| University
Studies: Field Production & Soil Analysis Correlations |
|---|
| University
Studies: Field Production & Soil Analysis Correlations |
|---|
In 1992, Dr. Lloyd Murdock, and Dr. Fred Magdoff (University of Vermont at Burlington), signed Confidentiality Agreements. Those agreements enabled CTI to reveal every last detail of CTI's proprietary on-the-go soil sensing technique --its Trade Secrets-- to both scientists. Many have asked since then for "full disclosure", but none have signed a Confidentiality Agreement which agrees that they will not use the secrets to enrich themselves or others and that they will not divulge these secrets to others. The Agreement was/is to ensure that CTI trade secrets remain CTI-property.
Although despised and berated by some in government and academia, both U.S. Patent protection and such protection of Trade Secrets are fully-protected by the U.S. Constitution and commonly relied upon by private U.S. citizens and companies. This protection of ones intellectual property from both gang mentality and from big-corporation takeover is what enables the U.S. of America to maintain its world lead in creativity and innovation.
Originally called the "June Nitrate Test", the PreSidedress Nitrate Testis the Agronomic Basis for the Soil DoctorŽ System's required Nitrate testing and one major soil assay used to command the operation of the Soil Doctor Nitrogen Applicator. Domestically, it is a test to determine the level of nitrogen fertilizer which should be applied --alternatively, can be reduced-- based upon in situ soil nitrate levels. It was discovered over twenty years ago at the University of Vermont at Burlington, by Dr. Fred Magdoff.
The PreSidedress Nitrate Test protocols include:
This test has slowly made its way westward from Vermont to the Midwest, having been proven, state by state, to be an effective method for managing nitrogen fertilizer, in corn, in the non-irrigated regions of the U.S. One aspect of the test incorporates the mineralization of organic matter into nitrate. Irrigation interferes with that process for assessment of nitrate in the plow layer.
One of the factors determined by Dr. Magdoff's PSNT research is the critical soil level above which it is unlikely that a corn crop will respond to added N. Test plots are therefore analyzed for both soil NO3-N and end of season corn yield after treatment. Analysis of this data (usually by the Cate-Nelson statistical procedure) results in a determination of this pivotal nutrient level.
In 1991, Dr. Magdoff confirmed unequivocally that any field test where outside forces come in and alter yield-affecting factors (such as when plant population stands are severely altered by rain, pests, or substandard soil (e.g., saline)) should be thrown out for correlation --regardless of the countless hours (a whole year's worth of work) invested in preparation for the tests. In short, one cannot determine (for criteria purposes) an agronomic parameter, unless the test conditions are accurately specified and followed. A poorly reviewed set of data can preclude a researcher from validating the PSNT.
The success of the test is due in part to taking advantage of the seasonal availability of soil nitrate, based on what is actually in the soil when it counts, rather than what is supposed to be in the soil. The crop growth stage criterion is key to knowing when to take soil samples. This growth stage coincides with a singular peak in the mineralization of organic N to NO3-N, just prior to rapid uptake of this N by the crop. The graph below shows how the timing of taking these measurements is key to optimizing use of current soil nitrate tests.
Using the proprietary technique described by CTI, in 1992 the University of Vermont examined 391 soil samples (obtained at the proper growth stage). The analysis results are graphed below. The data results in a 94.7% coefficient of determination --out of a possible maximum of 100%.
|
The vast majority of these samples were collected
under the supervision of representatives of the United States
Department of Energy, soil samples were pulled at (0-4") and
(0-12") depths at randomized plot locations in several
sidedress corn fields in Illinois. These samples were
supplemented by additional sidedress samples taken in Iowa and
Indiana, totaling 391 Midwest soil samples altogether.
All soil samples were immediately cooled in the field,
transported to a freezer, and hard frozen. In 1992, all were
transported in this frozen state to the University of Vermont for
detailed soil analyses.
The University of Vermont was chosen because of its nationally
recognized expertise in soil nitrate analyses and the
availability of specialized equipment for analysis of soil
solutions.
All soil samples were analyzed for soil and soil water nitrates, chlorides,
sulfates, pH, moisture content, saturated paste extract (soil solution)
conductivity, as well as CEC, organic matter, calcium, magnesium, potassium,
aluminum, sodium, and sensor output response.
Non-dimensional sensor readings from analysis of all 391 soil
samples tested in this series are summarized above. As
demonstrated, sensor output was found to be directly related to
soil solution nitrate concentration (R^2 =94.7%). (An R^2 value
of 100% is a "perfect" correlation.) Normal laboratory
soil sample repeatability infrequently approaches only R^2=72%.
With this background, which relates test data obtained on actual soil samples, we can place researcher concerns for anion interferences into perspective. See: Speculating about Chlorides
However, others have conducted independent research on these issues as well. In 1993, Candanian researchers did extensive tests of both Canadian and domestic U.S. soils, analyzing soil solutions for both nitrates and soil solution conductivity. They also concluded that nitrate was the dominant anion in the soil solution, directly related to soil soil solution conductivity. (Download ECofSoilExtract 293 KB).
And, USDA itself confirmed the utility of the same scientific observation when analyzing the response of the Geonics EM38 to manure nutrient variablility (See Papers by Eigenberger, et. al., including TEKTRAN )
