library(tidyverse)
library(ggpubr)
library(readxl)
library(lme4)
library(performance)
library(haven)
library(robustlmm)
library(emmeans)
library(extrafont)
#font_import()
#loadfonts(device = "all")Altitude Script
Load packages
Data
data<- data %>%
mutate(RPEdif = (RPELeg - RPEBreathe))
data2 = data %>%
mutate(Altitude_code = recode(Altitude_code, "1" = "Sea", "2" = "Altitude", "3" = "Altitude")) %>%
mutate(Altitude_code = factor(Altitude_code, levels = c("Sea", "Altitude")) %>%
fct_explicit_na())
head(data)# A tibble: 6 × 14
ID Replication GD TimeinEnvironment Altitude Altitude_code Duration
<fct> <fct> <fct> <dbl> <dbl> <fct> <dbl>
1 8 1 0 168 11.9 1 98
2 8 2 0 168 11.9 1 98
3 10 1 0 168 11.9 1 95
4 10 2 0 168 11.9 1 98
5 10 3 0 168 11.9 1 98
6 27 1 0 168 11.9 1 95
# ℹ 7 more variables: Distance <dbl>, Distancemin <dbl>,
# HighSpeedRunning <dbl>, HSRmin <dbl>, RPELeg <dbl>, RPEBreathe <dbl>,
# RPEdif <dbl>Filter data into match and training
Game2<-subset(data2 ,GD == "0" )
Game2<-data.frame(Game2)
Train<-subset(data2 ,GD == c("1","2" ))Example code for match data
Visualization of Match data: RPE
b <- ggplot(Game2) +
aes(x = Altitude_code, y = RPEBreathe, fill = Altitude_code) +
geom_boxplot() +
scale_fill_manual(values = c(Sea = "#cbd6e4", Altitude = "#b04238")) +
geom_jitter(aes(fill = Altitude_code), color = "grey2", alpha = 0.5, size = 2, shape = 21) +
theme_minimal() +
theme(legend.position = "none",
axis.text.y = element_text(size = 12, family = 'Arial', color = 'black'),
axis.text.x = element_text(size = 12, family = 'Arial', color = 'black'),
plot.subtitle = element_text(face = "bold.italic", size = 14, family = "Arial"),
axis.title.y = element_text(size = 14, family = 'Arial',face = "bold")) +
scale_y_continuous(
limits = c(0, 120),
breaks = c(2, 5, 13, 25, 35, 50, 70, 90, 100, 120),
labels = c("Minimal", "Very Easy", "Easy", "Moderate", "Somewhat hard", "Hard", "Very hard", "Extremely hard", "Almost Maximal", "Absolute Max")
) +
labs(y = "RPE: CR-100 scale (AU)", x = "", subtitle = "Breathlessness exertion")
l <- ggplot(Game2) +
aes(x = Altitude_code, y = RPELeg, fill = Altitude_code) +
geom_boxplot() +
scale_fill_manual(values = c(Sea = "#cbd6e4", Altitude = "#b04238")) +
geom_jitter(color = "grey2", alpha = 0.8, size = 1.5) +
theme_minimal() +
theme(legend.position = "none",
axis.text.y = element_text(size = 12, family = 'Arial', color = 'black'),
axis.text.x = element_text(size = 12, family = 'Arial', color = 'black')) +
theme(plot.subtitle = element_text(
face = "bold.italic",
size = 14,
family = "Arial"
))+
scale_y_continuous(
limits = c(0, 120),
breaks = c(2, 5, 13, 25, 35, 50, 70, 90, 100, 120),
labels = c("Minimal", "Very Easy", "Easy", "Moderate", "Somewhat hard", "Hard", "Very hard", "Extremely hard", "Almost Maximal", "Absolute Max")
) +
labs(y = "", x = "", subtitle = "Leg exertion")
d <- ggplot(Game2) +
aes(x = Altitude_code, y = RPEdif, fill = Altitude_code) +
geom_boxplot() +
scale_fill_manual(values = c(Sea = "#cbd6e4", Altitude = "#b04238")) +
geom_jitter(aes(fill = Altitude_code), color = "grey2", alpha = 0.5, size = 1.5, shape = 21) +
theme_minimal() +
theme(legend.position = "none",
axis.text.y = element_text(size = 12, family = 'Arial', color = 'black'),
axis.text.x = element_text(size = 12, family = 'Arial', color = 'black'),
plot.subtitle = element_text(face = "bold.italic", size = 14, family = "Arial"),
axis.title.y = element_text(size = 14, family = 'Arial',face = "bold")) +
labs(y = "RPE difference (AU)", x = "", subtitle = "Leg - Breathlessness exertion")
fig <- ggarrange(b, l, d, nrow = 1)
fig
Visualization of match data: Running data
td <- ggplot(Game2) +
aes(x = Altitude_code, y = Distancemin , fill = Altitude_code) +
geom_boxplot() +
scale_fill_manual(values = c(Sea = "#cbd6e4", Altitude = "#b04238")) +
geom_jitter(aes(fill = Altitude_code), color = "grey2", alpha = 0.5, size = 2, shape = 21) +
theme_minimal() +
theme(legend.position = "none",
axis.text.y = element_text(size = 12, family = 'Arial', color = 'black'),
axis.text.x = element_text(size = 12, family = 'Arial', color = 'black'),
plot.subtitle = element_text(face = "bold.italic", size = 14, family = "Arial"),
axis.title.y = element_text(size = 14, family = 'Arial',face = "bold")) +
labs(
y = "Distance (m·min)",
x = "", title = "",
subtitle = "Total distance (m·min)"
)
hsr <-ggplot(Game2) +
aes(x = Altitude_code, y = HSRmin , fill = Altitude_code) +
geom_boxplot() +
scale_fill_manual(values = c(Sea = "#cbd6e4", Altitude = "#b04238")) +
geom_jitter(aes(fill = Altitude_code), color = "grey2", alpha = 0.5, size = 2, shape = 21) +
theme_minimal() +
theme(legend.position = "none",
axis.text.y = element_text(size = 12, family = 'Arial', color = 'black'),
axis.text.x = element_text(size = 12, family = 'Arial', color = 'black'),
plot.subtitle = element_text(face = "bold.italic", size = 14, family = "Arial"),
axis.title.y = element_text(size = 14, family = 'Arial',face = "bold")) +
labs(
y = "",
x = "", title = "",
subtitle = "High-speed running distance (m·min)"
)
fig2<-ggarrange(td, hsr, nrow = 1)
fig2
Analysis of match data
Model development (example for RPE Legs)
null<- lmer(RPELeg ~ 1 + (1| ID), Game2)
mod1<- lmer(RPELeg ~ 1 + Altitude_code + (1| ID), Game2)
mod2<- lmer(RPELeg ~ 1 + Altitude_code + Replication + (1| ID), Game2)
mod3<- lmer(RPELeg ~ 1 + Altitude_code + Replication + Duration + (1| ID), Game2)
mod4<- lmer(RPELeg~ 1 + Altitude_code + Duration + (1| ID), Game2)
compare_performance(null, mod1,mod2,mod3,mod4)# Comparison of Model Performance Indices
Name | Model | AIC (weights) | AICc (weights) | BIC (weights) | R2 (cond.)
----------------------------------------------------------------------------
null | lmerMod | 289.7 (0.023) | 290.5 (0.047) | 294.5 (0.117) | 0.426
mod1 | lmerMod | 290.4 (0.017) | 291.7 (0.026) | 296.7 (0.038) | 0.442
mod2 | lmerMod | 293.2 (0.004) | 296.1 (0.003) | 302.7 (0.002) | 0.427
mod3 | lmerMod | 285.6 (0.180) | 289.6 (0.072) | 296.7 (0.038) | 0.679
mod4 | lmerMod | 282.7 (0.777) | 284.7 (0.851) | 290.6 (0.805) | 0.686
Name | R2 (marg.) | ICC | RMSE | Sigma
------------------------------------------
null | 0.000 | 0.426 | 8.631 | 10.215
mod1 | 0.025 | 0.428 | 8.428 | 10.160
mod2 | 0.047 | 0.399 | 8.424 | 10.463
mod3 | 0.232 | 0.582 | 6.169 | 8.320
mod4 | 0.223 | 0.596 | 6.266 | 8.104 anova(null, mod1,mod2,mod3,mod4)refitting model(s) with ML (instead of REML)Data: Game2
Models:
null: RPELeg ~ 1 + (1 | ID)
mod1: RPELeg ~ 1 + Altitude_code + (1 | ID)
mod4: RPELeg ~ 1 + Altitude_code + Duration + (1 | ID)
mod2: RPELeg ~ 1 + Altitude_code + Replication + (1 | ID)
mod3: RPELeg ~ 1 + Altitude_code + Replication + Duration + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
null 3 289.71 294.46 -141.86 283.71
mod1 4 290.39 296.73 -141.19 282.39 1.3209 1 0.250433
mod4 5 282.70 290.62 -136.35 272.70 9.6881 1 0.001855 **
mod2 6 293.15 302.65 -140.58 281.15 0.0000 1 1.000000
mod3 7 285.56 296.65 -135.78 271.56 9.5920 1 0.001954 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1compare_performance(null, mod1,mod2,mod3,mod4, rank = TRUE)# Comparison of Model Performance Indices
Name | Model | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma | AIC weights
-------------------------------------------------------------------------------
mod4 | lmerMod | 0.686 | 0.223 | 0.596 | 6.266 | 8.104 | 0.777
mod3 | lmerMod | 0.679 | 0.232 | 0.582 | 6.169 | 8.320 | 0.180
mod1 | lmerMod | 0.442 | 0.025 | 0.428 | 8.428 | 10.160 | 0.017
null | lmerMod | 0.426 | 0.000 | 0.426 | 8.631 | 10.215 | 0.023
mod2 | lmerMod | 0.427 | 0.047 | 0.399 | 8.424 | 10.463 | 0.004
Name | AICc weights | BIC weights | Performance-Score
-----------------------------------------------------
mod4 | 0.851 | 0.805 | 99.02%
mod3 | 0.072 | 0.038 | 64.57%
mod1 | 0.026 | 0.038 | 7.71%
null | 0.047 | 0.117 | 5.74%
mod2 | 0.003 | 0.002 | 3.66%check_heteroscedasticity(mod4)Warning: Heteroscedasticity (non-constant error variance) detected (p < .001).check_normality(mod4)OK: residuals appear as normally distributed (p = 0.078).check_collinearity(mod4)# Check for MulticollinearityWarning: Argument `pattern` is deprecated. Please use `select` instead.
Low Correlation
Term VIF VIF 95% CI Increased SE Tolerance Tolerance 95% CI
Altitude_code 1.02 [1.00, 4.50e+05] 1.01 0.98 [0.00, 1.00]
Duration 1.02 [1.00, 4.50e+05] 1.01 0.98 [0.00, 1.00]Moving to a robust model
Total Distance
rlm_GD_TD <- rlmer(Distance ~ 1 + Duration + Altitude_code + (1| ID), Game2)
residuals <- residuals(rlm_GD_TD )
# Perform normality tests (e.g., Shapiro-Wilk test)
shapiro.test(residuals )
Shapiro-Wilk normality test
data: residuals
W = 0.95827, p-value = 0.128# Alternatively, you can create a QQ plot
qqnorm(residuals )
qqline(residuals )
summary(rlm_GD_TD )Robust linear mixed model fit by DAStau
Formula: Distance ~ 1 + Duration + Altitude_code + (1 | ID)
Data: Game2
Scaled residuals:
Min 1Q Median 3Q Max
-2.75977 -0.63898 0.01019 0.58416 1.38970
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 150834 388.4
Residual 433951 658.7
Number of obs: 42, groups: ID, 20
Fixed effects:
Estimate Std. Error t value
(Intercept) 2072.337 896.292 2.312
Duration 80.273 9.736 8.245
Altitude_codeAltitude -387.588 219.536 -1.765
Correlation of Fixed Effects:
(Intr) Duratn
Duration -0.979
Alttd_cdAlt -0.235 0.113
Robustness weights for the residuals:
37 weights are ~= 1. The remaining 5 ones are
11 22 23 27 32
0.997 0.487 0.922 0.997 0.943
Robustness weights for the random effects:
All 20 weights are ~= 1.
Rho functions used for fitting:
Residuals:
eff: smoothed Huber (k = 1.345, s = 10)
sig: smoothed Huber, Proposal 2 (k = 1.345, s = 10)
Random Effects, variance component 1 (ID):
eff: smoothed Huber (k = 1.345, s = 10)
vcp: smoothed Huber, Proposal 2 (k = 1.345, s = 10) emm_rlm_GD_TD <- emmeans(rlm_GD_TD, pairwise ~ Altitude_code)
emm_rlm_GD_TD $emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 9326 182 Inf 8969 9682
Altitude 8938 181 Inf 8584 9292
Confidence level used: 0.95
$contrasts
contrast estimate SE df z.ratio p.value
Sea - Altitude 388 220 Inf 1.765 0.0775confint(emm_rlm_GD_TD , level = 0.95)$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 9326 182 Inf 8969 9682
Altitude 8938 181 Inf 8584 9292
Confidence level used: 0.95
$contrasts
contrast estimate SE df asymp.LCL asymp.UCL
Sea - Altitude 388 220 Inf -42.7 818
Confidence level used: 0.95 HSR
rlm_GD_HSR <- rlmer(HighSpeedRunning ~ 1 + Duration + Altitude_code + (1| ID), Game2)
residuals <- residuals(rlm_GD_HSR )
# Perform normality tests (e.g., Shapiro-Wilk test)
shapiro.test(residuals )
Shapiro-Wilk normality test
data: residuals
W = 0.96125, p-value = 0.1637# Alternatively, you can create a QQ plot
qqnorm(residuals )
qqline(residuals )
summary(rlm_GD_HSR )Robust linear mixed model fit by DAStau
Formula: HighSpeedRunning ~ 1 + Duration + Altitude_code + (1 | ID)
Data: Game2
Scaled residuals:
Min 1Q Median 3Q Max
-1.2151 -0.4828 -0.1034 0.4249 2.1681
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 27224 165.0
Residual 16657 129.1
Number of obs: 42, groups: ID, 20
Fixed effects:
Estimate Std. Error t value
(Intercept) 137.662 211.821 0.650
Duration 4.486 2.332 1.923
Altitude_codeAltitude -56.699 45.379 -1.249
Correlation of Fixed Effects:
(Intr) Duratn
Duration -0.971
Alttd_cdAlt -0.164 0.054
Robustness weights for the residuals:
40 weights are ~= 1. The remaining 2 ones are
2 35
0.620 0.792
Robustness weights for the random effects:
17 weights are ~= 1. The remaining 3 ones are
11 12 19
0.580 0.971 0.945
Rho functions used for fitting:
Residuals:
eff: smoothed Huber (k = 1.345, s = 10)
sig: smoothed Huber, Proposal 2 (k = 1.345, s = 10)
Random Effects, variance component 1 (ID):
eff: smoothed Huber (k = 1.345, s = 10)
vcp: smoothed Huber, Proposal 2 (k = 1.345, s = 10) emm_rlm_GD_HSR <- emmeans(rlm_GD_HSR, pairwise ~ Altitude_code)
emm_rlm_GD_HSR$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 543 50.6 Inf 444 642
Altitude 486 50.0 Inf 388 584
Confidence level used: 0.95
$contrasts
contrast estimate SE df z.ratio p.value
Sea - Altitude 56.7 45.4 Inf 1.249 0.2115confint(emm_rlm_GD_HSR , level = 0.95)$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 543 50.6 Inf 444 642
Altitude 486 50.0 Inf 388 584
Confidence level used: 0.95
$contrasts
contrast estimate SE df asymp.LCL asymp.UCL
Sea - Altitude 56.7 45.4 Inf -32.2 146
Confidence level used: 0.95 sRPE Legs
rlm_GD_Leg <- rlmer(RPELeg ~ 1 + Duration + Altitude_code + (1| ID), Game2)
residuals <- residuals(rlm_GD_Leg )
# Perform normality tests (e.g., Shapiro-Wilk test)
shapiro.test(residuals )
Shapiro-Wilk normality test
data: residuals
W = 0.94101, p-value = 0.05462# Alternatively, you can create a QQ plot
qqnorm(residuals )
qqline(residuals )
summary(rlm_GD_Leg )Robust linear mixed model fit by DAStau
Formula: RPELeg ~ 1 + Duration + Altitude_code + (1 | ID)
Data: Game2
Scaled residuals:
Min 1Q Median 3Q Max
-2.56274 -0.27252 0.09252 0.47114 1.48035
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 92.45 9.615
Residual 56.53 7.519
Number of obs: 36, groups: ID, 18
Fixed effects:
Estimate Std. Error t value
(Intercept) 34.8051 14.1179 2.465
Duration 0.5339 0.1518 3.517
Altitude_codeAltitude -4.0301 2.9823 -1.351
Correlation of Fixed Effects:
(Intr) Duratn
Duration -0.974
Alttd_cdAlt -0.239 0.127
Robustness weights for the residuals:
32 weights are ~= 1. The remaining 4 ones are
6 7 19 33
0.525 0.896 0.997 0.987
Robustness weights for the random effects:
17 weights are ~= 1. The remaining one are
16
0.475
Rho functions used for fitting:
Residuals:
eff: smoothed Huber (k = 1.345, s = 10)
sig: smoothed Huber, Proposal 2 (k = 1.345, s = 10)
Random Effects, variance component 1 (ID):
eff: smoothed Huber (k = 1.345, s = 10)
vcp: smoothed Huber, Proposal 2 (k = 1.345, s = 10) emm_rlm_GD_Leg <- emmeans(rlm_GD_Leg, pairwise ~ Altitude_code)
emm_rlm_GD_Leg $emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 83.5 3.20 Inf 77.3 89.8
Altitude 79.5 3.08 Inf 73.5 85.5
Confidence level used: 0.95
$contrasts
contrast estimate SE df z.ratio p.value
Sea - Altitude 4.03 2.98 Inf 1.351 0.1766confint(emm_rlm_GD_Leg , level = 0.95)$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 83.5 3.20 Inf 77.3 89.8
Altitude 79.5 3.08 Inf 73.5 85.5
Confidence level used: 0.95
$contrasts
contrast estimate SE df asymp.LCL asymp.UCL
Sea - Altitude 4.03 2.98 Inf -1.82 9.88
Confidence level used: 0.95 sRPE BReath
rlm_GD_Breath <- rlmer(RPEBreathe ~ 1 + Duration + Altitude_code + (1| ID), Game2)
residuals <- residuals(rlm_GD_Breath )
# Perform normality tests (e.g., Shapiro-Wilk test)
shapiro.test(residuals )
Shapiro-Wilk normality test
data: residuals
W = 0.75789, p-value = 2.629e-06# Alternatively, you can create a QQ plot
qqnorm(residuals )
qqline(residuals )
summary(rlm_GD_Breath )Robust linear mixed model fit by DAStau
Formula: RPEBreathe ~ 1 + Duration + Altitude_code + (1 | ID)
Data: Game2
Scaled residuals:
Min 1Q Median 3Q Max
-4.6792 -0.4374 0.1299 0.5179 1.1947
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 53.37 7.305
Residual 55.83 7.472
Number of obs: 36, groups: ID, 18
Fixed effects:
Estimate Std. Error t value
(Intercept) 48.5922 13.0475 3.724
Duration 0.3695 0.1401 2.637
Altitude_codeAltitude 5.0635 2.8869 1.754
Correlation of Fixed Effects:
(Intr) Duratn
Duration -0.977
Alttd_cdAlt -0.253 0.137
Robustness weights for the residuals:
33 weights are ~= 1. The remaining 3 ones are
6 11 17
0.287 0.409 0.410
Robustness weights for the random effects:
17 weights are ~= 1. The remaining one are
16
0.934
Rho functions used for fitting:
Residuals:
eff: smoothed Huber (k = 1.345, s = 10)
sig: smoothed Huber, Proposal 2 (k = 1.345, s = 10)
Random Effects, variance component 1 (ID):
eff: smoothed Huber (k = 1.345, s = 10)
vcp: smoothed Huber, Proposal 2 (k = 1.345, s = 10) emm_rlm_GD_Breath <- emmeans(rlm_GD_Breath , pairwise ~ Altitude_code)
emm_rlm_GD_Breath $emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 82.3 2.76 Inf 76.9 87.7
Altitude 87.4 2.64 Inf 82.2 92.6
Confidence level used: 0.95
$contrasts
contrast estimate SE df z.ratio p.value
Sea - Altitude -5.06 2.89 Inf -1.754 0.0794confint(emm_rlm_GD_Breath , level = 0.95)$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 82.3 2.76 Inf 76.9 87.7
Altitude 87.4 2.64 Inf 82.2 92.6
Confidence level used: 0.95
$contrasts
contrast estimate SE df asymp.LCL asymp.UCL
Sea - Altitude -5.06 2.89 Inf -10.7 0.595
Confidence level used: 0.95 sRPE dif
rlm_GD_dif <- rlmer(RPEdif ~ 1 + Duration + Altitude_code + (1| ID), Game2)boundary (singular) fit: see help('isSingular')residuals <- residuals(rlm_GD_dif)
# Perform normality tests (e.g., Shapiro-Wilk test)
shapiro.test(residuals )
Shapiro-Wilk normality test
data: residuals
W = 0.88833, p-value = 0.001662# Alternatively, you can create a QQ plot
qqnorm(residuals )
qqline(residuals )
summary(rlm_GD_dif )Robust linear mixed model fit by DAStau
Formula: RPEdif ~ 1 + Duration + Altitude_code + (1 | ID)
Data: Game2
Scaled residuals:
Min 1Q Median 3Q Max
-3.0934 -0.5319 -0.1768 0.4018 3.9028
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.00 0.000
Residual 20.49 4.527
Number of obs: 36, groups: ID, 18
Fixed effects:
Estimate Std. Error t value
(Intercept) -7.06232 6.24827 -1.130
Duration 0.09890 0.06572 1.505
Altitude_codeAltitude -10.26481 1.57766 -6.506
Correlation of Fixed Effects:
(Intr) Duratn
Duration -0.985
Alttd_cdAlt -0.313 0.194
Robustness weights for the residuals:
27 weights are ~= 1. The remaining 9 ones are
6 11 18 24 26 29 30 32 36
0.481 0.345 0.822 0.911 0.435 0.722 0.408 0.802 0.815
Robustness weights for the random effects:
All 18 weights are ~= 1.
Rho functions used for fitting:
Residuals:
eff: smoothed Huber (k = 1.345, s = 10)
sig: smoothed Huber, Proposal 2 (k = 1.345, s = 10)
Random Effects, variance component 1 (ID):
eff: smoothed Huber (k = 1.345, s = 10)
vcp: smoothed Huber, Proposal 2 (k = 1.345, s = 10) emm_rlm_GD_dif <- emmeans(rlm_GD_dif, pairwise ~ Altitude_code)
emm_rlm_GD_dif$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 1.97 1.1 Inf -0.2 4.13
Altitude -8.30 1.1 Inf -10.5 -6.13
Confidence level used: 0.95
$contrasts
contrast estimate SE df z.ratio p.value
Sea - Altitude 10.3 1.58 Inf 6.506 <.0001confint(emm_rlm_GD_dif , level = 0.95)$emmeans
Altitude_code emmean SE df asymp.LCL asymp.UCL
Sea 1.97 1.1 Inf -0.2 4.13
Altitude -8.30 1.1 Inf -10.5 -6.13
Confidence level used: 0.95
$contrasts
contrast estimate SE df asymp.LCL asymp.UCL
Sea - Altitude 10.3 1.58 Inf 7.17 13.4
Confidence level used: 0.95