O emprego da superficie de resposta na interpretacao de experimentos de adubacao na cultura do cacau na Bahia
Nine year yields of five NPK fertilizer field experiments with "Catongo" cacao were used for the calculations of the polynomial regression with and without interactions. The composite design proposed by Box was used, with five levels of nitrogen, phosphorus and potassium and two replicates for each experiment. The statistical analysis was carried out for each year, for all experiments in the same year and for each experiment in all years. In general the estimated parameters of the regression equations have presented large confidence intervals and the lineas interactions were not significant for each locality, each different analysis with a predominance of saddle points. The economic dosages of P2O5 for Agua Sumida (Rio Branco) and Itabuna (Ipiaú) soils lie between 90 and 135 kg ha exponent -1. These were calculated from a surface section in which N and K mainteined at constant doses. Finally, the response surface methodology was not suitable for the experimental data, but might serve for other polynomial functions.
| Main Authors: | , , |
|---|---|
| Format: | biblioteca |
| Published: |
1983
|
| Subjects: | THEOBROMA CACAO, NUTRICION MINERAL, FERTILIZANTES NPK, ANALISIS ESTADISTICO, CULTIVAR CATONGO, DOSIS DE APLICACION, |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Nine year yields of five NPK fertilizer field experiments with "Catongo" cacao were used for the calculations of the polynomial regression with and without interactions. The composite design proposed by Box was used, with five levels of nitrogen, phosphorus and potassium and two replicates for each experiment. The statistical analysis was carried out for each year, for all experiments in the same year and for each experiment in all years. In general the estimated parameters of the regression equations have presented large confidence intervals and the lineas interactions were not significant for each locality, each different analysis with a predominance of saddle points. The economic dosages of P2O5 for Agua Sumida (Rio Branco) and Itabuna (Ipiaú) soils lie between 90 and 135 kg ha exponent -1. These were calculated from a surface section in which N and K mainteined at constant doses. Finally, the response surface methodology was not suitable for the experimental data, but might serve for other polynomial functions. |
|---|