The highest returns for nitrogen applications on corn occur when soil color is used to create management zones for variable-rate applications, according to a Colorado State University study. The method achieved higher net returns than both a blanket application of nitrogen and when grid sampling was used to create management zones for variable-rate applications.
Ironically, a blanket application of nitrogen had better returns than variable-rate applications on corn based on grid sampling. The results of the study were presented at the National Alliance of Independent Crop Consultants, Tucson, Ariz., by Dwayne Westfall, professor, soil and crop sciences, CSU.
Westfall says there are three types of field variability — natural variability from soils and topography, random variability, which is weather-related, and managed variability, which is the variability growers can manage — inputs, tillage, traffic patterns, irrigation practices, etc.
Since the emergence of precision agriculture, grid sampling has been used to locate areas of manageable variability, Westfall noted. “But early on, we realized that it had some major limitations, namely the cost and labor associated with it. An important point is that the current grid size being used — 2.5 acres — is often too large to capture the spatial variability that exists within fields.”
The objective of the study is to “develop techniques for developing management zones that will optimize crop production and economic returns. The management zone is a sub-region of a field that expresses some commonality that affects yield. We are trying to identify those areas of commonality in the field and manage them.”
The researchers first looked at several techniques to establish three management zones for the fields in the study — corresponding to high, medium and low production potential.
“The first technique requires very little learning time and is the one we like the best because of its simplicity — the soil color management zone. It uses three data layers, black and white bare soil imagery, the farmer's perception of topography and the farmer's perception of his experience with his yield.
In this technique the bare soil imagery is color-enhanced to differentiate organic matter, soil texture, etc., and borders of management zones are automatically drawn. The farmer has the ability to come in and tweak those zones, to move the boundaries around.
There are two reasons for including the farmer's experience, according to Westfall. “He gives us previous management information, intuition, observations he's made. It also means that the farmer has bought into the concept. Immediately, he has more ownership.”
Another management zone delineation technique is using a Veris mapping cart to map the electrical conductivity of the soil.
“We also evaluated a more complex technique — a yield-based management zone. It has six data layers — bare soil imagery, topography, the previous year's yield map plus organic matter, texture and CEC. One limitation in this design is that we should have used more yield maps.”
Scientists evaluated the techniques through yield mapping. They discovered that even the most complicated management zone techniques failed to capture most of the variability in the field. “Only 51 percent of our predicted yields matched up with the actual yield.”
Maps based on soil color showed a 42 percent association with actual yield. In most of the techniques, the researchers had success in separating the low-potential areas from high areas, but not always the low from the medium or the medium from the high.
One reason why the zones weren't more accurate is that “small-scale variability is not captured with management zone delineation we use. Grid-sampling has the same problem, only on a larger scale. On the micro-scale, we have different production potential within different zones.
“Another point is that smoothing the data for the purpose of using large equipment will cause you to lose some accuracy. There is different yield potential within the zones. There may be an area of high yield potential within a low zone, but we can't manage that with equipment that is 60-feet to 90-feet wide. We have to smooth over it.”
Nonetheless, the economics of the using the commercially available soil color method of delineation indicated a good net return, according to Westfall.
Variable-rate nitrogen applications on corn based on grid-based sampling had a net return of $19 less than a uniform application. VR applications based on soil color showed an increase in net return due to variable-rate nitrogen management of $12 an acre compared to the uniform nitrogen application and a $32 an acre increase in returns compared to variable-rate applications based on grid-based zones.