2007
DOI: 10.1016/j.geoderma.2007.04.022
|View full text |Cite
|
Sign up to set email alerts
|

Methods to interpolate soil categorical variables from profile observations: Lessons from Iran

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

6
66
0
20

Year Published

2011
2011
2023
2023

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 156 publications
(92 citation statements)
references
References 24 publications
6
66
0
20
Order By: Relevance
“…Soils in the NV class were the less representatives both in the RA and in Carajazinho map, thus, they showed a lower number of samples in the model training, which resulted in a relatively small area predicted as NV. This fact agrees with Hengl et al (2007), who showed that predictive mapping should be performed with a minimal representation from each soil class, in order to allow that the models capture the complexity of soil formation in the area to be mapped, ensuring that soils of all classes be represented. In the error matrix for DSM (Table 1), UA values were as follows: LV showed the highest one, 77.5%, RL and GM had intermediate values, 40.0 and 37.4%, respectively; and RR, the lowest one, 25.0%.…”
Section: Resultssupporting
confidence: 87%
“…Soils in the NV class were the less representatives both in the RA and in Carajazinho map, thus, they showed a lower number of samples in the model training, which resulted in a relatively small area predicted as NV. This fact agrees with Hengl et al (2007), who showed that predictive mapping should be performed with a minimal representation from each soil class, in order to allow that the models capture the complexity of soil formation in the area to be mapped, ensuring that soils of all classes be represented. In the error matrix for DSM (Table 1), UA values were as follows: LV showed the highest one, 77.5%, RL and GM had intermediate values, 40.0 and 37.4%, respectively; and RR, the lowest one, 25.0%.…”
Section: Resultssupporting
confidence: 87%
“…evaluated. Because of the clear advantages, several authors applied tree-based methods for soil mapping problems (Hengl, T. et al 2007;Grinand, C. et al 2008;Cambule, A.H. et al 2013;Sun, X.L. et al 2011;Häring, T. et al 2012;Pásztor, L. et al 2013).…”
Section: Resultsmentioning
confidence: 99%
“…A ocorrência de multicolinearidade leva a elevados erros-padrão na predição dos parâmetros do modelo ou, até mesmo, inviabiliza o seu ajuste (Manly, 2008). A ausência de multicolinearidade seria uma das premissas para o estabelecimento de um modelo adequado (Hengl et al, 2007).…”
Section: Introductionunclassified
“…agropec. bras., Brasília, v.46, n.5, p.554-562, maio 2011 Entre as metodologias de predição espacial de propriedades e classes de solo está a regressão logística múltipla (RLM) (Giasson et al, 2006;Hengl et al, 2007;Kempen et al, 2009), que estima a probabilidade de ocorrência da variável dependente a partir de variáveis independentes, as quais podem ser discretas, contínuas ou binárias. No entanto, um dos problemas de aplicação dessa técnica de modelos lineares generalizados é que não são verificados efeitos de multicolinearidade entre os dados na fase de ajuste dos modelos (Hengl et al, 2007), sendo essa a maior limitação desses modelos, quando aplicados à predição de propriedades e classes de solos (Kempen et al, 2009).…”
Section: Introductionunclassified
See 1 more Smart Citation