively grid soil sampled using 100 by 100 foot spacing.  All sampling locations will be geo-referenced and the elevation recorded to the nearest 0.1 foot.  We plan to use one of two methods to determine the elevation, a laser plane device or a “real time” GPS survey instrument.  These data will be used to produce a topographic map of elevation with a 0.5-foot interval for the rolling topographies and a 0.1-foot interval for the creek or river bottom settings.  Soil sampling topographic data collection will be required two of the three years of the project since both growers use a corn-soybean rotation.

At the grid sampling locations, seven sub samples will be taken to form a composite soil sample.  A radius of 15 feet will be used with six samples collected around the perimeter of the circle and one at the middle.  These samples will be taken from the upper six inches of the soil.  In the fields where more than one soil type exists, and especially for the rolling upland positions, the soil will be investigated at a deeper depth to verify the accuracy of the soil-mapping unit.  This will be done with either a hand probe or a tractor-mounted Giddings soil sampler.  The soil series at each sampling point will be noted.  Samples will be taken between the 100 foot grid points where the soil type changes or there is a break in the slope class.  These extra sites will also be geo-referenced.

The composite soil samples will be characterized for plant nutrients, i.e., pH, lime requirement, P, K, Zn, (and possibly other minor elements), OM, texture, and water holding capacity.  Moist soil color will also be noted when the samples are collected if it appears this parameter would assist in the identification of the soil type.  Soil data will be evaluated using Surferâ or some other mapping software such as SSToolboxâ and GIS methods used to formulate relationships between topographic or soil properties and corn yield.

Water samples will be collected following major rainfall events from the tile drains.  Flow values will be estimated by collecting flow within a timed interval.  Samples will be placed in a cooler over ice and nitrogen components determined using accepted EPA procedures.  At least NO₃, Total N and P will be determined for the samples collected.

The yield data will be collected from separate six or eight rows along the edges of the fields where forest or critter impacts may occur.  In these regions, the data will be saved as a separate load by the yield monitor.  Economic analyses will be done to determine if greater net return would be realized if these regions were not planted.  If other regions within the fields occur, where yields are low due to topographic or soil type attributes, similar economic analyses will be done.  In these cases, the regions will be isolated by the mapping software, the area and yield contributions determined, and economic evaluation conducted to determine if not planting these regions would increase net return from the field. 

Expected Benefits

Precision agriculture has been described as placing the right nutrients (or number of seeds) at the right place at the right time.  This study evaluates the impact on net return on placing seed and nutrients in the right place.  Specifically, accounting for in-field variations in soil type due to landscape position (topsoil depth) has been shown to be economically beneficial (Barnhisel et al., 1999).  Similarly, it is expected that accounting for other sources of in-field variation such as tile drainage areas, field edges, and micro-depressions will also be economically beneficial.  These regions may be eligible for government substitutes such as CRP payments or riparian set aside.  The economic analyses of these situations will assist the grower to make decisions to adopt these practices compared to the cost of implementation.

The underlying principal is that seed and nutrients should be placed where physical field conditions are such that economic net returns (not yields) are maximized.  Anecdotal evidence suggests that some farmers are considering converting field edges to grass in order to receive federal and state conservation compliance transfers for doing so.  The implication is that the value of the conservation compliance transfer is greater than the net return that can be earned on that acreage in crops.  This study will be able to determine the extent of economic, perhaps yield, losses along field edges to assess if conversion to grass or some other crop is cost effective.  It is possible that simply lowering the rate of seed and N applications is sufficient to offset losses along field edges.  Either way, this information is important to producers in that it provides information that allows them to maximize whole field net return.

In a similar manner, the installation of tile drainage is expensive.  Clearly, such expenditures are made only after careful evaluation of potential yield increases associated with increased drainage.  But it is also possible that drainage sufficiently increases soil productivity that increased seed and N application in these drained areas are justified.  If this is the case, then field net returns can be increased.  This study will evaluate yield response and economic net return to variable rates of seed and nutrient applications in sub-field areas with tile drainage.  The information obtained from this investigation is important to producers seeking information that can be used to justify to lenders the cost of installing tile drainage systems in fields.  This information is also important to producers who have tile draining and are able to utilize variable seeding and N rate technologies to maximize whole field net return.

Deliverables

It is anticipated in addition to the benefits presented above, that data for at least four journal papers will be gathered, one from the seeding-nitrogen rate study and one from the tiling study.  It is anticipated that two of these articles will have a precision agricultural journal approach and one from the economic analyses of the two major objectives.  Materials will be available for an equal number of extension publications and workshops.  Workshops or extension meetings will be used to inform the Kentucky growers of our findings.