Year: 2003
Type: article
Abstract: Most agricultural fields naturally have within-field variations that may require site-specific attention for optimum crop performance. These variations are often contiguous, extensive and not easy to manage since every spot in a paddock is unique. Delineating a paddock into a few `management zones' is therefore a practical management approach. In this study, yield maps of various crops and seasons were used to delineate `management zones'. Map standardisation and the application of geographic information systems (GIS) were necessary to combine various yield maps for `management zones' delineation. A user-defined fuzzy set membership function was employed to standardise yield maps. The membership function used control points derived from the percentile calculation to rescale yield data continuously between 0 and 1. Map standardization and processing was performed within Idrisi-32 GIS and image processing software program. Standardised maps were combined within raster GIS to identify areas of low, medium and high yield performances. The classified areas were identified as `management zones' requiring different level of management attention. It was found that the continuous rescaling of yield data, using percentile based control points, was suitable to standardise yield maps. This process derived control points from within the yield data and minimized the influences of extreme yield values. Overlaying standardised yield maps to identify areas of low, medium and high yield potential was a straightforward process within GIS. The investigation revealed that most area performed moderately and the areas requiring highest management attention were around the periphery of the investigated paddock.
Institution: The University of Queensland
Language: English
Subfield: Soil Science
Domain: Life Sciences
SDG: Zero hunger
Open Access status: green
OpenAlex ID: W85996193