Statistical analysis of microclimate and ectotherm outputs
Urban Dajcman
2023-10-26
set.seed(18041993)
using <- function(...) {
libs <- unlist(list(...))
req <- unlist(lapply(libs, require, character.only = TRUE))
need <- libs[req == FALSE]
if (length(need) > 0) {
install.packages(need)
lapply(need, require, character.only = TRUE)
}
}
using("dplyr", "ggplot2", "ggpubr", "rstatix", "NCmisc", "car", "lme4", "DHARMa",
"emmeans", "performance", "sjPlot", "rphylopic", "jpeg", "flexplot", "report",
"broom", "ggfortify", "MASS", "jtools", "svglite", "multcomp", "vegan", "showtext",
"formatR")
rm(list = ls())Prepare modeling data.frames
load("../../Results/ECTOTHERM_results/results_all_ecto2.Rda")
res_ecto_stats <- results_all
# Get yearly values
res_ecto_stats$Forage_year <- res_ecto_stats$Forage/res_ecto_stats$Year
res_ecto_stats$Bask_year <- res_ecto_stats$Bask/res_ecto_stats$Year
res_ecto_stats$Fecundity_year <- res_ecto_stats$Fecundity/res_ecto_stats$Year
res_ecto_stats$Clutches_year <- res_ecto_stats$Clutches/res_ecto_stats$Year
# Rename values
res_ecto_stats$Species[res_ecto_stats$Species == "Ihor"] <- "I. horvathi"
res_ecto_stats$Species[res_ecto_stats$Species == "Pmur"] <- "P. muralis"
res_ecto_stats$Sintopy[res_ecto_stats$Sintopy == "sint"] <- "Sintopy"
res_ecto_stats$Sintopy[res_ecto_stats$Sintopy == "aloPmur"] <- "Muralis allotopy"
res_ecto_stats$Sintopy[res_ecto_stats$Sintopy == "aloIhor"] <- "Horvathi allotopy"
# Turn to factor
res_ecto_stats$Locality <- as.factor(res_ecto_stats$Locality)
res_ecto_stats$Sintopy <- as.factor(res_ecto_stats$Sintopy)
res_ecto_stats$Species <- as.factor(res_ecto_stats$Species)
res_ecto_stats <- res_ecto_stats %>%
rename(Loc_type = Sintopy)
res_ecto_stats <- res_ecto_stats[, c("Elevation", "Loc_type", "Species", "MeanTALOC",
"MeanRHLOC", "Solar", "NO_Snow", "egg_dev", "LifeSpan", "Years_repro", "Bask_year",
"Forage_year", "Fecundity_year")]
head(res_ecto_stats)## Elevation Loc_type Species MeanTALOC MeanRHLOC Solar NO_Snow
## 1 237.8280 Sintopy I. horvathi 14.555508 67.54956 25.63949 5650.208
## 2 945.6031 Muralis allotopy I. horvathi 8.673618 84.47199 39.68655 4750.208
## 3 377.6037 Sintopy I. horvathi 12.335269 66.66661 27.52628 5479.042
## 4 1052.0754 Sintopy I. horvathi 9.401028 79.55288 38.91069 4894.500
## 5 577.6978 Muralis allotopy I. horvathi 10.966352 79.44083 36.43622 5298.875
## 6 830.2592 Muralis allotopy I. horvathi 8.865953 76.12646 36.58389 5134.167
## egg_dev LifeSpan Years_repro Bask_year Forage_year Fecundity_year
## 1 35 7 4 1821.857 463.2857 1.428571
## 2 59 8 2 1521.500 309.6250 0.500000
## 3 39 7 3 1966.571 461.4286 1.428571
## 4 51 8 3 1575.500 341.1250 0.750000
## 5 50 8 3 1501.500 339.3750 0.750000
## 6 53 8 3 1562.750 336.8750 0.500000str(res_ecto_stats)## 'data.frame': 30 obs. of 13 variables:
## $ Elevation : num 238 946 378 1052 578 ...
## $ Loc_type : Factor w/ 3 levels "Horvathi allotopy",..: 3 2 3 3 2 2 3 1 1 2 ...
## $ Species : Factor w/ 2 levels "I. horvathi",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ MeanTALOC : num 14.56 8.67 12.34 9.4 10.97 ...
## $ MeanRHLOC : num 67.5 84.5 66.7 79.6 79.4 ...
## $ Solar : num 25.6 39.7 27.5 38.9 36.4 ...
## $ NO_Snow : num 5650 4750 5479 4894 5299 ...
## $ egg_dev : num 35 59 39 51 50 53 51 66 45 51 ...
## $ LifeSpan : num 7 8 7 8 8 8 8 9 8 8 ...
## $ Years_repro : int 4 2 3 3 3 3 4 2 4 3 ...
## $ Bask_year : num 1822 1522 1967 1576 1502 ...
## $ Forage_year : num 463 310 461 341 339 ...
## $ Fecundity_year: num 1.43 0.5 1.43 0.75 0.75 ...## Microclimate analysis table
loc_stats <- res_ecto_stats %>%
distinct(MeanTALOC, .keep_all = TRUE)Models and plots for elevation, species and sintopy analysis
Mean Temp at location of animal
shiny test
## Elevation Loc_type Species MeanTALOC MeanRHLOC Solar NO_Snow
## 1 237.8280 Sintopy I. horvathi 14.555508 67.54956 25.63949 5650.208
## 2 945.6031 Muralis allotopy I. horvathi 8.673618 84.47199 39.68655 4750.208
## 3 377.6037 Sintopy I. horvathi 12.335269 66.66661 27.52628 5479.042
## 4 1052.0754 Sintopy I. horvathi 9.401028 79.55288 38.91069 4894.500
## 5 577.6978 Muralis allotopy I. horvathi 10.966352 79.44083 36.43622 5298.875
## 6 830.2592 Muralis allotopy I. horvathi 8.865953 76.12646 36.58389 5134.167
## egg_dev LifeSpan Years_repro Bask_year Forage_year Fecundity_year
## 1 35 7 4 1821.857 463.2857 1.428571
## 2 59 8 2 1521.500 309.6250 0.500000
## 3 39 7 3 1966.571 461.4286 1.428571
## 4 51 8 3 1575.500 341.1250 0.750000
## 5 50 8 3 1501.500 339.3750 0.750000
## 6 53 8 3 1562.750 336.8750 0.500000Shiny applications not supported in static R Markdown documents
# check data distribution
gghistogram(loc_stats$MeanTALOC, bins = 15)
ggboxplot(loc_stats$MeanTALOC)
# Check lm model
lm_MeanTALOC <- lm(MeanTALOC ~ Elevation + Loc_type, data = loc_stats)
lm_MeanTALOC_int <- lm(MeanTALOC ~ Elevation * Loc_type, data = loc_stats)
AIC(lm_MeanTALOC, lm_MeanTALOC_int)## df AIC
## lm_MeanTALOC 5 29.31994
## lm_MeanTALOC_int 7 31.00826compare.fits(MeanTALOC ~ Elevation | Loc_type, data = loc_stats, lm_MeanTALOC, lm_MeanTALOC_int)
# Get summary and p values
summary(lm_MeanTALOC)##
## Call:
## lm(formula = MeanTALOC ~ Elevation + Loc_type, data = loc_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.62319 -0.38025 -0.02903 0.27615 0.92012
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 14.1146391 0.5812257 24.284 6.60e-11 ***
## Elevation -0.0054589 0.0005167 -10.564 4.26e-07 ***
## Loc_typeMuralis allotopy -0.2500794 0.3841958 -0.651 0.5285
## Loc_typeSintopy 0.8190153 0.4038880 2.028 0.0675 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.538 on 11 degrees of freedom
## Multiple R-squared: 0.9468, Adjusted R-squared: 0.9322
## F-statistic: 65.2 on 3 and 11 DF, p-value: 2.722e-07# Plot
lm_Plot_eTemp <- lm(MeanTALOC ~ Elevation, data = loc_stats)
MeanTemp_plot <- visualize(lm_Plot_eTemp, MeanTALOC ~ Elevation, plot = "model") +
scale_color_manual(values = c("black")) + theme_classic() + labs_pubr() + labs(title = NULL,
y = "Mean temperature (°C)", x = NULL) + theme(legend.position = "null")
MeanTemp_plot
# Check model fit, residuals etc.
autoplot(lm_MeanTALOC) + theme_classic()
shapiro.test(residuals(lm_MeanTALOC))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_MeanTALOC)
## W = 0.94825, p-value = 0.4972Mean relative humidity
# check data distribution
gghistogram(loc_stats$MeanRHLOC, bins = 15)
ggboxplot(loc_stats$MeanRHLOC)
# Check lm model first
lm_MeanRHLOC <- lm(MeanRHLOC ~ Elevation + Loc_type, data = loc_stats)
lm_MeanRHLOC_int <- lm(MeanRHLOC ~ Elevation * Loc_type, data = loc_stats)
compare.fits(MeanRHLOC ~ Elevation | Loc_type, data = loc_stats, lm_MeanRHLOC, lm_MeanRHLOC_int)
AIC(lm_MeanRHLOC, lm_MeanRHLOC_int)## df AIC
## lm_MeanRHLOC 5 77.75286
## lm_MeanRHLOC_int 7 79.55719# Get summary and p values
summary(lm_MeanRHLOC)##
## Call:
## lm(formula = MeanRHLOC ~ Elevation + Loc_type, data = loc_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.3674 -1.8245 0.2331 2.3209 3.1476
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 64.994519 2.920674 22.253 1.7e-10 ***
## Elevation 0.014790 0.002597 5.696 0.000139 ***
## Loc_typeMuralis allotopy 3.219854 1.930594 1.668 0.123539
## Loc_typeSintopy -1.232829 2.029548 -0.607 0.555891
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.704 on 11 degrees of freedom
## Multiple R-squared: 0.8282, Adjusted R-squared: 0.7814
## F-statistic: 17.68 on 3 and 11 DF, p-value: 0.0001618# Plot
lm_plot_eHum <- lm(MeanRHLOC ~ Elevation, data = loc_stats)
Mean_humid <- visualize(lm_plot_eHum, MeanRHLOC ~ Elevation, plot = "model") + scale_color_manual(values = c("black")) +
theme_classic() + labs_pubr() + labs(title = NULL, y = "Mean humidity", x = NULL) +
theme(legend.position = "null")
Mean_humid
# Check model fit, residuals etc.
autoplot(lm_MeanRHLOC) + theme_classic()
shapiro.test(residuals(lm_MeanRHLOC))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_MeanRHLOC)
## W = 0.91829, p-value = 0.1814Mean solar radiation
# check data distribution
gghistogram(loc_stats$Solar, bins = 5)
ggboxplot(loc_stats$Solar)
# Check lm model first
lm_Solar <- lm(Solar ~ Elevation + Loc_type, data = loc_stats)
lm_Solar_int <- lm(Solar ~ Elevation * Loc_type, data = loc_stats)
compare.fits(Solar ~ Elevation | Loc_type, data = loc_stats, lm_Solar, lm_Solar_int)
AIC(lm_Solar, lm_Solar_int)## df AIC
## lm_Solar 5 65.70929
## lm_Solar_int 7 59.61318anova(lm_Solar, lm_Solar_int)## Analysis of Variance Table
##
## Model 1: Solar ~ Elevation + Loc_type
## Model 2: Solar ~ Elevation * Loc_type
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 11 36.022
## 2 9 18.376 2 17.646 4.3211 0.04837 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Get summary and p values
summary(lm_Solar_int) #AIC delta <4, include interaction##
## Call:
## lm(formula = Solar ~ Elevation * Loc_type, data = loc_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8615 -0.7580 0.0894 0.4437 2.6319
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 28.606198 2.334521 12.254 6.44e-07 ***
## Elevation 0.009192 0.002193 4.192 0.00234 **
## Loc_typeMuralis allotopy 4.378561 3.112470 1.407 0.19307
## Loc_typeSintopy -6.323805 2.766069 -2.286 0.04807 *
## Elevation:Loc_typeMuralis allotopy -0.003747 0.003622 -1.034 0.32790
## Elevation:Loc_typeSintopy 0.006400 0.003121 2.050 0.07056 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.429 on 9 degrees of freedom
## Multiple R-squared: 0.9329, Adjusted R-squared: 0.8956
## F-statistic: 25.01 on 5 and 9 DF, p-value: 5.007e-05# Plot
Mean_solar <- visualize(lm_Solar_int, Solar ~ Elevation + Loc_type, plot = "model") +
scale_color_manual(values = c("firebrick", "gold", "navyblue")) + theme_classic() +
labs_pubr() + labs(title = NULL, y = "Mean solar radiation", x = NULL) + theme(legend.position = "null")
Mean_solar
# Check model fit, residuals etc.
autoplot(lm_Solar) + theme_classic()
shapiro.test(residuals(lm_Solar))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_Solar)
## W = 0.91047, p-value = 0.1377# Check comparisons
summary(glht(lm_Solar_int, linfct = mcp(Loc_type = "Tukey")))##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: lm(formula = Solar ~ Elevation * Loc_type, data = loc_stats)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## Muralis allotopy - Horvathi allotopy == 0 4.379 3.112 1.407 0.37564
## Sintopy - Horvathi allotopy == 0 -6.324 2.766 -2.286 0.10847
## Sintopy - Muralis allotopy == 0 -10.702 2.537 -4.218 0.00541
##
## Muralis allotopy - Horvathi allotopy == 0
## Sintopy - Horvathi allotopy == 0
## Sintopy - Muralis allotopy == 0 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)Days without snow
# check data distribution
gghistogram(loc_stats$NO_Snow, bins = 50)
ggboxplot(loc_stats$NO_Snow)
# Check lm model first
lm_NO_Snow <- lm(NO_Snow ~ Elevation + Loc_type, data = loc_stats)
lm_NO_Snow_int <- lm(NO_Snow ~ Elevation * Loc_type, data = loc_stats)
compare.fits(NO_Snow ~ Elevation | Loc_type, data = loc_stats, lm_NO_Snow, lm_NO_Snow_int)
AIC(lm_NO_Snow, lm_NO_Snow_int)## df AIC
## lm_NO_Snow 5 188.9979
## lm_NO_Snow_int 7 191.8064# Get summary and p values
summary(lm_NO_Snow)##
## Call:
## lm(formula = NO_Snow ~ Elevation + Loc_type, data = loc_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -147.40 -41.31 -14.50 61.21 191.42
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5824.6413 119.1022 48.905 3.21e-14 ***
## Elevation -1.0529 0.1059 -9.944 7.82e-07 ***
## Loc_typeMuralis allotopy 48.4005 78.7277 0.615 0.551
## Loc_typeSintopy 89.2488 82.7629 1.078 0.304
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 110.2 on 11 degrees of freedom
## Multiple R-squared: 0.9368, Adjusted R-squared: 0.9196
## F-statistic: 54.34 on 3 and 11 DF, p-value: 6.964e-07# Plot
lm_plot_eSnow <- lm(NO_Snow ~ Elevation, data = loc_stats)
SnowDays_plot <- visualize(lm_plot_eSnow, NO_Snow ~ Elevation, plot = "model") +
scale_color_manual(values = c("black")) + theme_classic() + labs_pubr() + labs(title = NULL,
y = "Days without snow", x = NULL) + theme(legend.position = "null")
SnowDays_plot
# Check model fit, residuals etc.
autoplot(lm_NO_Snow) + theme_classic()
shapiro.test(residuals(lm_NO_Snow))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_NO_Snow)
## W = 0.95194, p-value = 0.5555Arrange loc plots
plot_micro <- ggarrange(MeanTemp_plot, SnowDays_plot, Mean_solar, Mean_humid, nrow = 2,
ncol = 2, labels = "AUTO", common.legend = TRUE, legend = "none", font.label = list(size = 13))
plot_micro
ggsave("plot_micro_final.pdf", plot = plot_micro, path = "../../Figures/Ectotherm_plots/",
units = "cm", width = 20, height = 20, dpi = 300)
ggsave("plot_microfinal.svg", plot = plot_micro, path = "../../Figures/Ectotherm_plots/",
units = "cm", width = 20, height = 20, dpi = 300)
ggsave("plot_micro_final.png", plot = plot_micro, path = "../../Figures/Ectotherm_plots/",
units = "cm", width = 20, height = 20, dpi = 300, bg = "white")Models and plots for elevation, species and sintopy analysis
Egg development time
# check data distribution
gghistogram(res_ecto_stats$egg_dev, bins = 15) + theme_pubclean()
ggboxplot(res_ecto_stats$egg_dev) + theme_pubclean() + labs(y = NULL, x = NULL)
# Check lm model first
lm_eggDev <- lm(egg_dev ~ Elevation + Species + Loc_type, data = res_ecto_stats)
lm_eggDev_int <- lm(egg_dev ~ Elevation * Species + Loc_type, data = res_ecto_stats)
lm_eggDev_int2 <- lm(egg_dev ~ Elevation * Loc_type + Species, data = res_ecto_stats)
lm_eggDev_int3 <- lm(egg_dev ~ Elevation + Loc_type * Species, data = res_ecto_stats)
compare.fits(egg_dev ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_eggDev,
lm_eggDev_int)
compare.fits(egg_dev ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_eggDev,
lm_eggDev_int2)
compare.fits(egg_dev ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_eggDev,
lm_eggDev_int3)
AIC(lm_eggDev, lm_eggDev_int, lm_eggDev_int2, lm_eggDev_int3)## df AIC
## lm_eggDev 6 149.5496
## lm_eggDev_int 7 151.4352
## lm_eggDev_int2 8 153.3362
## lm_eggDev_int3 8 153.2577anova(lm_eggDev, lm_eggDev_int, lm_eggDev_int2, lm_eggDev_int3)## Analysis of Variance Table
##
## Model 1: egg_dev ~ Elevation + Species + Loc_type
## Model 2: egg_dev ~ Elevation * Species + Loc_type
## Model 3: egg_dev ~ Elevation * Loc_type + Species
## Model 4: egg_dev ~ Elevation + Loc_type * Species
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 25 172.14
## 2 24 171.49 1 0.65488 0.0881 0.7692
## 3 23 170.92 1 0.56505 0.0760 0.7852
## 4 23 170.47 0 0.44673summary(lm_eggDev) ## NO SIGNIFICANT INTERACTIONS##
## Call:
## lm(formula = egg_dev ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.2834 -1.4421 -0.6405 0.9118 6.2974
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 33.939118 2.060944 16.468 6.25e-15 ***
## Elevation 0.021286 0.001782 11.944 7.94e-12 ***
## SpeciesP. muralis 8.533333 0.958164 8.906 3.15e-09 ***
## Loc_typeMuralis allotopy 2.848066 1.324986 2.150 0.0415 *
## Loc_typeSintopy -3.337468 1.392899 -2.396 0.0244 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.624 on 25 degrees of freedom
## Multiple R-squared: 0.9305, Adjusted R-squared: 0.9193
## F-statistic: 83.62 on 4 and 25 DF, p-value: 4.264e-14# summary(lm_eggDev_int) summary(lm_eggDev_int2) summary(lm_eggDev_int3)
# Get summary and p values
visualize(lm_eggDev)
EggDev_plot <- visualize(lm_eggDev, egg_dev ~ Elevation + Loc_type | Species, plot = "model") +
scale_color_manual(values = c("firebrick", "gold", "navyblue")) + theme_classic() +
labs_pubr() + labs(title = NULL, y = "Egg development time (days)", x = NULL) +
theme(strip.text = element_text(face = "italic"), axis.text = element_text(size = 11),
axis.title = element_text(size = 12, face = "bold")) + facet_wrap(~Species)
EggDev_plot
# Check model fit, residuals etc.
autoplot(lm_eggDev) + theme_classic()
shapiro.test(residuals(lm_eggDev))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_eggDev)
## W = 0.94739, p-value = 0.1438# Check comparisons
summary(glht(lm_eggDev, linfct = mcp(Loc_type = "Tukey")))##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: lm(formula = egg_dev ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## Muralis allotopy - Horvathi allotopy == 0 2.848 1.325 2.150 0.0996
## Sintopy - Horvathi allotopy == 0 -3.337 1.393 -2.396 0.0607
## Sintopy - Muralis allotopy == 0 -6.186 1.181 -5.236 <0.001
##
## Muralis allotopy - Horvathi allotopy == 0 .
## Sintopy - Horvathi allotopy == 0 .
## Sintopy - Muralis allotopy == 0 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)Lifespan
# check data
gghistogram(res_ecto_stats$LifeSpan, bins = 4)
ggboxplot(res_ecto_stats$LifeSpan)
# Check lm model first
lm_Lifespan <- lm(LifeSpan ~ Elevation + Species + Loc_type, data = res_ecto_stats)
lm_Lifespan_int <- lm(LifeSpan ~ Elevation * Species + Loc_type, data = res_ecto_stats)
lm_Lifespan_int2 <- lm(LifeSpan ~ Elevation * Loc_type + Species, data = res_ecto_stats)
lm_Lifespan_int3 <- lm(LifeSpan ~ Elevation + Loc_type * Species, data = res_ecto_stats)
compare.fits(LifeSpan ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_Lifespan,
lm_Lifespan_int)
compare.fits(LifeSpan ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_Lifespan,
lm_Lifespan_int2)
compare.fits(LifeSpan ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_Lifespan,
lm_Lifespan_int3)
AIC(lm_Lifespan, lm_Lifespan_int, lm_Lifespan_int2, lm_Lifespan_int3)## df AIC
## lm_Lifespan 6 24.40601
## lm_Lifespan_int 7 25.53875
## lm_Lifespan_int2 8 26.08489
## lm_Lifespan_int3 8 27.64341anova(lm_Lifespan, lm_Lifespan_int, lm_Lifespan_int2, lm_Lifespan_int3)## Analysis of Variance Table
##
## Model 1: LifeSpan ~ Elevation + Species + Loc_type
## Model 2: LifeSpan ~ Elevation * Species + Loc_type
## Model 3: LifeSpan ~ Elevation * Loc_type + Species
## Model 4: LifeSpan ~ Elevation + Loc_type * Species
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 25 2.6561
## 2 24 2.5804 1 0.075685 0.7081 0.4087
## 3 23 2.4583 1 0.122070 1.1421 0.2963
## 4 23 2.5894 0 -0.131088summary(lm_Lifespan) ## NO SIGNIFICANT INTERACTIONS, CHECKED MODELS BY AIC AND COMPARED FITS##
## Call:
## lm(formula = LifeSpan ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.56118 -0.28673 0.01384 0.33310 0.42537
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.9824824 0.2560040 27.275 < 2e-16 ***
## Elevation 0.0012542 0.0002214 5.666 6.73e-06 ***
## SpeciesP. muralis 1.0666667 0.1190201 8.962 2.79e-09 ***
## Loc_typeMuralis allotopy -0.0670286 0.1645856 -0.407 0.6873
## Loc_typeSintopy -0.3718992 0.1730215 -2.149 0.0415 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.326 on 25 degrees of freedom
## Multiple R-squared: 0.8577, Adjusted R-squared: 0.8349
## F-statistic: 37.67 on 4 and 25 DF, p-value: 3.046e-10# summary(lm_Lifespan_int) summary(lm_Lifespan_int2) summary(lm_Lifespan_int3)
# Get summary and p values
visualize(lm_Lifespan_int2)
Lifespan_plot <- visualize(lm_Lifespan, LifeSpan ~ Elevation + Loc_type | Species,
plot = "model") + scale_color_manual(values = c("firebrick", "gold", "navyblue")) +
theme_classic() + labs_pubr() + labs(title = NULL, y = "Lifespan (years)", x = NULL) +
theme(strip.text = element_text(face = "italic"), axis.text = element_text(size = 11),
axis.title = element_text(size = 12, face = "bold")) + facet_wrap(~Species)
Lifespan_plot
# Check model fit, residuals etc.
autoplot(lm_Lifespan_int2) + theme_classic()
shapiro.test(residuals(lm_Lifespan_int2))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_Lifespan_int2)
## W = 0.95247, p-value = 0.1968# Check comparisons
summary(glht(lm_Lifespan, linfct = mcp(Loc_type = "Tukey")))##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: lm(formula = LifeSpan ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## Muralis allotopy - Horvathi allotopy == 0 -0.06703 0.16459 -0.407 0.9126
## Sintopy - Horvathi allotopy == 0 -0.37190 0.17302 -2.149 0.0997
## Sintopy - Muralis allotopy == 0 -0.30487 0.14673 -2.078 0.1143
##
## Muralis allotopy - Horvathi allotopy == 0
## Sintopy - Horvathi allotopy == 0 .
## Sintopy - Muralis allotopy == 0
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)Years reproducing
# check data
gghistogram(res_ecto_stats$Years_repro, bins = 8)
ggboxplot(res_ecto_stats$Years_repro)
# Check lm model first
lm_YearsRepro_noLoc <- lm(Years_repro ~ Elevation + Species, data = res_ecto_stats)
lm_YearsRepro <- lm(Years_repro ~ Elevation + Species + Loc_type, data = res_ecto_stats)
lm_YearsRepro_int <- lm(Years_repro ~ Elevation * Species + Loc_type, data = res_ecto_stats)
lm_YearsRepro_int2 <- lm(Years_repro ~ Elevation * Loc_type + Species, data = res_ecto_stats)
lm_YearsRepro_int3 <- lm(Years_repro ~ Elevation + Loc_type * Species, data = res_ecto_stats)
compare.fits(Years_repro ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_YearsRepro, lm_YearsRepro_int)
compare.fits(Years_repro ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_YearsRepro, lm_YearsRepro_int2)
compare.fits(Years_repro ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_YearsRepro, lm_YearsRepro_int3)
AIC(lm_YearsRepro, lm_YearsRepro_int, lm_YearsRepro_int2, lm_YearsRepro_int3, lm_YearsRepro_noLoc)## df AIC
## lm_YearsRepro 6 79.17173
## lm_YearsRepro_int 7 79.64740
## lm_YearsRepro_int2 8 79.43249
## lm_YearsRepro_int3 8 82.80551
## lm_YearsRepro_noLoc 4 78.61733anova(lm_YearsRepro_noLoc, lm_YearsRepro, lm_YearsRepro_int, lm_YearsRepro_int2,
lm_YearsRepro_int3)## Analysis of Variance Table
##
## Model 1: Years_repro ~ Elevation + Species
## Model 2: Years_repro ~ Elevation + Species + Loc_type
## Model 3: Years_repro ~ Elevation * Species + Loc_type
## Model 4: Years_repro ~ Elevation * Loc_type + Species
## Model 5: Years_repro ~ Elevation + Loc_type * Species
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 27 18.490
## 2 25 16.484 2 2.00623 1.5854 0.2264
## 3 24 15.667 1 0.81663 1.2907 0.2676
## 4 23 14.552 1 1.11505 1.7624 0.1974
## 5 23 16.284 0 -1.73168summary(lm_YearsRepro_noLoc) ## OPT FOR SIMPLER MODEL##
## Call:
## lm(formula = Years_repro ~ Elevation + Species, data = res_ecto_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.8976 -0.2866 0.1232 0.5074 1.0660
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.3575515 0.4159431 10.476 5.18e-11 ***
## Elevation -0.0017666 0.0004644 -3.804 0.000741 ***
## SpeciesP. muralis 0.8000000 0.3021734 2.647 0.013371 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8275 on 27 degrees of freedom
## Multiple R-squared: 0.4431, Adjusted R-squared: 0.4018
## F-statistic: 10.74 on 2 and 27 DF, p-value: 0.0003701# summary(lm_YearsRepro_int) summary(lm_YearsRepro_int2)
# summary(lm_YearsRepro_int3)
# Get summary and p values
visualize(lm_YearsRepro_noLoc)
YearsRepro_plot <- visualize(lm_YearsRepro_noLoc, Years_repro ~ Elevation | Species,
plot = "model") + scale_color_manual(values = c("black")) + theme_classic() +
labs_pubr() + labs(title = NULL, y = "Years reproducing", x = NULL) + theme(strip.background = element_blank(),
strip.text.x = element_blank(), axis.text = element_text(size = 11), axis.title = element_text(size = 12,
face = "bold")) + facet_wrap(~Species)
YearsRepro_plot
# Check model fit, residuals etc.
autoplot(lm_YearsRepro) + theme_classic()
shapiro.test(residuals(lm_YearsRepro))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_YearsRepro)
## W = 0.80519, p-value = 8.164e-05Yearly basking time
# check data
gghistogram(res_ecto_stats$Bask_year, bins = 8)
ggboxplot(res_ecto_stats$Bask_year)
# Check lm model first
lm_BaskYear <- lm(Bask_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
lm_BaskYear_int <- lm(Bask_year ~ Elevation * Species + Loc_type, data = res_ecto_stats)
lm_BaskYear_int2 <- lm(Bask_year ~ Elevation * Loc_type + Species, data = res_ecto_stats)
lm_BaskYear_int3 <- lm(Bask_year ~ Elevation + Loc_type * Species, data = res_ecto_stats)
compare.fits(Bask_year ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_BaskYear,
lm_BaskYear_int)
compare.fits(Bask_year ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_BaskYear,
lm_BaskYear_int2)
compare.fits(Bask_year ~ Elevation | Species + Loc_type, data = res_ecto_stats, lm_BaskYear,
lm_BaskYear_int3)
AIC(lm_BaskYear, lm_BaskYear_int, lm_BaskYear_int2, lm_BaskYear_int3)## df AIC
## lm_BaskYear 6 366.5616
## lm_BaskYear_int 7 368.5616
## lm_BaskYear_int2 8 368.9511
## lm_BaskYear_int3 8 370.5466anova(lm_BaskYear, lm_BaskYear_int, lm_BaskYear_int2, lm_BaskYear_int3)## Analysis of Variance Table
##
## Model 1: Bask_year ~ Elevation + Species + Loc_type
## Model 2: Bask_year ~ Elevation * Species + Loc_type
## Model 3: Bask_year ~ Elevation * Loc_type + Species
## Model 4: Bask_year ~ Elevation + Loc_type * Species
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 25 238480
## 2 24 238480 1 0.2 0.0000 0.9965
## 3 23 226015 1 12464.7 1.2684 0.2717
## 4 23 238361 0 -12345.7summary(lm_BaskYear) ## NO SIGNIFICANT INTERACTIONS##
## Call:
## lm(formula = Bask_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -181.902 -62.389 -4.095 74.524 158.334
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1846.36601 76.70997 24.069 < 2e-16 ***
## Elevation -0.34636 0.06633 -5.222 2.10e-05 ***
## SpeciesP. muralis -361.50730 35.66363 -10.137 2.44e-10 ***
## Loc_typeMuralis allotopy -33.78520 49.31702 -0.685 0.4996
## Loc_typeSintopy 122.38856 51.84480 2.361 0.0263 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 97.67 on 25 degrees of freedom
## Multiple R-squared: 0.8714, Adjusted R-squared: 0.8508
## F-statistic: 42.34 on 4 and 25 DF, p-value: 8.748e-11# summary(lm_BaskYear_int) summary(lm_BaskYear_int2) summary(lm_BaskYear_int3)
# Get summary and p values
visualize(lm_BaskYear)
Basking_plot <- visualize(lm_BaskYear, Bask_year ~ Elevation + Loc_type | Species,
plot = "model") + scale_color_manual(values = c("firebrick", "gold", "navyblue")) +
theme_classic() + labs_pubr() + labs(title = NULL, y = "Mean yearly basking time",
x = NULL) + theme(strip.background = element_blank(), strip.text.x = element_blank(),
axis.text = element_text(size = 11), axis.title = element_text(size = 12, face = "bold")) +
facet_wrap(~Species)
Basking_plot
# Check model fit, residuals etc.
autoplot(lm_BaskYear) + theme_classic()
shapiro.test(residuals(lm_BaskYear))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_BaskYear)
## W = 0.97735, p-value = 0.7517# Check comparisons
summary(glht(lm_BaskYear, linfct = mcp(Loc_type = "Tukey")))##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: lm(formula = Bask_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## Muralis allotopy - Horvathi allotopy == 0 -33.79 49.32 -0.685 0.77345
## Sintopy - Horvathi allotopy == 0 122.39 51.84 2.361 0.06532
## Sintopy - Muralis allotopy == 0 156.17 43.97 3.552 0.00414
##
## Muralis allotopy - Horvathi allotopy == 0
## Sintopy - Horvathi allotopy == 0 .
## Sintopy - Muralis allotopy == 0 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)Forage time
# check data
gghistogram(res_ecto_stats$Forage_year, bins = 8)
ggboxplot(res_ecto_stats$Forage_year)
# Check lm model first
lm_ForageYear <- lm(Forage_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
lm_ForageYear_int <- lm(Forage_year ~ Elevation * Species + Loc_type, data = res_ecto_stats)
lm_ForageYear_int2 <- lm(Forage_year ~ Elevation * Loc_type + Species, data = res_ecto_stats)
lm_ForageYear_int3 <- lm(Forage_year ~ Elevation + Loc_type * Species, data = res_ecto_stats)
compare.fits(Forage_year ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_ForageYear, lm_ForageYear_int)
compare.fits(Forage_year ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_ForageYear, lm_ForageYear_int2)
compare.fits(Forage_year ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_ForageYear, lm_ForageYear_int3)
AIC(lm_ForageYear, lm_ForageYear_int, lm_ForageYear_int2, lm_ForageYear_int3)## df AIC
## lm_ForageYear 6 303.6498
## lm_ForageYear_int 7 302.5191
## lm_ForageYear_int2 8 304.0072
## lm_ForageYear_int3 8 305.6708anova(lm_ForageYear, lm_ForageYear_int, lm_ForageYear_int2, lm_ForageYear_int3)## Analysis of Variance Table
##
## Model 1: Forage_year ~ Elevation + Species + Loc_type
## Model 2: Forage_year ~ Elevation * Species + Loc_type
## Model 3: Forage_year ~ Elevation * Loc_type + Species
## Model 4: Forage_year ~ Elevation + Loc_type * Species
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 25 29289
## 2 24 26387 1 2902.51 2.5735 0.1223
## 3 23 25941 1 446.37 0.3958 0.5355
## 4 23 27420 0 -1479.12summary(lm_ForageYear) ## NO SIGNIFICANT INTERACTION##
## Call:
## lm(formula = Forage_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -71.75 -19.80 1.03 19.75 51.40
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 482.33872 26.88320 17.942 8.61e-16 ***
## Elevation -0.16908 0.02325 -7.274 1.27e-07 ***
## SpeciesP. muralis 94.91675 12.49840 7.594 5.99e-08 ***
## Loc_typeMuralis allotopy -18.68100 17.28327 -1.081 0.2901
## Loc_typeSintopy 41.14568 18.16913 2.265 0.0325 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 34.23 on 25 degrees of freedom
## Multiple R-squared: 0.8691, Adjusted R-squared: 0.8482
## F-statistic: 41.5 on 4 and 25 DF, p-value: 1.086e-10# summary(lm_ForageYear_int) summary(lm_ForageYear_int2)
# summary(lm_ForageYear_int3)
# Plot
visualize(lm_ForageYear)
Forage_plot <- visualize(lm_ForageYear, Forage_year ~ Elevation + Loc_type | Species,
plot = "model") + scale_color_manual(values = c("firebrick", "gold", "navyblue")) +
theme_classic() + labs_pubr() + labs(title = NULL, y = "Mean yearly foraging time",
x = "Elevation (m)") + theme(strip.background = element_blank(), strip.text.x = element_blank(),
axis.text = element_text(size = 11), axis.title = element_text(size = 12, face = "bold")) +
facet_wrap(~Species)
Forage_plot
# Check model fit, residuals etc.
autoplot(lm_ForageYear) + theme_classic()
shapiro.test(residuals(lm_ForageYear))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_ForageYear)
## W = 0.97172, p-value = 0.5872# Check comparisons
summary(glht(lm_ForageYear, linfct = mcp(Loc_type = "Tukey")))##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: lm(formula = Forage_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## Muralis allotopy - Horvathi allotopy == 0 -18.68 17.28 -1.081 0.53326
## Sintopy - Horvathi allotopy == 0 41.15 18.17 2.265 0.07934
## Sintopy - Muralis allotopy == 0 59.83 15.41 3.883 0.00186
##
## Muralis allotopy - Horvathi allotopy == 0
## Sintopy - Horvathi allotopy == 0 .
## Sintopy - Muralis allotopy == 0 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)Fecundity per year
# check data distribution
gghistogram(res_ecto_stats$Fecundity_year, bins = 15)
ggboxplot(res_ecto_stats$Fecundity_year)
# Check lm model first
lm_FecundityYear <- lm(Fecundity_year ~ Elevation + Species + Loc_type, data = res_ecto_stats)
lm_FecundityYear_int <- lm(Fecundity_year ~ Elevation * Species + Loc_type, data = res_ecto_stats)
lm_FecundityYear_int2 <- lm(Fecundity_year ~ Elevation * Loc_type + Species, data = res_ecto_stats)
lm_FecundityYear_int3 <- lm(Fecundity_year ~ Elevation + Loc_type * Species, data = res_ecto_stats)
compare.fits(Fecundity_year ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_FecundityYear, lm_FecundityYear_int)
compare.fits(Fecundity_year ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_FecundityYear, lm_FecundityYear_int2)
compare.fits(Fecundity_year ~ Elevation | Species + Loc_type, data = res_ecto_stats,
lm_FecundityYear, lm_FecundityYear_int3)
AIC(lm_FecundityYear, lm_FecundityYear_int, lm_FecundityYear_int2, lm_FecundityYear_int3)## df AIC
## lm_FecundityYear 6 36.27660
## lm_FecundityYear_int 7 29.19254
## lm_FecundityYear_int2 8 37.83507
## lm_FecundityYear_int3 8 34.53056anova(lm_FecundityYear, lm_FecundityYear_int, lm_FecundityYear_int2, lm_FecundityYear_int3)## Analysis of Variance Table
##
## Model 1: Fecundity_year ~ Elevation + Species + Loc_type
## Model 2: Fecundity_year ~ Elevation * Species + Loc_type
## Model 3: Fecundity_year ~ Elevation * Loc_type + Species
## Model 4: Fecundity_year ~ Elevation + Loc_type * Species
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 25 3.9454
## 2 24 2.9146 1 1.03074 6.5183 0.01778 *
## 3 23 3.6370 1 -0.72237
## 4 23 3.2576 0 0.37934
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# summary(lm_FecundityYear)
summary(lm_FecundityYear_int) ## significant interaction##
## Call:
## lm(formula = Fecundity_year ~ Elevation * Species + Loc_type,
## data = res_ecto_stats)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.5756 -0.2397 0.0506 0.2124 0.7352
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.4635603 0.3122470 4.687 9.2e-05 ***
## Elevation -0.0008098 0.0003070 -2.638 0.01441 *
## SpeciesP. muralis 1.6057455 0.3263897 4.920 5.1e-05 ***
## Loc_typeMuralis allotopy -0.3450186 0.1759648 -1.961 0.06162 .
## Loc_typeSintopy 0.2443910 0.1849840 1.321 0.19891
## Elevation:SpeciesP. muralis -0.0011395 0.0003911 -2.913 0.00762 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3485 on 24 degrees of freedom
## Multiple R-squared: 0.8179, Adjusted R-squared: 0.7799
## F-statistic: 21.56 on 5 and 24 DF, p-value: 3.666e-08# summary(lm_FecundityYear_int2) summary(lm_FecundityYear_int3)
# Plot
visualize(lm_FecundityYear_int)
Fecundity_plot <- visualize(lm_FecundityYear_int, Fecundity_year ~ Elevation + Loc_type |
Species, plot = "model") + scale_color_manual(values = c("firebrick", "gold",
"navyblue")) + theme_classic() + labs_pubr() + labs(title = NULL, y = "Yearly fecundity",
x = "Elevation (m)") + theme(strip.background = element_blank(), strip.text.x = element_blank(),
axis.text = element_text(size = 11), axis.title = element_text(size = 12, face = "bold")) +
facet_wrap(~Species)
Fecundity_plot
# Check model fit, residuals etc.
autoplot(lm_FecundityYear_int) + theme_classic()
shapiro.test(residuals(lm_FecundityYear_int))##
## Shapiro-Wilk normality test
##
## data: residuals(lm_FecundityYear_int)
## W = 0.97448, p-value = 0.6674# Check comparisons
summary(glht(lm_FecundityYear_int, linfct = mcp(Loc_type = "Tukey")))##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: lm(formula = Fecundity_year ~ Elevation * Species + Loc_type,
## data = res_ecto_stats)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## Muralis allotopy - Horvathi allotopy == 0 -0.3450 0.1760 -1.961 0.1431
## Sintopy - Horvathi allotopy == 0 0.2444 0.1850 1.321 0.3963
## Sintopy - Muralis allotopy == 0 0.5894 0.1569 3.757 0.0027
##
## Muralis allotopy - Horvathi allotopy == 0
## Sintopy - Horvathi allotopy == 0
## Sintopy - Muralis allotopy == 0 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)Arrange plots
plot_ecto <- ggarrange(EggDev_plot, Lifespan_plot, YearsRepro_plot, Basking_plot,
Forage_plot, Fecundity_plot, nrow = 3, ncol = 2, labels = "AUTO", common.legend = FALSE,
legend = "none", font.label = list(size = 13, color = "black", face = "bold"))
plot_ecto
ggsave("plot_ecto_final.pdf", plot = plot_ecto, path = "../../Figures/Ectotherm_plots/",
units = "cm", width = 20, height = 20, dpi = 300)
ggsave("plot_ecto_final.svg", plot = plot_ecto, path = "../../Figures/Ectotherm_plots/",
units = "cm", width = 20, height = 20, dpi = 300)
ggsave("plot_ecto_final.png", plot = plot_ecto, path = "../../Figures/Ectotherm_plots/",
units = "cm", width = 20, height = 20, dpi = 300, bg = "white")