Determination of management zones for a tobacco field based on soil fertility
Computers and Electronics in Agriculture
Computers and Electronics in Agriculture
Exploratory hierarchical clustering for management zone delineation in precision agriculture
ICDM'11 Proceedings of the 11th international conference on Advances in data mining: applications and theoretical aspects
Rectangular shape management zone delineation using integer linear programming
Computers and Electronics in Agriculture
| Hi-index | 0.00 |
In Chile, the use of homogeneous management zones has demonstrated good potential for the site-specific management of traditional crops and vineyards. However, there is not enough information regarding the most efficient method to delineate management zones based on soil fertility variables. The present study had the following objectives: (1) to evaluate three zoning methods based on soil fertility variables, and (2) to quantitatively determine the relationships among the evaluated methods. To achieve these objectives, during the 2002/2003 and 2003/2004 growing seasons, in the geographical area between Metropolitan and Sixth regions of Chile, corn yield and pre-seeding soil fertility data were collected from 13 commercial fields. Data collection was performed spatially, demarking field boundaries and obtaining soil and yield samples at a proper intensity to achieve spatial dependence. A differential GPS unit and appropriate software were used for these purposes. Soils were analyzed for six chemical properties (pH, electrical conductivity (EC), organic matter (OM), available nitrogen (N), extractable phosphorus (P), and potassium (K)) according to standard laboratory procedures. For each site, yield samples were processed and corn yield expressed in terms of dry matter. Measured soil fertility variables were interpolated to each yield sampling point using punctual kriging and linear variograms. Three zoning methods were compared: (1) cluster analysis, (2) principal component analysis (SIPC), and (3) coefficient of variation method (SICV). Regression and variance analyses as well as relative variance were used as comparison tools. Results indicated that the three zoning methods were similar and adequate to delineate homogeneous management zones in terms of soil fertility. However, in most of the experimental locations (77%), zones defined by any method were unable to discriminate yield levels. It is concluded that to properly delineate management zones, it is more important to identify those soil variables responsible for yield variation than it is to select the most robust zoning method.