Abstract
A field study was conducted on hummocky terrain at the Manitoba Zero Tillage Association Research Farm to determine the variability of crop yield as related to landscape position, soil properties, weed populations and plant disease. This information was also used to evaluate technology required for delineation of management units related to precision farming. Variable rate fertiliser management systems can improve efficiency of fertilizer use and environmental sustainability. Adoption of this technology has been hampered due to the difficulty of classifying fields into management units, the high cost of sampling soils on a grid basis, and the variability of soil and plant properties in the landscape. Technology for variable rate fertilizer systems is available, but there is little information available related to yield response in clay soils on hummocky terrain, and the relationship of plant tissue test levels in relation to soil fertility as measured by soil test nitrate nitrogen. Current soil test recommendations for nitrogen are based on soil test nitrate nitrogen from samples bulked from samples in several locations in the field preferably grouped according to topography. Yield data for 1997-2001 were classified into groups with the fuzzy k means, normal mixtures and self-organizing map variants of cluster analysis. Although fuzzy k means commonly used for classification of crop yield and soil properties, a method based on self-organizing maps provided consistent classes when compared across years. Soil nitrate nitrogen varied considerably across the landscape at the site, but was not significantly different (P<0.05) between classes based on crop yield. Yield data can be used to delineate zones for variable management, although fertilizer inputs may be a function of spring soil moisture, runoff and growing season precipitation as they affect seeding, crop emergence and establishment.
Abstract from the paper’s OpenAlex record, shown for indexing and research. Copyright remains with the authors or the publisher. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 4
Claim · model-written summary
Soil nitrate nitrogen varied considerably across the landscape at the site.
Verbatim
t0“crop yield zones on hummocky terrain”
e0“A field study was conducted on hummocky terrain at the Manitoba Zero Tillage Association Research Farm”
e1“Soil nitrate nitrogen varied considerably across the landscape at the site”
Claim · model-written summary
Soil nitrate nitrogen was not significantly different between classes based on crop yield.
Verbatim
t0“crop yield zones on hummocky terrain”
e2“Soil nitrate nitrogen varied considerably across the landscape at the site, but was not significantly different (P<0.05) between classes based on crop yield”
Claim · model-written summary
A self-organizing map method gave more consistent crop yield classes across years than fuzzy k means.
Verbatim
t0“crop yield zones on hummocky terrain”
e3“Although fuzzy k means commonly used for classification of crop yield and soil properties, a method based on self-organizing maps provided consistent classes when compared across years”
Claim · model-written summary
Yield data can be used to delineate zones for variable management.
Verbatim
t0“crop yield zones on hummocky terrain”
e4“Yield data can be used to delineate zones for variable management”