# Executar com `Rscript -e 'rmarkdown::render("2_regressao_simples.R")'`
#///////////////////////////////////////////////////////////////////////
# Pacotes --------------------------------------------------------------
rm(list = ls())
suppressPackageStartupMessages({
library(car)
library(DataExplorer)
library(corrplot)
library(tidyverse)
})
# options(scipen = 4, width = 120)
options(width = 120)
#///////////////////////////////////////////////////////////////////////
# Importação dos dados -------------------------------------------------
tb <- read_csv("Parecis_wrc_params.csv")
## Rows: 318 Columns: 42
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): Local, Profundidade
## dbl (39): ID, Estrutura, Carbono, Tensao, Ds, Umid, Argila, Areia, Silte, Muito_grossa, Grossa, Media, Fina, Muito_f...
## lgl (1): SOC
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
attr(tb, "spec") <- NULL
str(tb)
## spc_tbl_ [318 × 42] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
## $ ID : num [1:318] 1 2 3 4 5 6 7 8 9 10 ...
## $ Local : chr [1:318] "IF CP01" "IF CP01" "IF CP01" "IF CP01" ...
## $ Profundidade: chr [1:318] "0-0,05" "0-0,05" "0-0,05" "0,15-0,20" ...
## $ Estrutura : num [1:318] 1 2 2 1 2 2 1 2 2 1 ...
## $ Carbono : num [1:318] 1 1 2 1 1 2 1 1 2 1 ...
## $ Tensao : num [1:318] 0 0 0 0 0 0 0 0 0 0 ...
## $ Ds : num [1:318] 0.969 NA NA 1.357 NA ...
## $ Umid : num [1:318] 0.653 0.513 0.553 0.386 0.515 ...
## $ Argila : num [1:318] 508 508 508 568 568 ...
## $ Areia : num [1:318] 335 335 335 331 331 ...
## $ Silte : num [1:318] 157 157 157 101 101 ...
## $ Muito_grossa: num [1:318] 0 0 0 0 0 0 0 0 0 0 ...
## $ Grossa : num [1:318] 5.5 5.5 5.5 6.5 6.5 6.5 35.5 35.5 35.5 31 ...
## $ Media : num [1:318] 122 122 122 118 118 ...
## $ Fina : num [1:318] 158 158 158 166 166 ...
## $ Muito_fina : num [1:318] 40.5 40.5 40.5 35 35 35 68 68 68 69 ...
## $ SOC : logi [1:318] NA NA NA NA NA NA ...
## $ N : num [1:318] 0.1436 0.1436 0.0748 0.0897 0.0897 ...
## $ C : num [1:318] 2.224 2.224 0.341 1.578 1.578 ...
## $ Mg : num [1:318] 0.478 0.478 NA NA NA ...
## $ Al : num [1:318] 10.2 10.2 10.8 11 11 ...
## $ Si : num [1:318] 17.1 17.1 17.2 17.6 17.6 ...
## $ P : num [1:318] 0.0302 0.0302 0.0459 0.0347 0.0347 0.0393 0.032 0.032 0.069 0.021 ...
## $ Ca : num [1:318] 0.736 0.736 0.611 0.552 0.552 ...
## $ Ti : num [1:318] 1.26 1.26 1.31 1.29 1.29 ...
## $ V : num [1:318] 0.0627 0.0627 0.0644 0.0616 0.0616 0.0686 0.0118 0.0118 0.0114 0.0143 ...
## $ Cr : num [1:318] 0.0162 0.0162 0.0238 0.0188 0.0188 0.02 0.0085 0.0085 0.007 0.0044 ...
## $ Mn : num [1:318] 0.0113 0.0113 0.014 0.006 0.006 0.0088 0.0018 0.0018 0.0015 NA ...
## $ Fe : num [1:318] 7.48 7.48 8.05 7.54 7.54 ...
## $ Cu : num [1:318] 0.0032 0.0032 0.0034 0.0031 0.0031 0.0043 0.0014 0.0014 0.0014 0.001 ...
## $ Zn : num [1:318] 0.0059 0.0059 0.0074 0.0058 0.0058 0.0063 0.003 0.003 0.0032 0.0017 ...
## $ As : num [1:318] 0.0017 0.0017 0.0021 0.0016 0.0016 0.0019 NA NA 0.0007 0.0003 ...
## $ Zr : num [1:318] 0.0407 0.0407 0.0502 0.0503 0.0503 0.0488 0.0152 0.0152 0.0199 0.0198 ...
## $ thr : num [1:318] 0.229 0.219 0.214 0.141 NA ...
## $ ths : num [1:318] 0.653 0.527 0.565 0.396 NA ...
## $ alp : num [1:318] 1.891 0.1 0.675 0.582 NA ...
## $ nscal : num [1:318] 9.202 0.733 1.041 0.718 NA ...
## $ mscal : num [1:318] 0.041 1.439 0.56 0.325 NA ...
## $ I : num [1:318] -0.29 1.802 0.949 2.105 NA ...
## $ U : num [1:318] 0.6 0.363 0.412 0.302 NA ...
## $ S : num [1:318] -0.1345 -0.0623 -0.0737 -0.0284 NA ...
## $ CAD : num [1:318] 2.89e-01 1.12e-07 2.05e-01 2.07e-01 NA ...
## - attr(*, "problems")=<externalptr>
tb <-
tb |>
mutate(Estrutura = factor(Estrutura,
levels = c(1, 2),
labels = c("EstPres", "EstAus")),
Carbono = factor(Carbono,
levels = c(1, 2),
labels = c("CarPres", "CarOxi")),
Trat = interaction(Estrutura, Carbono,
drop = TRUE, sep = "_"),
Profundidade = paste0(Profundidade, " m") |>
str_replace_all(",", ".")
)
tb |>
count(Trat, Estrutura, Carbono)
## # A tibble: 3 × 4
## Trat Estrutura Carbono n
## <fct> <fct> <fct> <int>
## 1 EstPres_CarPres EstPres CarPres 106
## 2 EstAus_CarPres EstAus CarPres 106
## 3 EstAus_CarOxi EstAus CarOxi 106
tb |>
count(Profundidade)
## # A tibble: 2 × 2
## Profundidade n
## <chr> <int>
## 1 0-0.05 m 159
## 2 0.15-0.20 m 159
# tb |>
# count(Local, Profundidade) |>
# print(n = Inf)
# tb <-
# tb |>
# # filter(Profundidade != "0-0,05")
# filter(Profundidade == "0-0,05")
#///////////////////////////////////////////////////////////////////////
# Carbono incrementa a água disponível? --------------------------------
# Comparar `EstAus_CarPres` vs `EstAus_CarOxi`.
tb |>
count(Trat, Profundidade)
## # A tibble: 6 × 3
## Trat Profundidade n
## <fct> <chr> <int>
## 1 EstPres_CarPres 0-0.05 m 53
## 2 EstPres_CarPres 0.15-0.20 m 53
## 3 EstAus_CarPres 0-0.05 m 53
## 4 EstAus_CarPres 0.15-0.20 m 53
## 5 EstAus_CarOxi 0-0.05 m 53
## 6 EstAus_CarOxi 0.15-0.20 m 53
tb_sel <-
tb |>
filter(
Trat %in% c("EstAus_CarPres", "EstAus_CarOxi"),
# Profundidade == "0-0,05",
# Profundidade == "0,15-0,20",
between(CAD, 0.000001, 1)) |>
# mutate(CAD = CAD^(1/3)) |>
select(Local, Profundidade, Trat, CAD)
# DANGER: O comando acima exclui locais que tiveram CAD = 0.
tb_sel |>
count(Local) |>
arrange(n) |>
print(n = Inf)
## # A tibble: 50 × 2
## Local n
## <chr> <int>
## 1 IF CP01 1
## 2 P06 1
## 3 P07 1
## 4 P10 1
## 5 P14 1
## 6 P30 1
## 7 P36 1
## 8 P41 1
## 9 P42 1
## 10 P45 1
## 11 P47 1
## 12 P48 1
## 13 P09 2
## 14 P12 2
## 15 P13 2
## 16 P18 2
## 17 P19 2
## 18 P23 2
## 19 P25 2
## 20 P28 2
## 21 P35 2
## 22 P37 2
## 23 P38 2
## 24 P39 2
## 25 P40 2
## 26 P46 2
## 27 P51 2
## 28 P03 3
## 29 P08 3
## 30 P15 3
## 31 P16 3
## 32 P21 3
## 33 P24 3
## 34 P27 3
## 35 P31 3
## 36 P32 3
## 37 P34 3
## 38 P43 3
## 39 P44.1 3
## 40 P44.2 3
## 41 P49 3
## 42 P50 3
## 43 P52 3
## 44 P02 4
## 45 P04 4
## 46 P05 4
## 47 P17 4
## 48 P20 4
## 49 P26 4
## 50 P33 4
tb_sel_wide <-
tb_sel |>
# arrange(Local, Trat) |>
pivot_wider(names_from = Trat, values_from = CAD) |>
drop_na()
tb_sel |>
count(Profundidade)
## # A tibble: 2 × 2
## Profundidade n
## <chr> <int>
## 1 0-0.05 m 49
## 2 0.15-0.20 m 69
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = c(4, 3),
names_to = "Trat",
values_to = "CAD") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
# ATTENTION: Os NA foram removidos para manter apenas observações com
# pares completos.
legend <-
list(trat_levels = c(
"EstAus_CarPres" = expression(S^{"-"}~C^{"+"}),
"EstAus_CarOxi" = expression(S^{"-"}~C^{"-"}),
"EstPres_CarPres" = expression(S^{"+"}~C^{"+"})
),
areia_levels = c(
"Grossa" = "Coarse sand",
"Media" = "Medium sand",
"Fina" = "Fine sand"
),
argila_axis = expression("Clay content"~(g~kg^{-1})),
trat_axis = "Treatments",
profundidade_axis = "Depth (m)",
concentracao_axis = "Total elemental content (%)",
awc_label = expression("Available water capacity"~(cm^3~cm^{-3})),
rwc = expression("Residual water content"~(theta[r] * ", " ~ cm^3~cm^{-3})),
wci = expression("Water content at the inflection point" ~ (theta[I] * ", " ~ cm^3~cm^{-3})),
wcs = expression("Water content at saturation" ~ (theta[s] * ", " ~ cm^3~cm^{-3})),
alpha = expression("Shape parameter" ~ (alpha * ", " ~ kPa^{-1})),
n = expression("Shape parameter" ~ (italic(n))),
m = expression("Shape parameter" ~ (italic(m))),
I = expression("Tension at the inflection point" ~ (kPa)),
# S = expression("Slope at the inflection point" ~ ("S - index," ~ cm^3~cm^{-3}~kPa^{-1}))
S = "Slope at the inflection point (S - index)"
)
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = CAD)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
# Segmentos que conectam pontos do mesmo local.
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstAus_CarPres,
yend = EstAus_CarOxi),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$awc_label)
## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
gg
dir.create("figs", showWarnings = FALSE)
# pdf("figs/awc_carbono_estrutura.pdf",
# width = 8, height = 6)
# print(gg)
# dev.off()
postscript("figs/awc_carbono_estrutura.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(CAD ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 30 0.309631 0.010321 1.4621 0.1264931
## Profundidade 1 0.000327 0.000327 0.0464 0.8305310
## Trat 1 0.116371 0.116371 16.4849 0.0002092 ***
## Profundidade:Trat 1 0.001838 0.001838 0.2604 0.6125169
## Residuals 42 0.296486 0.007059
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
par(mfrow = c(2, 2))
plot(m0)
layout(1)
emmeans::emmeans(m0, "Trat")
## NOTE: Results may be misleading due to involvement in interactions
## Trat emmean SE df lower.CL upper.CL
## EstAus_CarPres 0.198 0.0156 42 0.1663 0.229
## EstAus_CarOxi 0.123 0.0156 42 0.0913 0.154
##
## Results are averaged over the levels of: Local, Profundidade
## Confidence level used: 0.95
# im <- influence.measures(m0)
# tb_sel[im$infmat[, "hat"] > 0.9, ]
#///////////////////////////////////////////////////////////////////////
# Estrutura interfere na água disponível? ------------------------------
# Comparar `EstAus_CarPres` vs `EstAus_CarOxi`.
tb |>
count(Trat, Profundidade)
## # A tibble: 6 × 3
## Trat Profundidade n
## <fct> <chr> <int>
## 1 EstPres_CarPres 0-0.05 m 53
## 2 EstPres_CarPres 0.15-0.20 m 53
## 3 EstAus_CarPres 0-0.05 m 53
## 4 EstAus_CarPres 0.15-0.20 m 53
## 5 EstAus_CarOxi 0-0.05 m 53
## 6 EstAus_CarOxi 0.15-0.20 m 53
tb_sel <-
tb |>
filter(
Trat %in% c("EstPres_CarPres", "EstAus_CarPres"),
# Profundidade == "0-0,05",
# Profundidade == "0,15-0,20",
between(CAD, 0.000001, 1)) |>
# mutate(CAD = CAD^(1/3)) |>
select(Local, Profundidade, Trat, CAD)
tb_sel_wide <-
tb_sel |>
# arrange(Local, Trat) |>
pivot_wider(names_from = Trat, values_from = CAD) |>
drop_na()
tb_sel |>
count(Profundidade)
## # A tibble: 2 × 2
## Profundidade n
## <chr> <int>
## 1 0-0.05 m 73
## 2 0.15-0.20 m 85
# ATTENTION: Os NA foram removidos para manter apenas observações com
# pares completos.
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = c(4, 3),
names_to = "Trat",
values_to = "CAD") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = CAD)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
# Segmentos que conectam pontos do mesmo local.
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$awc_label)
gg
postscript("figs/awc_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(CAD ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 41 0.68364 0.016674 3.4060 2.070e-06 ***
## Profundidade 1 0.00270 0.002699 0.5513 0.4601
## Trat 1 0.11300 0.113000 23.0821 7.772e-06 ***
## Profundidade:Trat 1 0.00639 0.006386 1.3045 0.2570
## Residuals 75 0.36717 0.004896
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
par(mfrow = c(2, 2))
plot(m0)
layout(1)
emmeans::emmeans(m0, "Trat")
## NOTE: Results may be misleading due to involvement in interactions
## Trat emmean SE df lower.CL upper.CL
## EstPres_CarPres 0.147 0.00948 75 0.128 0.165
## EstAus_CarPres 0.206 0.00948 75 0.187 0.225
##
## Results are averaged over the levels of: Local, Profundidade
## Confidence level used: 0.95
#///////////////////////////////////////////////////////////////////////
# Qual a importância da Argila para a água disponível? -----------------
tb |>
count(Trat, Profundidade)
## # A tibble: 6 × 3
## Trat Profundidade n
## <fct> <chr> <int>
## 1 EstPres_CarPres 0-0.05 m 53
## 2 EstPres_CarPres 0.15-0.20 m 53
## 3 EstAus_CarPres 0-0.05 m 53
## 4 EstAus_CarPres 0.15-0.20 m 53
## 5 EstAus_CarOxi 0-0.05 m 53
## 6 EstAus_CarOxi 0.15-0.20 m 53
tb_sel <-
tb |>
filter(
Trat %in% c("EstPres_CarPres"),
# Profundidade == "0-0,05",
# Profundidade == "0,15-0,20",
between(CAD, 0.000001, 1)) |>
# mutate(CAD = CAD^(1/3)) |>
select(Local, Profundidade, Trat, Argila, CAD, thr, U)
gg <-
ggplot(tb_sel, aes(x = Argila, y = CAD)) +
facet_wrap(~ Profundidade) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(x = legend$argila_axis,
y = legend$awc_label)
gg
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 2 rows containing non-finite outside the scale range (`stat_smooth()`).
## Warning: Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).
postscript("figs/awc_argila_preservado.eps",
width = 8, height = 6)
print(gg)
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 2 rows containing non-finite outside the scale range (`stat_smooth()`).
## Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).
dev.off()
## png
## 2
m0 <- lm(CAD ~ Profundidade * Argila, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.048803 0.048803 13.8697 0.0003561 ***
## Argila 1 0.155877 0.155877 44.2996 2.835e-09 ***
## Profundidade:Argila 1 0.022999 0.022999 6.5363 0.0123902 *
## Residuals 83 0.292052 0.003519
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
m1 <- update(m0, . ~ 0 + Profundidade/Argila)
summary(m1)$coefficients
## Estimate Std. Error t value Pr(>|t|)
## Profundidade0-0.05 m 0.0918009506 1.588907e-02 5.777615 1.282448e-07
## Profundidade0.15-0.20 m 0.0718919956 1.583686e-02 4.539536 1.888807e-05
## Profundidade0-0.05 m:Argila 0.0003169155 5.296993e-05 5.982931 5.345369e-08
## Profundidade0.15-0.20 m:Argila 0.0001494467 3.853511e-05 3.878196 2.098373e-04
gg <-
ggplot(tb_sel, aes(x = Argila, y = thr)) +
facet_wrap(~ Profundidade) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(x = legend$argila_axis,
y = legend$rwc)
gg
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 2 rows containing non-finite outside the scale range (`stat_smooth()`).
## Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).
postscript("figs/rwc_argila_preservado.eps",
width = 8, height = 6)
print(gg)
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 2 rows containing non-finite outside the scale range (`stat_smooth()`).
## Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).
dev.off()
## png
## 2
m0 <- lm(thr ~ Profundidade * Argila, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: thr
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.007520 0.007520 2.1015 0.1509
## Argila 1 0.247312 0.247312 69.1112 1.544e-12 ***
## Profundidade:Argila 1 0.001157 0.001157 0.3232 0.5712
## Residuals 83 0.297013 0.003578
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
m1 <- update(m0, . ~ 0 + Profundidade/Argila)
summary(m1)$coefficients
## Estimate Std. Error t value Pr(>|t|)
## Profundidade0-0.05 m 0.0474492491 1.602344e-02 2.9612405 3.994479e-03
## Profundidade0.15-0.20 m 0.0151328474 1.597078e-02 0.9475334 3.461177e-01
## Profundidade0-0.05 m:Argila 0.0002858043 5.341786e-05 5.3503505 7.625783e-07
## Profundidade0.15-0.20 m:Argila 0.0002482488 3.886097e-05 6.3881275 9.223422e-09
gg <-
ggplot(tb_sel, aes(x = Argila, y = U)) +
facet_wrap(~ Profundidade) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(x = legend$argila_axis,
y = legend$wci)
gg
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 2 rows containing non-finite outside the scale range (`stat_smooth()`).
## Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).
postscript("figs/wci_argila_preservado.eps",
width = 8, height = 6)
print(gg)
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 2 rows containing non-finite outside the scale range (`stat_smooth()`).
## Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).
dev.off()
## png
## 2
m0 <- lm(U ~ Profundidade * Argila, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: U
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.17410 0.17410 21.4001 1.358e-05 ***
## Argila 1 0.64387 0.64387 79.1410 1.054e-13 ***
## Profundidade:Argila 1 0.03290 0.03290 4.0441 0.04757 *
## Residuals 83 0.67526 0.00814
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
m1 <- update(m0, . ~ 0 + Profundidade/Argila)
summary(m1)$coefficients
## Estimate Std. Error t value Pr(>|t|)
## Profundidade0-0.05 m 0.1961506759 2.416040e-02 8.118683 3.774783e-12
## Profundidade0.15-0.20 m 0.1234639528 2.408101e-02 5.127026 1.891807e-06
## Profundidade0-0.05 m:Argila 0.0005525066 8.054434e-05 6.859657 1.144825e-09
## Profundidade0.15-0.20 m:Argila 0.0003522061 5.859522e-05 6.010833 4.742038e-08
#///////////////////////////////////////////////////////////////////////
# Qual a importância das frações de areia? -----------------------------
tb |>
count(Trat, Profundidade)
## # A tibble: 6 × 3
## Trat Profundidade n
## <fct> <chr> <int>
## 1 EstPres_CarPres 0-0.05 m 53
## 2 EstPres_CarPres 0.15-0.20 m 53
## 3 EstAus_CarPres 0-0.05 m 53
## 4 EstAus_CarPres 0.15-0.20 m 53
## 5 EstAus_CarOxi 0-0.05 m 53
## 6 EstAus_CarOxi 0.15-0.20 m 53
tb_sel <-
tb |>
filter(
Trat %in% c("EstPres_CarPres"),
# Profundidade == "0-0,05",
# Profundidade == "0,15-0,20",
between(CAD, 0.000001, 1)) |>
# mutate(CAD = CAD^(1/3)) |>
select(Local, Profundidade, Trat,
Muito_grossa, Grossa, Media, Fina, Muito_fina, CAD)
frac_areia <- c("Muito_grossa", "Grossa", "Media", "Fina", "Muito_fina")
gg <-
tb_sel |>
pivot_longer(cols = frac_areia,
names_to = "Fracao",
values_to = "Valor") |>
mutate(Fracao = factor(Fracao, levels = frac_areia)) |>
mutate(Fracao = fct_collapse(Fracao,
"Grossa" = c("Muito_grossa", "Grossa"),
"Fina" = c("Fina", "Muito_fina"))) |>
ggplot(aes(x = Valor, y = CAD, color = Profundidade)) +
facet_wrap(~Fracao, scales = "free_x",
labeller = labeller(Fracao = legend$areia_levels)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(x = legend$argila_axis,
y = legend$awc_label,
color = legend$profundidade_axis)
## Warning: Using an external vector in selections was deprecated in tidyselect 1.1.0.
## ℹ Please use `all_of()` or `any_of()` instead.
## # Was:
## data %>% select(frac_areia)
##
## # Now:
## data %>% select(all_of(frac_areia))
##
## See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
postscript("figs/awc_fracoes_areia_preservado.eps",
width = 8, height = 6)
print(gg)
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 10 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 10 rows containing missing values or values outside the scale range
## (`geom_point()`).
dev.off()
## png
## 2
tb_sel[["Grossa"]] <- tb_sel[["Muito_grossa"]] + tb_sel[["Grossa"]]
tb_sel[["Fina"]] <- tb_sel[["Fina"]] + tb_sel[["Muito_fina"]]
tb_sel[["Muito_grossa"]] <- NULL
tb_sel[["Muito_fina"]] <- NULL
frac_areia <- c("Grossa", "Media", "Fina")
fits <-
map(frac_areia,
function(x) {
formula <- sprintf("CAD ~ Profundidade * %s", x)
tb_sel |>
select(Profundidade, all_of(x), CAD) |>
lm(formula, data = _)
})
names(fits) <- frac_areia
map_dbl(fits, ~summary(.)$r.squared)
## Grossa Media Fina
## 0.4852980 0.5581455 0.4996979
map(fits, anova)
## $Grossa
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.048803 0.048803 15.1423 0.0002005 ***
## Grossa 1 0.194208 0.194208 60.2574 1.929e-11 ***
## Profundidade:Grossa 1 0.009214 0.009214 2.8587 0.0946312 .
## Residuals 83 0.267507 0.003223
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Media
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.048803 0.048803 17.6388 6.680e-05 ***
## Media 1 0.221140 0.221140 79.9258 8.605e-14 ***
## Profundidade:Media 1 0.020143 0.020143 7.2801 0.008444 **
## Residuals 83 0.229646 0.002767
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Fina
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.048803 0.048803 15.578 0.0001651 ***
## Fina 1 0.176245 0.176245 56.258 6.366e-11 ***
## Profundidade:Fina 1 0.034661 0.034661 11.064 0.0013129 **
## Residuals 83 0.260023 0.003133
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
map(fits, ~summary(.)$coefficients)
## $Grossa
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.258110066 0.0156222294 16.521974 5.688544e-28
## Profundidade0.15-0.20 m -0.096200171 0.0201853412 -4.765843 7.913281e-06
## Grossa -0.003996542 0.0005895279 -6.779226 1.638682e-09
## Profundidade0.15-0.20 m:Grossa 0.001442379 0.0008530856 1.690779 9.463117e-02
##
## $Media
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2748301109 1.505140e-02 18.259439 8.314554e-31
## Profundidade0.15-0.20 m -0.0987559442 2.090379e-02 -4.724308 9.298217e-06
## Media -0.0004370271 5.307088e-05 -8.234782 2.214651e-12
## Profundidade0.15-0.20 m:Media 0.0002027491 7.514347e-05 2.698160 8.444471e-03
##
## $Fina
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3001459951 1.927534e-02 15.571506 2.375329e-26
## Profundidade0.15-0.20 m -0.1258312119 2.637956e-02 -4.770027 7.785450e-06
## Fina -0.0004298257 5.705745e-05 -7.533210 5.487078e-11
## Profundidade0.15-0.20 m:Fina 0.0002584012 7.768582e-05 3.326234 1.312890e-03
#///////////////////////////////////////////////////////////////////////
# Os nutrientes conseguem prever a CAD? --------------------------------
# str(tb)
# names(tb) |>
# dput()
qui <- c(
"N", "C", "Mg", "Al", "Si", "P", "Ca", "Ti", "V", "Cr", "Mn",
"Fe", "Cu", "Zn", "As", "Zr"
)
tb_sel <-
tb |>
filter(
Trat %in% c("EstPres_CarPres"),
# Profundidade == "0-0,05",
# Profundidade == "0,15-0,20",
between(CAD, 0.000001, 1)) |>
# mutate(CAD = CAD^(1/3)) |>
select(Local, Profundidade, Trat, CAD, any_of(qui))
# Retirar valor extremo de Carbono.
tb_sel$C <- ifelse(tb_sel$C > 100, NA, tb_sel$C)
gg <-
tb_sel |>
pivot_longer(cols = qui,
names_to = "Nutriente",
values_to = "Valor") |>
mutate(Fracao = factor(Nutriente, levels = qui)) |>
ggplot(aes(x = Valor, y = CAD, color = Profundidade)) +
facet_wrap(~Nutriente, scales = "free_x") +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(x = legend$concentracao_axis,
y = legend$awc_label,
color = legend$profundidade_axis)
## Warning: Using an external vector in selections was deprecated in tidyselect 1.1.0.
## ℹ Please use `all_of()` or `any_of()` instead.
## # Was:
## data %>% select(qui)
##
## # Now:
## data %>% select(all_of(qui))
##
## See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
gg
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 76 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 76 rows containing missing values or values outside the scale range
## (`geom_point()`).
postscript("figs/awc_nutrientes_preservado.eps",
width = 8, height = 9)
print(gg)
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 76 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 76 rows containing missing values or values outside the scale range
## (`geom_point()`).
dev.off()
## png
## 2
# DANGER: Nem todos os nutrientes possuem valores em %.
fits <-
map(qui, function(x) {
formula <- sprintf("CAD ~ Profundidade * %s", x)
tb_sel |>
select(Profundidade, all_of(x), CAD) |>
lm(formula, data = _)
})
names(fits) <- qui
map_dbl(fits, ~summary(.)$r.squared)
## N C Mg Al Si P Ca Ti V Cr Mn Fe
## 0.5213386 0.5596402 0.2127714 0.5364417 0.5889933 0.1394291 0.1371322 0.6022130 0.6256020 0.5817285 0.1606472 0.6112882
## Cu Zn As Zr
## 0.4553027 0.3750770 0.5900188 0.5728256
map(fits, anova)
## $N
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.048803 0.048803 16.2824 0.0001208 ***
## N 1 0.221499 0.221499 73.8997 4.192e-13 ***
## Profundidade:N 1 0.000654 0.000654 0.2181 0.6416885
## Residuals 83 0.248776 0.002997
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $C
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050675 0.050675 18.1823 5.338e-05 ***
## C 1 0.233479 0.233479 83.7722 3.576e-14 ***
## Profundidade:C 1 0.006290 0.006290 2.2569 0.1369
## Residuals 82 0.228540 0.002787
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Mg
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.027833 0.027833 5.7886 0.01946 *
## Mg 1 0.033661 0.033661 7.0008 0.01056 *
## Profundidade:Mg 1 0.011281 0.011281 2.3462 0.13122
## Residuals 56 0.269260 0.004808
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Al
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 17.271 7.836e-05 ***
## Al 1 0.189860 0.189860 65.324 4.460e-12 ***
## Profundidade:Al 1 0.039103 0.039103 13.454 0.0004305 ***
## Residuals 83 0.241232 0.002906
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Si
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 19.480 3.036e-05 ***
## Si 1 0.225730 0.225730 87.596 1.248e-14 ***
## Profundidade:Si 1 0.030581 0.030581 11.867 0.0008984 ***
## Residuals 83 0.213885 0.002577
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $P
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.05020 0.050197 9.3034 0.00307 **
## P 1 0.02038 0.020375 3.7763 0.05537 .
## Profundidade:P 1 0.00199 0.001985 0.3679 0.54578
## Residuals 83 0.44783 0.005396
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Ca
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.05020 0.050197 9.2786 0.003107 **
## Ca 1 0.01015 0.010152 1.8766 0.174421
## Profundidade:Ca 1 0.01101 0.011013 2.0357 0.157394
## Residuals 83 0.44903 0.005410
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Ti
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 20.127 2.310e-05 ***
## Ti 1 0.229632 0.229632 92.072 4.211e-15 ***
## Profundidade:Ti 1 0.033558 0.033558 13.455 0.0004303 ***
## Residuals 83 0.207005 0.002494
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $V
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 21.384 1.367e-05 ***
## V 1 0.242335 0.242335 103.236 3.157e-16 ***
## Profundidade:V 1 0.033026 0.033026 14.069 0.0003252 ***
## Residuals 83 0.194834 0.002347
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Cr
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 19.141 3.505e-05 ***
## Cr 1 0.223507 0.223507 85.227 2.244e-14 ***
## Profundidade:Cr 1 0.029023 0.029023 11.067 0.001311 **
## Residuals 83 0.217665 0.002622
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Mn
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.02479 0.024791 4.9002 0.029978 *
## Mn 1 0.04503 0.045031 8.9008 0.003874 **
## Profundidade:Mn 1 0.00086 0.000864 0.1708 0.680620
## Residuals 73 0.36932 0.005059
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Fe
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 20.597 1.897e-05 ***
## Fe 1 0.234691 0.234691 96.298 1.549e-15 ***
## Profundidade:Fe 1 0.033221 0.033221 13.631 0.000397 ***
## Residuals 83 0.202283 0.002437
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Cu
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 14.6985 0.0002447 ***
## Cu 1 0.180020 0.180020 52.7123 1.891e-10 ***
## Profundidade:Cu 1 0.006719 0.006719 1.9674 0.1644509
## Residuals 83 0.283456 0.003415
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Zn
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.05020 0.050197 12.8115 0.0005787 ***
## Zn 1 0.14340 0.143404 36.5999 4.011e-08 ***
## Profundidade:Zn 1 0.00159 0.001586 0.4049 0.5263184
## Residuals 83 0.32521 0.003918
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $As
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.060502 0.060502 23.215 7.116e-06 ***
## As 1 0.188897 0.188897 72.481 1.040e-12 ***
## Profundidade:As 1 0.039400 0.039400 15.118 0.0002127 ***
## Residuals 77 0.200675 0.002606
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $Zr
## Analysis of Variance Table
##
## Response: CAD
## Df Sum Sq Mean Sq F value Pr(>F)
## Profundidade 1 0.050197 0.050197 18.742 4.155e-05 ***
## Zr 1 0.215209 0.215209 80.353 7.709e-14 ***
## Profundidade:Zr 1 0.032688 0.032688 12.205 0.0007672 ***
## Residuals 83 0.222298 0.002678
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
map(fits, ~summary(.)$coefficients)
## $N
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.01350543 0.02243159 0.6020720 5.487674e-01
## Profundidade0.15-0.20 m 0.04141896 0.02882025 1.4371476 1.544353e-01
## N 1.45344684 0.19342321 7.5143354 5.978894e-11
## Profundidade0.15-0.20 m:N -0.16896426 0.36176973 -0.4670492 6.416885e-01
##
## $C
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.02619601 0.01905926 1.374451 1.730466e-01
## Profundidade0.15-0.20 m 0.03742788 0.02504040 1.494700 1.388313e-01
## C 0.08599089 0.01033355 8.321527 1.608940e-12
## Profundidade0.15-0.20 m:C -0.02671301 0.01778129 -1.502310 1.368585e-01
##
## $Mg
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3132459 0.05508927 5.686151 4.896859e-07
## Profundidade0.15-0.20 m -0.1682615 0.08087765 -2.080444 4.207199e-02
## Mg -0.2613972 0.08726710 -2.995369 4.077405e-03
## Profundidade0.15-0.20 m:Mg 0.2012633 0.13139645 1.531726 1.312203e-01
##
## $Al
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.12546195 0.037696143 -3.328244 1.304572e-03
## Profundidade0.15-0.20 m 0.12737423 0.049942447 2.550420 1.259612e-02
## Al 0.03439764 0.004283605 8.030068 5.668474e-12
## Profundidade0.15-0.20 m:Al -0.02062487 0.005622940 -3.667986 4.305115e-04
##
## $Si
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.59284273 0.048295648 12.275283 2.384253e-20
## Profundidade0.15-0.20 m -0.27146391 0.065625436 -4.136565 8.402035e-05
## Si -0.02115072 0.002384266 -8.870957 1.183565e-13
## Profundidade0.15-0.20 m:Si 0.01115844 0.003239133 3.444884 8.984006e-04
##
## $P
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2485639 0.04205314 5.9107084 7.281404e-08
## Profundidade0.15-0.20 m -0.1075318 0.05334467 -2.0157926 4.705685e-02
## P -1.7586149 0.90679609 -1.9393719 5.585338e-02
## Profundidade0.15-0.20 m:P 0.9581386 1.57955322 0.6065884 5.457797e-01
##
## $Ca
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.23550498 0.03505847 6.717491 2.156421e-09
## Profundidade0.15-0.20 m -0.11711425 0.04536183 -2.581780 1.158567e-02
## Ca -0.06384028 0.03234875 -1.973501 5.176592e-02
## Profundidade0.15-0.20 m:Ca 0.06854568 0.04804220 1.426781 1.573941e-01
##
## $Ti
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.003808733 0.02031747 -0.187461 8.517567e-01
## Profundidade0.15-0.20 m 0.044843085 0.02810127 1.595767 1.143416e-01
## Ti 0.187534051 0.02032896 9.224972 2.315958e-14
## Profundidade0.15-0.20 m:Ti -0.101922551 0.02778606 -3.668118 4.303214e-04
##
## $V
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.03228826 0.01599758 2.018322 4.678724e-02
## Profundidade0.15-0.20 m 0.02372997 0.02228528 1.064827 2.900421e-01
## V 3.19495745 0.32930337 9.702170 2.578570e-15
## Profundidade0.15-0.20 m:V -1.69858156 0.45284485 -3.750913 3.251726e-04
##
## $Cr
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.00571906 0.0209361 0.2731674 7.854031e-01
## Profundidade0.15-0.20 m 0.04167767 0.0268645 1.5514031 1.246100e-01
## Cr 10.35193906 1.2243872 8.4547916 8.050400e-13
## Profundidade0.15-0.20 m:Cr -5.29241049 1.5908729 -3.3267338 1.310816e-03
##
## $Mn
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.12462622 0.02091067 5.9599329 8.225652e-08
## Profundidade0.15-0.20 m -0.02647168 0.03430232 -0.7717168 4.427732e-01
## Mn 3.51987176 1.35058843 2.6061764 1.109211e-02
## Profundidade0.15-0.20 m:Mn 1.47738685 3.57486732 0.4132704 6.806200e-01
##
## $Fe
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.03712734 0.016003752 2.3199148 2.280168e-02
## Profundidade0.15-0.20 m 0.02049070 0.022451019 0.9126847 3.640518e-01
## Fe 0.02362469 0.002511831 9.4053682 1.009281e-14
## Profundidade0.15-0.20 m:Fe -0.01271258 0.003443236 -3.6920430 3.970096e-04
##
## $Cu
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.04590651 0.02090908 2.1955292 3.091710e-02
## Profundidade0.15-0.20 m 0.01048837 0.02970898 0.3530369 7.249558e-01
## Cu 43.12071625 6.57505690 6.5582271 4.365432e-09
## Profundidade0.15-0.20 m:Cu -14.83164723 10.57402482 -1.4026492 1.644509e-01
##
## $Zn
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.06476343 0.02291332 2.8264540 5.895615e-03
## Profundidade0.15-0.20 m -0.01411057 0.03248779 -0.4343346 6.651726e-01
## Zn 17.06862181 3.38000782 5.0498764 2.578831e-06
## Profundidade0.15-0.20 m:Zn 4.59899767 7.22748437 0.6363207 5.263184e-01
##
## $As
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.05371852 0.01665850 3.2246908 1.851272e-03
## Profundidade0.15-0.20 m 0.02123262 0.02256904 0.9407854 3.497570e-01
## As 89.80551572 10.46357064 8.5826836 7.652269e-13
## Profundidade0.15-0.20 m:As -54.70010286 14.06828380 -3.8881859 2.127042e-04
##
## $Zr
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.01405273 0.01968376 0.7139249 4.772765e-01
## Profundidade0.15-0.20 m 0.03253912 0.02769165 1.1750520 2.433339e-01
## Zr 4.94960507 0.57249693 8.6456447 3.342782e-13
## Profundidade0.15-0.20 m:Zr -2.72096417 0.77885980 -3.4935224 7.671559e-04
#///////////////////////////////////////////////////////////////////////
# Quais os erros em estimar a CRAS usando a amostra sem estrutura? -----
# cras_params <- c("thr", "ths", "alp", "nscal", "mscal", "I", "U", "S", "CAD")
# Umidade residual (thr) ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "thr"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"),
y > 0)
# tb_sel[["y"]] <- tb_sel[["y"]] |>
# log10()
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$rwc)
gg
postscript("figs/rwc_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 45 0.77631 0.0172513 6.7621 3.248e-15 ***
## Profundidade 1 0.01179 0.0117897 4.6213 0.03412 *
## Trat 1 0.00045 0.0004497 0.1763 0.67554
## Profundidade:Trat 1 0.00594 0.0059350 2.3264 0.13052
## Residuals 95 0.24236 0.0025512
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.118 0.00911 95 0.0998 0.136 a
## EstAus_CarPres 0.135 0.00911 95 0.1166 0.153 a
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstAus_CarPres 0.101 0.00876 95 0.0831 0.118 a
## EstPres_CarPres 0.109 0.00876 95 0.0921 0.127 a
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Umidade de saturação (ths) ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "ths"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"),
y > 0, y < 1)
# tb_sel[["y"]] <- tb_sel[["y"]] |>
# log10()
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$wcs)
gg
postscript("figs/rws_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 49 2.13133 0.04350 6.3269 1.006e-15 ***
## Profundidade 1 0.14664 0.14664 21.3296 1.082e-05 ***
## Trat 1 0.35748 0.35748 51.9987 8.290e-11 ***
## Profundidade:Trat 1 0.08321 0.08321 12.1032 0.0007292 ***
## Residuals 107 0.73561 0.00687
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.412 0.0151 107 0.382 0.442 a
## EstAus_CarPres 0.455 0.0151 107 0.425 0.485 b
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.297 0.0127 107 0.271 0.322 a
## EstAus_CarPres 0.431 0.0127 107 0.406 0.456 b
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Umidade no ponto de inflexão (I) ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "U"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"),
y > 0, y < 1)
# tb_sel[["y"]] <- tb_sel[["y"]] |>
# log10()
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$wci)
gg
postscript("figs/awi_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 50 2.44817 0.048963 6.2955 < 2.2e-16 ***
## Profundidade 1 0.23701 0.237008 30.4735 1.728e-07 ***
## Trat 1 0.20508 0.205079 26.3683 9.885e-07 ***
## Profundidade:Trat 1 0.02281 0.022811 2.9329 0.08914 .
## Residuals 132 1.02663 0.007778
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.331 0.0135 132 0.304 0.358 a
## EstAus_CarPres 0.375 0.0135 132 0.348 0.402 b
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.234 0.0129 132 0.208 0.260 a
## EstAus_CarPres 0.322 0.0129 132 0.296 0.347 b
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Parâmetro de forma alpha ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "alp"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"))
# tb_sel[["y"]] <- tb_sel[["y"]] |>
# log10()
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$alpha)
gg
postscript("figs/alpha_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 50 52.889 1.05778 0.9896 0.5037
## Profundidade 1 0.049 0.04911 0.0459 0.8306
## Trat 1 0.047 0.04715 0.0441 0.8340
## Profundidade:Trat 1 0.002 0.00212 0.0020 0.9645
## Residuals 132 141.099 1.06893
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.825 0.159 132 0.511 1.14 a
## EstAus_CarPres 0.850 0.159 132 0.536 1.16 a
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.852 0.151 132 0.553 1.15 a
## EstAus_CarPres 0.891 0.151 132 0.591 1.19 a
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Parâmetro de forma n ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "nscal"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"))
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_y_log10(
breaks = scales::trans_breaks("log10", \(x) 10^x),
labels = scales::trans_format("log10", scales::math_format(10^.x))
) +
annotation_logticks(
sides = "l",
short = unit(0.05, "cm"),
mid = unit(0.1, "cm"),
long = unit(0.15, "cm"),
colour = "black",
linewidth = 0.25,
outside = TRUE) +
coord_cartesian(clip = "off") +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$n)
gg
postscript("figs/n_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
tb_sel[["y"]] <- tb_sel[["y"]] |>
log10()
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 50 25.519 0.5104 1.6198 0.01575 *
## Profundidade 1 0.450 0.4498 1.4274 0.23433
## Trat 1 18.781 18.7810 59.6050 2.55e-12 ***
## Profundidade:Trat 1 0.814 0.8136 2.5821 0.11047
## Residuals 132 41.592 0.3151
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstAus_CarPres -0.2592 0.0861 132 -0.430 -0.0888 a
## EstPres_CarPres 0.5129 0.0861 132 0.343 0.6833 b
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstAus_CarPres -0.0234 0.0822 132 -0.186 0.1393 a
## EstPres_CarPres 0.4841 0.0822 132 0.321 0.6467 b
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Parâmetro de forma m ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "mscal"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"))
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_y_log10(
breaks = scales::trans_breaks("log10", \(x) 10^x),
labels = scales::trans_format("log10", scales::math_format(10^.x))
) +
annotation_logticks(
sides = "l",
short = unit(0.05, "cm"),
mid = unit(0.1, "cm"),
long = unit(0.15, "cm"),
colour = "black",
linewidth = 0.25,
outside = TRUE) +
coord_cartesian(clip = "off") +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$m)
gg
postscript("figs/m_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
tb_sel[["y"]] <- tb_sel[["y"]] |>
log10()
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 50 30.152 0.6030 1.5165 0.0318492 *
## Profundidade 1 0.599 0.5993 1.5071 0.2217661
## Trat 1 5.176 5.1765 13.0173 0.0004367 ***
## Profundidade:Trat 1 0.033 0.0328 0.0824 0.7745321
## Residuals 132 52.491 0.3977
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres -0.653 0.0968 132 -0.845 -0.462 a
## EstAus_CarPres -0.292 0.0968 132 -0.484 -0.101 b
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres -0.746 0.0924 132 -0.929 -0.564 a
## EstAus_CarPres -0.438 0.0924 132 -0.621 -0.256 b
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Tensão do ponto de inflexão ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "I"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"))
# tb_sel[["y"]] <- tb_sel[["y"]] |>
# log10()
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$I)
gg
postscript("figs/i_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 50 212.67 4.2533 1.0538 0.3980
## Profundidade 1 4.14 4.1363 1.0248 0.3132
## Trat 1 4.75 4.7531 1.1776 0.2798
## Profundidade:Trat 1 1.45 1.4501 0.3593 0.5499
## Residuals 132 532.80 4.0364
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.721 0.308 132 0.111 1.33 a
## EstAus_CarPres 0.859 0.308 132 0.249 1.47 a
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres 0.859 0.294 132 0.276 1.44 a
## EstAus_CarPres 1.349 0.294 132 0.767 1.93 a
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
# Inclinação no ponto de inflexão (S) ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬
y <- "S"
tb_sel <-
tb |>
select(Local, Profundidade, Trat, y = any_of(y)) |>
filter(Trat %in% c("EstPres_CarPres", "EstAus_CarPres"))
# tb_sel[["y"]] <- -tb_sel[["y"]] |>
# log10()
tb_sel_wide <-
tb_sel |>
pivot_wider(names_from = Trat, values_from = y) |>
drop_na()
tb_sel <-
tb_sel_wide |>
pivot_longer(cols = -c("Local", "Profundidade"),
names_to = "Trat",
values_to = "y") |>
mutate(Trat = factor(Trat, levels = levels(tb$Trat)))
gg <-
tb_sel |>
ggplot(data = _,
aes(x = Trat, y = y)) +
facet_wrap(~ Profundidade) +
geom_boxplot() +
geom_jitter(width = 0.05, color = "orange") +
geom_segment(data = tb_sel_wide,
aes(x = 1,
xend = 2,
y = EstPres_CarPres,
yend = EstAus_CarPres),
color = "gray", size = 0.5) +
scale_x_discrete(labels = legend$trat_levels) +
labs(x = legend$trat_axis,
y = legend$S)
gg
postscript("figs/s_index_estrutura_carbono_preservado.eps",
width = 8, height = 6)
print(gg)
dev.off()
## png
## 2
m0 <- lm(y ~ Local + Profundidade * Trat, data = tb_sel)
anova(m0)
## Analysis of Variance Table
##
## Response: y
## Df Sum Sq Mean Sq F value Pr(>F)
## Local 50 0.118988 0.0023798 2.5896 8.219e-06 ***
## Profundidade 1 0.007257 0.0072569 7.8968 0.005706 **
## Trat 1 0.026571 0.0265708 28.9138 3.331e-07 ***
## Profundidade:Trat 1 0.024232 0.0242321 26.3688 9.882e-07 ***
## Residuals 132 0.121304 0.0009190
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# par(mfrow = c(2, 2))
# plot(m0)
# layout(1)
# MASS::boxcox(m0)
emmeans::emmeans(m0, ~Trat | Profundidade) |>
multcomp::cld(Letters = letters)
## Profundidade = 0-0.05 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres -0.0956 0.00465 132 -0.1048 -0.0863 a
## EstAus_CarPres -0.0481 0.00465 132 -0.0573 -0.0389 b
##
## Profundidade = 0.15-0.20 m:
## Trat emmean SE df lower.CL upper.CL .group
## EstPres_CarPres -0.0596 0.00444 132 -0.0684 -0.0508 a
## EstAus_CarPres -0.0578 0.00444 132 -0.0666 -0.0490 a
##
## Results are averaged over the levels of: Local
## Confidence level used: 0.95
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.
#///////////////////////////////////////////////////////////////////////