MODELOS NÃO LINEARES MISTOS PARA DESCREVER O TEOR DE CARBONO ORGÂNICO NO SOLO

Main Article Content

Marcello Neiva de MELLO
Carlos Tadeu dos Santos DIAS
Lucieta Guerreiro MARTORANO
Siglea Sanna de Freitas CHAVES
Paulo Campos Christo FERNANDES

Abstract

The aim of this study was to model organic carbon dataset along with a soil profile in different land use systems using mixed models theory. The samples were collected in layers 0-10, 10-20, 20-30, 30-40, 40-60, 60-80, 80-100, 100-130, and 130-150 cm deep in three areas. High carbon content variability was found in the surface layers due to greater incorporation of organic matter. By changing the structure containing the fixed effect and the one containing the random effect, models could be generated and tested by the maximum likelihood method. The model that performed validation through residue analysis expressed significant effects, which indicates improvements compared to the initial fit. The soil carbon data estimation was highly precise along the soil profile. It is concluded that the mixed model was efficient in simulating the carbon dynamics along the soil profile in different soil use and coverage patterns.

Article Details

How to Cite
MELLO, M. N. de, DIAS, C. T. dos S., MARTORANO, L. G., CHAVES, S. S. de F., & FERNANDES, P. C. C. (2018). MODELOS NÃO LINEARES MISTOS PARA DESCREVER O TEOR DE CARBONO ORGÂNICO NO SOLO. Brazilian Journal of Biometrics, 36(1), 230–240. https://doi.org/10.28951/rbb.v36i1.176
Section
Articles

Most read articles by the same author(s)