## package 'tidyverse' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'dplyr' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'dplyr'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\dplyr\libs\x64\dplyr.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\dplyr\libs\x64\dplyr.dll:
## Permission denied
## Warning: restored 'dplyr'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'detect' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'MASS' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'MASS'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\MASS\libs\x64\MASS.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\MASS\libs\x64\MASS.dll:
## Permission denied
## Warning: restored 'MASS'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'ggpubr' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'rstatix' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'ggcorrplot' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'lme4' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'lme4'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\lme4\libs\x64\lme4.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\lme4\libs\x64\lme4.dll:
## Permission denied
## Warning: restored 'lme4'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'FSA' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'factoextra' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'pheatmap' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'dendextend' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'vegan' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'vegan'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\vegan\libs\x64\vegan.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\vegan\libs\x64\vegan.dll:
## Permission denied
## Warning: restored 'vegan'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'caret' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'caret'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\caret\libs\x64\caret.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\caret\libs\x64\caret.dll:
## Permission denied
## Warning: restored 'caret'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'glmnet' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'glmnet'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\glmnet\libs\x64\glmnet.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\glmnet\libs\x64\glmnet.dll:
## Permission denied
## Warning: restored 'glmnet'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'pheatmap' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
## package 'mgcv' successfully unpacked and MD5 sums checked
## Warning: cannot remove prior installation of package 'mgcv'
## Warning in file.copy(savedcopy, lib, recursive = TRUE): problem copying
## C:\Users\aste490\AppData\Local\R\win-library\4.4\00LOCK\mgcv\libs\x64\mgcv.dll
## to C:\Users\aste490\AppData\Local\R\win-library\4.4\mgcv\libs\x64\mgcv.dll:
## Permission denied
## Warning: restored 'mgcv'
##
## The downloaded binary packages are in
## C:\Users\aste490\AppData\Local\Temp\Rtmpmyt8Zc\downloaded_packages
library(tidyverse)
## Warning: package 'tidyverse' was built under R version 4.4.3
## Warning: package 'ggplot2' was built under R version 4.4.2
## Warning: package 'tibble' was built under R version 4.4.2
## Warning: package 'tidyr' was built under R version 4.4.2
## Warning: package 'readr' was built under R version 4.4.3
## Warning: package 'purrr' was built under R version 4.4.2
## Warning: package 'dplyr' was built under R version 4.4.3
## Warning: package 'stringr' was built under R version 4.4.2
## Warning: package 'forcats' was built under R version 4.4.2
## Warning: package 'lubridate' was built under R version 4.4.2
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.4 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(dplyr)
library(ggplot2)
library(MASS)
## Warning: package 'MASS' was built under R version 4.4.3
##
## Attaching package: 'MASS'
##
## The following object is masked from 'package:dplyr':
##
## select
library(lme4)
## Warning: package 'lme4' was built under R version 4.4.3
## Loading required package: Matrix
##
## Attaching package: 'Matrix'
##
## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpack
library(car)
## Warning: package 'car' was built under R version 4.4.3
## Loading required package: carData
## Warning: package 'carData' was built under R version 4.4.2
##
## Attaching package: 'car'
##
## The following object is masked from 'package:dplyr':
##
## recode
##
## The following object is masked from 'package:purrr':
##
## some
library(ggpubr)
## Warning: package 'ggpubr' was built under R version 4.4.3
library(rstatix)
## Warning: package 'rstatix' was built under R version 4.4.3
##
## Attaching package: 'rstatix'
##
## The following object is masked from 'package:MASS':
##
## select
##
## The following object is masked from 'package:stats':
##
## filter
library(tidyr)
library(corrplot)
## Warning: package 'corrplot' was built under R version 4.4.2
## corrplot 0.95 loaded
library(ggcorrplot)
## Warning: package 'ggcorrplot' was built under R version 4.4.3
##
## Attaching package: 'ggcorrplot'
##
## The following object is masked from 'package:rstatix':
##
## cor_pmat
library(vegan)
## Warning: package 'vegan' was built under R version 4.4.3
## Loading required package: permute
## Warning: package 'permute' was built under R version 4.4.2
## Loading required package: lattice
library(caret)
## Warning: package 'caret' was built under R version 4.4.3
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:vegan':
##
## tolerance
##
## The following object is masked from 'package:purrr':
##
## lift
library(glmnet)
## Warning: package 'glmnet' was built under R version 4.4.3
## Loaded glmnet 4.1-8
library(pheatmap)
## Warning: package 'pheatmap' was built under R version 4.4.3
library(mgcv)
## Warning: package 'mgcv' was built under R version 4.4.3
## Loading required package: nlme
##
## Attaching package: 'nlme'
##
## The following object is masked from 'package:lme4':
##
## lmList
##
## The following object is masked from 'package:dplyr':
##
## collapse
##
## This is mgcv 1.9-1. For overview type 'help("mgcv-package")'.
# Load the data
setwd("C:/Users/aste490/OneDrive - Washington State Executive Branch Agencies/Documents/Data/2025 Project/This One")
fish_2023 <- read.csv("2023_fish.csv")
fish_2024 <- read.csv("2024_fish.csv")
HU_23 <- read.csv("HU_23.csv")
HU_24 <- read.csv("HU_24.csv")
canopy <- read.csv("canopy.csv")
peri_harvest <- read.csv("peri_prescription.csv")
macros <- read.csv("Macros.csv")
twoyr_peri <- read.csv("Biyearly_peri.csv")
benthic <- read.csv("Benthic.csv")
env <- read.csv("Flow_Year.csv")
all_biomass <- read.csv("all_biomass.csv")
#NOTE THAT BP IS OMITTED DUE TO CONSISTENT DRY 60M BOTH YEARS - NOT USABLE
#periphyton <- ug/100m^2
#identify structure, outliers, distribution, normality
str(fish_2023)
## 'data.frame': 1250 obs. of 5 variables:
## $ Basin : chr "AA1_REACH" "AA1_REACH" "AA1_REACH" "AA1_REACH" ...
## $ Pass : int 1 1 1 1 1 1 1 1 1 1 ...
## $ Species: chr "CTT" "CTT" "TRT" "CTT" ...
## $ Length : num 99 81 75 92 89 114 101 94 62 81 ...
## $ Weight : num 9.6 6.2 5.9 7.9 8.8 16.6 13.3 10.3 3.8 6.9 ...
str(fish_2024)
## 'data.frame': 1172 obs. of 5 variables:
## $ Basin : chr "AA1_REACH" "AA1_REACH" "AA1_REACH" "AA1_REACH" ...
## $ Pass : int 1 1 1 2 3 1 1 1 1 1 ...
## $ Species: chr "CTT" "CTT" "CTT" "CTT" ...
## $ Length : int 100 111 96 120 99 53 48 100 39 62 ...
## $ Weight : num 10.3 15.4 11.3 19.1 12.2 2.5 1.7 13.1 0.9 3.8 ...
str(HU_23)
## 'data.frame': 18 obs. of 5 variables:
## $ Basin : chr "AA1_REACH" "AA2_REACH" "AC_REACH" "AP_REACH" ...
## $ Reach_Length: num 80.7 119.7 111.2 86.7 98.5 ...
## $ Total_Area : num 9.35 38.79 91.28 14.7 69.63 ...
## $ Area_m2 : num 92.4 271.2 376.1 137.3 360.8 ...
## $ Yearly_Avg : num 12166 26491 39297 8264 54385 ...
str(HU_24)
## 'data.frame': 18 obs. of 5 variables:
## $ Basin : chr "AA1_REACH" "AA2_REACH" "AC_REACH" "AP_REACH" ...
## $ Reach_Length: num 74.2 112.2 105 94.5 103.7 ...
## $ Total_Area : num 9.88 44.4 86.61 50.08 178.98 ...
## $ Area_m2 : num 133 313 349 255 594 ...
## $ Yearly_Avg : num 11911 14564 30512 4994 12700 ...
colSums(is.na(fish_2023))
## Basin Pass Species Length Weight
## 0 0 0 0 0
colSums(is.na(fish_2024))
## Basin Pass Species Length Weight
## 0 0 0 0 0
colSums(is.na(HU_23))
## Basin Reach_Length Total_Area Area_m2 Yearly_Avg
## 0 0 0 0 0
colSums(is.na(HU_24))
## Basin Reach_Length Total_Area Area_m2 Yearly_Avg
## 0 0 0 0 0
# Boxplots for fish length and weight
ggplot(fish_2023, aes(y = Length)) + geom_boxplot() + ggtitle("Fish Lengths (2023)")
ggplot(fish_2023, aes(y = Weight)) + geom_boxplot() + ggtitle("Fish Weights (2023)")
ggplot(fish_2024, aes(y = Length)) + geom_boxplot() + ggtitle("Fish Lengths (2024)")
ggplot(fish_2024, aes(y = Weight)) + geom_boxplot() + ggtitle("Fish Weights (2024)")
ggplot(HU_23, aes(y = Yearly_Avg)) + geom_boxplot() + ggtitle("Chlorophyll_Yearly (2023)")
ggplot(HU_24, aes(y = Yearly_Avg)) + geom_boxplot() + ggtitle("Chlorophyll_Yearly (2024)")
#Results: Chlorophyll have a few outliers for both years except for 2024 yearly averages. Fish are all over the place, but are likely true outliers. Checking with z-scores.
# Calculate Z-scores
fish_2023 <- fish_2023 %>% mutate(
z_length = as.numeric(scale(Length)),
z_weight = as.numeric(scale(Weight))
)
fish_2024 <- fish_2024 %>% mutate(
z_length = as.numeric(scale(Length)),
z_weight = as.numeric(scale(Weight))
)
HU_23 <- HU_23 %>% mutate(z_chl_yr = scale(Yearly_Avg))
HU_24 <- HU_24 %>% mutate(z_chl_yr = scale(Yearly_Avg))
# Find extreme outliers (Z-score > 3 or < -3)
fish_2023 %>% filter(abs(z_length) > 3 | abs(z_weight) > 3)
## Basin Pass Species Length Weight z_length z_weight
## 1 AC_REACH 1 CTT 157 34.1 2.994786 3.6686916
## 2 AZ_REACH 1 CTT 156 36.9 2.959988 4.0335582
## 3 AZ_REACH 1 CTT 162 37.6 3.168773 4.1247749
## 4 AZ_REACH 1 CTT 144 31.0 2.542419 3.2647322
## 5 AZ_REACH 1 CTT 162 47.6 3.168773 5.4278697
## 6 AZ_REACH 1 CTT 190 69.2 4.143102 8.2425547
## 7 AZ_REACH 1 CTT 137 30.6 2.298837 3.2126084
## 8 AZ_REACH 2 CTT 159 43.7 3.064381 4.9196627
## 9 AZ_REACH 3 CTT 150 33.7 2.751203 3.6165678
## 10 BA2_REACH 1 CTT 168 37.9 3.377558 4.1638677
## 11 CA2_REACH 1 CTT 159 38.9 3.064381 4.2941772
## 12 CA2_REACH 1 CTT 194 80.0 4.282291 9.6498972
## 13 CA2_REACH 1 SCP 161 2.7 3.133975 -0.4230263
## 14 CA2_REACH 1 CTT 154 37.8 2.890393 4.1508367
## 15 CA2_REACH 4 CTT 139 30.8 2.368431 3.2386703
## 16 CP_REACH 1 CTT 147 35.4 2.646811 3.8380940
## 17 CP_REACH 1 CTT 143 30.1 2.507621 3.1474537
## 18 CP_REACH 1 CTT 160 47.3 3.099178 5.3887769
## 19 CP_REACH 1 CTT 168 46.1 3.377558 5.2324055
## 20 CP_REACH 1 CTT 144 35.4 2.542419 3.8380940
## 21 CP_REACH 1 CTT 145 34.8 2.577216 3.7599083
## 22 DC_REACH 1 CTT 158 36.0 3.029583 3.9162797
## 23 DZ_REACH 1 CTT 157 44.1 2.994786 4.9717865
## 24 DZ_REACH 1 CTT 184 71.8 3.934317 8.5813594
fish_2024 %>% filter(abs(z_length) > 3 | abs(z_weight) > 3)
## Basin Pass Species Length Weight z_length z_weight
## 1 AC_REACH 1 CTT 156 38.0 3.007443 3.380516
## 2 AZ_REACH 1 CTT 150 43.5 2.784609 3.980993
## 3 AZ_REACH 1 CTT 180 73.3 3.898782 7.234483
## 4 AZ_REACH 1 CTT 150 37.8 2.784609 3.358681
## 5 AZ_REACH 1 CTT 189 112.8 4.233034 11.546996
## 6 AZ_REACH 2 CTT 185 68.2 4.084477 6.677678
## 7 AZ_REACH 2 CTT 236 134.0 5.978572 13.861559
## 8 BA2_REACH 1 CTT 152 39.9 2.858887 3.587954
## 9 CA2_REACH 1 CTT 156 38.0 3.007443 3.380516
## 10 CA2_REACH 1 CTT 167 50.0 3.415973 4.690647
## 11 CC_REACH 1 CTT 159 37.2 3.118861 3.293174
## 12 CP_REACH 1 CTT 155 37.5 2.970304 3.325928
## 13 CP_REACH 1 CTT 150 37.3 2.784609 3.304092
## 14 CP_REACH 1 CTT 152 37.7 2.858887 3.347763
## 15 CZ2_REACH 1 CTT 162 49.9 3.230278 4.679729
## 16 DA2_REACH 1 CTT 148 37.3 2.710331 3.304092
## 17 DZ_REACH 1 CTT 175 72.6 3.713086 7.158059
HU_23 %>% filter(abs(z_chl_yr) > 3)
## Warning: Using one column matrices in `filter()` was deprecated in dplyr 1.1.0.
## ℹ Please use one dimensional logical vectors instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
## [1] Basin Reach_Length Total_Area Area_m2 Yearly_Avg
## [6] z_chl_yr
## <0 rows> (or 0-length row.names)
HU_24 %>% filter(abs(z_chl_yr) > 3)
## Basin Reach_Length Total_Area Area_m2 Yearly_Avg z_chl_yr
## 1 BZ_REACH 86.1 30.4 194.7 54641.64 3.014104
#All extreme outliers for fish data are true biological observations - no transformation or exclusion needed. No extreme outliers for chlorophyll.
#histograms for distribution
ggplot(fish_2023, aes(x = Length)) + geom_histogram(bins = 30, fill = "blue", alpha = 0.5) + ggtitle("Fish Lengths (2023)")
ggplot(fish_2023, aes(x = Weight)) + geom_histogram(bins = 30, fill = "red", alpha = 0.5) + ggtitle("Fish Weights (2023)")
ggplot(fish_2024, aes(x = Length)) + geom_histogram(bins = 30, fill = "lightblue", alpha = 0.5) + ggtitle("Fish Lengths (2024)")
ggplot(fish_2024, aes(x = Weight)) + geom_histogram(bins = 30, fill = "pink", alpha = 0.5) + ggtitle("Fish Weights (2024)")
ggplot(HU_23, aes(x = Yearly_Avg)) + geom_histogram(bins = 30, fill = "purple", alpha = 0.5) + ggtitle("Chlorophyll_Yearly (2023)")
ggplot(HU_24, aes(x = Yearly_Avg)) + geom_histogram(bins = 30, fill = "orange", alpha = 0.5) + ggtitle("Chlorophyll_Yearly (2024)")
#Results: Fish weights and lengths for both years are right skewed, understandably (majority juveniles). Chlorophyll yearly averages seem to be right skewed as well in 2023, indicating lower values were more common. 2024 chlorophyll data looks a little all over the place - which makes sense due to harvest treatments being completed and disturbances abound.
#Shapiro-wilk normality tests
shapiro.test(fish_2023$Length)
##
## Shapiro-Wilk normality test
##
## data: fish_2023$Length
## W = 0.91638, p-value < 2.2e-16
shapiro.test(fish_2023$Weight)
##
## Shapiro-Wilk normality test
##
## data: fish_2023$Weight
## W = 0.67106, p-value < 2.2e-16
shapiro.test(fish_2024$Length)
##
## Shapiro-Wilk normality test
##
## data: fish_2024$Length
## W = 0.92613, p-value < 2.2e-16
shapiro.test(fish_2024$Weight)
##
## Shapiro-Wilk normality test
##
## data: fish_2024$Weight
## W = 0.5922, p-value < 2.2e-16
shapiro.test(HU_23$Yearly_Avg)
##
## Shapiro-Wilk normality test
##
## data: HU_23$Yearly_Avg
## W = 0.78261, p-value = 0.0008716
shapiro.test(HU_24$Yearly_Avg)
##
## Shapiro-Wilk normality test
##
## data: HU_24$Yearly_Avg
## W = 0.86414, p-value = 0.01427
#Results:All fish data and 2023 chlorophyll data reject null hypothesis of normally distributed data. Normally distributed data is necessary for likely used tests in this analysis - ANOVA, linear correlation, linear regression, etc.
#Quantile plots, why not?
qqnorm(fish_2023$Length)
qqline(fish_2023$Length, col = "blue")
qqnorm(fish_2023$Weight)
qqline(fish_2023$Weight, col = "red")
qqnorm(fish_2024$Length)
qqline(fish_2024$Length, col = "lightblue")
qqnorm(fish_2024$Weight)
qqline(fish_2024$Weight, col = "pink")
qqnorm(HU_23$Yearly_Avg)
qqline(HU_23$Yearly_Avg, col = "purple")
qqnorm(HU_24$Yearly_Avg)
qqline(HU_24$Yearly_Avg, col = "orange" )
#Results: All fish data is right skewed, Chlorophyll 2023 data is also slightly right skewed with a few high outliers, 2024 chlorophyll data looks the closest to normal, although the summed values per reach per 100m shows some outliers at the high. Outliers, although not identified as extreme, are consistent. Az and Cp reaches in 2023 did have consistently high fluorescence values, so this is an ecologically sound representation.
# Log-transforming all chlorophyll columns for both HU_23 and HU_24 datasets
HU_23 <- HU_23 %>%
mutate(
Yearly_Avg = log(Yearly_Avg) # Log transform of Yearly_Avg
)
HU_24 <- HU_24 %>%
mutate(
Yearly_Avg = log(Yearly_Avg) # Log transform of Yearly_Avg
)
#check distribution
ggplot(HU_23, aes(x = Yearly_Avg)) + geom_histogram(bins = 30, fill = "purple", alpha = 0.5) + ggtitle("Chlorophyll_Yearly_log (2023)")
ggplot(HU_24, aes(x = Yearly_Avg)) + geom_histogram(bins = 30, fill = "orange", alpha = 0.5) + ggtitle("Chlorophyll_Yearly_log (2024)")
shapiro.test(HU_23$Yearly_Avg)
##
## Shapiro-Wilk normality test
##
## data: HU_23$Yearly_Avg
## W = 0.94407, p-value = 0.3397
shapiro.test(HU_24$Yearly_Avg)
##
## Shapiro-Wilk normality test
##
## data: HU_24$Yearly_Avg
## W = 0.98096, p-value = 0.9594
#scale periphyton data for comparability.
HU_23$Yearly_Avg <- scale(HU_23$Yearly_Avg)
HU_24$Yearly_Avg <- scale(HU_24$Yearly_Avg)
# Function to calculate length-weight coefficients
coefficients23 <- function(df) {
# Ensure length and weight are numeric before applying log()
df$log_lengths <- log(as.numeric(df$Length))
df$log_weights <- log(as.numeric(df$Weight))
# Fit linear regression
model <- lm(log_weights ~ log_lengths, data = df)
log_a <- coef(model)[1] # Intercept
b <- coef(model)[2] # Slope
# Transform intercept back from log scale
a <- exp(log_a)
# Return both coefficients and transformed data
return(list(coefficients = data.frame(a = a, b = b), transformed_data = df))
}
# Apply the function to the dataset
result <- coefficients23(fish_2023)
# Calculate coefficients for each species within each basin
coef_sp_reach23 <- fish_2023 %>%
group_by(Basin, Species) %>%
summarise(
a = coefficients23(cur_data())$coefficients$a,
b = coefficients23(cur_data())$coefficients$b,
.groups = "drop"
)
## Warning: There was 1 warning in `summarise()`.
## ℹ In argument: `a = coefficients23(cur_data())$coefficients$a`.
## ℹ In group 1: `Basin = "AA1_REACH"` and `Species = "CTT"`.
## Caused by warning:
## ! `cur_data()` was deprecated in dplyr 1.1.0.
## ℹ Please use `pick()` instead.
# Ensure length and weight are numeric before performing calculations
fish_2023_wbiomass <- fish_2023 %>%
mutate(
Length = as.numeric(Length), # Convert length to numeric
Weight = as.numeric(Weight) # Convert weight to numeric
) %>%
left_join(coef_sp_reach23, by = c("Basin", "Species")) %>% # Add coefficients (a and b)
mutate(
predicted_biomass = a * (Length^b) # Predict weight
)
# Summarize fish biomass by Basin and Species FIRST
avg_biomass23 <- fish_2023_wbiomass %>%
group_by(Basin, Species) %>%
summarise(
avg_biomass = mean(predicted_biomass, na.rm = TRUE), # Average biomass
total_biomass = sum(predicted_biomass, na.rm = TRUE), # Total biomass
.groups = "drop"
)
# Merge summarized biomass data with HU_24
avg_biomass23 <- avg_biomass23 %>%
left_join(HU_23, by = "Basin") %>% # Left join ensures no duplicate expansion
mutate(biomass_reach = (total_biomass / Area_m2)) # Biomass per m^2
# Print to verify unique rows
print(avg_biomass23)
## # A tibble: 49 × 10
## Basin Species avg_biomass total_biomass Reach_Length Total_Area Area_m2
## <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AA1_REACH CTT 9.36 122. 80.7 9.35 92.4
## 2 AA1_REACH TRT 4.85 9.70 80.7 9.35 92.4
## 3 AA2_REACH CTT 10.1 294. 120. 38.8 271.
## 4 AA2_REACH SCP 2.26 15.8 120. 38.8 271.
## 5 AA2_REACH TRT 2.39 115. 120. 38.8 271.
## 6 AC_REACH CTT 13.4 576. 111. 91.3 376.
## 7 AC_REACH TRT 0.927 25.0 111. 91.3 376.
## 8 AP_REACH CTT 7.58 106. 86.7 14.7 137.
## 9 AP_REACH TRT 3.38 16.9 86.7 14.7 137.
## 10 AZ_REACH CTT 22.1 706. 98.5 69.6 361.
## # ℹ 39 more rows
## # ℹ 3 more variables: Yearly_Avg <dbl[,1]>, z_chl_yr <dbl[,1]>,
## # biomass_reach <dbl>
#Repeat with 2024 data
# Function to calculate length-weight coefficients
coefficients24 <- function(df) {
# Ensure length and weight are numeric before applying log()
df$log_lengths <- log(as.numeric(df$Length))
df$log_weights <- log(as.numeric(df$Weight))
# Fit linear regression
model <- lm(log_weights ~ log_lengths, data = df)
log_a <- coef(model)[1] # Intercept
b <- coef(model)[2] # Slope
# Transform intercept back from log scale
a <- exp(log_a)
# Return both coefficients and transformed data
return(list(coefficients = data.frame(a = a, b = b), transformed_data = df))
}
# Apply the function to the dataset
result <- coefficients24(fish_2024)
# Calculate coefficients for each species within each basin
coef_sp_reach24 <- fish_2024 %>%
group_by(Basin, Species) %>%
summarise(
a = coefficients24(cur_data())$coefficients$a,
b = coefficients24(cur_data())$coefficients$b,
.groups = "drop"
)
# Ensure length and weight are numeric before performing calculations
fish_2024_wbiomass <- fish_2024 %>%
mutate(
Length = as.numeric(Length), # Convert length to numeric
Weight = as.numeric(Weight) # Convert weight to numeric
) %>%
left_join(coef_sp_reach24, by = c("Basin", "Species")) %>% # Add coefficients (a and b)
mutate(
predicted_biomass = a * (Length^b) # Predict weight
)
# Summarize fish biomass by Basin and Species FIRST
avg_biomass24 <- fish_2024_wbiomass %>%
group_by(Basin, Species) %>%
summarise(
avg_biomass = mean(predicted_biomass, na.rm = TRUE), # Average biomass
total_biomass = sum(predicted_biomass, na.rm = TRUE), # Total biomass
.groups = "drop"
)
# Merge summarized biomass data with HU_24
avg_biomass24 <- avg_biomass24 %>%
left_join(HU_24, by = "Basin") %>% # Left join ensures no duplicate expansion
mutate(biomass_reach = (total_biomass / Area_m2)) # Biomass per m2
# Print to verify unique rowsprint(avg_biomass24)
# Perform Chi-square test to compare catchability between first and second pass
pass_data23 <- fish_2023 %>%
filter(Pass %in% c(1, 2, 3, 4)) %>%
group_by(Basin, Species) %>%
summarise(catch1 = sum(Pass == 1), catch2 = sum(Pass == 2), catch3 = sum(Pass == 3), catch4 = sum(Pass == 4), .groups = "drop")
# Perform the chi-square test for each combination of Basin, Species, and habitat unit
pass_comparison23 <- pass_data23 %>%
mutate(chi_square_test = mapply(function(x, y) chisq.test(c(x, y))$p.value, catch1, catch2))
## Warning: There were 20 warnings in `mutate()`.
## The first warning was:
## ℹ In argument: `chi_square_test = mapply(function(x, y) chisq.test(c(x,
## y))$p.value, catch1, catch2)`.
## Caused by warning in `chisq.test()`:
## ! Chi-squared approximation may be incorrect
## ℹ Run `dplyr::last_dplyr_warnings()` to see the 19 remaining warnings.
# View the results with p-values
print(pass_comparison23)
## # A tibble: 49 × 7
## Basin Species catch1 catch2 catch3 catch4 chi_square_test
## <chr> <chr> <int> <int> <int> <int> <dbl>
## 1 AA1_REACH CTT 12 1 0 0 2.28e- 3
## 2 AA1_REACH TRT 2 0 0 0 1.57e- 1
## 3 AA2_REACH CTT 24 4 1 0 1.57e- 4
## 4 AA2_REACH SCP 5 2 0 0 2.57e- 1
## 5 AA2_REACH TRT 36 11 1 0 2.66e- 4
## 6 AC_REACH CTT 42 0 1 0 9.13e-11
## 7 AC_REACH TRT 22 2 2 1 4.46e- 5
## 8 AP_REACH CTT 9 3 2 0 8.33e- 2
## 9 AP_REACH TRT 3 2 0 0 6.55e- 1
## 10 AZ_REACH CTT 22 8 2 0 1.06e- 2
## # ℹ 39 more rows
#Repeat with 2024 data
pass_data24 <- fish_2024 %>%
filter(Pass %in% c(1, 2, 3)) %>%
group_by(Basin, Species) %>%
summarise(catch1 = sum(Pass == 1), catch2 = sum(Pass == 2), catch3 = sum(Pass == 3), .groups = "drop")
pass_comparison24 <- pass_data24 %>%
mutate(chi_square_test = mapply(function(x, y) chisq.test(c(x, y))$p.value, catch1, catch2))
## Warning: There were 21 warnings in `mutate()`.
## The first warning was:
## ℹ In argument: `chi_square_test = mapply(function(x, y) chisq.test(c(x,
## y))$p.value, catch1, catch2)`.
## Caused by warning in `chisq.test()`:
## ! Chi-squared approximation may be incorrect
## ℹ Run `dplyr::last_dplyr_warnings()` to see the 20 remaining warnings.
print(pass_comparison24)
## # A tibble: 50 × 6
## Basin Species catch1 catch2 catch3 chi_square_test
## <chr> <chr> <int> <int> <int> <dbl>
## 1 AA1_REACH CTT 3 1 1 3.17e- 1
## 2 AA2_REACH CTT 16 3 1 2.86e- 3
## 3 AA2_REACH SCP 5 0 0 2.53e- 2
## 4 AA2_REACH TRT 77 7 3 2.21e-14
## 5 AC_REACH CTT 59 14 3 1.39e- 7
## 6 AC_REACH TRT 43 9 0 2.42e- 6
## 7 AP_REACH CTT 7 3 0 2.06e- 1
## 8 AP_REACH TRT 7 2 0 9.56e- 2
## 9 AZ_REACH CTT 21 8 1 1.58e- 2
## 10 AZ_REACH SCP 17 8 5 7.19e- 2
## # ℹ 40 more rows
#using FSA package, see https://derekogle.com/fishR/examples/oldFishRVignettes/Depletion.pdf
library(FSA)
## Warning: package 'FSA' was built under R version 4.4.3
## Registered S3 methods overwritten by 'FSA':
## method from
## confint.boot car
## hist.boot car
## ## FSA v0.9.6. See citation('FSA') if used in publication.
## ## Run fishR() for related website and fishR('IFAR') for related book.
##
## Attaching package: 'FSA'
## The following object is masked from 'package:car':
##
## bootCase
library(detect)
## Warning: package 'detect' was built under R version 4.4.3
## Loading required package: Formula
## Loading required package: stats4
## Loading required package: pbapply
## Warning: package 'pbapply' was built under R version 4.4.2
## detect 0.4-6 2023-03-08
#create standalone reach length dataframe for both years
# 2023 Reach area DataFrame
Reach_area_23 <- data.frame(
Basin = c("AA1_REACH", "AA2_REACH", "AC_REACH", "AP_REACH", "AZ_REACH",
"BA1_REACH", "BA2_REACH", "BC_REACH", "BZ_REACH", "CA1_REACH",
"CA2_REACH", "CP_REACH", "CZ2_REACH", "DA1_REACH", "DA2_REACH",
"DC_REACH", "DP_REACH", "DZ_REACH"),
Reach_area = c(92.37, 271.25, 376.09, 137.32, 360.84, 158.78, 338.36, 98.65, 157.84, 57.04, 580.36, 277.7, 266.65, 56.83, 263.67, 314.49, 119.66, 323.37
)
)
# 2024 Reach area DataFrame
Reach_area_24 <- data.frame(
Basin = c("AA1_REACH", "AA2_REACH", "AC_REACH", "AP_REACH", "AZ_REACH",
"BA1_REACH", "BA2_REACH", "BC_REACH", "BZ_REACH", "CA1_REACH",
"CA2_REACH", "CP_REACH", "CZ2_REACH", "DA1_REACH", "DA2_REACH",
"DC_REACH", "DP_REACH", "DZ_REACH"),
Reach_area = c(132.53, 312.94, 348.72, 255.27, 593.88, 178.08, 414.52, 153.61, 86.91, 358.36, 399.94, 247.11, 499.68, 101.21, 353.91, 327.62, 111.36, 410.58
)
)
# View to verify
print(Reach_area_23)
## Basin Reach_area
## 1 AA1_REACH 92.37
## 2 AA2_REACH 271.25
## 3 AC_REACH 376.09
## 4 AP_REACH 137.32
## 5 AZ_REACH 360.84
## 6 BA1_REACH 158.78
## 7 BA2_REACH 338.36
## 8 BC_REACH 98.65
## 9 BZ_REACH 157.84
## 10 CA1_REACH 57.04
## 11 CA2_REACH 580.36
## 12 CP_REACH 277.70
## 13 CZ2_REACH 266.65
## 14 DA1_REACH 56.83
## 15 DA2_REACH 263.67
## 16 DC_REACH 314.49
## 17 DP_REACH 119.66
## 18 DZ_REACH 323.37
print(Reach_area_24)
## Basin Reach_area
## 1 AA1_REACH 132.53
## 2 AA2_REACH 312.94
## 3 AC_REACH 348.72
## 4 AP_REACH 255.27
## 5 AZ_REACH 593.88
## 6 BA1_REACH 178.08
## 7 BA2_REACH 414.52
## 8 BC_REACH 153.61
## 9 BZ_REACH 86.91
## 10 CA1_REACH 358.36
## 11 CA2_REACH 399.94
## 12 CP_REACH 247.11
## 13 CZ2_REACH 499.68
## 14 DA1_REACH 101.21
## 15 DA2_REACH 353.91
## 16 DC_REACH 327.62
## 17 DP_REACH 111.36
## 18 DZ_REACH 410.58
#merge with pass catches
depletion_2023 <- merge(Reach_area_23, pass_data23, by = "Basin")
#Summarize captures
capture_summary23 <- depletion_2023 %>%
group_by(Basin, Species, Reach_area) %>%
summarise(catch1 = sum(catch1), catch2 = sum(catch2), catch3 = sum(catch3), catch4 = sum(catch4), .groups = "drop")
#Estimate population by removal() method
population_estimates23 <- capture_summary23 %>% rowwise() %>%
mutate(removal_result = list(removal(c(catch1, catch2, catch3, catch4), type = "CarleStrub"))) %>%
ungroup()
# Create a data frame to store the 'No' estimate from each removal result
population_estimates23 <- data.frame(
estimated_population = sapply(population_estimates23$removal_result, function(x) x$est["No"]),
No_se = sapply(population_estimates23$removal_result, function(x) x$est["No.se"]),
No_LCI = sapply(population_estimates23$removal_result, function(x) x$est["No.LCI"]),
No_UCI = sapply(population_estimates23$removal_result, function(x) x$est["No.UCI"]),
p = sapply(population_estimates23$removal_result, function(x) x$est["p"]),
p_se = sapply(population_estimates23$removal_result, function(x) x$est["p.se"]),
p_LCI = sapply(population_estimates23$removal_result, function(x) x$est["p.LCI"]),
p_UCI = sapply(population_estimates23$removal_result, function(x) x$est["p.UCI"])
)
# Combine the original data with the new 'estimated_population' column
population_estimates23_summary <- cbind(capture_summary23, population_estimates23) %>%
mutate(
abundance_per_100m = (estimated_population / Reach_area) # Normalize to m2
)
print(population_estimates23_summary)
## Basin Species Reach_area catch1 catch2 catch3 catch4
## 1 AA1_REACH CTT 92.37 12 1 0 0
## 2 AA1_REACH TRT 92.37 2 0 0 0
## 3 AA2_REACH CTT 271.25 24 4 1 0
## 4 AA2_REACH SCP 271.25 5 2 0 0
## 5 AA2_REACH TRT 271.25 36 11 1 0
## 6 AC_REACH CTT 376.09 42 0 1 0
## 7 AC_REACH TRT 376.09 22 2 2 1
## 8 AP_REACH CTT 137.32 9 3 2 0
## 9 AP_REACH TRT 137.32 3 2 0 0
## 10 AZ_REACH CTT 360.84 22 8 2 0
## 11 AZ_REACH SCP 360.84 6 0 0 0
## 12 AZ_REACH TRT 360.84 12 3 0 0
## 13 BA1_REACH CTT 158.78 14 2 1 0
## 14 BA1_REACH TRT 158.78 15 4 1 0
## 15 BA2_REACH COH 338.36 112 26 7 0
## 16 BA2_REACH CTT 338.36 24 5 0 0
## 17 BA2_REACH SCP 338.36 9 0 0 0
## 18 BA2_REACH TRT 338.36 6 0 0 0
## 19 BC_REACH CTT 98.65 5 0 0 0
## 20 BC_REACH TRT 98.65 101 14 9 2
## 21 BZ_REACH CTT 157.84 12 2 0 0
## 22 BZ_REACH SCP 157.84 3 0 0 0
## 23 BZ_REACH TRT 157.84 12 7 0 0
## 24 CA2_REACH CTT 580.36 37 6 3 4
## 25 CA2_REACH SCP 580.36 6 0 1 0
## 26 CA2_REACH TRT 580.36 25 5 5 2
## 27 CP_REACH CTT 277.70 71 7 2 0
## 28 CP_REACH SCP 277.70 7 0 0 0
## 29 CP_REACH TRT 277.70 23 5 2 0
## 30 CZ2_REACH CTT 266.65 20 9 2 0
## 31 CZ2_REACH TRT 266.65 17 3 1 0
## 32 DA2_REACH COH 263.67 73 15 6 0
## 33 DA2_REACH CTT 263.67 8 2 1 0
## 34 DA2_REACH RBT 263.67 1 0 0 0
## 35 DA2_REACH SCP 263.67 5 0 0 0
## 36 DA2_REACH TRT 263.67 9 1 1 0
## 37 DC_REACH COH 314.49 44 5 5 3
## 38 DC_REACH CTT 314.49 3 1 0 0
## 39 DC_REACH SCP 314.49 5 0 0 0
## 40 DC_REACH TRT 314.49 6 2 0 0
## 41 DP_REACH CTT 119.66 2 1 0 0
## 42 DP_REACH SCP 119.66 5 0 0 0
## 43 DP_REACH TRT 119.66 8 2 0 0
## 44 DZ_REACH COH 323.37 38 19 7 7
## 45 DZ_REACH CTT 323.37 3 3 2 0
## 46 DZ_REACH RBT 323.37 1 0 0 0
## 47 DZ_REACH SCP 323.37 7 0 0 0
## estimated_population No_se No_LCI No_UCI p p_se
## 1 13 0.01844233 12.963854 13.036146 0.9285714 0.06900488
## 2 2 0.00000000 2.000000 2.000000 1.0000000 NaN
## 3 29 0.16290596 28.680710 29.319290 0.8285714 0.06557588
## 4 7 0.13840904 6.728723 7.271277 0.7777778 0.14680496
## 5 48 0.33134565 47.350574 48.649426 0.7868852 0.05521171
## 6 43 0.01296130 42.974596 43.025404 0.9555556 0.03074043
## 7 27 0.35123667 26.311589 27.688411 0.7500000 0.07805259
## 8 14 0.48961825 13.040366 14.959634 0.6666667 0.12114910
## 9 5 0.20386998 4.600422 5.399578 0.7142857 0.19070290
## 10 32 0.46412871 31.090324 32.909676 0.7272727 0.07406157
## 11 6 0.00000000 6.000000 6.000000 1.0000000 NaN
## 12 15 0.11050874 14.783407 15.216593 0.8333333 0.09023001
## 13 17 0.15544273 16.695338 17.304662 0.8095238 0.08904101
## 14 20 0.25381094 19.502540 20.497460 0.7692308 0.08805827
## 15 145 0.59396820 143.835844 146.164156 0.7837838 0.03193442
## 16 29 0.11870851 28.767336 29.232664 0.8529412 0.06191078
## 17 9 0.00000000 9.000000 9.000000 1.0000000 NaN
## 18 6 0.00000000 6.000000 6.000000 1.0000000 NaN
## 19 5 0.00000000 5.000000 5.000000 1.0000000 NaN
## 20 126 0.64268573 124.740359 127.259641 0.7682927 0.03513542
## 21 14 0.05918317 13.884003 14.115997 0.8750000 0.08369765
## 22 3 0.00000000 3.000000 3.000000 1.0000000 NaN
## 23 19 0.34740979 18.319089 19.680911 0.7307692 0.09564927
## 24 50 0.86641003 48.301868 51.698132 0.6756757 0.06339025
## 25 7 0.13840904 6.728723 7.271277 0.7777778 0.14680496
## 26 37 0.97539631 35.088258 38.911742 0.6379310 0.07719091
## 27 80 0.13215734 79.740976 80.259024 0.8791209 0.03455596
## 28 7 0.00000000 7.000000 7.000000 1.0000000 NaN
## 29 30 0.31085365 29.390738 30.609262 0.7692308 0.07189927
## 30 31 0.54822575 29.925497 32.074503 0.7045455 0.07758375
## 31 21 0.17627764 20.654502 21.345498 0.8076923 0.08039497
## 32 94 0.51172108 92.997045 95.002955 0.7768595 0.04012179
## 33 11 0.25873003 10.492898 11.507102 0.7333333 0.12525714
## 34 1 0.00000000 1.000000 1.000000 1.0000000 NaN
## 35 5 0.00000000 5.000000 5.000000 1.0000000 NaN
## 36 11 0.16048788 10.685450 11.314550 0.7857143 0.11556448
## 37 57 0.74828670 55.533385 58.466615 0.7037037 0.05727877
## 38 4 0.08357177 3.836202 4.163798 0.8000000 0.18687216
## 39 5 0.00000000 5.000000 5.000000 1.0000000 NaN
## 40 8 0.11818833 7.768355 8.231645 0.8000000 0.13213857
## 41 3 0.11707889 2.770530 3.229470 0.7500000 0.23415778
## 42 5 0.00000000 5.000000 5.000000 1.0000000 NaN
## 43 10 0.09023001 9.823152 10.176848 0.8333333 0.11050874
## 44 75 3.33044284 68.472452 81.527548 0.5035461 0.06392365
## 45 8 0.90134784 6.233391 9.766609 0.5333333 0.18849116
## 46 1 0.00000000 1.000000 1.000000 1.0000000 NaN
## 47 7 0.00000000 7.000000 7.000000 1.0000000 NaN
## p_LCI p_UCI abundance_per_100m
## 1 0.7933243 1.0638185 0.140738335
## 2 NaN NaN 0.021652052
## 3 0.7000451 0.9570978 0.106912442
## 4 0.4900453 1.0655102 0.025806452
## 5 0.6786723 0.8950982 0.176958525
## 6 0.8953054 1.0158057 0.114334335
## 7 0.5970197 0.9029803 0.071791327
## 8 0.4292188 0.9041145 0.101951646
## 9 0.3405149 1.0880565 0.036411302
## 10 0.5821147 0.8724307 0.088681964
## 11 NaN NaN 0.016627868
## 12 0.6564858 1.0101809 0.041569671
## 13 0.6350066 0.9840410 0.107066381
## 14 0.5966397 0.9418218 0.125960448
## 15 0.7211935 0.8463741 0.428537652
## 16 0.7315983 0.9742841 0.085707530
## 17 NaN NaN 0.026598889
## 18 NaN NaN 0.017732593
## 19 NaN NaN 0.050684237
## 20 0.6994285 0.8371568 1.277242777
## 21 0.7109556 1.0390444 0.088697415
## 22 NaN NaN 0.019006589
## 23 0.5433001 0.9182384 0.120375063
## 24 0.5514331 0.7999183 0.086153422
## 25 0.4900453 1.0655102 0.012061479
## 26 0.4866396 0.7892224 0.063753532
## 27 0.8113924 0.9468493 0.288080663
## 28 NaN NaN 0.025207058
## 29 0.6283108 0.9101508 0.108030248
## 30 0.5524841 0.8566068 0.116257266
## 31 0.6501211 0.9652636 0.078754922
## 32 0.6982222 0.8554968 0.356506239
## 33 0.4878339 0.9788328 0.041718815
## 34 NaN NaN 0.003792620
## 35 NaN NaN 0.018963098
## 36 0.5592121 1.0122165 0.041718815
## 37 0.5914394 0.8159680 0.181245827
## 38 0.4337373 1.1662627 0.012719005
## 39 NaN NaN 0.015898757
## 40 0.5410132 1.0589868 0.025438011
## 41 0.2910592 1.2089408 0.025071035
## 42 NaN NaN 0.041785058
## 43 0.6167402 1.0499265 0.083570115
## 44 0.3782580 0.6288342 0.231932461
## 45 0.1638974 0.9027692 0.024739463
## 46 NaN NaN 0.003092433
## 47 NaN NaN 0.021647030
#Repeat with 2024 data
#merge with pass catches
depletion_2024 <- merge(Reach_area_24, pass_data24, by = "Basin")
#Summarize captures
capture_summary24 <- depletion_2024 %>%
group_by(Basin, Species, Reach_area) %>%
summarise(catch1 = sum(catch1), catch2 = sum(catch2), catch3 = sum(catch3), .groups = "drop")
#Estimate population by removal() method
population_estimates24 <- capture_summary24 %>% rowwise() %>%
mutate(removal_result = list(removal(c(catch1, catch2, catch3), type = "CarleStrub"))) %>%
ungroup()
# Create a data frame to store the 'No' estimate from each removal result
population_estimates24 <- data.frame(
estimated_population = sapply(population_estimates24$removal_result, function(x) x$est["No"]),
No_se = sapply(population_estimates24$removal_result, function(x) x$est["No.se"]),
No_LCI = sapply(population_estimates24$removal_result, function(x) x$est["No.LCI"]),
No_UCI = sapply(population_estimates24$removal_result, function(x) x$est["No.UCI"]),
p = sapply(population_estimates24$removal_result, function(x) x$est["p"]),
p_se = sapply(population_estimates24$removal_result, function(x) x$est["p.se"]),
p_LCI = sapply(population_estimates24$removal_result, function(x) x$est["p.LCI"]),
p_UCI = sapply(population_estimates24$removal_result, function(x) x$est["p.UCI"])
)
# Combine the original data with the new 'estimated_population' column
population_estimates24_summary <- cbind(capture_summary24, population_estimates24) %>%
mutate(
abundance_per_100m = (estimated_population / Reach_area) # m2
)
print(population_estimates24_summary)
## Basin Species Reach_area catch1 catch2 catch3 estimated_population
## 1 AA1_REACH CTT 132.53 3 1 1 5
## 2 AA2_REACH CTT 312.94 16 3 1 20
## 3 AA2_REACH SCP 312.94 5 0 0 5
## 4 AA2_REACH TRT 312.94 77 7 3 87
## 5 AC_REACH CTT 348.72 59 14 3 76
## 6 AC_REACH TRT 348.72 43 9 0 52
## 7 AP_REACH CTT 255.27 7 3 0 10
## 8 AP_REACH TRT 255.27 7 2 0 9
## 9 AZ_REACH CTT 593.88 21 8 1 30
## 10 AZ_REACH SCP 593.88 17 8 5 33
## 11 AZ_REACH TRT 593.88 29 14 0 43
## 12 BA1_REACH CTT 178.08 14 2 0 16
## 13 BA1_REACH TRT 178.08 14 0 0 14
## 14 BA2_REACH COH 414.52 83 24 7 116
## 15 BA2_REACH CTT 414.52 24 4 1 29
## 16 BA2_REACH RBT 414.52 1 0 1 2
## 17 BA2_REACH SCP 414.52 9 0 0 9
## 18 BA2_REACH TRT 414.52 16 4 0 20
## 19 BC_REACH CTT 153.61 5 0 0 5
## 20 BC_REACH TRT 153.61 123 18 14 157
## 21 BZ_REACH CTT 86.91 18 2 0 20
## 22 BZ_REACH SCP 86.91 6 0 0 6
## 23 BZ_REACH TRT 86.91 20 5 1 26
## 24 CA2_REACH CTT 399.94 31 2 2 35
## 25 CA2_REACH SCP 399.94 6 1 0 7
## 26 CA2_REACH TRT 399.94 37 7 3 47
## 27 CP_REACH CTT 247.11 25 4 3 32
## 28 CP_REACH SCP 247.11 8 0 0 8
## 29 CP_REACH TRT 247.11 10 3 1 14
## 30 CZ2_REACH CTT 499.68 26 2 0 28
## 31 CZ2_REACH TRT 499.68 16 1 1 18
## 32 DA2_REACH COH 353.91 30 9 2 41
## 33 DA2_REACH CTT 353.91 4 4 0 8
## 34 DA2_REACH RBT 353.91 1 1 0 2
## 35 DA2_REACH SCP 353.91 5 0 1 6
## 36 DA2_REACH TRT 353.91 3 2 1 6
## 37 DC_REACH COH 327.62 17 3 0 20
## 38 DC_REACH CTT 327.62 10 1 0 11
## 39 DC_REACH RBT 327.62 1 0 0 1
## 40 DC_REACH SCP 327.62 7 0 0 7
## 41 DC_REACH TRT 327.62 7 1 1 9
## 42 DP_REACH CTT 111.36 2 0 0 2
## 43 DP_REACH SCP 111.36 11 0 0 11
## 44 DP_REACH TRT 111.36 2 0 1 3
## 45 DZ_REACH COH 410.58 18 4 0 22
## 46 DZ_REACH CTT 410.58 4 0 0 4
## 47 DZ_REACH SCP 410.58 5 0 0 5
## 48 DZ_REACH TRT 410.58 5 2 1 8
## No_se No_LCI No_UCI p p_se p_LCI
## 1 0.78732564 3.45687011 6.543130 0.6250000 0.26244188 0.11062337
## 2 0.45890939 19.10055413 20.899446 0.8000000 0.09178188 0.62011083
## 3 0.00000000 5.00000000 5.000000 1.0000000 NaN NaN
## 4 0.46541004 86.08781309 87.912187 0.8700000 0.03580077 0.79983178
## 5 0.96113897 74.11620223 77.883798 0.7916667 0.04805695 0.69747678
## 6 0.44223360 51.13323808 52.866762 0.8524590 0.04913707 0.75615214
## 7 0.41883071 9.17910690 10.820893 0.7692308 0.13961024 0.49559974
## 8 0.26107864 8.48829526 9.511705 0.8181818 0.13053932 0.56232945
## 9 0.84087497 28.35191535 31.648085 0.7500000 0.08408750 0.58519154
## 10 3.61894328 25.90700151 40.092998 0.5263158 0.12184994 0.28749429
## 11 0.97403420 41.09092805 44.909072 0.7543860 0.06957387 0.61802368
## 12 0.15524038 15.69573444 16.304266 0.8888889 0.07762019 0.73675611
## 13 0.00000000 14.00000000 14.000000 1.0000000 NaN NaN
## 14 2.03150305 112.01832718 119.981673 0.7215190 0.04537448 0.63258664
## 15 0.42405500 28.16886747 29.831133 0.8285714 0.07067583 0.69004934
## 16 1.03774904 -0.03395075 4.033951 0.5000000 0.51887452 -0.51697537
## 17 0.00000000 9.00000000 9.000000 1.0000000 NaN NaN
## 18 0.33583759 19.34177042 20.658230 0.8333333 0.08395940 0.66877594
## 19 0.00000000 5.00000000 5.000000 1.0000000 NaN NaN
## 20 1.94105630 153.19559956 160.804400 0.7487923 0.03685249 0.67656272
## 21 0.12658632 19.75189537 20.248105 0.9090909 0.06329316 0.78503860
## 22 0.00000000 6.00000000 6.000000 1.0000000 NaN NaN
## 23 0.58040495 24.86242721 27.137573 0.7878788 0.08291499 0.62536839
## 24 0.35797829 34.29837544 35.701625 0.8536585 0.05966305 0.73672111
## 25 0.12393031 6.75710106 7.242899 0.8750000 0.12393031 0.63210106
## 26 0.81039555 45.41165391 48.588346 0.7833333 0.06233812 0.66115287
## 27 0.79332514 30.44511131 33.554889 0.7619048 0.07933251 0.60641589
## 28 0.00000000 8.00000000 8.000000 1.0000000 NaN NaN
## 29 0.63282718 12.75968152 15.240318 0.7368421 0.12656544 0.48877841
## 30 0.09272834 27.81825580 28.181744 0.9333333 0.04636417 0.84246123
## 31 0.24677406 17.51633173 18.483668 0.8571429 0.08225802 0.69592010
## 32 0.91645386 39.20378345 42.796217 0.7592593 0.07049645 0.62108876
## 33 0.76870611 6.49336370 9.506636 0.6666667 0.19217653 0.29000759
## 34 0.38435306 1.24668185 2.753318 0.6666667 0.38435306 -0.08665148
## 35 0.37605072 5.26295414 6.737046 0.7500000 0.18802536 0.38147707
## 36 1.00241255 4.03530751 7.964692 0.6000000 0.25060314 0.10882688
## 37 0.22435411 19.56027403 20.439726 0.8695652 0.07478470 0.72298989
## 38 0.08199332 10.83929604 11.160704 0.9166667 0.08199332 0.75596271
## 39 0.00000000 1.00000000 1.000000 1.0000000 NaN NaN
## 40 0.00000000 7.00000000 7.000000 1.0000000 NaN NaN
## 41 0.46056619 8.09730686 9.902693 0.7500000 0.15352206 0.44910229
## 42 0.00000000 2.00000000 2.000000 1.0000000 NaN NaN
## 43 0.00000000 11.00000000 11.000000 1.0000000 NaN NaN
## 44 0.70881271 1.61075262 4.389247 0.6000000 0.35440636 -0.09462369
## 45 0.30821513 21.39590944 22.604091 0.8461538 0.07705378 0.69513121
## 46 0.00000000 4.00000000 4.000000 1.0000000 NaN NaN
## 47 0.00000000 5.00000000 5.000000 1.0000000 NaN NaN
## 48 0.76870611 6.49336370 9.506636 0.6666667 0.19217653 0.29000759
## p_UCI abundance_per_100m
## 1 1.1393766 0.037727307
## 2 0.9798892 0.063910015
## 3 NaN 0.015977504
## 4 0.9401682 0.278008564
## 5 0.8858566 0.217939894
## 6 0.9487659 0.149116770
## 7 1.0428618 0.039174208
## 8 1.0740342 0.035256787
## 9 0.9148085 0.050515256
## 10 0.7651373 0.055566781
## 11 0.8907482 0.072405200
## 12 1.0410217 0.089847260
## 13 NaN 0.078616352
## 14 0.8104513 0.279841745
## 15 0.9670935 0.069960436
## 16 1.5169754 0.004824858
## 17 NaN 0.021711860
## 18 0.9978907 0.048248577
## 19 NaN 0.032549964
## 20 0.8210218 1.022068876
## 21 1.0331432 0.230123116
## 22 NaN 0.069036935
## 23 0.9503892 0.299160051
## 24 0.9705960 0.087513127
## 25 1.1178989 0.017502625
## 26 0.9055138 0.117517628
## 27 0.9173936 0.129496985
## 28 NaN 0.032374246
## 29 0.9849058 0.056654931
## 30 1.0242054 0.056035863
## 31 1.0183656 0.036023055
## 32 0.8974298 0.115848662
## 33 1.0433257 0.022604617
## 34 1.4199848 0.005651154
## 35 1.1185229 0.016953463
## 36 1.0911731 0.016953463
## 37 1.0161405 0.061046334
## 38 1.0773706 0.033575484
## 39 NaN 0.003052317
## 40 NaN 0.021366217
## 41 1.0508977 0.027470850
## 42 NaN 0.017959770
## 43 NaN 0.098778736
## 44 1.2946237 0.026939655
## 45 0.9971765 0.053582737
## 46 NaN 0.009742316
## 47 NaN 0.012177895
## 48 1.0433257 0.019484631
#note that abundance per 100m is really per m2. its just everywhere and im too lazy to go changing every time its mentioned
#scale abundance/m^2
population_estimates23_summary$abundance_per_100m <- scale(population_estimates23_summary$abundance_per_100m)
population_estimates24_summary$abundance_per_100m <- scale(population_estimates24_summary$abundance_per_100m)
# Correlation test for biomass vs. periphyton with no species differentiation
# scale biomass data for comparability. not NECESSARY right now but makes the future stuff way better to handle
# Exclude rows with zero or negative biomass
avg_biomass23 <- avg_biomass23 %>%
filter(biomass_reach > 0)
avg_biomass24 <- avg_biomass24 %>%
filter(biomass_reach > 0)
avg_biomass23$biomass_reach_scale <- scale(log(avg_biomass23$biomass_reach))
avg_biomass24$biomass_reach_scale <- scale(log(avg_biomass24$biomass_reach))
#omitting CC REACH due to lack of periphyton data in one of the years
# Aggregate 2023 data by Basin
avg_biomass23_summarized <- avg_biomass23 %>%
filter(Basin != "CC_REACH")%>%
group_by(Basin) %>%
summarize(total_biomass_23 = sum(biomass_reach_scale), Yearly_Avg_23 = mean(Yearly_Avg), biomass_reach = sum(biomass_reach))
# Aggregate 2024 data by Basin
avg_biomass24_summarized <- avg_biomass24 %>%
filter(Basin != "CC_REACH")%>%
group_by(Basin) %>%
summarize(total_biomass_24 = sum(biomass_reach_scale), Yearly_Avg_24 = mean(Yearly_Avg), biomass_reach = sum(biomass_reach))
# Scatter plot for 2023 biomass vs. Yearly_Avg
ggplot(avg_biomass23_summarized, aes(x = Yearly_Avg_23, y = total_biomass_23)) +
geom_point() +
geom_smooth(method = "lm", color = "green") +
theme_minimal() +
labs(title = "2023 Biomass vs. Yearly_Avg Chlorophyll", x = "Yearly_Avg", y = "Total Fish Biomass")
## `geom_smooth()` using formula = 'y ~ x'
# Scatter plot for 2024 biomass vs. Yearly_Avg
ggplot(avg_biomass24_summarized, aes(x = Yearly_Avg_24, y = total_biomass_24)) +
geom_point() +
geom_smooth(method = "lm", color = "purple") +
theme_minimal() +
labs(title = "2024 Biomass vs. Yearly_Avg Chlorophyll", x = "Yearly_Avg", y = "Total Fish Biomass")
## `geom_smooth()` using formula = 'y ~ x'
# Correlation test for 2023 biomass vs. Yearly_Avg
cor_2023_avg <- cor.test(avg_biomass23_summarized$total_biomass_23, avg_biomass23_summarized$Yearly_Avg_23, method = "pearson")
print(cor_2023_avg)
##
## Pearson's product-moment correlation
##
## data: avg_biomass23_summarized$total_biomass_23 and avg_biomass23_summarized$Yearly_Avg_23
## t = 1.9171, df = 14, p-value = 0.07587
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.05131063 0.77623868
## sample estimates:
## cor
## 0.4559929
# Correlation test for 2024 biomass vs. Yearly_Avg
cor_2024_avg <- cor.test(avg_biomass24_summarized$total_biomass_24, avg_biomass24_summarized$Yearly_Avg_24, method = "pearson")
print(cor_2024_avg)
##
## Pearson's product-moment correlation
##
## data: avg_biomass24_summarized$total_biomass_24 and avg_biomass24_summarized$Yearly_Avg_24
## t = 1.0793, df = 14, p-value = 0.2987
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.2533554 0.6795069
## sample estimates:
## cor
## 0.2771582
#Statistical significance of correlation test in 2023 but not 2024. is it due to fish or periphyton? What about per species?
# Correlation test with species differentiation
cor_biomassspecies_2023 <- avg_biomass23 %>%
filter(Species != "RBT")%>%
group_by(Species) %>%
summarize(correlation = cor.test(biomass_reach, Yearly_Avg, method = "pearson")$estimate,
p_value = cor.test(biomass_reach, Yearly_Avg, method = "pearson")$p.value)
cor_biomassspecies_2024 <- avg_biomass24 %>%
filter(Species != "RBT")%>%
group_by(Species) %>%
summarize(correlation = cor.test(biomass_reach, Yearly_Avg, method = "pearson")$estimate,
p_value = cor.test(biomass_reach, Yearly_Avg, method = "pearson")$p.value)
print(cor_biomassspecies_2023)
## # A tibble: 4 × 3
## Species correlation p_value
## <chr> <dbl> <dbl>
## 1 COH -0.546 0.454
## 2 CTT 0.759 0.000643
## 3 SCP 0.133 0.714
## 4 TRT 0.0302 0.915
print(cor_biomassspecies_2024)
## # A tibble: 4 × 3
## Species correlation p_value
## <chr> <dbl> <dbl>
## 1 COH 0.421 0.579
## 2 CTT 0.509 0.0442
## 3 SCP -0.339 0.338
## 4 TRT 0.212 0.449
ggplot(avg_biomass23, aes(x = biomass_reach, y = Yearly_Avg, color = Species)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm") + # Adds regression lines per species
labs(
x = "Fish Biomass per Reach",
y = "Chlorophyll Biomass",
title = "Relationship Between Fish Biomass and Chlorophyll by Species - 2023"
) +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
ggplot(avg_biomass24, aes(x = biomass_reach, y = Yearly_Avg, color = Species)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm") + # Adds regression lines per species
labs(
x = "Fish Biomass per Reach",
y = "Chlorophyll Biomass",
title = "Relationship Between Fish Biomass and Chlorophyll by Species - 2024"
) +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
# Left join the 2023 and 2024 aggregated data by Basin
combined_biomass <- left_join(avg_biomass23_summarized %>% mutate(Year = 2023),
avg_biomass24_summarized %>% mutate(Year = 2024),
by = "Basin")
# Correlation test for fish biomass between 2023 and 2024
cor_biomass_fish <- cor.test(combined_biomass$total_biomass_23, combined_biomass$total_biomass_24, method = "pearson")
print(cor_biomass_fish)
##
## Pearson's product-moment correlation
##
## data: combined_biomass$total_biomass_23 and combined_biomass$total_biomass_24
## t = 1.1023, df = 14, p-value = 0.2889
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.2478289 0.6826682
## sample estimates:
## cor
## 0.282593
# Run the correlation test for periphyton biomass between 2023 and 2024
cor_biomass_peri <- cor.test(combined_biomass$Yearly_Avg_23, combined_biomass$Yearly_Avg_24, method = "pearson")
print(cor_biomass_peri)
##
## Pearson's product-moment correlation
##
## data: combined_biomass$Yearly_Avg_23 and combined_biomass$Yearly_Avg_24
## t = 1.676, df = 14, p-value = 0.1159
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1090071 0.7520904
## sample estimates:
## cor
## 0.4087872
# Scatter plot for fish
ggplot(combined_biomass, aes(x = total_biomass_23, y = total_biomass_24)) +
geom_point() +
geom_smooth(method = "lm", color = "darkblue") +
theme_minimal() +
labs(title = "Fish Biomass '23 vs. Fish Biomass '24", x = "Biomass 2023", y = "Biomass 2024")
## `geom_smooth()` using formula = 'y ~ x'
# Scatter plot for periphyton
ggplot(combined_biomass, aes(x = Yearly_Avg_23, y = Yearly_Avg_24)) +
geom_point() +
geom_smooth(method = "lm", color = "darkgreen") +
theme_minimal() +
labs(title = "Yearly_Avg Chlorophyll '23 vs. Yearly_Avg Chlorophyll '24", x = "Yearly_Avg 2023", y = "Yearly_Avg 2024")
## `geom_smooth()` using formula = 'y ~ x'
#Results: Seems to be that periphyton is not related to the trend of decreasing fish biomass between 2023 and 2024. Wondering if next year we'll see a delayed effect.
# Plot 2023/2024 data
ggplot() +
geom_point(data = combined_biomass, aes(x = Yearly_Avg_23, y = total_biomass_23, color = "2023")) +
geom_point(data = combined_biomass, aes(x = Yearly_Avg_24, y = total_biomass_24, color = "2024")) +
geom_smooth(data = combined_biomass, aes(x = Yearly_Avg_23, y = total_biomass_23, color = "2023"), method = "lm") +
geom_smooth(data = combined_biomass, aes(x = Yearly_Avg_24, y = total_biomass_24, color = "2024"), method = "lm") +
labs(x = "Yearly Average Chlorophyll", y = "Total Biomass", color = "Year") +
scale_color_manual(values = c("2023" = "blue", "2024" = "red")) +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
# regressions for significance on year based data
# 2023 fish v. peri
lm_2023_avg <- lm(total_biomass_23 ~ Yearly_Avg_23, data = avg_biomass23_summarized)
summary(lm_2023_avg)
##
## Call:
## lm(formula = total_biomass_23 ~ Yearly_Avg_23, data = avg_biomass23_summarized)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.4993 -0.5324 -0.0287 0.7314 1.5917
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.006846 0.248357 -0.028 0.9784
## Yearly_Avg_23 0.466841 0.243517 1.917 0.0759 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9933 on 14 degrees of freedom
## Multiple R-squared: 0.2079, Adjusted R-squared: 0.1514
## F-statistic: 3.675 on 1 and 14 DF, p-value: 0.07587
#2024 fish v. peri
lm_2024_avg <- lm(total_biomass_24 ~ Yearly_Avg_24, data = avg_biomass24_summarized)
summary(lm_2024_avg)
##
## Call:
## lm(formula = total_biomass_24 ~ Yearly_Avg_24, data = avg_biomass24_summarized)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.2930 -0.7870 0.1922 0.9776 1.9296
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.06399 0.40541 -0.158 0.877
## Yearly_Avg_24 0.45964 0.42587 1.079 0.299
##
## Residual standard error: 1.604 on 14 degrees of freedom
## Multiple R-squared: 0.07682, Adjusted R-squared: 0.01087
## F-statistic: 1.165 on 1 and 14 DF, p-value: 0.2987
#2023 v 2024 fish
lm_fish_biomass <- lm(total_biomass_24 ~ total_biomass_23, data = combined_biomass)
summary(lm_fish_biomass)
##
## Call:
## lm(formula = total_biomass_24 ~ total_biomass_23, data = combined_biomass)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.4774 -0.6283 0.0431 0.9874 2.1127
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -4.249e-16 4.004e-01 0.000 1.000
## total_biomass_23 4.227e-01 3.835e-01 1.102 0.289
##
## Residual standard error: 1.602 on 14 degrees of freedom
## Multiple R-squared: 0.07986, Adjusted R-squared: 0.01413
## F-statistic: 1.215 on 1 and 14 DF, p-value: 0.2889
#2023 v. 2024 peri
lm_peri_biomass <- lm(Yearly_Avg_24 ~ Yearly_Avg_23, data = combined_biomass)
summary(lm_peri_biomass)
##
## Call:
## lm(formula = Yearly_Avg_24 ~ Yearly_Avg_23, data = combined_biomass)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.64294 -0.51689 -0.00269 0.46245 1.90240
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.1337 0.2297 0.582 0.570
## Yearly_Avg_23 0.3775 0.2252 1.676 0.116
##
## Residual standard error: 0.9188 on 14 degrees of freedom
## Multiple R-squared: 0.1671, Adjusted R-squared: 0.1076
## F-statistic: 2.809 on 1 and 14 DF, p-value: 0.1159
#Combine species population estimate data with periphyton data but summarizing to keep chlorophyll values at reach level
peri_abundance_23 <- population_estimates23_summary %>%
filter(Species != "RBT")%>%
group_by(Basin, Species) %>%
left_join(HU_23, by = "Basin") # Join with habitat data
peri_abundance_24 <- population_estimates24_summary %>%
filter(Species != "RBT")%>%
group_by(Basin, Species) %>%
left_join(HU_24, by = "Basin") # Join with habitat data
#only using yearly summarized data
cor_abun_2023 <- peri_abundance_23 %>%
group_by(Species) %>%
summarize(correlation = cor.test(abundance_per_100m, Yearly_Avg, method = "pearson")$estimate,
p_value = cor.test(abundance_per_100m, Yearly_Avg, method = "pearson")$p.value)
cor_abun_2024 <- peri_abundance_24 %>%
group_by(Species) %>%
summarize(correlation = cor.test(abundance_per_100m, Yearly_Avg, method = "pearson")$estimate,
p_value = cor.test(abundance_per_100m, Yearly_Avg, method = "pearson")$p.value)
print(cor_abun_2023)
## # A tibble: 4 × 3
## Species correlation p_value
## <chr> <dbl> <dbl>
## 1 COH -0.430 0.570
## 2 CTT 0.634 0.00831
## 3 SCP -0.221 0.539
## 4 TRT 0.0221 0.938
print(cor_abun_2024)
## # A tibble: 4 × 3
## Species correlation p_value
## <chr> <dbl> <dbl>
## 1 COH 0.614 0.386
## 2 CTT 0.712 0.00199
## 3 SCP -0.114 0.755
## 4 TRT 0.300 0.278
#plot correlations
ggplot(peri_abundance_23, aes(x = Yearly_Avg, y = abundance_per_100m)) +
geom_point() +
geom_smooth(method = "lm") +
facet_wrap(~Species) +
theme_minimal() +
labs(title = "2023 Species-Specific Abundance vs. Chlorophyll", x = "Yearly_Avg", y = "Abundance")
## `geom_smooth()` using formula = 'y ~ x'
ggplot(peri_abundance_24, aes(x = Yearly_Avg, y = abundance_per_100m)) +
geom_point() +
geom_smooth(method = "lm") +
facet_wrap(~Species) +
theme_minimal() +
labs(title = "2024 Species-Specific Abundance vs. Chlorophyll", x = "Yearly_Avg", y = "Abundance")
## `geom_smooth()` using formula = 'y ~ x'
#regressions
lm_abun_2023 <- peri_abundance_23 %>%
group_by(Species) %>%
do(model = lm(abundance_per_100m ~ Yearly_Avg, data = .))
lm_abun_2024 <- peri_abundance_24 %>%
group_by(Species) %>%
do(model = lm(abundance_per_100m ~ Yearly_Avg, data = .))
# View model summaries
summary(lm_abun_2023$model[[1]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## 1 2 3 4
## 0.627525 0.030777 -0.651409 -0.006894
## attr(,"scaled:center")
## [1] 0.1106
## attr(,"scaled:scale")
## [1] 0.1956
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.6067 0.6219 0.976 0.432
## Yearly_Avg -0.4449 0.6605 -0.673 0.570
##
## Residual standard error: 0.64 on 2 degrees of freedom
## Multiple R-squared: 0.1849, Adjusted R-squared: -0.2227
## F-statistic: 0.4536 on 1 and 2 DF, p-value: 0.57
summary(lm_abun_2023$model[[2]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.38354 -0.16012 -0.00678 0.06387 0.61754
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.09570 0.06579 -1.455 0.16786
## Yearly_Avg 0.19808 0.06451 3.070 0.00831 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2632 on 14 degrees of freedom
## Multiple R-squared: 0.4024, Adjusted R-squared: 0.3597
## F-statistic: 9.428 on 1 and 14 DF, p-value: 0.008306
summary(lm_abun_2023$model[[3]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.053721 -0.025284 -0.009305 0.021348 0.089088
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.451360 0.013873 -32.534 8.68e-10 ***
## Yearly_Avg -0.007768 0.012117 -0.641 0.539
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.04387 on 8 degrees of freedom
## Multiple R-squared: 0.04886, Adjusted R-squared: -0.07003
## F-statistic: 0.411 on 1 and 8 DF, p-value: 0.5394
summary(lm_abun_2023$model[[4]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.6609 -0.5897 -0.4491 -0.2279 5.7538
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.21467 0.43056 0.499 0.626
## Yearly_Avg 0.03255 0.40879 0.080 0.938
##
## Residual standard error: 1.667 on 13 degrees of freedom
## Multiple R-squared: 0.0004874, Adjusted R-squared: -0.0764
## F-statistic: 0.00634 on 1 and 13 DF, p-value: 0.9378
summary(lm_abun_2024$model[[1]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## 1 2 3 4
## 0.4790 0.4409 -0.4236 -0.4963
## attr(,"scaled:center")
## [1] 0.09079
## attr(,"scaled:scale")
## [1] 0.1562
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2233 0.3261 0.685 0.564
## Yearly_Avg 0.5631 0.5124 1.099 0.386
##
## Residual standard error: 0.6518 on 2 degrees of freedom
## Multiple R-squared: 0.3764, Adjusted R-squared: 0.06463
## F-statistic: 1.207 on 1 and 2 DF, p-value: 0.3865
summary(lm_abun_2024$model[[2]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.42335 -0.23121 -0.00212 0.08400 0.63141
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.14870 0.07771 -1.913 0.07636 .
## Yearly_Avg 0.30931 0.08164 3.789 0.00199 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3075 on 14 degrees of freedom
## Multiple R-squared: 0.5063, Adjusted R-squared: 0.471
## F-statistic: 14.36 on 1 and 14 DF, p-value: 0.001994
summary(lm_abun_2024$model[[3]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.15655 -0.12845 -0.08782 0.08018 0.37086
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.34549 0.06319 -5.467 0.000596 ***
## Yearly_Avg -0.02204 0.06819 -0.323 0.754778
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1951 on 8 degrees of freedom
## Multiple R-squared: 0.01289, Adjusted R-squared: -0.1105
## F-statistic: 0.1045 on 1 and 8 DF, p-value: 0.7548
summary(lm_abun_2024$model[[4]])
##
## Call:
## lm(formula = abundance_per_100m ~ Yearly_Avg, data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.0247 -0.7547 -0.4627 -0.0558 5.3400
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3040 0.4273 0.711 0.489
## Yearly_Avg 0.4966 0.4381 1.133 0.278
##
## Residual standard error: 1.626 on 13 degrees of freedom
## Multiple R-squared: 0.08992, Adjusted R-squared: 0.01991
## F-statistic: 1.284 on 1 and 13 DF, p-value: 0.2775
#check abundance trends
combined_abundance <- left_join(peri_abundance_23 %>% rename(abundance_per_100m_23 = abundance_per_100m),
peri_abundance_24 %>% rename(abundance_per_100m_24 = abundance_per_100m),
by = c("Basin", "Species"))
ggplot(combined_abundance, aes(x = abundance_per_100m_23, y = abundance_per_100m_24, color = Species)) +
geom_point() +
geom_smooth(method = "lm") +
theme_minimal() +
labs(title = "Species-Specific Abundance 2023 vs. 2024", x = "Abundance 2023", y = "Abundance 2024")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 1 row containing missing values or values outside the scale range
## (`geom_point()`).
# Note that basins Aa1, Bp, Ca1, Cz, Da2, Dp, and Dz are omitted due to missing data
canopy_peri <- merge(canopy, combined_biomass, by = "Basin")
cor_canopy_peri_23 <- canopy_peri %>%
summarize(correlation = cor.test(average_closure_23, Yearly_Avg_23, method = "pearson")$estimate,
p_value = cor.test(average_closure_23, Yearly_Avg_23, method = "pearson")$p.value)
cor_canopy_peri_24 <- canopy_peri %>%
summarize(correlation = cor.test(average_closure_24, Yearly_Avg_24, method = "pearson")$estimate,
p_value = cor.test(average_closure_24, Yearly_Avg_24, method = "pearson")$p.value)
print(cor_canopy_peri_23)
## correlation p_value
## 1 0.02125142 0.9477331
print(cor_canopy_peri_24)
## correlation p_value
## 1 -0.3060507 0.3333091
ggplot(canopy_peri, aes(x = Yearly_Avg_23, y = average_closure_23)) +
geom_point() +
geom_smooth(method = "lm") +
theme_minimal() +
labs(title = "2023 Canopy Closure % vs. Chlorophyll", x = "Yearly Avg Chl-A '23", y = "Canopy Closure '23 (%)")
## `geom_smooth()` using formula = 'y ~ x'
ggplot(canopy_peri, aes(x = Yearly_Avg_24, y = average_closure_24)) +
geom_point() +
geom_smooth(method = "lm") +
theme_minimal() +
labs(title = "2024 Canopy Closure % vs. Chlorophyll", x = "Yearly Avg Chl-A '24", y = "Canopy Closure '24 (%)")
## `geom_smooth()` using formula = 'y ~ x'
lm_canopy23 <- lm(average_closure_23 ~ Yearly_Avg_23, data = canopy_peri)
lm_canopy24 <- lm(average_closure_24 ~ Yearly_Avg_24, data = canopy_peri)
summary(lm_canopy23)
##
## Call:
## lm(formula = average_closure_23 ~ Yearly_Avg_23, data = canopy_peri)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8424 -1.0224 0.2846 0.7567 1.4360
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 92.09484 0.31531 292.073 <2e-16 ***
## Yearly_Avg_23 0.01932 0.28746 0.067 0.948
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.084 on 10 degrees of freedom
## Multiple R-squared: 0.0004516, Adjusted R-squared: -0.0995
## F-statistic: 0.004518 on 1 and 10 DF, p-value: 0.9477
summary(lm_canopy24)
##
## Call:
## lm(formula = average_closure_24 ~ Yearly_Avg_24, data = canopy_peri)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.7978 -0.6950 0.5899 1.3685 3.3491
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 92.2451 0.8134 113.408 <2e-16 ***
## Yearly_Avg_24 -0.8086 0.7954 -1.017 0.333
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.718 on 10 degrees of freedom
## Multiple R-squared: 0.09367, Adjusted R-squared: 0.003034
## F-statistic: 1.033 on 1 and 10 DF, p-value: 0.3333
# Histogram of chlorophyll values - using Avg_Chl for the reach since we don't need to scale it for comparison to other variables. This is just intersite of one variable.
ggplot(peri_harvest, aes(x = Avg_Chl, fill = Harvest_Status)) +
geom_histogram(bins = 20, alpha = 0.7, position = "identity") +
theme_minimal()
# Shapiro-Wilk test for normality
shapiro.test(peri_harvest$Avg_Chl) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: peri_harvest$Avg_Chl
## W = 0.70975, p-value = 6.742e-12
# QQ Plot
qqnorm(peri_harvest$Avg_Chl)
qqline(peri_harvest$Avg_Chl, col = "red")
#normality test miserably fails - lots of low values in general with some outliers. Let's find those outliers.
# Calculate Z-scores
z_peri_harvest <- peri_harvest %>% mutate(z_Avg_Chl = scale(Avg_Chl))
# Find extreme outliers (Z-score > 3 or < -3)
z_peri_harvest %>% filter(abs(z_Avg_Chl) > 3)
## Site Sample.Date Season Year Avg_Chl Reach.length Chl_100 Harvest_Status
## 1 AZ_Reach 7/3/2023 Summer 2023 233.35 98.5 236.90 PRE
## 2 CP_Reach 6/26/2023 Summer 2023 346.33 111.3 311.17 PRE
## Prescription z_Avg_Chl
## 1 Thin w/ gaps 3.273518
## 2 VRH 5.321246
#Looks like we need to remove only Bz from 9/2024 - makes sense, this was a mistaken filtration of all 125mL. The other two are true outliers.
peri_harvest <- peri_harvest[-c(20),]
#Tried both sqrt and log transformations, and data still fails normality test.
#Need to account for random effects by basin and investigate a little more closely. using mixed effects due to data not being normal. can handle the uneven replications as well. Both a Kruskal-Wallis and a Wilcox test fail because of lack of consistent pairs and some basins still being in PRE or CON status.
# Ensuring both Harvest_Status and Prescription are factors with appropriate levels
peri_harvest$Harvest_Status <- factor(peri_harvest$Harvest_Status, levels = c("PRE", "POST", "CON"))
peri_harvest$Prescription <- factor(peri_harvest$Prescription, levels = c("Wide thin", "VRH", "Thin w/ gaps", "Control"))
#Setting reference parameters for both harvest status and prescription as the Control sites
peri_harvest$Prescription <- relevel(peri_harvest$Prescription, ref = "Control")
peri_harvest$Harvest_Status <- relevel(peri_harvest$Harvest_Status, ref = "CON")
# LMM with random effect for Basin
lmm_model <- lmer(Avg_Chl ~ Harvest_Status * Prescription + (1 | Site), data = peri_harvest)
## fixed-effect model matrix is rank deficient so dropping 5 columns / coefficients
summary(lmm_model)
## Linear mixed model fit by REML ['lmerMod']
## Formula: Avg_Chl ~ Harvest_Status * Prescription + (1 | Site)
## Data: peri_harvest
##
## REML criterion at convergence: 878.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.4119 -0.5269 -0.2396 0.2112 5.2226
##
## Random effects:
## Groups Name Variance Std.Dev.
## Site (Intercept) 294.2 17.15
## Residual 2584.9 50.84
## Number of obs: 87, groups: Site, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 54.303 18.550 2.927
## Harvest_StatusPRE 8.321 23.906 0.348
## Harvest_StatusPOST -1.024 29.376 -0.035
## PrescriptionWide thin -18.123 32.829 -0.552
## PrescriptionVRH -4.974 31.068 -0.160
## Harvest_StatusPRE:PrescriptionWide thin -5.165 36.247 -0.143
## Harvest_StatusPRE:PrescriptionVRH -6.285 33.569 -0.187
##
## Correlation of Fixed Effects:
## (Intr) Hr_SPRE H_SPOS PrscWt PrsVRH H_SPRt
## Hrvst_StPRE -0.776
## Hrvst_SPOST -0.631 0.594
## PrscrptnWdt 0.000 -0.093 -0.538
## PrscrptnVRH 0.000 -0.098 -0.569 0.509
## Hr_SPRE:PWt 0.000 -0.178 0.419 -0.793 -0.396
## H_SPRE:PVRH 0.000 -0.193 0.452 -0.405 -0.830 0.494
## fit warnings:
## fixed-effect model matrix is rank deficient so dropping 5 columns / coefficients
# Visualize the fixed effects using ggplot
ggplot(peri_harvest, aes(x = Harvest_Status, y = Avg_Chl, color = Prescription)) +
geom_boxplot() +
labs(title = "Chlorophyll Levels by Harvest Status and Prescription",
x = "Harvest Status", y = "Average Chlorophyll") +
theme_minimal()
#results indicate that chlorophyll varies significantly across sites (529.2 ug/L, SD 23.00 ug/L), with quite a lot that is not really explainable by the model.
#control chl values are low (54.41 ug/L), which seems ecologically sound considering the heavy canopy coverage at most of these sites (~90%)
#Post prescription, we're seeing a very small drop in average chl in the first year - this will likely become more evident as populations adjust to disturbance and new light regimes
#Thin with gaps has the greatest drop in average chl. This is likely more due to immediate disturbance and scouring of periphyton populations in the reach, as well as drastic habitat unit changes. However not statistically significant
#Despite these observations of patterns, all t-values are low except for the intercept, suggesting that there hasn't been a significant effect of treatment or harvest type on chlorophyll values in the first year.
# Histogram of macro biomass by feeding group
ggplot(macros, aes(x = Biomass, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 30 rows containing non-finite outside the scale range
## (`stat_bin()`).
## Warning: Removed 16 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(macros$Biomass) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: macros$Biomass
## W = 0.2247, p-value < 2.2e-16
# QQ Plot
qqnorm(macros$Biomass)
qqline(macros$Biomass, col = "forestgreen")
#Needs transformation
macros <- macros %>%
mutate(
Biomass_log = log(Biomass)
)
# Histogram of macro biomass by feeding group
ggplot(macros, aes(x = Biomass_log, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 862 rows containing non-finite outside the scale range (`stat_bin()`).
## Removed 16 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(macros$Biomass_log) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: macros$Biomass_log
## W = 0.99794, p-value = 0.2112
# QQ Plot
qqnorm(macros$Biomass_log)
qqline(macros$Biomass_log, col = "forestgreen")
# Calculate Z-scores
z_macros <- macros %>% mutate(z_Biomass_log = scale(Biomass_log))
# Find extreme outliers (Z-score > 3 or < -3)
z_macros %>% filter(abs(z_Biomass_log) > 3)
## Basin Location Date Taxon Stage Insect Origin
## 1 CP_REACH reach 45510 Ephemerella tibialis L insect Aquatic
## 2 DA1_REACH reach 45526 Cinygma L insect Aquatic
## 3 DA1_REACH reach 45526 Cinygma L insect Aquatic
## 4 DA2_REACH reach 45526 Rhabdomastix (Rhabdomastix) L insect Aquatic
## Higher.classification Order Family Common.name Length
## 1 Arthropoda: Insecta Ephemeroptera Ephemerellidae mayflies 7
## 2 Arthropoda: Insecta Ephemeroptera Heptageniidae mayflies 12
## 3 Arthropoda: Insecta Ephemeroptera Heptageniidae mayflies 8
## 4 Arthropoda: Insecta Diptera Tipulidae crane flies 20
## Abundance Biomass Stage.1 BCG.Attribute Feeding.Group CA.feeding.group Habit
## 1 17.636684 45.19368 L 3 CG CG CL
## 2 6.783340 68.69617 L 4 SC SC CL
## 3 13.566680 44.97873 L 4 SC SC CL
## 4 3.956948 35.30367 L 2 SH SH BU
## Tolerance WY.HBI PSSB.tolerance CA.tolerance Metals.tolerance BCI.TV
## 1 0 2 0 2 11 48
## 2 0 2 0 2 11 48
## 3 0 2 0 2 11 48
## 4 MI 3 0 3 11 72
## PSSB.long.lived Voltinism Development Occurrence.in.drift Size.at.maturity
## 1 0 2 2 2 2
## 2 0 2 1 2 2
## 3 0 2 1 2 2
## 4 0 1 2 1 3
## Rheophily Thermal.preference a b Biomass_log z_Biomass_log
## 1 2 2 0.0104 2.830 3.810957 3.214113
## 2 2 2 0.0108 2.754 4.229693 3.477871
## 3 2 2 0.0108 2.754 3.806190 3.211110
## 4 2 1 0.0029 2.681 3.563987 3.058548
#relationship with periphyton biomass?
#visualize biomass by feeding type per basin
ggplot(macros, aes(x = Basin, y = Biomass, fill = Feeding.Group)) +
geom_col() +
labs(title = "Biomass by Feeding Category and Basin", y = "Biomass", x = "Feeding Category") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
library(ggplot2)
library(dplyr)
library(scales)
## Warning: package 'scales' was built under R version 4.4.2
##
## Attaching package: 'scales'
## The following object is masked from 'package:purrr':
##
## discard
## The following object is masked from 'package:readr':
##
## col_factor
# Calculate relative abundance and plot
ggplot(macros, aes(x = Basin, y = Biomass, fill = Feeding.Group)) +
geom_col(position = "fill") + # Stacks bars and normalizes to 1
scale_y_continuous(labels = percent) + # Y-axis as percentages
labs(title = "Relative Biomass of Drift Feeding Groups by Basin",
y = "Relative Biomass",
x = "Basin") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Calculate relative abundance for each Feeding.Group in each Basin
drift_biosummary <- macros %>%
group_by(Basin, Feeding.Group) %>%
summarise(Biomass = sum(Biomass), .groups = "drop") %>%
group_by(Basin) %>%
mutate(Relative_Biomass = paste0(round((Biomass / sum(Biomass))*100, 0), "%")) %>%
select(Basin, Feeding.Group, Relative_Biomass)
# Create a table with relative abundance in percentage format
drift_biosummary_table <- drift_biosummary
print(drift_biosummary_table)
## # A tibble: 114 × 3
## # Groups: Basin [20]
## Basin Feeding.Group Relative_Biomass
## <chr> <chr> <chr>
## 1 AA1_REACH CF 9%
## 2 AA1_REACH CG 46%
## 3 AA1_REACH PA 1%
## 4 AA1_REACH PR 3%
## 5 AA1_REACH SC 8%
## 6 AA1_REACH SH 4%
## 7 AA1_REACH UN 27%
## 8 AA2_REACH CF 21%
## 9 AA2_REACH CG 26%
## 10 AA2_REACH PA 2%
## # ℹ 104 more rows
# Aggregate by Basin and Feeding Category
macro_aggregated <- macros %>%
group_by(Basin, Feeding.Group) %>%
summarise(
total_biomass = sum(Biomass),
mean_biomass = mean(Biomass))
## `summarise()` has grouped output by 'Basin'. You can override using the
## `.groups` argument.
peri_macro <- macro_aggregated %>%
left_join(twoyr_peri, by = "Basin")
#scatter plot of total macroinvertebrate biomass vs periphyton biomass by Basin
ggplot(peri_macro, aes(x = Total_Avg, y = total_biomass, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Macroinvertebrate Biomass vs Periphyton Biomass by Basin",
x = "Periphyton Biomass", y = "Total Macroinvertebrate Biomass")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
library(lme4)
peri_macro <- peri_macro %>%
mutate(
total_biomass_log = log(total_biomass),
mean_biomass_log = log(mean_biomass),
Total_Avg_log = log(Total_Avg)
)
#scale
peri_macro$total_biomass_log <- scale(peri_macro$total_biomass_log)
peri_macro$Total_Avg_log <- scale(peri_macro$Total_Avg_log)
peri_macro$mean_biomass_log <- scale(peri_macro$mean_biomass_log)
library(lme4)
# Model with Feeding.Group as a fixed effect and random effects for Basin
model_peri_macro <- lmer(
total_biomass_log ~ Total_Avg_log * Feeding.Group + (1 | Basin),
data = peri_macro
)
summary(model_peri_macro)
## Linear mixed model fit by REML ['lmerMod']
## Formula: total_biomass_log ~ Total_Avg_log * Feeding.Group + (1 | Basin)
## Data: peri_macro
##
## REML criterion at convergence: 228.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.92708 -0.58988 -0.04268 0.60981 1.91015
##
## Random effects:
## Groups Name Variance Std.Dev.
## Basin (Intercept) 0.4344 0.6591
## Residual 0.3911 0.6254
## Number of obs: 101, groups: Basin, 18
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) -0.02427 0.23913 -0.101
## Total_Avg_log -0.33249 0.25412 -1.308
## Feeding.GroupCG 0.72455 0.23677 3.060
## Feeding.GroupMH -1.20817 0.62748 -1.925
## Feeding.GroupPA -0.91222 0.24143 -3.778
## Feeding.GroupPR 0.07458 0.26173 0.285
## Feeding.GroupSC 0.33937 0.24230 1.401
## Feeding.GroupSH -0.64930 0.26188 -2.479
## Feeding.GroupUN 0.14576 0.24064 0.606
## Total_Avg_log:Feeding.GroupCG 0.32607 0.24933 1.308
## Total_Avg_log:Feeding.GroupMH 0.50823 0.73948 0.687
## Total_Avg_log:Feeding.GroupPA 0.51515 0.24932 2.066
## Total_Avg_log:Feeding.GroupPR 0.17930 0.26828 0.668
## Total_Avg_log:Feeding.GroupSC 0.25923 0.25723 1.008
## Total_Avg_log:Feeding.GroupSH -0.29625 0.27887 -1.062
## Total_Avg_log:Feeding.GroupUN 0.32742 0.25748 1.272
##
## Correlation matrix not shown by default, as p = 16 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it
# Plot the relationship for each Feeding Group
ggplot(peri_macro, aes(x = Total_Avg_log, y = total_biomass_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Feeding.Group) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Total Drift Macro Biomass by Feeding Group",
x = "Total Average Periphyton Biomass",
y = "Drift Total Biomass") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
ggplot(peri_macro, aes(x = Total_Avg_log, y = total_biomass_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Basin) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Total Drift Macro Biomass by Basin",
x = "Total Average Periphyton Biomass",
y = "Drift Total Biomass") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
# Histogram of macro biomass by feeding group
ggplot(macros, aes(x = Abundance, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 457 rows containing non-finite outside the scale range
## (`stat_bin()`).
## Warning: Removed 16 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(macros$Abundance) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: macros$Abundance
## W = 0.34602, p-value < 2.2e-16
# QQ Plot
qqnorm(macros$Abundance)
qqline(macros$Abundance, col = "lightblue3")
#Needs transformation
macros2 <- macros %>%
mutate(
Abundance_log = log(Abundance)
)
# Histogram of macro biomass by feeding group
ggplot(macros2, aes(x = Abundance_log, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 4 rows containing non-finite outside the scale range (`stat_bin()`).
## Removed 16 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(macros2$Abundance_log) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: macros2$Abundance_log
## W = 0.9357, p-value < 2.2e-16
# QQ Plot
qqnorm(macros2$Abundance_log)
qqline(macros2$Abundance_log, col = "lightblue3")
# Calculate Z-scores
z_macros2 <- macros2 %>% mutate(z_Abundance_log = scale(Abundance_log))
# Find extreme outliers (Z-score > 3 or < -3)
z_macros2 %>% filter(abs(z_Abundance_log) > 3)
## Basin Location Date Taxon Stage Insect Origin
## 1 AZ_REACH reach 45572 Baetis tricaudatus complex L insect Aquatic
## 2 AZ_REACH reach 45572 Brachycera L insect Terrestrial
## 3 AZ_REACH reach 45572 Brachycera L insect Terrestrial
## 4 BC_REACH reach 45512 Glossosoma L insect Aquatic
## 5 BC_REACH reach 45512 Glossosoma L insect Aquatic
## 6 CP_REACH reach 45510 Baetis tricaudatus complex L insect Aquatic
## 7 CP_REACH reach 45510 Baetis tricaudatus complex L insect Aquatic
## 8 CP_REACH reach 45510 Baetis tricaudatus complex L insect Aquatic
## 9 CP_REACH reach 45510 Tvetenia bavarica group L insect Aquatic
## 10 CP_REACH reach 45510 Tvetenia bavarica group L insect Aquatic
## 11 CP_REACH reach 45510 Tvetenia bavarica group L insect Aquatic
## 12 CZ2_REACH reach 45504 Baetis tricaudatus complex L insect Aquatic
## Higher.classification Order Family
## 1 Arthropoda: Insecta Ephemeroptera Baetidae
## 2 Arthropoda: Insecta Diptera x
## 3 Arthropoda: Insecta Diptera x
## 4 Arthropoda: Insecta Trichoptera Glossosomatidae
## 5 Arthropoda: Insecta Trichoptera Glossosomatidae
## 6 Arthropoda: Insecta Ephemeroptera Baetidae
## 7 Arthropoda: Insecta Ephemeroptera Baetidae
## 8 Arthropoda: Insecta Ephemeroptera Baetidae
## 9 Arthropoda: Insecta Diptera Chironomidae: Orthocladiinae
## 10 Arthropoda: Insecta Diptera Chironomidae: Orthocladiinae
## 11 Arthropoda: Insecta Diptera Chironomidae: Orthocladiinae
## 12 Arthropoda: Insecta Ephemeroptera Baetidae
## Common.name Length Abundance Biomass Stage.1 BCG.Attribute Feeding.Group
## 1 mayflies 1.5 249.07949 4.2352222 L 4 CG
## 2 higher flies 2.0 162.44314 3.9014401 A 0 UN
## 3 higher flies 2.0 162.44314 3.9014401 L 0 UN
## 4 caddisflies 1.5 88.18342 2.3992679 L 3 SC
## 5 caddisflies 1.5 88.18342 2.3992679 P 3 SC
## 6 mayflies 1.0 84.65608 0.4486772 L 4 CG
## 7 mayflies 1.5 229.27690 3.8985089 L 4 CG
## 8 mayflies 2.0 84.65608 3.2915108 L 4 CG
## 9 midges 2.0 98.76543 1.0906188 L 4 CG
## 10 midges 2.5 112.87478 2.2350046 L 4 CG
## 11 midges 3.0 123.45679 3.9392524 L 4 CG
## 12 mayflies 1.5 77.16049 1.3119982 L 4 CG
## CA.feeding.group Habit Tolerance WY.HBI PSSB.tolerance CA.tolerance
## 1 CG CL 0 6 0 6
## 2 UN BU HT 0 0 11
## 3 UN BU HT 0 0 11
## 4 SC CL 0 1 0 1
## 5 SC CL 0 1 0 1
## 6 CG CL 0 6 0 6
## 7 CG CL 0 6 0 6
## 8 CG CL 0 6 0 6
## 9 CG SP 0 5 0 5
## 10 CG SP 0 5 0 5
## 11 CG SP 0 5 0 5
## 12 CG CL 0 6 0 6
## Metals.tolerance BCI.TV PSSB.long.lived Voltinism Development
## 1 11 72 0 3 1
## 2 11 110 0 3 2
## 3 11 110 0 3 2
## 4 11 24 0 3 3
## 5 11 24 0 3 3
## 6 11 72 0 3 1
## 7 11 72 0 3 1
## 8 11 72 0 3 1
## 9 4 108 0 3 1
## 10 4 108 0 3 1
## 11 4 108 0 3 1
## 12 11 72 0 3 1
## Occurrence.in.drift Size.at.maturity Rheophily Thermal.preference a
## 1 3 1 2 2 0.0053
## 2 1 2 1 2 0.0060
## 3 1 2 1 2 0.0040
## 4 1 2 3 2 0.0082
## 5 1 2 3 2 0.0082
## 6 3 1 2 2 0.0053
## 7 3 1 2 2 0.0053
## 8 3 1 2 2 0.0053
## 9 3 1 2 2 0.0018
## 10 3 1 2 2 0.0018
## 11 3 1 2 2 0.0018
## 12 3 1 2 2 0.0053
## b Biomass_log Abundance_log z_Abundance_log
## 1 2.875 1.44343581 5.517772 4.574188
## 2 3.050 1.36134575 5.090328 4.064580
## 3 2.586 1.36134575 5.090328 4.064580
## 4 2.958 0.87516365 4.479419 3.336241
## 5 2.958 0.87516365 4.479419 3.336241
## 6 2.875 -0.80145147 4.438597 3.287573
## 7 2.875 1.36059415 5.434930 4.475422
## 8 2.875 1.19134667 4.438597 3.287573
## 9 2.617 0.08674522 4.592748 3.471354
## 10 2.617 0.80424329 4.726279 3.630553
## 11 2.617 1.37099096 4.815891 3.737391
## 12 2.875 0.27155132 4.345888 3.177042
#visualize abundance by feeding type per basin
ggplot(macros2, aes(x = Basin, y = Abundance, fill = Feeding.Group)) +
geom_col() +
labs(title = "Abundance by Feeding Category and Basin", y = "Abundance", x = "Feeding Category") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
library(ggplot2)
library(dplyr)
library(scales)
# Calculate relative abundance and plot
ggplot(macros, aes(x = Basin, y = Abundance, fill = Feeding.Group)) +
geom_col(position = "fill") + # Stacks bars and normalizes to 1
scale_y_continuous(labels = percent) + # Y-axis as percentages
labs(title = "Relative Abundance of Drift Feeding Groups by Basin",
y = "Relative Abundance",
x = "Basin") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Calculate relative abundance for each Feeding.Group in each Basin
drift_absummary <- macros %>%
group_by(Basin, Feeding.Group) %>%
summarise(Abundance = sum(Abundance), .groups = "drop") %>%
group_by(Basin) %>%
mutate(Relative_Abundance = paste0(round((Abundance / sum(Abundance))*100, 0), "%")) %>%
select(Basin, Feeding.Group, Relative_Abundance)
# Create a table with relative abundance in percentage format
drift_absummary_table <- drift_absummary
print(drift_absummary_table)
## # A tibble: 114 × 3
## # Groups: Basin [20]
## Basin Feeding.Group Relative_Abundance
## <chr> <chr> <chr>
## 1 AA1_REACH CF 5%
## 2 AA1_REACH CG 62%
## 3 AA1_REACH PA 5%
## 4 AA1_REACH PR 4%
## 5 AA1_REACH SC 7%
## 6 AA1_REACH SH 6%
## 7 AA1_REACH UN 12%
## 8 AA2_REACH CF 16%
## 9 AA2_REACH CG 42%
## 10 AA2_REACH PA 11%
## # ℹ 104 more rows
# Aggregate by Basin and Feeding Category
macro_aggregated2 <- macros2 %>%
group_by(Basin, Feeding.Group) %>%
summarise(
total_abundance = sum(Abundance))
## `summarise()` has grouped output by 'Basin'. You can override using the
## `.groups` argument.
peri_macro2 <- macro_aggregated2 %>%
left_join(twoyr_peri, by = "Basin")
#scatter plot of total macroinvertebrate abundance vs periphyton abundance by Basin
ggplot(peri_macro2, aes(x = Total_Avg, y = total_abundance, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Macroinvertebrate Abundance vs Periphyton Biomass by Basin",
x = "Periphyton Biomass", y = "Total Macroinvertebrate Abundance")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
peri_macro2 <- peri_macro2 %>%
mutate(
total_abundance_log = log(total_abundance),
Total_Avg_log = log(Total_Avg)
)
#scale abundance
peri_macro2$total_abundance_log <- scale(peri_macro2$total_abundance_log)
peri_macro2$Total_Avg_log <- scale(peri_macro2$Total_Avg_log)
# Model with Feeding.Group as a fixed effect and random effects for Basin
model_peri_macro2 <- lmer(
total_abundance_log ~ Total_Avg_log * Feeding.Group + (1 | Basin),
data = peri_macro2
)
summary(model_peri_macro2)
## Linear mixed model fit by REML ['lmerMod']
## Formula: total_abundance_log ~ Total_Avg_log * Feeding.Group + (1 | Basin)
## Data: peri_macro2
##
## REML criterion at convergence: 217.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.19154 -0.65683 0.04784 0.53789 2.63320
##
## Random effects:
## Groups Name Variance Std.Dev.
## Basin (Intercept) 0.4059 0.6371
## Residual 0.3402 0.5833
## Number of obs: 101, groups: Basin, 18
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 0.003149 0.226603 0.014
## Total_Avg_log -0.044310 0.240501 -0.184
## Feeding.GroupCG 1.295045 0.220847 5.864
## Feeding.GroupMH -1.623826 0.585451 -2.774
## Feeding.GroupPA -0.414475 0.225209 -1.840
## Feeding.GroupPR -0.488028 0.244150 -1.999
## Feeding.GroupSC -0.119888 0.226150 -0.530
## Feeding.GroupSH -0.216593 0.244290 -0.887
## Feeding.GroupUN -0.028409 0.224463 -0.127
## Total_Avg_log:Feeding.GroupCG 0.122203 0.232590 0.525
## Total_Avg_log:Feeding.GroupMH 0.155456 0.689947 0.225
## Total_Avg_log:Feeding.GroupPA 0.086776 0.232580 0.373
## Total_Avg_log:Feeding.GroupPR -0.161271 0.250265 -0.644
## Total_Avg_log:Feeding.GroupSC 0.137122 0.240000 0.571
## Total_Avg_log:Feeding.GroupSH -0.401843 0.260159 -1.545
## Total_Avg_log:Feeding.GroupUN 0.084468 0.240221 0.352
##
## Correlation matrix not shown by default, as p = 16 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it
# Plot the relationship for each Feeding Group
ggplot(peri_macro2, aes(x = Total_Avg_log, y = total_abundance_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Feeding.Group) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Total Drift Macro Abundance by Feeding Group",
x = "Total Average Periphyton Biomass",
y = "Drift Abundance") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
ggplot(peri_macro2, aes(x = Total_Avg_log, y = total_abundance_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Basin) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Total Drift Macro Abundance by Basin",
x = "Total Average Periphyton Biomass",
y = "Drift Abundance") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
# Histogram of macro biomass by feeding group
ggplot(benthic, aes(x = Biomass, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 572 rows containing non-finite outside the scale range
## (`stat_bin()`).
## Warning: Removed 18 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(benthic$Biomass) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: benthic$Biomass
## W = 0.01987, p-value < 2.2e-16
# QQ Plot
qqnorm(benthic$Biomass)
qqline(benthic$Biomass, col = "forestgreen")
#Needs transformation
benthic <- benthic %>%
mutate(
Biomass_log = log(Biomass)
)
# Histogram of macro biomass by feeding group
ggplot(benthic, aes(x = Biomass_log, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 1907 rows containing non-finite outside the scale range (`stat_bin()`).
## Removed 18 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(benthic$Biomass_log) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: benthic$Biomass_log
## W = 0.99327, p-value = 2.609e-11
# QQ Plot
qqnorm(benthic$Biomass_log)
qqline(benthic$Biomass_log, col = "blueviolet")
# Calculate Z-scores
z_benthic <- benthic %>% mutate(z_Biomass_log = scale(Biomass_log))
# Find extreme outliers (Z-score > 3 or < -3)
z_benthic %>% filter(abs(z_Biomass_log) > 3)
## Basin Location Date Taxon Stage Insect Origin
## 1 CC_REACH reach 45524 Calineuria californica L insect Aquatic
## 2 CP_REACH reach 45510 Pteronarcys princeps L insect Aquatic
## 3 CZ2_REACH reach 45504 Calineuria californica L insect Aquatic
## 4 DA1_REACH reach 45526 Pacifastacus U non-insect Aquatic
## 5 DA2_REACH reach 45526 Pacifastacus U non-insect Aquatic
## 6 DC_REACH reach 45505 Pacifastacus U non-insect Aquatic
## 7 DZ_REACH reach 45502 Pacifastacus U non-insect Aquatic
## Higher.classification Order Family Common.name Length Abundance
## 1 Arthropoda: Insecta Plecoptera Perlidae stoneflies 21 6.66000
## 2 Arthropoda: Insecta Plecoptera Pteronarcyidae stoneflies 37 2.22000
## 3 Arthropoda: Insecta Plecoptera Perlidae stoneflies 25 4.44000
## 4 Crustacea: Decapoda x Astacidae crayfish 30 2.22000
## 5 Crustacea: Decapoda x Astacidae crayfish 28 2.22000
## 6 Crustacea: Decapoda x Astacidae crayfish 12 2.65164
## 7 Crustacea: Decapoda x Astacidae crayfish 30 2.32600
## Biomass Table3.BCG.Attribute Feeding.Group Table3.CA.feeding.group
## 1 422.4534 2 PR PR
## 2 779.6165 2 OM OM
## 3 465.2511 2 PR PR
## 4 7408.1877 4 OM OM
## 5 5768.5323 4 OM OM
## 6 319.1120 4 OM OM
## 7 7761.9120 4 OM OM
## Table3.Habit Table3.Tolerance Table3.WY.HBI Table3.PSSB.tolerance
## 1 CL 0 2 0
## 2 SP MI 1 0
## 3 CL 0 2 0
## 4 SP 0 6 0
## 5 SP 0 6 0
## 6 SP 0 6 0
## 7 SP 0 6 0
## Table3.CA.tolerance Table3.Metals.tolerance Table3.BCI.TV
## 1 2 11 24
## 2 0 1 24
## 3 2 11 24
## 4 6 3 108
## 5 6 3 108
## 6 6 3 108
## 7 6 3 108
## Table3.PSSB.long.lived Table3.Voltinism Table3.Development
## 1 0 1 2
## 2 0 1 2
## 3 0 1 2
## 4 0 1 3
## 5 0 1 3
## 6 0 1 3
## 7 0 1 3
## Table3.Occurrence.in.drift Table3.Size.at.maturity Table3.Rheophily
## 1 1 3 3
## 2 1 3 2
## 3 1 3 3
## 4 1 3 2
## 5 1 3 2
## 6 1 3 2
## 7 1 3 2
## Table3.Thermal.preference Table3.a Table3.b Biomass_log z_Biomass_log
## 1 2 0.0099 2.879 6.046079 3.254630
## 2 1 0.0324 2.573 6.658802 3.569277
## 3 2 0.0099 2.879 6.142577 3.304184
## 4 2 0.0147 3.626 8.910341 4.725491
## 5 2 0.0147 3.626 8.660173 4.597024
## 6 2 0.0147 3.626 5.765542 3.110568
## 7 2 0.0147 3.626 8.956984 4.749443
#visualize biomass by feeding type per basin
ggplot(benthic, aes(x = Basin, y = Biomass, fill = Feeding.Group)) +
geom_col() +
labs(title = "Benthic Biomass by Feeding Category and Basin", y = "Biomass", x = "Feeding Category") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Calculate relative abundance and plot
ggplot(benthic, aes(x = Basin, y = Biomass, fill = Feeding.Group)) +
geom_col(position = "fill") + # Stacks bars and normalizes to 1
scale_y_continuous(labels = percent) + # Y-axis as percentages
labs(title = "Relative Biomass of Feeding Benthic Groups by Basin",
y = "Relative Biomass",
x = "Basin") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Calculate relative abundance for each Feeding.Group in each Basin
benthic_biosummary <- benthic %>%
group_by(Basin, Feeding.Group) %>%
summarise(Biomass = sum(Biomass), .groups = "drop") %>%
group_by(Basin) %>%
mutate(Relative_Biomass = paste0(round((Biomass / sum(Biomass))*100, 0), "%")) %>%
select(Basin, Feeding.Group, Relative_Biomass)
# Create a table with relative abundance in percentage format
benthic_biosummary_table <- benthic_biosummary
print(benthic_biosummary_table)
## # A tibble: 134 × 3
## # Groups: Basin [20]
## Basin Feeding.Group Relative_Biomass
## <chr> <chr> <chr>
## 1 AA1_REACH CF 4%
## 2 AA1_REACH CG 8%
## 3 AA1_REACH DSH 10%
## 4 AA1_REACH PA 0%
## 5 AA1_REACH PR 73%
## 6 AA1_REACH SC 6%
## 7 AA2_REACH CF 2%
## 8 AA2_REACH CG 29%
## 9 AA2_REACH DSH 1%
## 10 AA2_REACH PA 1%
## # ℹ 124 more rows
# Aggregate by Basin and Feeding Category
benthic_aggregated <- benthic %>%
group_by(Basin, Feeding.Group) %>%
summarise(
total_biomass = sum(Biomass),
mean_biomass = mean(Biomass))
## `summarise()` has grouped output by 'Basin'. You can override using the
## `.groups` argument.
peri_benthic <- benthic_aggregated %>%
left_join(twoyr_peri, by = "Basin")
peri_benthic$total
## Warning: Unknown or uninitialised column: `total`.
## NULL
#scatter plot of total macroinvertebrate biomass vs periphyton biomass by Basin
ggplot(peri_benthic, aes(x = Total_Avg, y = total_biomass, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Benthic Macroinvertebrate Biomass vs Periphyton Biomass by Basin",
x = "Periphyton Biomass", y = "Total Benthic Biomass")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
peri_benthic <- peri_benthic %>%
mutate(
total_biomass_log = log(total_biomass),
mean_biomass_log = log(mean_biomass),
Total_Avg_log = log(Total_Avg)
)
#scale biomass
peri_benthic$total_biomass_log <- scale(peri_benthic$total_biomass_log)
peri_benthic$mean_biomass_log <- scale(peri_benthic$mean_biomass_log)
peri_benthic$Total_Avg_log <- scale(peri_benthic$Total_Avg_log)
# Model with Feeding.Group as a fixed effect and random effects for Basin
model_peri_benthic <- lmer(
total_biomass_log ~ Total_Avg_log * Feeding.Group + (1 | Basin),
data = peri_benthic
)
summary(model_peri_benthic)
## Linear mixed model fit by REML ['lmerMod']
## Formula: total_biomass_log ~ Total_Avg_log * Feeding.Group + (1 | Basin)
## Data: peri_benthic
##
## REML criterion at convergence: 167.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.35094 -0.48172 -0.03527 0.48575 2.74866
##
## Random effects:
## Groups Name Variance Std.Dev.
## Basin (Intercept) 0.06861 0.2619
## Residual 0.15970 0.3996
## Number of obs: 121, groups: Basin, 18
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) -0.32510 0.11271 -2.884
## Total_Avg_log 0.24210 0.11415 2.121
## Feeding.GroupCG 0.79975 0.13331 5.999
## Feeding.GroupDSH 0.06041 0.13331 0.453
## Feeding.GroupHSH -0.68419 0.31094 -2.200
## Feeding.GroupMH -1.68420 0.22036 -7.643
## Feeding.GroupOM 2.05497 0.21329 9.635
## Feeding.GroupPA -0.71464 0.13331 -5.361
## Feeding.GroupPR 1.23052 0.13331 9.230
## Feeding.GroupSC 0.94421 0.13331 7.083
## Total_Avg_log:Feeding.GroupCG 0.01562 0.13501 0.116
## Total_Avg_log:Feeding.GroupDSH -0.25048 0.13501 -1.855
## Total_Avg_log:Feeding.GroupHSH -0.75785 0.23718 -3.195
## Total_Avg_log:Feeding.GroupMH -0.11267 0.23110 -0.488
## Total_Avg_log:Feeding.GroupOM -0.47899 0.20359 -2.353
## Total_Avg_log:Feeding.GroupPA 0.04621 0.13501 0.342
## Total_Avg_log:Feeding.GroupPR -0.03789 0.13501 -0.281
## Total_Avg_log:Feeding.GroupSC 0.01055 0.13501 0.078
##
## Correlation matrix not shown by default, as p = 18 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it
# Plot the relationship for each Feeding Group
ggplot(peri_benthic, aes(x = Total_Avg_log, y = total_biomass_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Feeding.Group) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Benthic Macro Biomass by Feeding Group",
x = "Total Average Periphyton Biomass",
y = "Benthic Biomass") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
ggplot(peri_benthic, aes(x = Total_Avg_log, y = total_biomass_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Basin) + # Create separate plots for each Basin
labs(title = "Relationship between Biennial Periphyton Biomass and Benthic Macro Biomass by Basin",
x = "Total Average Periphyton Biomass",
y = "Benthic Biomass") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
#Histogram of benthic abundance by feeding group
ggplot(benthic, aes(x = Abundance, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 1433 rows containing non-finite outside the scale range
## (`stat_bin()`).
## Warning: Removed 18 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(benthic$Abundance) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: benthic$Abundance
## W = 0.32183, p-value < 2.2e-16
# QQ Plot
qqnorm(benthic$Abundance)
qqline(benthic$Abundance, col = "lightblue3")
#Needs transformation
benthic2 <- benthic %>%
mutate(
Abundance_log = log(Abundance)
)
# Histogram of macro biomass by feeding group
ggplot(benthic2, aes(x = Abundance_log, fill = Feeding.Group)) +
geom_histogram(bins = 10, binwidth = 0.25, alpha = 0.7, position = "identity") +
xlim(0,5) +
theme_minimal()
## Warning: Removed 29 rows containing non-finite outside the scale range (`stat_bin()`).
## Removed 18 rows containing missing values or values outside the scale range
## (`geom_bar()`).
# Shapiro-Wilk test for normality
shapiro.test(benthic2$Abundance_log) # If p < 0.05, data is not normal
##
## Shapiro-Wilk normality test
##
## data: benthic2$Abundance_log
## W = 0.86442, p-value < 2.2e-16
# QQ Plot
qqnorm(benthic2$Abundance_log)
qqline(benthic2$Abundance_log, col = "lightblue3")
# Calculate Z-scores
z_benthic2 <- benthic2 %>% mutate(z_Abundance_log = scale(Abundance_log))
# Find extreme outliers (Z-score > 3 or < -3)
z_benthic2 %>% filter(abs(z_Abundance_log) > 3)
## Basin Location Date Taxon Stage Insect Origin
## 1 AA2_REACH reach 45516 Baetis tricaudatus complex L insect Aquatic
## 2 AA2_REACH reach 45516 Glossosoma L insect Aquatic
## 3 AA2_REACH reach 45516 Glossosoma L insect Aquatic
## 4 AZ_REACH reach 45572 Cinygmula L insect Aquatic
## 5 BA1_REACH reach 45528 Micropsectra L insect Aquatic
## 6 BA1_REACH reach 45528 Micropsectra L insect Aquatic
## 7 BA2_REACH reach 45559 Baetis tricaudatus complex L insect Aquatic
## 8 BA2_REACH reach 45559 Baetis tricaudatus complex L insect Aquatic
## 9 BA2_REACH reach 45559 Baetis tricaudatus complex L insect Aquatic
## 10 BA2_REACH reach 45559 Zapada cinctipes L insect Aquatic
## 11 BC_REACH reach 45512 Glossosoma L insect Aquatic
## 12 BC_REACH reach 45512 Glossosoma L insect Aquatic
## 13 BP_REACH reach 45517 Glossosoma L insect Aquatic
## 14 BP_REACH reach 45517 Glossosoma L insect Aquatic
## 15 BP_REACH reach 45517 Glossosoma L insect Aquatic
## 16 BP_REACH reach 45517 Glossosoma L insect Aquatic
## 17 BP_REACH reach 45517 Glossosoma L insect Aquatic
## 18 BP_REACH reach 45517 Glossosoma L insect Aquatic
## 19 BZ_REACH reach 45511 Baetis tricaudatus complex L insect Aquatic
## 20 BZ_REACH reach 45511 Zapada cinctipes L insect Aquatic
## 21 CA2_REACH reach 45573 Optioservus A insect Aquatic
## 22 CA2_REACH reach 45573 Baetis tricaudatus complex L insect Aquatic
## 23 CA2_REACH reach 45573 Cinygmula L insect Aquatic
## 24 CA2_REACH reach 45573 Cinygmula L insect Aquatic
## 25 CA2_REACH reach 45573 Glossosoma L insect Aquatic
## 26 CA2_REACH reach 45573 Glossosoma L insect Aquatic
## 27 CC_REACH reach 45524 Micropsectra L insect Aquatic
## 28 CC_REACH reach 45524 Micropsectra L insect Aquatic
## 29 CC_REACH reach 45524 Baetis tricaudatus complex L insect Aquatic
## 30 CC_REACH reach 45524 Calineuria californica L insect Aquatic
## 31 CC_REACH reach 45524 Cinygmula L insect Aquatic
## 32 CP_REACH reach 45510 Baetis tricaudatus complex L insect Aquatic
## 33 CP_REACH reach 45510 Baetis tricaudatus complex L insect Aquatic
## 34 CP_REACH reach 45510 Glossosoma L insect Aquatic
## 35 CP_REACH reach 45510 Glossosoma L insect Aquatic
## 36 CZ2_REACH reach 45504 Baetis tricaudatus complex L insect Aquatic
## 37 CZ2_REACH reach 45504 Baetis tricaudatus complex L insect Aquatic
## 38 CZ2_REACH reach 45504 Baetis tricaudatus complex L insect Aquatic
## 39 CZ2_REACH reach 45504 Calineuria californica L insect Aquatic
## 40 DC_REACH reach 45505 Glossosoma L insect Aquatic
## 41 DC_REACH reach 45505 Glossosoma L insect Aquatic
## 42 DZ_REACH reach 45502 Baetis tricaudatus complex L insect Aquatic
## Higher.classification Order Family
## 1 Arthropoda: Insecta Ephemeroptera Baetidae
## 2 Arthropoda: Insecta Trichoptera Glossosomatidae
## 3 Arthropoda: Insecta Trichoptera Glossosomatidae
## 4 Arthropoda: Insecta Ephemeroptera Heptageniidae
## 5 Arthropoda: Insecta Diptera Chironomidae: Chironominae: Tanytarsini
## 6 Arthropoda: Insecta Diptera Chironomidae: Chironominae: Tanytarsini
## 7 Arthropoda: Insecta Ephemeroptera Baetidae
## 8 Arthropoda: Insecta Ephemeroptera Baetidae
## 9 Arthropoda: Insecta Ephemeroptera Baetidae
## 10 Arthropoda: Insecta Plecoptera Nemouridae
## 11 Arthropoda: Insecta Trichoptera Glossosomatidae
## 12 Arthropoda: Insecta Trichoptera Glossosomatidae
## 13 Arthropoda: Insecta Trichoptera Glossosomatidae
## 14 Arthropoda: Insecta Trichoptera Glossosomatidae
## 15 Arthropoda: Insecta Trichoptera Glossosomatidae
## 16 Arthropoda: Insecta Trichoptera Glossosomatidae
## 17 Arthropoda: Insecta Trichoptera Glossosomatidae
## 18 Arthropoda: Insecta Trichoptera Glossosomatidae
## 19 Arthropoda: Insecta Ephemeroptera Baetidae
## 20 Arthropoda: Insecta Plecoptera Nemouridae
## 21 Arthropoda: Insecta Coleoptera Elmidae
## 22 Arthropoda: Insecta Ephemeroptera Baetidae
## 23 Arthropoda: Insecta Ephemeroptera Heptageniidae
## 24 Arthropoda: Insecta Ephemeroptera Heptageniidae
## 25 Arthropoda: Insecta Trichoptera Glossosomatidae
## 26 Arthropoda: Insecta Trichoptera Glossosomatidae
## 27 Arthropoda: Insecta Diptera Chironomidae: Chironominae: Tanytarsini
## 28 Arthropoda: Insecta Diptera Chironomidae: Chironominae: Tanytarsini
## 29 Arthropoda: Insecta Ephemeroptera Baetidae
## 30 Arthropoda: Insecta Plecoptera Perlidae
## 31 Arthropoda: Insecta Ephemeroptera Heptageniidae
## 32 Arthropoda: Insecta Ephemeroptera Baetidae
## 33 Arthropoda: Insecta Ephemeroptera Baetidae
## 34 Arthropoda: Insecta Trichoptera Glossosomatidae
## 35 Arthropoda: Insecta Trichoptera Glossosomatidae
## 36 Arthropoda: Insecta Ephemeroptera Baetidae
## 37 Arthropoda: Insecta Ephemeroptera Baetidae
## 38 Arthropoda: Insecta Ephemeroptera Baetidae
## 39 Arthropoda: Insecta Plecoptera Perlidae
## 40 Arthropoda: Insecta Trichoptera Glossosomatidae
## 41 Arthropoda: Insecta Trichoptera Glossosomatidae
## 42 Arthropoda: Insecta Ephemeroptera Baetidae
## Common.name Length Abundance Biomass Table3.BCG.Attribute
## 1 mayflies 1.5 160.8662 2.7352872 4
## 2 caddisflies 1.5 238.2057 6.4810276 3
## 3 caddisflies 1.5 238.2057 6.4810276 3
## 4 mayflies 1.5 142.1244 4.6886491 3
## 5 midges 3.0 115.1514 3.6742445 4
## 6 midges 3.0 115.1514 3.6742445 4
## 7 mayflies 1.5 185.7252 3.1579778 4
## 8 mayflies 2.0 248.9508 9.6794489 4
## 9 mayflies 2.5 138.3060 10.2139662 4
## 10 stoneflies 2.0 165.9672 6.3045652 4
## 11 caddisflies 1.5 159.8400 4.3488784 3
## 12 caddisflies 1.5 159.8400 4.3488784 3
## 13 caddisflies 1.0 111.6482 0.9155156 3
## 14 caddisflies 1.0 111.6482 0.9155156 3
## 15 caddisflies 1.5 451.6679 12.2888432 3
## 16 caddisflies 1.5 451.6679 12.2888432 3
## 17 caddisflies 2.0 380.6190 24.2521945 3
## 18 caddisflies 2.0 380.6190 24.2521945 3
## 19 mayflies 2.0 142.1244 5.5259347 4
## 20 stoneflies 1.5 258.4080 4.4346366 4
## 21 riffle beetles 3.0 114.5964 94.6810962 4
## 22 mayflies 1.5 201.5316 3.4267419 4
## 23 mayflies 1.5 438.6276 14.4702168 3
## 24 mayflies 2.5 189.6768 25.5484616 3
## 25 caddisflies 2.0 213.3864 13.5965059 3
## 26 caddisflies 2.0 213.3864 13.5965059 3
## 27 midges 5.0 187.3458 22.7573122 4
## 28 midges 5.0 187.3458 22.7573122 4
## 29 mayflies 2.0 125.9739 4.8979876 4
## 30 stoneflies 2.5 151.8147 21.0193027 2
## 31 mayflies 1.5 119.5137 3.9427276 3
## 32 mayflies 2.0 462.3372 17.9761195 4
## 33 mayflies 3.0 189.6768 23.6599428 4
## 34 caddisflies 2.0 165.9672 10.5750602 3
## 35 caddisflies 2.0 165.9672 10.5750602 3
## 36 mayflies 1.5 197.5800 3.3595509 4
## 37 mayflies 2.0 217.3380 8.4503126 4
## 38 mayflies 2.5 162.0156 11.9649318 4
## 39 stoneflies 3.0 134.3544 31.4426529 2
## 40 caddisflies 1.5 111.3689 3.0300908 3
## 41 caddisflies 1.5 111.3689 3.0300908 3
## 42 mayflies 1.5 148.8640 2.5312085 4
## Feeding.Group Table3.CA.feeding.group Table3.Habit Table3.Tolerance
## 1 CG CG CL 0
## 2 SC SC CL 0
## 3 SC SC CL 0
## 4 SC SC CL 0
## 5 CG CG CL 0
## 6 CG CG CL 0
## 7 CG CG CL 0
## 8 CG CG CL 0
## 9 CG CG CL 0
## 10 DSH SH SP 0
## 11 SC SC CL 0
## 12 SC SC CL 0
## 13 SC SC CL 0
## 14 SC SC CL 0
## 15 SC SC CL 0
## 16 SC SC CL 0
## 17 SC SC CL 0
## 18 SC SC CL 0
## 19 CG CG CL 0
## 20 DSH SH SP 0
## 21 CG SC CL MT
## 22 CG CG CL 0
## 23 SC SC CL 0
## 24 SC SC CL 0
## 25 SC SC CL 0
## 26 SC SC CL 0
## 27 CG CG CL 0
## 28 CG CG CL 0
## 29 CG CG CL 0
## 30 PR PR CL 0
## 31 SC SC CL 0
## 32 CG CG CL 0
## 33 CG CG CL 0
## 34 SC SC CL 0
## 35 SC SC CL 0
## 36 CG CG CL 0
## 37 CG CG CL 0
## 38 CG CG CL 0
## 39 PR PR CL 0
## 40 SC SC CL 0
## 41 SC SC CL 0
## 42 CG CG CL 0
## Table3.WY.HBI Table3.PSSB.tolerance Table3.CA.tolerance
## 1 6 0 6
## 2 1 0 1
## 3 1 0 1
## 4 4 0 4
## 5 7 0 7
## 6 7 0 7
## 7 6 0 6
## 8 6 0 6
## 9 6 0 6
## 10 2 0 2
## 11 1 0 1
## 12 1 0 1
## 13 1 0 1
## 14 1 0 1
## 15 1 0 1
## 16 1 0 1
## 17 1 0 1
## 18 1 0 1
## 19 6 0 6
## 20 2 0 2
## 21 4 0 4
## 22 6 0 6
## 23 4 0 4
## 24 4 0 4
## 25 1 0 1
## 26 1 0 1
## 27 7 0 7
## 28 7 0 7
## 29 6 0 6
## 30 2 0 2
## 31 4 0 4
## 32 6 0 6
## 33 6 0 6
## 34 1 0 1
## 35 1 0 1
## 36 6 0 6
## 37 6 0 6
## 38 6 0 6
## 39 2 0 2
## 40 1 0 1
## 41 1 0 1
## 42 6 0 6
## Table3.Metals.tolerance Table3.BCI.TV Table3.PSSB.long.lived
## 1 11 72 0
## 2 11 24 0
## 3 11 24 0
## 4 11 21 0
## 5 1 108 0
## 6 1 108 0
## 7 11 72 0
## 8 11 72 0
## 9 11 72 0
## 10 3 16 0
## 11 11 24 0
## 12 11 24 0
## 13 11 24 0
## 14 11 24 0
## 15 11 24 0
## 16 11 24 0
## 17 11 24 0
## 18 11 24 0
## 19 11 72 0
## 20 3 16 0
## 21 3 104 0
## 22 11 72 0
## 23 11 21 0
## 24 11 21 0
## 25 11 24 0
## 26 11 24 0
## 27 1 108 0
## 28 1 108 0
## 29 11 72 0
## 30 11 24 0
## 31 11 21 0
## 32 11 72 0
## 33 11 72 0
## 34 11 24 0
## 35 11 24 0
## 36 11 72 0
## 37 11 72 0
## 38 11 72 0
## 39 11 24 0
## 40 11 24 0
## 41 11 24 0
## 42 11 72 0
## Table3.Voltinism Table3.Development Table3.Occurrence.in.drift
## 1 3 1 3
## 2 3 3 1
## 3 3 3 1
## 4 2 1 2
## 5 3 1 3
## 6 3 1 3
## 7 3 1 3
## 8 3 1 3
## 9 3 1 3
## 10 2 1 1
## 11 3 3 1
## 12 3 3 1
## 13 3 3 1
## 14 3 3 1
## 15 3 3 1
## 16 3 3 1
## 17 3 3 1
## 18 3 3 1
## 19 3 1 3
## 20 2 1 1
## 21 1 3 2
## 22 3 1 3
## 23 2 1 2
## 24 2 1 2
## 25 3 3 1
## 26 3 3 1
## 27 3 1 3
## 28 3 1 3
## 29 3 1 3
## 30 1 2 1
## 31 2 1 2
## 32 3 1 3
## 33 3 1 3
## 34 3 3 1
## 35 3 3 1
## 36 3 1 3
## 37 3 1 3
## 38 3 1 3
## 39 1 2 1
## 40 3 3 1
## 41 3 3 1
## 42 3 1 3
## Table3.Size.at.maturity Table3.Rheophily Table3.Thermal.preference Table3.a
## 1 1 2 2 0.00530
## 2 2 3 2 0.00820
## 3 2 3 2 0.00820
## 4 2 3 1 0.01080
## 5 1 2 2 0.00180
## 6 1 2 2 0.02100
## 7 1 2 2 0.00530
## 8 1 2 2 0.00530
## 9 1 2 2 0.00530
## 10 2 2 2 0.00560
## 11 2 3 2 0.00820
## 12 2 3 2 0.00820
## 13 2 3 2 0.00820
## 14 2 3 2 0.00820
## 15 2 3 2 0.00820
## 16 2 3 2 0.00820
## 17 2 3 2 0.00820
## 18 2 3 2 0.00820
## 19 1 2 2 0.00530
## 20 2 2 2 0.00560
## 21 1 2 2 0.00093
## 22 1 2 2 0.00530
## 23 2 3 1 0.01080
## 24 2 3 1 0.01080
## 25 2 3 2 0.00820
## 26 2 3 2 0.00820
## 27 1 2 2 0.00180
## 28 1 2 2 0.02100
## 29 1 2 2 0.00530
## 30 3 3 2 0.00990
## 31 2 3 1 0.01080
## 32 1 2 2 0.00530
## 33 1 2 2 0.00530
## 34 2 3 2 0.00820
## 35 2 3 2 0.00820
## 36 1 2 2 0.00530
## 37 1 2 2 0.00530
## 38 1 2 2 0.00530
## 39 3 3 2 0.00990
## 40 2 3 2 0.00820
## 41 2 3 2 0.00820
## 42 1 2 2 0.00530
## Table3.b Biomass_log Abundance_log z_Abundance_log
## 1 2.875 1.00623644 5.080573 3.407488
## 2 2.958 1.86887908 5.473134 3.811348
## 3 2.958 1.86887908 5.473134 3.811348
## 4 2.754 1.54514449 4.956703 3.280054
## 5 2.617 1.30134753 4.746248 3.063542
## 6 2.081 1.30134753 4.746248 3.063542
## 7 2.875 1.14993189 5.224268 3.555319
## 8 2.875 2.27000497 5.517255 3.856738
## 9 2.875 2.32375602 4.929469 3.252036
## 10 2.762 1.84127401 5.111790 3.439604
## 11 2.958 1.46991798 5.074173 3.400905
## 12 2.958 1.46991798 5.074173 3.400905
## 13 2.958 -0.08826791 4.715353 3.031758
## 14 2.958 -0.08826791 4.715353 3.031758
## 15 2.958 2.50869180 6.112947 4.469573
## 16 2.958 2.50869180 6.112947 4.469573
## 17 2.958 3.18850711 5.941799 4.293499
## 18 2.958 3.18850711 5.941799 4.293499
## 19 2.875 1.70945242 4.956703 3.280054
## 20 2.762 1.48944568 5.554540 3.895096
## 21 6.180 4.55051436 4.741416 3.058572
## 22 2.875 1.23160992 5.305946 3.639348
## 23 2.754 2.67209252 6.083651 4.439434
## 24 2.754 3.24057710 5.245322 3.576979
## 25 2.958 2.60981284 5.363105 3.698151
## 26 2.958 2.60981284 5.363105 3.698151
## 27 2.617 3.12488651 5.232956 3.564257
## 28 2.081 3.12488651 5.232956 3.564257
## 29 2.875 1.58882443 4.836075 3.155954
## 30 2.879 3.04544119 5.022661 3.347910
## 31 2.754 1.37187277 4.783431 3.101795
## 32 2.875 2.88904418 6.136294 4.493592
## 33 2.875 3.16378344 5.245322 3.576979
## 34 2.958 2.35849841 5.111790 3.439604
## 35 2.958 2.35849841 5.111790 3.439604
## 36 2.875 1.21180729 5.286144 3.618975
## 37 2.875 2.13420343 5.381454 3.717028
## 38 2.875 2.48198002 5.087693 3.414813
## 39 2.879 3.44816534 4.900481 3.222214
## 40 2.958 1.10859260 4.712848 3.029181
## 41 2.958 1.10859260 4.712848 3.029181
## 42 2.875 0.92869686 5.003033 3.327717
#visualize abundance by feeding type per basin
ggplot(benthic2, aes(x = Basin, y = Abundance, fill = Feeding.Group)) +
geom_col() +
labs(title = "Abundance by Feeding Category and Basin", y = "Abundance", x = "Feeding Category") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
library(ggplot2)
library(dplyr)
library(scales)
# Calculate relative abundance and plot
ggplot(benthic, aes(x = Basin, y = Abundance, fill = Feeding.Group)) +
geom_col(position = "fill") + # Stacks bars and normalizes to 1
scale_y_continuous(labels = percent) + # Y-axis as percentages
labs(title = "Relative Abundance of Benthic Feeding Groups by Basin",
y = "Relative Abundance",
x = "Basin") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Calculate relative abundance for each Feeding.Group in each Basin
benthic_absummary <- benthic %>%
group_by(Basin, Feeding.Group) %>%
summarise(Abundance = sum(Abundance), .groups = "drop") %>%
group_by(Basin) %>%
mutate(Relative_Abundance = paste0(round((Abundance / sum(Abundance))*100, 0), "%")) %>%
select(Basin, Feeding.Group, Relative_Abundance)
# Create a table with relative abundance in percentage format
benthic_absummary_table <- benthic_absummary
print(benthic_absummary_table)
## # A tibble: 134 × 3
## # Groups: Basin [20]
## Basin Feeding.Group Relative_Abundance
## <chr> <chr> <chr>
## 1 AA1_REACH CF 7%
## 2 AA1_REACH CG 28%
## 3 AA1_REACH DSH 21%
## 4 AA1_REACH PA 2%
## 5 AA1_REACH PR 33%
## 6 AA1_REACH SC 9%
## 7 AA2_REACH CF 7%
## 8 AA2_REACH CG 40%
## 9 AA2_REACH DSH 4%
## 10 AA2_REACH PA 5%
## # ℹ 124 more rows
# Aggregate by Basin and Feeding Category
benthic_aggregated2 <- benthic2 %>%
group_by(Basin, Feeding.Group) %>%
summarise(
total_abundance = sum(Abundance))
## `summarise()` has grouped output by 'Basin'. You can override using the
## `.groups` argument.
peri_benthic2 <- benthic_aggregated2 %>%
left_join(twoyr_peri, by = "Basin")
#scatter plot of total macroinvertebrate abundance vs periphyton abundance by Basin
ggplot(peri_benthic2, aes(x = Total_Avg, y = total_abundance, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Benthic Macroinvertebrate Abundance vs Periphyton Biomass by Basin",
x = "Periphyton Biomass", y = "Total Benthic Abundance")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
peri_benthic2 <- peri_benthic2 %>%
mutate(
total_abundance_log = log(total_abundance),
Total_Avg_log = log(Total_Avg)
)
peri_benthic2$total_abundance_log <- scale(peri_benthic2$total_abundance_log)
peri_benthic2$Total_Avg_log <- scale(peri_benthic2$Total_Avg_log)
# Model with Feeding.Group as a fixed effect and random effects for Basin
model_peri_benthic2 <- lmer(
total_abundance_log ~ Total_Avg_log * Feeding.Group + (1 | Basin),
data = peri_benthic2
)
summary(model_peri_benthic2)
## Linear mixed model fit by REML ['lmerMod']
## Formula: total_abundance_log ~ Total_Avg_log * Feeding.Group + (1 | Basin)
## Data: peri_benthic2
##
## REML criterion at convergence: 166.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.96268 -0.63459 -0.01771 0.53631 2.26551
##
## Random effects:
## Groups Name Variance Std.Dev.
## Basin (Intercept) 0.1517 0.3895
## Residual 0.1415 0.3761
## Number of obs: 121, groups: Basin, 18
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) -0.24323 0.12772 -1.904
## Total_Avg_log 0.24144 0.12936 1.866
## Feeding.GroupCG 1.28229 0.12548 10.219
## Feeding.GroupDSH 0.16715 0.12548 1.332
## Feeding.GroupHSH -1.51255 0.29492 -5.129
## Feeding.GroupMH -1.34619 0.20876 -6.449
## Feeding.GroupOM -1.80991 0.20199 -8.961
## Feeding.GroupPA -0.37831 0.12548 -3.015
## Feeding.GroupPR 0.64566 0.12548 5.146
## Feeding.GroupSC 0.91954 0.12548 7.328
## Total_Avg_log:Feeding.GroupCG 0.06344 0.12708 0.499
## Total_Avg_log:Feeding.GroupDSH -0.21790 0.12708 -1.715
## Total_Avg_log:Feeding.GroupHSH -0.06910 0.22464 -0.308
## Total_Avg_log:Feeding.GroupMH -0.27534 0.21899 -1.257
## Total_Avg_log:Feeding.GroupOM -0.32567 0.19273 -1.690
## Total_Avg_log:Feeding.GroupPA 0.14587 0.12708 1.148
## Total_Avg_log:Feeding.GroupPR -0.05496 0.12708 -0.432
## Total_Avg_log:Feeding.GroupSC 0.17954 0.12708 1.413
##
## Correlation matrix not shown by default, as p = 18 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it
# Plot the relationship for each Feeding Group
ggplot(peri_benthic2, aes(x = Total_Avg_log, y = total_abundance_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Feeding.Group) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Total Benthic Macro Abundance by Feeding Group",
x = "Total Average Periphyton Biomass",
y = "Benthic Abundance") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
ggplot(peri_benthic2, aes(x = Total_Avg_log, y = total_abundance_log, color = Feeding.Group)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) + # Linear regression line
facet_wrap(~ Basin) + # Create separate plots for each Feeding Group
labs(title = "Relationship between Biennial Periphyton Biomass and Total Benthic Macro Abundance by Basin",
x = "Total Average Periphyton Biomass",
y = "Benthic Abundance") +
theme_minimal()
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 13 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Removed 13 rows containing missing values or values outside the scale range
## (`geom_point()`).
# Log-transform the biomasses if they aren't already
library(ggplot2)
library(dplyr)
library(dendextend)
## Warning: package 'dendextend' was built under R version 4.4.3
## Registered S3 method overwritten by 'dendextend':
## method from
## rev.hclust vegan
##
## ---------------------
## Welcome to dendextend version 1.19.0
## Type citation('dendextend') for how to cite the package.
##
## Type browseVignettes(package = 'dendextend') for the package vignette.
## The github page is: https://github.com/talgalili/dendextend/
##
## Suggestions and bug-reports can be submitted at: https://github.com/talgalili/dendextend/issues
## You may ask questions at stackoverflow, use the r and dendextend tags:
## https://stackoverflow.com/questions/tagged/dendextend
##
## To suppress this message use: suppressPackageStartupMessages(library(dendextend))
## ---------------------
##
## Attaching package: 'dendextend'
## The following object is masked from 'package:permute':
##
## shuffle
## The following object is masked from 'package:ggpubr':
##
## rotate
## The following object is masked from 'package:stats':
##
## cutree
library(cluster)
library(factoextra)
## Warning: package 'factoextra' was built under R version 4.4.3
## Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBa
# Fish data - Add DA1/CA1 to line up
new_row <- data.frame(
Basin = "DA1_REACH",
total_biomass_24 = 0,
Yearly_Avg_24 = 0,
biomass_reach = 0
)
new_row2 <- data.frame(
Basin = "CA1_REACH",
total_biomass_24 = 0,
Yearly_Avg_24 = 0,
biomass_reach = 0
)
# Insert new_row2 at row 10
avg_biomass24_summarized2 <- bind_rows(
avg_biomass24_summarized[1:9, ], # Rows 1 to 9
new_row2, # Insert new_row2 at row 10
avg_biomass24_summarized[10:nrow(avg_biomass24_summarized), ] # Remaining rows
)
# Insert new_row at row 14 (after the shift)
avg_biomass24_summarized2 <- bind_rows(
avg_biomass24_summarized2[1:13, ], # Rows 1 to 13 (now shifted)
new_row, # Insert new_row at row 14
avg_biomass24_summarized2[14:nrow(avg_biomass24_summarized2), ] # Remaining rows
)
# Reset row indices
rownames(avg_biomass24_summarized2) <- NULL
fish_long <- avg_biomass24_summarized2 %>%
mutate(
Category = "Fish",
Biomass = sqrt(biomass_reach +1), #sqrt transform
Biomass = ((Biomass - min(Biomass, na.rm = TRUE)) / # Min-max scaling
(max(Biomass, na.rm = TRUE) - min(Biomass, na.rm = TRUE))) + 1 )
# Periphyton data
peri_long <- twoyr_peri %>%
mutate(
Category = "Periphyton",
Biomass = sqrt(Total_Avg+1), # sqrt transformation
Biomass = ((Biomass - min(Biomass, na.rm = TRUE)) / # Min-max scaling
(max(Biomass, na.rm = TRUE) - min(Biomass, na.rm = TRUE)))+1)
# Drift data
drift_long <- macros %>%
group_by(Basin) %>% # Group by Basin first!
summarize(
Biomass = sum(Biomass, na.rm = TRUE) # Summing original biomass first
) %>%
ungroup() %>%
mutate(
Biomass = sqrt(Biomass+1),
Biomass = ((Biomass - min(Biomass, na.rm = TRUE)) / # Min-max scaling
(max(Biomass, na.rm = TRUE) - min(Biomass, na.rm = TRUE))
) +1,
Category = "Drift" # Assign category
)
# Benthic data
benthic_long <- benthic %>%
group_by(Basin) %>% # Group by Basin first!
summarize(
Biomass = sum(Biomass, na.rm = TRUE) # Summing original biomass first
) %>%
ungroup() %>%
mutate(
Biomass = sqrt(Biomass+1),
Biomass = ((Biomass - min(Biomass, na.rm = TRUE)) / # Min-max scaling
(max(Biomass, na.rm = TRUE) - min(Biomass, na.rm = TRUE))
) +1,
Category = "Benthic" # Assign category
)
# Bind rows to create a single long-format dataset
all_data <- bind_rows(fish_long, peri_long, drift_long, benthic_long) %>%
group_by(Basin, Category)
all_data <- all_data %>%
select(-X, -X.1, -X.2) # Remove empty columns
# Create the bar plot
ggplot(all_data, aes(x = Basin, y = Biomass, fill = Category)) +
geom_col(position = "dodge", alpha = 0.8) + # Creates side-by-side bars for each biomass type
labs(
title = "Biomass Comparison Across Basins",
x = "Basin",
y = "Biomass",
fill = "Biomass Type"
) +
scale_fill_manual(values = c("darkblue", "forestgreen", "lightblue3", "orange")) + # Adjust colors
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x-axis labels for readability
ggplot(all_data, aes(x = Basin, y = Biomass, color = Category, group = Category)) +
geom_line(size = 1) + # Connects data points with lines
geom_point(size = 3, alpha = 0.8) + # Adds points for clarity
labs(
title = "Biomass Trends Across Basins",
x = "Basin",
y = "Biomass",
color = "Biomass Type"
) +
scale_color_manual(values = c("darkblue", "forestgreen", "lightblue3", "orange")) + # Adjust colors
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x-axis labels for readability
## 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.
ggplot(all_data, aes(x = Basin, y = Biomass, fill = Category)) +
geom_col(position = "dodge", alpha = 0.8) + # Creates side-by-side bars for each biomass type
facet_wrap(~Category, scales = "free_y") + # Creates a facet for each biomass type with its own y-axis
labs(
title = "Biomass Comparison Across Basins",
x = "Basin",
y = "Biomass",
fill = "Biomass Type"
) +
scale_fill_manual(values = c("darkblue", "forestgreen", "lightblue3", "orange")) + # Adjust colors
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x-axis labels for readability
ggplot(all_data, aes(x = Basin, y = Biomass, color = Category, group = Category)) +
geom_line(size = 1) + # Connects data points with lines
geom_point(size = 3, alpha = 0.8) + # Adds points for clarity
facet_wrap(~Category, scales = "free_y") + # Creates a facet for each biomass type with its own y-axis
labs(
title = "Biomass Trends Across Basins",
x = "Basin",
y = "Biomass",
color = "Biomass Type"
) +
scale_color_manual(values = c("darkblue", "forestgreen", "lightblue3", "orange")) + # Adjust colors
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x-axis labels for readability
# Reshape the data to wide format for clustering
for_clust <- all_data %>%
group_by(Basin, Category) %>% # Group by Basin and Category
summarize(Biomass = mean(Biomass, na.rm = TRUE)) %>% # Calculate the mean Biomass per group
pivot_wider(names_from = Category, values_from = Biomass) %>% # Pivot to wide format
drop_na() # Drop rows with NA values
## `summarise()` has grouped output by 'Basin'. You can override using the
## `.groups` argument.
# Ensure all columns except Basin are numeric for clustering
for_clust_numeric <- for_clust %>%
ungroup() %>%
select(-Basin) %>% # Remove Basin column
mutate(across(everything(), as.numeric)) # Convert all columns to numeric
# Remove rows with any NA values
for_clust_numeric <- na.omit(for_clust_numeric)
# Calculate the distance matrix (Euclidean distance)
dist_matrix <- dist(for_clust_numeric, method = "euclidean")
# Perform hierarchical clustering (Ward's method)
final_clust <- hclust(dist_matrix, method = "ward.D2")
# Determine the optimal number of clusters using the Gap Statistic
gap_stat <- clusGap(for_clust_numeric, FUN = hcut, nstart = 25, K.max = 10, B = 50)
fviz_gap_stat(gap_stat) # Visualize Gap Statistic to determine the optimal number of clusters
# Cut the dendrogram into the desired number of clusters (4 in this case)
groups <- cutree(final_clust, k = 4)
# Add cluster labels to the original data, preserving 'Basin'
final_data <- cbind(for_clust, cluster = groups)
# View the table of cluster assignments
table(final_data$cluster)
##
## 1 2 3 4
## 6 8 1 3
# Print the final data with Basin and cluster assignments
print(final_data)
## # A tibble: 18 × 6
## # Groups: Basin [18]
## Basin Benthic Drift Fish Periphyton cluster
## <chr> <dbl> <dbl> <dbl> <dbl> <int>
## 1 AA1_REACH 1.30 1.31 1.22 1.24 1
## 2 AA2_REACH 1.17 1.08 1.60 1.51 2
## 3 AC_REACH 1.16 1.18 2 1.88 2
## 4 AP_REACH 1.04 1.14 1.28 1 1
## 5 AZ_REACH 1.21 1.46 1.75 1.85 2
## 6 BA1_REACH 1.25 1.24 1.60 1.50 2
## 7 BA2_REACH 1.26 1.07 1.78 1.45 2
## 8 BC_REACH 1.24 1.08 1.73 1.48 2
## 9 BZ_REACH 1.28 1.02 1.56 1.89 2
## 10 CA1_REACH 1 1 1 1.30 1
## 11 CA2_REACH 1.28 1.24 1.77 1.54 2
## 12 CP_REACH 1.46 1.90 1.83 2 3
## 13 CZ2_REACH 1.32 1.53 1.36 1.57 1
## 14 DA1_REACH 1.91 2 1 1.16 4
## 15 DA2_REACH 1.82 1.87 1.56 1.07 4
## 16 DC_REACH 1.27 1.31 1.38 1.26 1
## 17 DP_REACH 1.08 1.12 1.19 1.04 1
## 18 DZ_REACH 2 1.42 1.24 1.38 4
# Plot the hierarchical clustering dendrogram
plot(final_clust, labels = final_data$Basin, main = "Trophic Biomass Dendrogram by Basin")
rect.hclust(final_clust, k = 4, border = 2:5) # Adds rectangles around the 4 clusters
# Create a heatmap of the clustered data
cluster_matrix <- final_data %>%
ungroup()%>%
select(-Basin, -cluster) %>% # Remove Basin and cluster columns for heatmap
as.data.frame()
# Ensure all columns are numeric
cluster_matrix <- cluster_matrix %>%
mutate(across(everything(), as.numeric))
# Calculate the distance matrix for the heatmap
dist_matrix <- dist(cluster_matrix, method = "euclidean")
#clustering
final_clust_heatmap <- hclust(dist_matrix, method = "ward.D2")
# Create a heatmap of the clustering result
pheatmap(cluster_matrix,
clustering_method = "ward.D2",
cluster_rows = final_clust_heatmap,
cluster_cols = TRUE,
show_rownames = TRUE,
show_colnames = TRUE,
main = "Heatmap of Trophic Biomass Clusters",
labels_row = final_data$Basin)
library(factoextra)
#visualize environmental variables for distribution - transformation needs
# Identify numeric columns
num_cols <- names(env)[sapply(env, is.numeric)]
# Exclude 'Basin' if it's numeric (e.g., an ID or name)
num_cols <- setdiff(num_cols, "Basin")
# Loop through and plot histograms
for (col in num_cols) {
print(
ggplot(env, aes_string(x = col)) +
geom_histogram(bins = 30, fill = "lightblue", color = "black") +
theme_minimal() +
ggtitle(paste("Histogram of", col))
)
}
## Warning: `aes_string()` was deprecated in ggplot2 3.0.0.
## ℹ Please use tidy evaluation idioms with `aes()`.
## ℹ See also `vignette("ggplot2-in-packages")` for more information.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
# Squareroot transformation to handle true zeroes and min-max scale all numeric columns except 'Basin'
env_scaled <- env %>%
mutate(across(
-Basin, # Exclude Basin column
~ sqrt(. + 1), # Apply square root transformation (adding 1 to handle zeros)
.names = "{.col}" # Keep the original column names after the transformation
)) %>%
mutate(across(
-Basin, # Exclude Basin column
~ (. - min(.)) / (max(.) - min(.)), # Apply min-max scaling
.names = "{.col}" # Keep the original column names after scaling
))
#Principal component analysis of environmental variables to determine if model reduction is needed/multicollinearity is present
pca_env <- env_scaled[,-c(1)]
pca_matrix <- cor(pca_env)
pca_result <- prcomp(pca_matrix)
summary(pca_result)
## Importance of components:
## PC1 PC2 PC3 PC4 PC5 PC6 PC7
## Standard deviation 1.4041 0.8590 0.53872 0.36318 0.24235 0.14574 0.11777
## Proportion of Variance 0.6094 0.2281 0.08971 0.04077 0.01816 0.00657 0.00429
## Cumulative Proportion 0.6094 0.8375 0.92718 0.96795 0.98611 0.99268 0.99696
## PC8 PC9 PC10 PC11 PC12 PC13 PC14
## Standard deviation 0.08057 0.05367 0.0178 0.01096 0.00413 0.001216 5.28e-05
## Proportion of Variance 0.00201 0.00089 0.0001 0.00004 0.00001 0.000000 0.00e+00
## Cumulative Proportion 0.99897 0.99986 1.0000 0.99999 1.00000 1.000000 1.00e+00
## PC15
## Standard deviation 1.466e-17
## Proportion of Variance 0.000e+00
## Cumulative Proportion 1.000e+00
pca_result$rotation[,1:2]
## PC1 PC2
## Avg_Flow -0.31834065 -0.12029552
## Flow_Variance -0.30652574 -0.11814831
## Avg_Sand 0.17357840 0.30709036
## Avg_Gravel -0.13149148 0.44400045
## Avg_Cobble -0.15650163 -0.27803500
## Avg_Boulder 0.10379102 -0.46689277
## Avg_Bedrock 0.16274930 -0.24502044
## Avg_Organic 0.27931778 0.15600948
## Mean_Slope_Degrees 0.24427454 -0.08039395
## Dist_Pour -0.31627013 0.20488476
## Dist_Head -0.36068004 -0.15179631
## Total_length -0.41021991 -0.02653216
## Reach_length 0.06034157 -0.29978199
## Relative_position 0.04853392 -0.37047483
## Percent_occupied 0.39172270 -0.03398395
fviz_eig(pca_result, addlabels = TRUE)
fviz_cos2(pca_result, choice = "var", axes = 1:2, title = "Combined contribution by variable")
fviz_pca_var(pca_result, col.var = "cos2", gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"), repel = TRUE, title = "PCA Biplot of Scaled Environmental Predictor Variables")
#Reach length can be removed. Only collinear variables are Flow Variance, Avg_Flow and Distance to Headwaters. Opting to remove Flow Variance since Distance to Headwaters significantly different.
# Prep data
merged <- merge(for_clust, env_scaled, by = "Basin")
# Split data ONCE and reuse
set.seed(123)
split_ind <- sample(1:2, size = nrow(merged), replace = TRUE)
# Train control parameters
custom <- trainControl(method = "repeatedcv",
number = 5,
repeats = 5,
verboseIter = TRUE)
### --- FISH BIOMASS MODEL --- ###
fishnet <- merged[,-c(1:3,5)]
train <- fishnet[split_ind == 1,]
test <- fishnet[split_ind == 2,]
set.seed(1234)
en_f <- train(Fish ~ .,
train,
method = 'glmnet',
tuneGrid = expand.grid(alpha = seq(0, 1, length = 10),
lambda = seq(0.0001, 0.2, length = 5)),
trControl = custom)
## + Fold1.Rep1: alpha=0.0000, lambda=0.2
## - Fold1.Rep1: alpha=0.0000, lambda=0.2
## + Fold1.Rep1: alpha=0.1111, lambda=0.2
## - Fold1.Rep1: alpha=0.1111, lambda=0.2
## + Fold1.Rep1: alpha=0.2222, lambda=0.2
## - Fold1.Rep1: alpha=0.2222, lambda=0.2
## + Fold1.Rep1: alpha=0.3333, lambda=0.2
## - Fold1.Rep1: alpha=0.3333, lambda=0.2
## + Fold1.Rep1: alpha=0.4444, lambda=0.2
## - Fold1.Rep1: alpha=0.4444, lambda=0.2
## + Fold1.Rep1: alpha=0.5556, lambda=0.2
## - Fold1.Rep1: alpha=0.5556, lambda=0.2
## + Fold1.Rep1: alpha=0.6667, lambda=0.2
## - Fold1.Rep1: alpha=0.6667, lambda=0.2
## + Fold1.Rep1: alpha=0.7778, lambda=0.2
## - Fold1.Rep1: alpha=0.7778, lambda=0.2
## + Fold1.Rep1: alpha=0.8889, lambda=0.2
## - Fold1.Rep1: alpha=0.8889, lambda=0.2
## + Fold1.Rep1: alpha=1.0000, lambda=0.2
## - Fold1.Rep1: alpha=1.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.0000, lambda=0.2
## - Fold2.Rep1: alpha=0.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.1111, lambda=0.2
## - Fold2.Rep1: alpha=0.1111, lambda=0.2
## + Fold2.Rep1: alpha=0.2222, lambda=0.2
## - Fold2.Rep1: alpha=0.2222, lambda=0.2
## + Fold2.Rep1: alpha=0.3333, lambda=0.2
## - Fold2.Rep1: alpha=0.3333, lambda=0.2
## + Fold2.Rep1: alpha=0.4444, lambda=0.2
## - Fold2.Rep1: alpha=0.4444, lambda=0.2
## + Fold2.Rep1: alpha=0.5556, lambda=0.2
## - Fold2.Rep1: alpha=0.5556, lambda=0.2
## + Fold2.Rep1: alpha=0.6667, lambda=0.2
## - Fold2.Rep1: alpha=0.6667, lambda=0.2
## + Fold2.Rep1: alpha=0.7778, lambda=0.2
## - Fold2.Rep1: alpha=0.7778, lambda=0.2
## + Fold2.Rep1: alpha=0.8889, lambda=0.2
## - Fold2.Rep1: alpha=0.8889, lambda=0.2
## + Fold2.Rep1: alpha=1.0000, lambda=0.2
## - Fold2.Rep1: alpha=1.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.0000, lambda=0.2
## - Fold3.Rep1: alpha=0.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.1111, lambda=0.2
## - Fold3.Rep1: alpha=0.1111, lambda=0.2
## + Fold3.Rep1: alpha=0.2222, lambda=0.2
## - Fold3.Rep1: alpha=0.2222, lambda=0.2
## + Fold3.Rep1: alpha=0.3333, lambda=0.2
## - Fold3.Rep1: alpha=0.3333, lambda=0.2
## + Fold3.Rep1: alpha=0.4444, lambda=0.2
## - Fold3.Rep1: alpha=0.4444, lambda=0.2
## + Fold3.Rep1: alpha=0.5556, lambda=0.2
## - Fold3.Rep1: alpha=0.5556, lambda=0.2
## + Fold3.Rep1: alpha=0.6667, lambda=0.2
## - Fold3.Rep1: alpha=0.6667, lambda=0.2
## + Fold3.Rep1: alpha=0.7778, lambda=0.2
## - Fold3.Rep1: alpha=0.7778, lambda=0.2
## + Fold3.Rep1: alpha=0.8889, lambda=0.2
## - Fold3.Rep1: alpha=0.8889, lambda=0.2
## + Fold3.Rep1: alpha=1.0000, lambda=0.2
## - Fold3.Rep1: alpha=1.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.0000, lambda=0.2
## - Fold4.Rep1: alpha=0.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.1111, lambda=0.2
## - Fold4.Rep1: alpha=0.1111, lambda=0.2
## + Fold4.Rep1: alpha=0.2222, lambda=0.2
## - Fold4.Rep1: alpha=0.2222, lambda=0.2
## + Fold4.Rep1: alpha=0.3333, lambda=0.2
## - Fold4.Rep1: alpha=0.3333, lambda=0.2
## + Fold4.Rep1: alpha=0.4444, lambda=0.2
## - Fold4.Rep1: alpha=0.4444, lambda=0.2
## + Fold4.Rep1: alpha=0.5556, lambda=0.2
## - Fold4.Rep1: alpha=0.5556, lambda=0.2
## + Fold4.Rep1: alpha=0.6667, lambda=0.2
## - Fold4.Rep1: alpha=0.6667, lambda=0.2
## + Fold4.Rep1: alpha=0.7778, lambda=0.2
## - Fold4.Rep1: alpha=0.7778, lambda=0.2
## + Fold4.Rep1: alpha=0.8889, lambda=0.2
## - Fold4.Rep1: alpha=0.8889, lambda=0.2
## + Fold4.Rep1: alpha=1.0000, lambda=0.2
## - Fold4.Rep1: alpha=1.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.0000, lambda=0.2
## - Fold5.Rep1: alpha=0.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.1111, lambda=0.2
## - Fold5.Rep1: alpha=0.1111, lambda=0.2
## + Fold5.Rep1: alpha=0.2222, lambda=0.2
## - Fold5.Rep1: alpha=0.2222, lambda=0.2
## + Fold5.Rep1: alpha=0.3333, lambda=0.2
## - Fold5.Rep1: alpha=0.3333, lambda=0.2
## + Fold5.Rep1: alpha=0.4444, lambda=0.2
## - Fold5.Rep1: alpha=0.4444, lambda=0.2
## + Fold5.Rep1: alpha=0.5556, lambda=0.2
## - Fold5.Rep1: alpha=0.5556, lambda=0.2
## + Fold5.Rep1: alpha=0.6667, lambda=0.2
## - Fold5.Rep1: alpha=0.6667, lambda=0.2
## + Fold5.Rep1: alpha=0.7778, lambda=0.2
## - Fold5.Rep1: alpha=0.7778, lambda=0.2
## + Fold5.Rep1: alpha=0.8889, lambda=0.2
## - Fold5.Rep1: alpha=0.8889, lambda=0.2
## + Fold5.Rep1: alpha=1.0000, lambda=0.2
## - Fold5.Rep1: alpha=1.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.0000, lambda=0.2
## - Fold1.Rep2: alpha=0.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.1111, lambda=0.2
## - Fold1.Rep2: alpha=0.1111, lambda=0.2
## + Fold1.Rep2: alpha=0.2222, lambda=0.2
## - Fold1.Rep2: alpha=0.2222, lambda=0.2
## + Fold1.Rep2: alpha=0.3333, lambda=0.2
## - Fold1.Rep2: alpha=0.3333, lambda=0.2
## + Fold1.Rep2: alpha=0.4444, lambda=0.2
## - Fold1.Rep2: alpha=0.4444, lambda=0.2
## + Fold1.Rep2: alpha=0.5556, lambda=0.2
## - Fold1.Rep2: alpha=0.5556, lambda=0.2
## + Fold1.Rep2: alpha=0.6667, lambda=0.2
## - Fold1.Rep2: alpha=0.6667, lambda=0.2
## + Fold1.Rep2: alpha=0.7778, lambda=0.2
## - Fold1.Rep2: alpha=0.7778, lambda=0.2
## + Fold1.Rep2: alpha=0.8889, lambda=0.2
## - Fold1.Rep2: alpha=0.8889, lambda=0.2
## + Fold1.Rep2: alpha=1.0000, lambda=0.2
## - Fold1.Rep2: alpha=1.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.0000, lambda=0.2
## - Fold2.Rep2: alpha=0.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.1111, lambda=0.2
## - Fold2.Rep2: alpha=0.1111, lambda=0.2
## + Fold2.Rep2: alpha=0.2222, lambda=0.2
## - Fold2.Rep2: alpha=0.2222, lambda=0.2
## + Fold2.Rep2: alpha=0.3333, lambda=0.2
## - Fold2.Rep2: alpha=0.3333, lambda=0.2
## + Fold2.Rep2: alpha=0.4444, lambda=0.2
## - Fold2.Rep2: alpha=0.4444, lambda=0.2
## + Fold2.Rep2: alpha=0.5556, lambda=0.2
## - Fold2.Rep2: alpha=0.5556, lambda=0.2
## + Fold2.Rep2: alpha=0.6667, lambda=0.2
## - Fold2.Rep2: alpha=0.6667, lambda=0.2
## + Fold2.Rep2: alpha=0.7778, lambda=0.2
## - Fold2.Rep2: alpha=0.7778, lambda=0.2
## + Fold2.Rep2: alpha=0.8889, lambda=0.2
## - Fold2.Rep2: alpha=0.8889, lambda=0.2
## + Fold2.Rep2: alpha=1.0000, lambda=0.2
## - Fold2.Rep2: alpha=1.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.0000, lambda=0.2
## - Fold3.Rep2: alpha=0.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.1111, lambda=0.2
## - Fold3.Rep2: alpha=0.1111, lambda=0.2
## + Fold3.Rep2: alpha=0.2222, lambda=0.2
## - Fold3.Rep2: alpha=0.2222, lambda=0.2
## + Fold3.Rep2: alpha=0.3333, lambda=0.2
## - Fold3.Rep2: alpha=0.3333, lambda=0.2
## + Fold3.Rep2: alpha=0.4444, lambda=0.2
## - Fold3.Rep2: alpha=0.4444, lambda=0.2
## + Fold3.Rep2: alpha=0.5556, lambda=0.2
## - Fold3.Rep2: alpha=0.5556, lambda=0.2
## + Fold3.Rep2: alpha=0.6667, lambda=0.2
## - Fold3.Rep2: alpha=0.6667, lambda=0.2
## + Fold3.Rep2: alpha=0.7778, lambda=0.2
## - Fold3.Rep2: alpha=0.7778, lambda=0.2
## + Fold3.Rep2: alpha=0.8889, lambda=0.2
## - Fold3.Rep2: alpha=0.8889, lambda=0.2
## + Fold3.Rep2: alpha=1.0000, lambda=0.2
## - Fold3.Rep2: alpha=1.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.0000, lambda=0.2
## - Fold4.Rep2: alpha=0.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.1111, lambda=0.2
## - Fold4.Rep2: alpha=0.1111, lambda=0.2
## + Fold4.Rep2: alpha=0.2222, lambda=0.2
## - Fold4.Rep2: alpha=0.2222, lambda=0.2
## + Fold4.Rep2: alpha=0.3333, lambda=0.2
## - Fold4.Rep2: alpha=0.3333, lambda=0.2
## + Fold4.Rep2: alpha=0.4444, lambda=0.2
## - Fold4.Rep2: alpha=0.4444, lambda=0.2
## + Fold4.Rep2: alpha=0.5556, lambda=0.2
## - Fold4.Rep2: alpha=0.5556, lambda=0.2
## + Fold4.Rep2: alpha=0.6667, lambda=0.2
## - Fold4.Rep2: alpha=0.6667, lambda=0.2
## + Fold4.Rep2: alpha=0.7778, lambda=0.2
## - Fold4.Rep2: alpha=0.7778, lambda=0.2
## + Fold4.Rep2: alpha=0.8889, lambda=0.2
## - Fold4.Rep2: alpha=0.8889, lambda=0.2
## + Fold4.Rep2: alpha=1.0000, lambda=0.2
## - Fold4.Rep2: alpha=1.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.0000, lambda=0.2
## - Fold5.Rep2: alpha=0.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.1111, lambda=0.2
## - Fold5.Rep2: alpha=0.1111, lambda=0.2
## + Fold5.Rep2: alpha=0.2222, lambda=0.2
## - Fold5.Rep2: alpha=0.2222, lambda=0.2
## + Fold5.Rep2: alpha=0.3333, lambda=0.2
## - Fold5.Rep2: alpha=0.3333, lambda=0.2
## + Fold5.Rep2: alpha=0.4444, lambda=0.2
## - Fold5.Rep2: alpha=0.4444, lambda=0.2
## + Fold5.Rep2: alpha=0.5556, lambda=0.2
## - Fold5.Rep2: alpha=0.5556, lambda=0.2
## + Fold5.Rep2: alpha=0.6667, lambda=0.2
## - Fold5.Rep2: alpha=0.6667, lambda=0.2
## + Fold5.Rep2: alpha=0.7778, lambda=0.2
## - Fold5.Rep2: alpha=0.7778, lambda=0.2
## + Fold5.Rep2: alpha=0.8889, lambda=0.2
## - Fold5.Rep2: alpha=0.8889, lambda=0.2
## + Fold5.Rep2: alpha=1.0000, lambda=0.2
## - Fold5.Rep2: alpha=1.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.0000, lambda=0.2
## - Fold1.Rep3: alpha=0.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.1111, lambda=0.2
## - Fold1.Rep3: alpha=0.1111, lambda=0.2
## + Fold1.Rep3: alpha=0.2222, lambda=0.2
## - Fold1.Rep3: alpha=0.2222, lambda=0.2
## + Fold1.Rep3: alpha=0.3333, lambda=0.2
## - Fold1.Rep3: alpha=0.3333, lambda=0.2
## + Fold1.Rep3: alpha=0.4444, lambda=0.2
## - Fold1.Rep3: alpha=0.4444, lambda=0.2
## + Fold1.Rep3: alpha=0.5556, lambda=0.2
## - Fold1.Rep3: alpha=0.5556, lambda=0.2
## + Fold1.Rep3: alpha=0.6667, lambda=0.2
## - Fold1.Rep3: alpha=0.6667, lambda=0.2
## + Fold1.Rep3: alpha=0.7778, lambda=0.2
## - Fold1.Rep3: alpha=0.7778, lambda=0.2
## + Fold1.Rep3: alpha=0.8889, lambda=0.2
## - Fold1.Rep3: alpha=0.8889, lambda=0.2
## + Fold1.Rep3: alpha=1.0000, lambda=0.2
## - Fold1.Rep3: alpha=1.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.0000, lambda=0.2
## - Fold2.Rep3: alpha=0.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.1111, lambda=0.2
## - Fold2.Rep3: alpha=0.1111, lambda=0.2
## + Fold2.Rep3: alpha=0.2222, lambda=0.2
## - Fold2.Rep3: alpha=0.2222, lambda=0.2
## + Fold2.Rep3: alpha=0.3333, lambda=0.2
## - Fold2.Rep3: alpha=0.3333, lambda=0.2
## + Fold2.Rep3: alpha=0.4444, lambda=0.2
## - Fold2.Rep3: alpha=0.4444, lambda=0.2
## + Fold2.Rep3: alpha=0.5556, lambda=0.2
## - Fold2.Rep3: alpha=0.5556, lambda=0.2
## + Fold2.Rep3: alpha=0.6667, lambda=0.2
## - Fold2.Rep3: alpha=0.6667, lambda=0.2
## + Fold2.Rep3: alpha=0.7778, lambda=0.2
## - Fold2.Rep3: alpha=0.7778, lambda=0.2
## + Fold2.Rep3: alpha=0.8889, lambda=0.2
## - Fold2.Rep3: alpha=0.8889, lambda=0.2
## + Fold2.Rep3: alpha=1.0000, lambda=0.2
## - Fold2.Rep3: alpha=1.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.0000, lambda=0.2
## - Fold3.Rep3: alpha=0.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.1111, lambda=0.2
## - Fold3.Rep3: alpha=0.1111, lambda=0.2
## + Fold3.Rep3: alpha=0.2222, lambda=0.2
## - Fold3.Rep3: alpha=0.2222, lambda=0.2
## + Fold3.Rep3: alpha=0.3333, lambda=0.2
## - Fold3.Rep3: alpha=0.3333, lambda=0.2
## + Fold3.Rep3: alpha=0.4444, lambda=0.2
## - Fold3.Rep3: alpha=0.4444, lambda=0.2
## + Fold3.Rep3: alpha=0.5556, lambda=0.2
## - Fold3.Rep3: alpha=0.5556, lambda=0.2
## + Fold3.Rep3: alpha=0.6667, lambda=0.2
## - Fold3.Rep3: alpha=0.6667, lambda=0.2
## + Fold3.Rep3: alpha=0.7778, lambda=0.2
## - Fold3.Rep3: alpha=0.7778, lambda=0.2
## + Fold3.Rep3: alpha=0.8889, lambda=0.2
## - Fold3.Rep3: alpha=0.8889, lambda=0.2
## + Fold3.Rep3: alpha=1.0000, lambda=0.2
## - Fold3.Rep3: alpha=1.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.0000, lambda=0.2
## - Fold4.Rep3: alpha=0.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.1111, lambda=0.2
## - Fold4.Rep3: alpha=0.1111, lambda=0.2
## + Fold4.Rep3: alpha=0.2222, lambda=0.2
## - Fold4.Rep3: alpha=0.2222, lambda=0.2
## + Fold4.Rep3: alpha=0.3333, lambda=0.2
## - Fold4.Rep3: alpha=0.3333, lambda=0.2
## + Fold4.Rep3: alpha=0.4444, lambda=0.2
## - Fold4.Rep3: alpha=0.4444, lambda=0.2
## + Fold4.Rep3: alpha=0.5556, lambda=0.2
## - Fold4.Rep3: alpha=0.5556, lambda=0.2
## + Fold4.Rep3: alpha=0.6667, lambda=0.2
## - Fold4.Rep3: alpha=0.6667, lambda=0.2
## + Fold4.Rep3: alpha=0.7778, lambda=0.2
## - Fold4.Rep3: alpha=0.7778, lambda=0.2
## + Fold4.Rep3: alpha=0.8889, lambda=0.2
## - Fold4.Rep3: alpha=0.8889, lambda=0.2
## + Fold4.Rep3: alpha=1.0000, lambda=0.2
## - Fold4.Rep3: alpha=1.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.0000, lambda=0.2
## - Fold5.Rep3: alpha=0.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.1111, lambda=0.2
## - Fold5.Rep3: alpha=0.1111, lambda=0.2
## + Fold5.Rep3: alpha=0.2222, lambda=0.2
## - Fold5.Rep3: alpha=0.2222, lambda=0.2
## + Fold5.Rep3: alpha=0.3333, lambda=0.2
## - Fold5.Rep3: alpha=0.3333, lambda=0.2
## + Fold5.Rep3: alpha=0.4444, lambda=0.2
## - Fold5.Rep3: alpha=0.4444, lambda=0.2
## + Fold5.Rep3: alpha=0.5556, lambda=0.2
## - Fold5.Rep3: alpha=0.5556, lambda=0.2
## + Fold5.Rep3: alpha=0.6667, lambda=0.2
## - Fold5.Rep3: alpha=0.6667, lambda=0.2
## + Fold5.Rep3: alpha=0.7778, lambda=0.2
## - Fold5.Rep3: alpha=0.7778, lambda=0.2
## + Fold5.Rep3: alpha=0.8889, lambda=0.2
## - Fold5.Rep3: alpha=0.8889, lambda=0.2
## + Fold5.Rep3: alpha=1.0000, lambda=0.2
## - Fold5.Rep3: alpha=1.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.0000, lambda=0.2
## - Fold1.Rep4: alpha=0.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.1111, lambda=0.2
## - Fold1.Rep4: alpha=0.1111, lambda=0.2
## + Fold1.Rep4: alpha=0.2222, lambda=0.2
## - Fold1.Rep4: alpha=0.2222, lambda=0.2
## + Fold1.Rep4: alpha=0.3333, lambda=0.2
## - Fold1.Rep4: alpha=0.3333, lambda=0.2
## + Fold1.Rep4: alpha=0.4444, lambda=0.2
## - Fold1.Rep4: alpha=0.4444, lambda=0.2
## + Fold1.Rep4: alpha=0.5556, lambda=0.2
## - Fold1.Rep4: alpha=0.5556, lambda=0.2
## + Fold1.Rep4: alpha=0.6667, lambda=0.2
## - Fold1.Rep4: alpha=0.6667, lambda=0.2
## + Fold1.Rep4: alpha=0.7778, lambda=0.2
## - Fold1.Rep4: alpha=0.7778, lambda=0.2
## + Fold1.Rep4: alpha=0.8889, lambda=0.2
## - Fold1.Rep4: alpha=0.8889, lambda=0.2
## + Fold1.Rep4: alpha=1.0000, lambda=0.2
## - Fold1.Rep4: alpha=1.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.0000, lambda=0.2
## - Fold2.Rep4: alpha=0.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.1111, lambda=0.2
## - Fold2.Rep4: alpha=0.1111, lambda=0.2
## + Fold2.Rep4: alpha=0.2222, lambda=0.2
## - Fold2.Rep4: alpha=0.2222, lambda=0.2
## + Fold2.Rep4: alpha=0.3333, lambda=0.2
## - Fold2.Rep4: alpha=0.3333, lambda=0.2
## + Fold2.Rep4: alpha=0.4444, lambda=0.2
## - Fold2.Rep4: alpha=0.4444, lambda=0.2
## + Fold2.Rep4: alpha=0.5556, lambda=0.2
## - Fold2.Rep4: alpha=0.5556, lambda=0.2
## + Fold2.Rep4: alpha=0.6667, lambda=0.2
## - Fold2.Rep4: alpha=0.6667, lambda=0.2
## + Fold2.Rep4: alpha=0.7778, lambda=0.2
## - Fold2.Rep4: alpha=0.7778, lambda=0.2
## + Fold2.Rep4: alpha=0.8889, lambda=0.2
## - Fold2.Rep4: alpha=0.8889, lambda=0.2
## + Fold2.Rep4: alpha=1.0000, lambda=0.2
## - Fold2.Rep4: alpha=1.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.0000, lambda=0.2
## - Fold3.Rep4: alpha=0.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.1111, lambda=0.2
## - Fold3.Rep4: alpha=0.1111, lambda=0.2
## + Fold3.Rep4: alpha=0.2222, lambda=0.2
## - Fold3.Rep4: alpha=0.2222, lambda=0.2
## + Fold3.Rep4: alpha=0.3333, lambda=0.2
## - Fold3.Rep4: alpha=0.3333, lambda=0.2
## + Fold3.Rep4: alpha=0.4444, lambda=0.2
## - Fold3.Rep4: alpha=0.4444, lambda=0.2
## + Fold3.Rep4: alpha=0.5556, lambda=0.2
## - Fold3.Rep4: alpha=0.5556, lambda=0.2
## + Fold3.Rep4: alpha=0.6667, lambda=0.2
## - Fold3.Rep4: alpha=0.6667, lambda=0.2
## + Fold3.Rep4: alpha=0.7778, lambda=0.2
## - Fold3.Rep4: alpha=0.7778, lambda=0.2
## + Fold3.Rep4: alpha=0.8889, lambda=0.2
## - Fold3.Rep4: alpha=0.8889, lambda=0.2
## + Fold3.Rep4: alpha=1.0000, lambda=0.2
## - Fold3.Rep4: alpha=1.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.0000, lambda=0.2
## - Fold4.Rep4: alpha=0.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.1111, lambda=0.2
## - Fold4.Rep4: alpha=0.1111, lambda=0.2
## + Fold4.Rep4: alpha=0.2222, lambda=0.2
## - Fold4.Rep4: alpha=0.2222, lambda=0.2
## + Fold4.Rep4: alpha=0.3333, lambda=0.2
## - Fold4.Rep4: alpha=0.3333, lambda=0.2
## + Fold4.Rep4: alpha=0.4444, lambda=0.2
## - Fold4.Rep4: alpha=0.4444, lambda=0.2
## + Fold4.Rep4: alpha=0.5556, lambda=0.2
## - Fold4.Rep4: alpha=0.5556, lambda=0.2
## + Fold4.Rep4: alpha=0.6667, lambda=0.2
## - Fold4.Rep4: alpha=0.6667, lambda=0.2
## + Fold4.Rep4: alpha=0.7778, lambda=0.2
## - Fold4.Rep4: alpha=0.7778, lambda=0.2
## + Fold4.Rep4: alpha=0.8889, lambda=0.2
## - Fold4.Rep4: alpha=0.8889, lambda=0.2
## + Fold4.Rep4: alpha=1.0000, lambda=0.2
## - Fold4.Rep4: alpha=1.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.0000, lambda=0.2
## - Fold5.Rep4: alpha=0.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.1111, lambda=0.2
## - Fold5.Rep4: alpha=0.1111, lambda=0.2
## + Fold5.Rep4: alpha=0.2222, lambda=0.2
## - Fold5.Rep4: alpha=0.2222, lambda=0.2
## + Fold5.Rep4: alpha=0.3333, lambda=0.2
## - Fold5.Rep4: alpha=0.3333, lambda=0.2
## + Fold5.Rep4: alpha=0.4444, lambda=0.2
## - Fold5.Rep4: alpha=0.4444, lambda=0.2
## + Fold5.Rep4: alpha=0.5556, lambda=0.2
## - Fold5.Rep4: alpha=0.5556, lambda=0.2
## + Fold5.Rep4: alpha=0.6667, lambda=0.2
## - Fold5.Rep4: alpha=0.6667, lambda=0.2
## + Fold5.Rep4: alpha=0.7778, lambda=0.2
## - Fold5.Rep4: alpha=0.7778, lambda=0.2
## + Fold5.Rep4: alpha=0.8889, lambda=0.2
## - Fold5.Rep4: alpha=0.8889, lambda=0.2
## + Fold5.Rep4: alpha=1.0000, lambda=0.2
## - Fold5.Rep4: alpha=1.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.0000, lambda=0.2
## - Fold1.Rep5: alpha=0.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.1111, lambda=0.2
## - Fold1.Rep5: alpha=0.1111, lambda=0.2
## + Fold1.Rep5: alpha=0.2222, lambda=0.2
## - Fold1.Rep5: alpha=0.2222, lambda=0.2
## + Fold1.Rep5: alpha=0.3333, lambda=0.2
## - Fold1.Rep5: alpha=0.3333, lambda=0.2
## + Fold1.Rep5: alpha=0.4444, lambda=0.2
## - Fold1.Rep5: alpha=0.4444, lambda=0.2
## + Fold1.Rep5: alpha=0.5556, lambda=0.2
## - Fold1.Rep5: alpha=0.5556, lambda=0.2
## + Fold1.Rep5: alpha=0.6667, lambda=0.2
## - Fold1.Rep5: alpha=0.6667, lambda=0.2
## + Fold1.Rep5: alpha=0.7778, lambda=0.2
## - Fold1.Rep5: alpha=0.7778, lambda=0.2
## + Fold1.Rep5: alpha=0.8889, lambda=0.2
## - Fold1.Rep5: alpha=0.8889, lambda=0.2
## + Fold1.Rep5: alpha=1.0000, lambda=0.2
## - Fold1.Rep5: alpha=1.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.0000, lambda=0.2
## - Fold2.Rep5: alpha=0.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.1111, lambda=0.2
## - Fold2.Rep5: alpha=0.1111, lambda=0.2
## + Fold2.Rep5: alpha=0.2222, lambda=0.2
## - Fold2.Rep5: alpha=0.2222, lambda=0.2
## + Fold2.Rep5: alpha=0.3333, lambda=0.2
## - Fold2.Rep5: alpha=0.3333, lambda=0.2
## + Fold2.Rep5: alpha=0.4444, lambda=0.2
## - Fold2.Rep5: alpha=0.4444, lambda=0.2
## + Fold2.Rep5: alpha=0.5556, lambda=0.2
## - Fold2.Rep5: alpha=0.5556, lambda=0.2
## + Fold2.Rep5: alpha=0.6667, lambda=0.2
## - Fold2.Rep5: alpha=0.6667, lambda=0.2
## + Fold2.Rep5: alpha=0.7778, lambda=0.2
## - Fold2.Rep5: alpha=0.7778, lambda=0.2
## + Fold2.Rep5: alpha=0.8889, lambda=0.2
## - Fold2.Rep5: alpha=0.8889, lambda=0.2
## + Fold2.Rep5: alpha=1.0000, lambda=0.2
## - Fold2.Rep5: alpha=1.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.0000, lambda=0.2
## - Fold3.Rep5: alpha=0.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.1111, lambda=0.2
## - Fold3.Rep5: alpha=0.1111, lambda=0.2
## + Fold3.Rep5: alpha=0.2222, lambda=0.2
## - Fold3.Rep5: alpha=0.2222, lambda=0.2
## + Fold3.Rep5: alpha=0.3333, lambda=0.2
## - Fold3.Rep5: alpha=0.3333, lambda=0.2
## + Fold3.Rep5: alpha=0.4444, lambda=0.2
## - Fold3.Rep5: alpha=0.4444, lambda=0.2
## + Fold3.Rep5: alpha=0.5556, lambda=0.2
## - Fold3.Rep5: alpha=0.5556, lambda=0.2
## + Fold3.Rep5: alpha=0.6667, lambda=0.2
## - Fold3.Rep5: alpha=0.6667, lambda=0.2
## + Fold3.Rep5: alpha=0.7778, lambda=0.2
## - Fold3.Rep5: alpha=0.7778, lambda=0.2
## + Fold3.Rep5: alpha=0.8889, lambda=0.2
## - Fold3.Rep5: alpha=0.8889, lambda=0.2
## + Fold3.Rep5: alpha=1.0000, lambda=0.2
## - Fold3.Rep5: alpha=1.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.0000, lambda=0.2
## - Fold4.Rep5: alpha=0.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.1111, lambda=0.2
## - Fold4.Rep5: alpha=0.1111, lambda=0.2
## + Fold4.Rep5: alpha=0.2222, lambda=0.2
## - Fold4.Rep5: alpha=0.2222, lambda=0.2
## + Fold4.Rep5: alpha=0.3333, lambda=0.2
## - Fold4.Rep5: alpha=0.3333, lambda=0.2
## + Fold4.Rep5: alpha=0.4444, lambda=0.2
## - Fold4.Rep5: alpha=0.4444, lambda=0.2
## + Fold4.Rep5: alpha=0.5556, lambda=0.2
## - Fold4.Rep5: alpha=0.5556, lambda=0.2
## + Fold4.Rep5: alpha=0.6667, lambda=0.2
## - Fold4.Rep5: alpha=0.6667, lambda=0.2
## + Fold4.Rep5: alpha=0.7778, lambda=0.2
## - Fold4.Rep5: alpha=0.7778, lambda=0.2
## + Fold4.Rep5: alpha=0.8889, lambda=0.2
## - Fold4.Rep5: alpha=0.8889, lambda=0.2
## + Fold4.Rep5: alpha=1.0000, lambda=0.2
## - Fold4.Rep5: alpha=1.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.0000, lambda=0.2
## - Fold5.Rep5: alpha=0.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.1111, lambda=0.2
## - Fold5.Rep5: alpha=0.1111, lambda=0.2
## + Fold5.Rep5: alpha=0.2222, lambda=0.2
## - Fold5.Rep5: alpha=0.2222, lambda=0.2
## + Fold5.Rep5: alpha=0.3333, lambda=0.2
## - Fold5.Rep5: alpha=0.3333, lambda=0.2
## + Fold5.Rep5: alpha=0.4444, lambda=0.2
## - Fold5.Rep5: alpha=0.4444, lambda=0.2
## + Fold5.Rep5: alpha=0.5556, lambda=0.2
## - Fold5.Rep5: alpha=0.5556, lambda=0.2
## + Fold5.Rep5: alpha=0.6667, lambda=0.2
## - Fold5.Rep5: alpha=0.6667, lambda=0.2
## + Fold5.Rep5: alpha=0.7778, lambda=0.2
## - Fold5.Rep5: alpha=0.7778, lambda=0.2
## + Fold5.Rep5: alpha=0.8889, lambda=0.2
## - Fold5.Rep5: alpha=0.8889, lambda=0.2
## + Fold5.Rep5: alpha=1.0000, lambda=0.2
## - Fold5.Rep5: alpha=1.0000, lambda=0.2
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 1, lambda = 0.15 on full training set
print(en_f)
## glmnet
##
## 9 samples
## 15 predictors
##
## No pre-processing
## Resampling: Cross-Validated (5 fold, repeated 5 times)
## Summary of sample sizes: 7, 8, 7, 7, 7, 7, ...
## Resampling results across tuning parameters:
##
## alpha lambda RMSE Rsquared MAE
## 0.0000000 0.000100 0.3080098 1 0.2823277
## 0.0000000 0.050075 0.3080098 1 0.2823277
## 0.0000000 0.100050 0.3080098 1 0.2823277
## 0.0000000 0.150025 0.3080098 1 0.2823277
## 0.0000000 0.200000 0.3080098 1 0.2823277
## 0.1111111 0.000100 0.3816175 1 0.3429264
## 0.1111111 0.050075 0.3561445 1 0.3211453
## 0.1111111 0.100050 0.3290367 1 0.2955102
## 0.1111111 0.150025 0.3187866 1 0.2886116
## 0.1111111 0.200000 0.3111702 1 0.2822383
## 0.2222222 0.000100 0.4135175 1 0.3692482
## 0.2222222 0.050075 0.3603217 1 0.3218363
## 0.2222222 0.100050 0.3328512 1 0.2993225
## 0.2222222 0.150025 0.3160757 1 0.2852699
## 0.2222222 0.200000 0.3077394 1 0.2773534
## 0.3333333 0.000100 0.4355176 1 0.3876573
## 0.3333333 0.050075 0.3664811 1 0.3259316
## 0.3333333 0.100050 0.3289844 1 0.2940809
## 0.3333333 0.150025 0.3149256 1 0.2837315
## 0.3333333 0.200000 0.3114864 1 0.2826361
## 0.4444444 0.000100 0.4550473 1 0.4044608
## 0.4444444 0.050075 0.3674237 1 0.3248749
## 0.4444444 0.100050 0.3273134 1 0.2939227
## 0.4444444 0.150025 0.3158526 1 0.2861151
## 0.4444444 0.200000 0.3118124 1 0.2844568
## 0.5555556 0.000100 0.4689209 1 0.4153287
## 0.5555556 0.050075 0.3669409 1 0.3246236
## 0.5555556 0.100050 0.3262598 1 0.2940403
## 0.5555556 0.150025 0.3135711 1 0.2853656
## 0.5555556 0.200000 0.3058679 1 0.2793134
## 0.6666667 0.000100 0.4833006 1 0.4265619
## 0.6666667 0.050075 0.3647250 1 0.3230683
## 0.6666667 0.100050 0.3223254 1 0.2922088
## 0.6666667 0.150025 0.3070589 1 0.2794447
## 0.6666667 0.200000 0.3033075 1 0.2776935
## 0.7777778 0.000100 0.4941193 1 0.4355600
## 0.7777778 0.050075 0.3601262 1 0.3213819
## 0.7777778 0.100050 0.3169979 1 0.2883673
## 0.7777778 0.150025 0.3006092 1 0.2732433
## 0.7777778 0.200000 0.3013959 1 0.2762056
## 0.8888889 0.000100 0.4962638 1 0.4391466
## 0.8888889 0.050075 0.3526819 1 0.3167135
## 0.8888889 0.100050 0.3108904 1 0.2829865
## 0.8888889 0.150025 0.2958961 1 0.2684388
## 0.8888889 0.200000 0.3023506 1 0.2778307
## 1.0000000 0.000100 0.5197809 1 0.4633824
## 1.0000000 0.050075 0.3443541 1 0.3100357
## 1.0000000 0.100050 0.3036718 1 0.2762264
## 1.0000000 0.150025 0.2919726 1 0.2653162
## 1.0000000 0.200000 0.3044357 1 0.2819776
##
## RMSE was used to select the optimal model using the smallest value.
## The final values used for the model were alpha = 1 and lambda = 0.150025.
plot(varImp(en_f, scale = TRUE))
# extract Coefficients and rank predictors
options(scipen = 99)
coefs <- coef(en_f$finalModel, en_f$bestTune$lambda)
coefs.zero <- coefs[which(coefs == 0),]
coef_order <- order(abs(coefs), decreasing = TRUE)
head(as.matrix(coefs[coef_order,]), 10)
## [,1]
## (Intercept) 1.2420957
## Dist_Pour 0.4075865
## Avg_Flow 0.0000000
## Flow_Variance 0.0000000
## Avg_Sand 0.0000000
## Avg_Gravel 0.0000000
## Avg_Cobble 0.0000000
## Avg_Boulder 0.0000000
## Avg_Bedrock 0.0000000
## Avg_Organic 0.0000000
### --- PERIPHYTON BIOMASS MODEL --- ###
perinet <- merged[,-c(1:4)]
train <- perinet[split_ind == 1,]
test <- perinet[split_ind == 2,]
set.seed(1234)
en_p <- train(Periphyton ~ .,
train,
method = 'glmnet',
tuneGrid = expand.grid(alpha = seq(0, 1, length = 10),
lambda = seq(0.0001, 0.2, length = 5)),
trControl = custom)
## + Fold1.Rep1: alpha=0.0000, lambda=0.2
## - Fold1.Rep1: alpha=0.0000, lambda=0.2
## + Fold1.Rep1: alpha=0.1111, lambda=0.2
## - Fold1.Rep1: alpha=0.1111, lambda=0.2
## + Fold1.Rep1: alpha=0.2222, lambda=0.2
## - Fold1.Rep1: alpha=0.2222, lambda=0.2
## + Fold1.Rep1: alpha=0.3333, lambda=0.2
## - Fold1.Rep1: alpha=0.3333, lambda=0.2
## + Fold1.Rep1: alpha=0.4444, lambda=0.2
## - Fold1.Rep1: alpha=0.4444, lambda=0.2
## + Fold1.Rep1: alpha=0.5556, lambda=0.2
## - Fold1.Rep1: alpha=0.5556, lambda=0.2
## + Fold1.Rep1: alpha=0.6667, lambda=0.2
## - Fold1.Rep1: alpha=0.6667, lambda=0.2
## + Fold1.Rep1: alpha=0.7778, lambda=0.2
## - Fold1.Rep1: alpha=0.7778, lambda=0.2
## + Fold1.Rep1: alpha=0.8889, lambda=0.2
## - Fold1.Rep1: alpha=0.8889, lambda=0.2
## + Fold1.Rep1: alpha=1.0000, lambda=0.2
## - Fold1.Rep1: alpha=1.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.0000, lambda=0.2
## - Fold2.Rep1: alpha=0.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.1111, lambda=0.2
## - Fold2.Rep1: alpha=0.1111, lambda=0.2
## + Fold2.Rep1: alpha=0.2222, lambda=0.2
## - Fold2.Rep1: alpha=0.2222, lambda=0.2
## + Fold2.Rep1: alpha=0.3333, lambda=0.2
## - Fold2.Rep1: alpha=0.3333, lambda=0.2
## + Fold2.Rep1: alpha=0.4444, lambda=0.2
## - Fold2.Rep1: alpha=0.4444, lambda=0.2
## + Fold2.Rep1: alpha=0.5556, lambda=0.2
## - Fold2.Rep1: alpha=0.5556, lambda=0.2
## + Fold2.Rep1: alpha=0.6667, lambda=0.2
## - Fold2.Rep1: alpha=0.6667, lambda=0.2
## + Fold2.Rep1: alpha=0.7778, lambda=0.2
## - Fold2.Rep1: alpha=0.7778, lambda=0.2
## + Fold2.Rep1: alpha=0.8889, lambda=0.2
## - Fold2.Rep1: alpha=0.8889, lambda=0.2
## + Fold2.Rep1: alpha=1.0000, lambda=0.2
## - Fold2.Rep1: alpha=1.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.0000, lambda=0.2
## - Fold3.Rep1: alpha=0.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.1111, lambda=0.2
## - Fold3.Rep1: alpha=0.1111, lambda=0.2
## + Fold3.Rep1: alpha=0.2222, lambda=0.2
## - Fold3.Rep1: alpha=0.2222, lambda=0.2
## + Fold3.Rep1: alpha=0.3333, lambda=0.2
## - Fold3.Rep1: alpha=0.3333, lambda=0.2
## + Fold3.Rep1: alpha=0.4444, lambda=0.2
## - Fold3.Rep1: alpha=0.4444, lambda=0.2
## + Fold3.Rep1: alpha=0.5556, lambda=0.2
## - Fold3.Rep1: alpha=0.5556, lambda=0.2
## + Fold3.Rep1: alpha=0.6667, lambda=0.2
## - Fold3.Rep1: alpha=0.6667, lambda=0.2
## + Fold3.Rep1: alpha=0.7778, lambda=0.2
## - Fold3.Rep1: alpha=0.7778, lambda=0.2
## + Fold3.Rep1: alpha=0.8889, lambda=0.2
## - Fold3.Rep1: alpha=0.8889, lambda=0.2
## + Fold3.Rep1: alpha=1.0000, lambda=0.2
## - Fold3.Rep1: alpha=1.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.0000, lambda=0.2
## - Fold4.Rep1: alpha=0.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.1111, lambda=0.2
## - Fold4.Rep1: alpha=0.1111, lambda=0.2
## + Fold4.Rep1: alpha=0.2222, lambda=0.2
## - Fold4.Rep1: alpha=0.2222, lambda=0.2
## + Fold4.Rep1: alpha=0.3333, lambda=0.2
## - Fold4.Rep1: alpha=0.3333, lambda=0.2
## + Fold4.Rep1: alpha=0.4444, lambda=0.2
## - Fold4.Rep1: alpha=0.4444, lambda=0.2
## + Fold4.Rep1: alpha=0.5556, lambda=0.2
## - Fold4.Rep1: alpha=0.5556, lambda=0.2
## + Fold4.Rep1: alpha=0.6667, lambda=0.2
## - Fold4.Rep1: alpha=0.6667, lambda=0.2
## + Fold4.Rep1: alpha=0.7778, lambda=0.2
## - Fold4.Rep1: alpha=0.7778, lambda=0.2
## + Fold4.Rep1: alpha=0.8889, lambda=0.2
## - Fold4.Rep1: alpha=0.8889, lambda=0.2
## + Fold4.Rep1: alpha=1.0000, lambda=0.2
## - Fold4.Rep1: alpha=1.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.0000, lambda=0.2
## - Fold5.Rep1: alpha=0.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.1111, lambda=0.2
## - Fold5.Rep1: alpha=0.1111, lambda=0.2
## + Fold5.Rep1: alpha=0.2222, lambda=0.2
## - Fold5.Rep1: alpha=0.2222, lambda=0.2
## + Fold5.Rep1: alpha=0.3333, lambda=0.2
## - Fold5.Rep1: alpha=0.3333, lambda=0.2
## + Fold5.Rep1: alpha=0.4444, lambda=0.2
## - Fold5.Rep1: alpha=0.4444, lambda=0.2
## + Fold5.Rep1: alpha=0.5556, lambda=0.2
## - Fold5.Rep1: alpha=0.5556, lambda=0.2
## + Fold5.Rep1: alpha=0.6667, lambda=0.2
## - Fold5.Rep1: alpha=0.6667, lambda=0.2
## + Fold5.Rep1: alpha=0.7778, lambda=0.2
## - Fold5.Rep1: alpha=0.7778, lambda=0.2
## + Fold5.Rep1: alpha=0.8889, lambda=0.2
## - Fold5.Rep1: alpha=0.8889, lambda=0.2
## + Fold5.Rep1: alpha=1.0000, lambda=0.2
## - Fold5.Rep1: alpha=1.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.0000, lambda=0.2
## - Fold1.Rep2: alpha=0.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.1111, lambda=0.2
## - Fold1.Rep2: alpha=0.1111, lambda=0.2
## + Fold1.Rep2: alpha=0.2222, lambda=0.2
## - Fold1.Rep2: alpha=0.2222, lambda=0.2
## + Fold1.Rep2: alpha=0.3333, lambda=0.2
## - Fold1.Rep2: alpha=0.3333, lambda=0.2
## + Fold1.Rep2: alpha=0.4444, lambda=0.2
## - Fold1.Rep2: alpha=0.4444, lambda=0.2
## + Fold1.Rep2: alpha=0.5556, lambda=0.2
## - Fold1.Rep2: alpha=0.5556, lambda=0.2
## + Fold1.Rep2: alpha=0.6667, lambda=0.2
## - Fold1.Rep2: alpha=0.6667, lambda=0.2
## + Fold1.Rep2: alpha=0.7778, lambda=0.2
## - Fold1.Rep2: alpha=0.7778, lambda=0.2
## + Fold1.Rep2: alpha=0.8889, lambda=0.2
## - Fold1.Rep2: alpha=0.8889, lambda=0.2
## + Fold1.Rep2: alpha=1.0000, lambda=0.2
## - Fold1.Rep2: alpha=1.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.0000, lambda=0.2
## - Fold2.Rep2: alpha=0.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.1111, lambda=0.2
## - Fold2.Rep2: alpha=0.1111, lambda=0.2
## + Fold2.Rep2: alpha=0.2222, lambda=0.2
## - Fold2.Rep2: alpha=0.2222, lambda=0.2
## + Fold2.Rep2: alpha=0.3333, lambda=0.2
## - Fold2.Rep2: alpha=0.3333, lambda=0.2
## + Fold2.Rep2: alpha=0.4444, lambda=0.2
## - Fold2.Rep2: alpha=0.4444, lambda=0.2
## + Fold2.Rep2: alpha=0.5556, lambda=0.2
## - Fold2.Rep2: alpha=0.5556, lambda=0.2
## + Fold2.Rep2: alpha=0.6667, lambda=0.2
## - Fold2.Rep2: alpha=0.6667, lambda=0.2
## + Fold2.Rep2: alpha=0.7778, lambda=0.2
## - Fold2.Rep2: alpha=0.7778, lambda=0.2
## + Fold2.Rep2: alpha=0.8889, lambda=0.2
## - Fold2.Rep2: alpha=0.8889, lambda=0.2
## + Fold2.Rep2: alpha=1.0000, lambda=0.2
## - Fold2.Rep2: alpha=1.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.0000, lambda=0.2
## - Fold3.Rep2: alpha=0.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.1111, lambda=0.2
## - Fold3.Rep2: alpha=0.1111, lambda=0.2
## + Fold3.Rep2: alpha=0.2222, lambda=0.2
## - Fold3.Rep2: alpha=0.2222, lambda=0.2
## + Fold3.Rep2: alpha=0.3333, lambda=0.2
## - Fold3.Rep2: alpha=0.3333, lambda=0.2
## + Fold3.Rep2: alpha=0.4444, lambda=0.2
## - Fold3.Rep2: alpha=0.4444, lambda=0.2
## + Fold3.Rep2: alpha=0.5556, lambda=0.2
## - Fold3.Rep2: alpha=0.5556, lambda=0.2
## + Fold3.Rep2: alpha=0.6667, lambda=0.2
## - Fold3.Rep2: alpha=0.6667, lambda=0.2
## + Fold3.Rep2: alpha=0.7778, lambda=0.2
## - Fold3.Rep2: alpha=0.7778, lambda=0.2
## + Fold3.Rep2: alpha=0.8889, lambda=0.2
## - Fold3.Rep2: alpha=0.8889, lambda=0.2
## + Fold3.Rep2: alpha=1.0000, lambda=0.2
## - Fold3.Rep2: alpha=1.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.0000, lambda=0.2
## - Fold4.Rep2: alpha=0.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.1111, lambda=0.2
## - Fold4.Rep2: alpha=0.1111, lambda=0.2
## + Fold4.Rep2: alpha=0.2222, lambda=0.2
## - Fold4.Rep2: alpha=0.2222, lambda=0.2
## + Fold4.Rep2: alpha=0.3333, lambda=0.2
## - Fold4.Rep2: alpha=0.3333, lambda=0.2
## + Fold4.Rep2: alpha=0.4444, lambda=0.2
## - Fold4.Rep2: alpha=0.4444, lambda=0.2
## + Fold4.Rep2: alpha=0.5556, lambda=0.2
## - Fold4.Rep2: alpha=0.5556, lambda=0.2
## + Fold4.Rep2: alpha=0.6667, lambda=0.2
## - Fold4.Rep2: alpha=0.6667, lambda=0.2
## + Fold4.Rep2: alpha=0.7778, lambda=0.2
## - Fold4.Rep2: alpha=0.7778, lambda=0.2
## + Fold4.Rep2: alpha=0.8889, lambda=0.2
## - Fold4.Rep2: alpha=0.8889, lambda=0.2
## + Fold4.Rep2: alpha=1.0000, lambda=0.2
## - Fold4.Rep2: alpha=1.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.0000, lambda=0.2
## - Fold5.Rep2: alpha=0.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.1111, lambda=0.2
## - Fold5.Rep2: alpha=0.1111, lambda=0.2
## + Fold5.Rep2: alpha=0.2222, lambda=0.2
## - Fold5.Rep2: alpha=0.2222, lambda=0.2
## + Fold5.Rep2: alpha=0.3333, lambda=0.2
## - Fold5.Rep2: alpha=0.3333, lambda=0.2
## + Fold5.Rep2: alpha=0.4444, lambda=0.2
## - Fold5.Rep2: alpha=0.4444, lambda=0.2
## + Fold5.Rep2: alpha=0.5556, lambda=0.2
## - Fold5.Rep2: alpha=0.5556, lambda=0.2
## + Fold5.Rep2: alpha=0.6667, lambda=0.2
## - Fold5.Rep2: alpha=0.6667, lambda=0.2
## + Fold5.Rep2: alpha=0.7778, lambda=0.2
## - Fold5.Rep2: alpha=0.7778, lambda=0.2
## + Fold5.Rep2: alpha=0.8889, lambda=0.2
## - Fold5.Rep2: alpha=0.8889, lambda=0.2
## + Fold5.Rep2: alpha=1.0000, lambda=0.2
## - Fold5.Rep2: alpha=1.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.0000, lambda=0.2
## - Fold1.Rep3: alpha=0.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.1111, lambda=0.2
## - Fold1.Rep3: alpha=0.1111, lambda=0.2
## + Fold1.Rep3: alpha=0.2222, lambda=0.2
## - Fold1.Rep3: alpha=0.2222, lambda=0.2
## + Fold1.Rep3: alpha=0.3333, lambda=0.2
## - Fold1.Rep3: alpha=0.3333, lambda=0.2
## + Fold1.Rep3: alpha=0.4444, lambda=0.2
## - Fold1.Rep3: alpha=0.4444, lambda=0.2
## + Fold1.Rep3: alpha=0.5556, lambda=0.2
## - Fold1.Rep3: alpha=0.5556, lambda=0.2
## + Fold1.Rep3: alpha=0.6667, lambda=0.2
## - Fold1.Rep3: alpha=0.6667, lambda=0.2
## + Fold1.Rep3: alpha=0.7778, lambda=0.2
## - Fold1.Rep3: alpha=0.7778, lambda=0.2
## + Fold1.Rep3: alpha=0.8889, lambda=0.2
## - Fold1.Rep3: alpha=0.8889, lambda=0.2
## + Fold1.Rep3: alpha=1.0000, lambda=0.2
## - Fold1.Rep3: alpha=1.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.0000, lambda=0.2
## - Fold2.Rep3: alpha=0.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.1111, lambda=0.2
## - Fold2.Rep3: alpha=0.1111, lambda=0.2
## + Fold2.Rep3: alpha=0.2222, lambda=0.2
## - Fold2.Rep3: alpha=0.2222, lambda=0.2
## + Fold2.Rep3: alpha=0.3333, lambda=0.2
## - Fold2.Rep3: alpha=0.3333, lambda=0.2
## + Fold2.Rep3: alpha=0.4444, lambda=0.2
## - Fold2.Rep3: alpha=0.4444, lambda=0.2
## + Fold2.Rep3: alpha=0.5556, lambda=0.2
## - Fold2.Rep3: alpha=0.5556, lambda=0.2
## + Fold2.Rep3: alpha=0.6667, lambda=0.2
## - Fold2.Rep3: alpha=0.6667, lambda=0.2
## + Fold2.Rep3: alpha=0.7778, lambda=0.2
## - Fold2.Rep3: alpha=0.7778, lambda=0.2
## + Fold2.Rep3: alpha=0.8889, lambda=0.2
## - Fold2.Rep3: alpha=0.8889, lambda=0.2
## + Fold2.Rep3: alpha=1.0000, lambda=0.2
## - Fold2.Rep3: alpha=1.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.0000, lambda=0.2
## - Fold3.Rep3: alpha=0.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.1111, lambda=0.2
## - Fold3.Rep3: alpha=0.1111, lambda=0.2
## + Fold3.Rep3: alpha=0.2222, lambda=0.2
## - Fold3.Rep3: alpha=0.2222, lambda=0.2
## + Fold3.Rep3: alpha=0.3333, lambda=0.2
## - Fold3.Rep3: alpha=0.3333, lambda=0.2
## + Fold3.Rep3: alpha=0.4444, lambda=0.2
## - Fold3.Rep3: alpha=0.4444, lambda=0.2
## + Fold3.Rep3: alpha=0.5556, lambda=0.2
## - Fold3.Rep3: alpha=0.5556, lambda=0.2
## + Fold3.Rep3: alpha=0.6667, lambda=0.2
## - Fold3.Rep3: alpha=0.6667, lambda=0.2
## + Fold3.Rep3: alpha=0.7778, lambda=0.2
## - Fold3.Rep3: alpha=0.7778, lambda=0.2
## + Fold3.Rep3: alpha=0.8889, lambda=0.2
## - Fold3.Rep3: alpha=0.8889, lambda=0.2
## + Fold3.Rep3: alpha=1.0000, lambda=0.2
## - Fold3.Rep3: alpha=1.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.0000, lambda=0.2
## - Fold4.Rep3: alpha=0.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.1111, lambda=0.2
## - Fold4.Rep3: alpha=0.1111, lambda=0.2
## + Fold4.Rep3: alpha=0.2222, lambda=0.2
## - Fold4.Rep3: alpha=0.2222, lambda=0.2
## + Fold4.Rep3: alpha=0.3333, lambda=0.2
## - Fold4.Rep3: alpha=0.3333, lambda=0.2
## + Fold4.Rep3: alpha=0.4444, lambda=0.2
## - Fold4.Rep3: alpha=0.4444, lambda=0.2
## + Fold4.Rep3: alpha=0.5556, lambda=0.2
## - Fold4.Rep3: alpha=0.5556, lambda=0.2
## + Fold4.Rep3: alpha=0.6667, lambda=0.2
## - Fold4.Rep3: alpha=0.6667, lambda=0.2
## + Fold4.Rep3: alpha=0.7778, lambda=0.2
## - Fold4.Rep3: alpha=0.7778, lambda=0.2
## + Fold4.Rep3: alpha=0.8889, lambda=0.2
## - Fold4.Rep3: alpha=0.8889, lambda=0.2
## + Fold4.Rep3: alpha=1.0000, lambda=0.2
## - Fold4.Rep3: alpha=1.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.0000, lambda=0.2
## - Fold5.Rep3: alpha=0.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.1111, lambda=0.2
## - Fold5.Rep3: alpha=0.1111, lambda=0.2
## + Fold5.Rep3: alpha=0.2222, lambda=0.2
## - Fold5.Rep3: alpha=0.2222, lambda=0.2
## + Fold5.Rep3: alpha=0.3333, lambda=0.2
## - Fold5.Rep3: alpha=0.3333, lambda=0.2
## + Fold5.Rep3: alpha=0.4444, lambda=0.2
## - Fold5.Rep3: alpha=0.4444, lambda=0.2
## + Fold5.Rep3: alpha=0.5556, lambda=0.2
## - Fold5.Rep3: alpha=0.5556, lambda=0.2
## + Fold5.Rep3: alpha=0.6667, lambda=0.2
## - Fold5.Rep3: alpha=0.6667, lambda=0.2
## + Fold5.Rep3: alpha=0.7778, lambda=0.2
## - Fold5.Rep3: alpha=0.7778, lambda=0.2
## + Fold5.Rep3: alpha=0.8889, lambda=0.2
## - Fold5.Rep3: alpha=0.8889, lambda=0.2
## + Fold5.Rep3: alpha=1.0000, lambda=0.2
## - Fold5.Rep3: alpha=1.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.0000, lambda=0.2
## - Fold1.Rep4: alpha=0.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.1111, lambda=0.2
## - Fold1.Rep4: alpha=0.1111, lambda=0.2
## + Fold1.Rep4: alpha=0.2222, lambda=0.2
## - Fold1.Rep4: alpha=0.2222, lambda=0.2
## + Fold1.Rep4: alpha=0.3333, lambda=0.2
## - Fold1.Rep4: alpha=0.3333, lambda=0.2
## + Fold1.Rep4: alpha=0.4444, lambda=0.2
## - Fold1.Rep4: alpha=0.4444, lambda=0.2
## + Fold1.Rep4: alpha=0.5556, lambda=0.2
## - Fold1.Rep4: alpha=0.5556, lambda=0.2
## + Fold1.Rep4: alpha=0.6667, lambda=0.2
## - Fold1.Rep4: alpha=0.6667, lambda=0.2
## + Fold1.Rep4: alpha=0.7778, lambda=0.2
## - Fold1.Rep4: alpha=0.7778, lambda=0.2
## + Fold1.Rep4: alpha=0.8889, lambda=0.2
## - Fold1.Rep4: alpha=0.8889, lambda=0.2
## + Fold1.Rep4: alpha=1.0000, lambda=0.2
## - Fold1.Rep4: alpha=1.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.0000, lambda=0.2
## - Fold2.Rep4: alpha=0.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.1111, lambda=0.2
## - Fold2.Rep4: alpha=0.1111, lambda=0.2
## + Fold2.Rep4: alpha=0.2222, lambda=0.2
## - Fold2.Rep4: alpha=0.2222, lambda=0.2
## + Fold2.Rep4: alpha=0.3333, lambda=0.2
## - Fold2.Rep4: alpha=0.3333, lambda=0.2
## + Fold2.Rep4: alpha=0.4444, lambda=0.2
## - Fold2.Rep4: alpha=0.4444, lambda=0.2
## + Fold2.Rep4: alpha=0.5556, lambda=0.2
## - Fold2.Rep4: alpha=0.5556, lambda=0.2
## + Fold2.Rep4: alpha=0.6667, lambda=0.2
## - Fold2.Rep4: alpha=0.6667, lambda=0.2
## + Fold2.Rep4: alpha=0.7778, lambda=0.2
## - Fold2.Rep4: alpha=0.7778, lambda=0.2
## + Fold2.Rep4: alpha=0.8889, lambda=0.2
## - Fold2.Rep4: alpha=0.8889, lambda=0.2
## + Fold2.Rep4: alpha=1.0000, lambda=0.2
## - Fold2.Rep4: alpha=1.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.0000, lambda=0.2
## - Fold3.Rep4: alpha=0.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.1111, lambda=0.2
## - Fold3.Rep4: alpha=0.1111, lambda=0.2
## + Fold3.Rep4: alpha=0.2222, lambda=0.2
## - Fold3.Rep4: alpha=0.2222, lambda=0.2
## + Fold3.Rep4: alpha=0.3333, lambda=0.2
## - Fold3.Rep4: alpha=0.3333, lambda=0.2
## + Fold3.Rep4: alpha=0.4444, lambda=0.2
## - Fold3.Rep4: alpha=0.4444, lambda=0.2
## + Fold3.Rep4: alpha=0.5556, lambda=0.2
## - Fold3.Rep4: alpha=0.5556, lambda=0.2
## + Fold3.Rep4: alpha=0.6667, lambda=0.2
## - Fold3.Rep4: alpha=0.6667, lambda=0.2
## + Fold3.Rep4: alpha=0.7778, lambda=0.2
## - Fold3.Rep4: alpha=0.7778, lambda=0.2
## + Fold3.Rep4: alpha=0.8889, lambda=0.2
## - Fold3.Rep4: alpha=0.8889, lambda=0.2
## + Fold3.Rep4: alpha=1.0000, lambda=0.2
## - Fold3.Rep4: alpha=1.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.0000, lambda=0.2
## - Fold4.Rep4: alpha=0.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.1111, lambda=0.2
## - Fold4.Rep4: alpha=0.1111, lambda=0.2
## + Fold4.Rep4: alpha=0.2222, lambda=0.2
## - Fold4.Rep4: alpha=0.2222, lambda=0.2
## + Fold4.Rep4: alpha=0.3333, lambda=0.2
## - Fold4.Rep4: alpha=0.3333, lambda=0.2
## + Fold4.Rep4: alpha=0.4444, lambda=0.2
## - Fold4.Rep4: alpha=0.4444, lambda=0.2
## + Fold4.Rep4: alpha=0.5556, lambda=0.2
## - Fold4.Rep4: alpha=0.5556, lambda=0.2
## + Fold4.Rep4: alpha=0.6667, lambda=0.2
## - Fold4.Rep4: alpha=0.6667, lambda=0.2
## + Fold4.Rep4: alpha=0.7778, lambda=0.2
## - Fold4.Rep4: alpha=0.7778, lambda=0.2
## + Fold4.Rep4: alpha=0.8889, lambda=0.2
## - Fold4.Rep4: alpha=0.8889, lambda=0.2
## + Fold4.Rep4: alpha=1.0000, lambda=0.2
## - Fold4.Rep4: alpha=1.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.0000, lambda=0.2
## - Fold5.Rep4: alpha=0.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.1111, lambda=0.2
## - Fold5.Rep4: alpha=0.1111, lambda=0.2
## + Fold5.Rep4: alpha=0.2222, lambda=0.2
## - Fold5.Rep4: alpha=0.2222, lambda=0.2
## + Fold5.Rep4: alpha=0.3333, lambda=0.2
## - Fold5.Rep4: alpha=0.3333, lambda=0.2
## + Fold5.Rep4: alpha=0.4444, lambda=0.2
## - Fold5.Rep4: alpha=0.4444, lambda=0.2
## + Fold5.Rep4: alpha=0.5556, lambda=0.2
## - Fold5.Rep4: alpha=0.5556, lambda=0.2
## + Fold5.Rep4: alpha=0.6667, lambda=0.2
## - Fold5.Rep4: alpha=0.6667, lambda=0.2
## + Fold5.Rep4: alpha=0.7778, lambda=0.2
## - Fold5.Rep4: alpha=0.7778, lambda=0.2
## + Fold5.Rep4: alpha=0.8889, lambda=0.2
## - Fold5.Rep4: alpha=0.8889, lambda=0.2
## + Fold5.Rep4: alpha=1.0000, lambda=0.2
## - Fold5.Rep4: alpha=1.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.0000, lambda=0.2
## - Fold1.Rep5: alpha=0.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.1111, lambda=0.2
## - Fold1.Rep5: alpha=0.1111, lambda=0.2
## + Fold1.Rep5: alpha=0.2222, lambda=0.2
## - Fold1.Rep5: alpha=0.2222, lambda=0.2
## + Fold1.Rep5: alpha=0.3333, lambda=0.2
## - Fold1.Rep5: alpha=0.3333, lambda=0.2
## + Fold1.Rep5: alpha=0.4444, lambda=0.2
## - Fold1.Rep5: alpha=0.4444, lambda=0.2
## + Fold1.Rep5: alpha=0.5556, lambda=0.2
## - Fold1.Rep5: alpha=0.5556, lambda=0.2
## + Fold1.Rep5: alpha=0.6667, lambda=0.2
## - Fold1.Rep5: alpha=0.6667, lambda=0.2
## + Fold1.Rep5: alpha=0.7778, lambda=0.2
## - Fold1.Rep5: alpha=0.7778, lambda=0.2
## + Fold1.Rep5: alpha=0.8889, lambda=0.2
## - Fold1.Rep5: alpha=0.8889, lambda=0.2
## + Fold1.Rep5: alpha=1.0000, lambda=0.2
## - Fold1.Rep5: alpha=1.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.0000, lambda=0.2
## - Fold2.Rep5: alpha=0.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.1111, lambda=0.2
## - Fold2.Rep5: alpha=0.1111, lambda=0.2
## + Fold2.Rep5: alpha=0.2222, lambda=0.2
## - Fold2.Rep5: alpha=0.2222, lambda=0.2
## + Fold2.Rep5: alpha=0.3333, lambda=0.2
## - Fold2.Rep5: alpha=0.3333, lambda=0.2
## + Fold2.Rep5: alpha=0.4444, lambda=0.2
## - Fold2.Rep5: alpha=0.4444, lambda=0.2
## + Fold2.Rep5: alpha=0.5556, lambda=0.2
## - Fold2.Rep5: alpha=0.5556, lambda=0.2
## + Fold2.Rep5: alpha=0.6667, lambda=0.2
## - Fold2.Rep5: alpha=0.6667, lambda=0.2
## + Fold2.Rep5: alpha=0.7778, lambda=0.2
## - Fold2.Rep5: alpha=0.7778, lambda=0.2
## + Fold2.Rep5: alpha=0.8889, lambda=0.2
## - Fold2.Rep5: alpha=0.8889, lambda=0.2
## + Fold2.Rep5: alpha=1.0000, lambda=0.2
## - Fold2.Rep5: alpha=1.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.0000, lambda=0.2
## - Fold3.Rep5: alpha=0.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.1111, lambda=0.2
## - Fold3.Rep5: alpha=0.1111, lambda=0.2
## + Fold3.Rep5: alpha=0.2222, lambda=0.2
## - Fold3.Rep5: alpha=0.2222, lambda=0.2
## + Fold3.Rep5: alpha=0.3333, lambda=0.2
## - Fold3.Rep5: alpha=0.3333, lambda=0.2
## + Fold3.Rep5: alpha=0.4444, lambda=0.2
## - Fold3.Rep5: alpha=0.4444, lambda=0.2
## + Fold3.Rep5: alpha=0.5556, lambda=0.2
## - Fold3.Rep5: alpha=0.5556, lambda=0.2
## + Fold3.Rep5: alpha=0.6667, lambda=0.2
## - Fold3.Rep5: alpha=0.6667, lambda=0.2
## + Fold3.Rep5: alpha=0.7778, lambda=0.2
## - Fold3.Rep5: alpha=0.7778, lambda=0.2
## + Fold3.Rep5: alpha=0.8889, lambda=0.2
## - Fold3.Rep5: alpha=0.8889, lambda=0.2
## + Fold3.Rep5: alpha=1.0000, lambda=0.2
## - Fold3.Rep5: alpha=1.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.0000, lambda=0.2
## - Fold4.Rep5: alpha=0.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.1111, lambda=0.2
## - Fold4.Rep5: alpha=0.1111, lambda=0.2
## + Fold4.Rep5: alpha=0.2222, lambda=0.2
## - Fold4.Rep5: alpha=0.2222, lambda=0.2
## + Fold4.Rep5: alpha=0.3333, lambda=0.2
## - Fold4.Rep5: alpha=0.3333, lambda=0.2
## + Fold4.Rep5: alpha=0.4444, lambda=0.2
## - Fold4.Rep5: alpha=0.4444, lambda=0.2
## + Fold4.Rep5: alpha=0.5556, lambda=0.2
## - Fold4.Rep5: alpha=0.5556, lambda=0.2
## + Fold4.Rep5: alpha=0.6667, lambda=0.2
## - Fold4.Rep5: alpha=0.6667, lambda=0.2
## + Fold4.Rep5: alpha=0.7778, lambda=0.2
## - Fold4.Rep5: alpha=0.7778, lambda=0.2
## + Fold4.Rep5: alpha=0.8889, lambda=0.2
## - Fold4.Rep5: alpha=0.8889, lambda=0.2
## + Fold4.Rep5: alpha=1.0000, lambda=0.2
## - Fold4.Rep5: alpha=1.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.0000, lambda=0.2
## - Fold5.Rep5: alpha=0.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.1111, lambda=0.2
## - Fold5.Rep5: alpha=0.1111, lambda=0.2
## + Fold5.Rep5: alpha=0.2222, lambda=0.2
## - Fold5.Rep5: alpha=0.2222, lambda=0.2
## + Fold5.Rep5: alpha=0.3333, lambda=0.2
## - Fold5.Rep5: alpha=0.3333, lambda=0.2
## + Fold5.Rep5: alpha=0.4444, lambda=0.2
## - Fold5.Rep5: alpha=0.4444, lambda=0.2
## + Fold5.Rep5: alpha=0.5556, lambda=0.2
## - Fold5.Rep5: alpha=0.5556, lambda=0.2
## + Fold5.Rep5: alpha=0.6667, lambda=0.2
## - Fold5.Rep5: alpha=0.6667, lambda=0.2
## + Fold5.Rep5: alpha=0.7778, lambda=0.2
## - Fold5.Rep5: alpha=0.7778, lambda=0.2
## + Fold5.Rep5: alpha=0.8889, lambda=0.2
## - Fold5.Rep5: alpha=0.8889, lambda=0.2
## + Fold5.Rep5: alpha=1.0000, lambda=0.2
## - Fold5.Rep5: alpha=1.0000, lambda=0.2
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 1, lambda = 0.1 on full training set
print(en_p)
## glmnet
##
## 9 samples
## 15 predictors
##
## No pre-processing
## Resampling: Cross-Validated (5 fold, repeated 5 times)
## Summary of sample sizes: 7, 8, 7, 7, 7, 7, ...
## Resampling results across tuning parameters:
##
## alpha lambda RMSE Rsquared MAE
## 0.0000000 0.000100 0.2884883 1 0.2731310
## 0.0000000 0.050075 0.2884883 1 0.2731310
## 0.0000000 0.100050 0.2884883 1 0.2731310
## 0.0000000 0.150025 0.2884883 1 0.2731310
## 0.0000000 0.200000 0.2884883 1 0.2731310
## 0.1111111 0.000100 0.4161650 1 0.3712960
## 0.1111111 0.050075 0.3779783 1 0.3363616
## 0.1111111 0.100050 0.3407692 1 0.3063865
## 0.1111111 0.150025 0.3159271 1 0.2850946
## 0.1111111 0.200000 0.3018028 1 0.2742277
## 0.2222222 0.000100 0.4482313 1 0.3987135
## 0.2222222 0.050075 0.3728569 1 0.3312740
## 0.2222222 0.100050 0.3223021 1 0.2869565
## 0.2222222 0.150025 0.2974051 1 0.2663367
## 0.2222222 0.200000 0.2888337 1 0.2626941
## 0.3333333 0.000100 0.4729963 1 0.4244349
## 0.3333333 0.050075 0.3619451 1 0.3214525
## 0.3333333 0.100050 0.3050714 1 0.2682782
## 0.3333333 0.150025 0.2893753 1 0.2604168
## 0.3333333 0.200000 0.2849683 1 0.2597520
## 0.4444444 0.000100 0.4979959 1 0.4477444
## 0.4444444 0.050075 0.3472966 1 0.3074893
## 0.4444444 0.100050 0.2966345 1 0.2624511
## 0.4444444 0.150025 0.2836646 1 0.2571966
## 0.4444444 0.200000 0.2861085 1 0.2659180
## 0.5555556 0.000100 0.5184818 1 0.4675372
## 0.5555556 0.050075 0.3365184 1 0.2980955
## 0.5555556 0.100050 0.2882350 1 0.2577133
## 0.5555556 0.150025 0.2819720 1 0.2595075
## 0.5555556 0.200000 0.2860976 1 0.2670695
## 0.6666667 0.000100 0.5380984 1 0.4848576
## 0.6666667 0.050075 0.3278270 1 0.2906837
## 0.6666667 0.100050 0.2814685 1 0.2541339
## 0.6666667 0.150025 0.2803094 1 0.2594379
## 0.6666667 0.200000 0.2865466 1 0.2671460
## 0.7777778 0.000100 0.5579192 1 0.5039367
## 0.7777778 0.050075 0.3251949 1 0.2905071
## 0.7777778 0.100050 0.2808168 1 0.2556296
## 0.7777778 0.150025 0.2791882 1 0.2577627
## 0.7777778 0.200000 0.2949911 1 0.2761257
## 0.8888889 0.000100 0.5575253 1 0.5034880
## 0.8888889 0.050075 0.3238583 1 0.2929500
## 0.8888889 0.100050 0.2759525 1 0.2519883
## 0.8888889 0.150025 0.2845154 1 0.2632881
## 0.8888889 0.200000 0.3021457 1 0.2840843
## 1.0000000 0.000100 0.5756223 1 0.5241637
## 1.0000000 0.050075 0.3194978 1 0.2902943
## 1.0000000 0.100050 0.2730480 1 0.2477272
## 1.0000000 0.150025 0.2902694 1 0.2701464
## 1.0000000 0.200000 0.3097939 1 0.2922539
##
## RMSE was used to select the optimal model using the smallest value.
## The final values used for the model were alpha = 1 and lambda = 0.10005.
plot(varImp(en_p, scale = TRUE))
# Coefficients
coefs <- coef(en_p$finalModel, en_p$bestTune$lambda)
coefs.zero <- coefs[which(coefs == 0),]
coef_order <- order(abs(coefs), decreasing = TRUE)
head(as.matrix(coefs[coef_order,]), 10)
## [,1]
## (Intercept) 1.3228105
## Dist_Pour 0.4047863
## Avg_Flow 0.0000000
## Flow_Variance 0.0000000
## Avg_Sand 0.0000000
## Avg_Gravel 0.0000000
## Avg_Cobble 0.0000000
## Avg_Boulder 0.0000000
## Avg_Bedrock 0.0000000
## Avg_Organic 0.0000000
### --- BENTHIC BIOMASS MODEL --- ###
bennet <- merged[,-c(1,3:5)]
train <- bennet[split_ind == 1,]
test <- bennet[split_ind == 2,]
set.seed(1234)
en_b <- train(Benthic ~ .,
train,
method = 'glmnet',
tuneGrid = expand.grid(alpha = seq(0, 1, length = 10),
lambda = seq(0.0001, 0.2, length = 5)),
trControl = custom)
## + Fold1.Rep1: alpha=0.0000, lambda=0.2
## - Fold1.Rep1: alpha=0.0000, lambda=0.2
## + Fold1.Rep1: alpha=0.1111, lambda=0.2
## - Fold1.Rep1: alpha=0.1111, lambda=0.2
## + Fold1.Rep1: alpha=0.2222, lambda=0.2
## - Fold1.Rep1: alpha=0.2222, lambda=0.2
## + Fold1.Rep1: alpha=0.3333, lambda=0.2
## - Fold1.Rep1: alpha=0.3333, lambda=0.2
## + Fold1.Rep1: alpha=0.4444, lambda=0.2
## - Fold1.Rep1: alpha=0.4444, lambda=0.2
## + Fold1.Rep1: alpha=0.5556, lambda=0.2
## - Fold1.Rep1: alpha=0.5556, lambda=0.2
## + Fold1.Rep1: alpha=0.6667, lambda=0.2
## - Fold1.Rep1: alpha=0.6667, lambda=0.2
## + Fold1.Rep1: alpha=0.7778, lambda=0.2
## - Fold1.Rep1: alpha=0.7778, lambda=0.2
## + Fold1.Rep1: alpha=0.8889, lambda=0.2
## - Fold1.Rep1: alpha=0.8889, lambda=0.2
## + Fold1.Rep1: alpha=1.0000, lambda=0.2
## - Fold1.Rep1: alpha=1.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.0000, lambda=0.2
## - Fold2.Rep1: alpha=0.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.1111, lambda=0.2
## - Fold2.Rep1: alpha=0.1111, lambda=0.2
## + Fold2.Rep1: alpha=0.2222, lambda=0.2
## - Fold2.Rep1: alpha=0.2222, lambda=0.2
## + Fold2.Rep1: alpha=0.3333, lambda=0.2
## - Fold2.Rep1: alpha=0.3333, lambda=0.2
## + Fold2.Rep1: alpha=0.4444, lambda=0.2
## - Fold2.Rep1: alpha=0.4444, lambda=0.2
## + Fold2.Rep1: alpha=0.5556, lambda=0.2
## - Fold2.Rep1: alpha=0.5556, lambda=0.2
## + Fold2.Rep1: alpha=0.6667, lambda=0.2
## - Fold2.Rep1: alpha=0.6667, lambda=0.2
## + Fold2.Rep1: alpha=0.7778, lambda=0.2
## - Fold2.Rep1: alpha=0.7778, lambda=0.2
## + Fold2.Rep1: alpha=0.8889, lambda=0.2
## - Fold2.Rep1: alpha=0.8889, lambda=0.2
## + Fold2.Rep1: alpha=1.0000, lambda=0.2
## - Fold2.Rep1: alpha=1.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.0000, lambda=0.2
## - Fold3.Rep1: alpha=0.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.1111, lambda=0.2
## - Fold3.Rep1: alpha=0.1111, lambda=0.2
## + Fold3.Rep1: alpha=0.2222, lambda=0.2
## - Fold3.Rep1: alpha=0.2222, lambda=0.2
## + Fold3.Rep1: alpha=0.3333, lambda=0.2
## - Fold3.Rep1: alpha=0.3333, lambda=0.2
## + Fold3.Rep1: alpha=0.4444, lambda=0.2
## - Fold3.Rep1: alpha=0.4444, lambda=0.2
## + Fold3.Rep1: alpha=0.5556, lambda=0.2
## - Fold3.Rep1: alpha=0.5556, lambda=0.2
## + Fold3.Rep1: alpha=0.6667, lambda=0.2
## - Fold3.Rep1: alpha=0.6667, lambda=0.2
## + Fold3.Rep1: alpha=0.7778, lambda=0.2
## - Fold3.Rep1: alpha=0.7778, lambda=0.2
## + Fold3.Rep1: alpha=0.8889, lambda=0.2
## - Fold3.Rep1: alpha=0.8889, lambda=0.2
## + Fold3.Rep1: alpha=1.0000, lambda=0.2
## - Fold3.Rep1: alpha=1.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.0000, lambda=0.2
## - Fold4.Rep1: alpha=0.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.1111, lambda=0.2
## - Fold4.Rep1: alpha=0.1111, lambda=0.2
## + Fold4.Rep1: alpha=0.2222, lambda=0.2
## - Fold4.Rep1: alpha=0.2222, lambda=0.2
## + Fold4.Rep1: alpha=0.3333, lambda=0.2
## - Fold4.Rep1: alpha=0.3333, lambda=0.2
## + Fold4.Rep1: alpha=0.4444, lambda=0.2
## - Fold4.Rep1: alpha=0.4444, lambda=0.2
## + Fold4.Rep1: alpha=0.5556, lambda=0.2
## - Fold4.Rep1: alpha=0.5556, lambda=0.2
## + Fold4.Rep1: alpha=0.6667, lambda=0.2
## - Fold4.Rep1: alpha=0.6667, lambda=0.2
## + Fold4.Rep1: alpha=0.7778, lambda=0.2
## - Fold4.Rep1: alpha=0.7778, lambda=0.2
## + Fold4.Rep1: alpha=0.8889, lambda=0.2
## - Fold4.Rep1: alpha=0.8889, lambda=0.2
## + Fold4.Rep1: alpha=1.0000, lambda=0.2
## - Fold4.Rep1: alpha=1.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.0000, lambda=0.2
## - Fold5.Rep1: alpha=0.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.1111, lambda=0.2
## - Fold5.Rep1: alpha=0.1111, lambda=0.2
## + Fold5.Rep1: alpha=0.2222, lambda=0.2
## - Fold5.Rep1: alpha=0.2222, lambda=0.2
## + Fold5.Rep1: alpha=0.3333, lambda=0.2
## - Fold5.Rep1: alpha=0.3333, lambda=0.2
## + Fold5.Rep1: alpha=0.4444, lambda=0.2
## - Fold5.Rep1: alpha=0.4444, lambda=0.2
## + Fold5.Rep1: alpha=0.5556, lambda=0.2
## - Fold5.Rep1: alpha=0.5556, lambda=0.2
## + Fold5.Rep1: alpha=0.6667, lambda=0.2
## - Fold5.Rep1: alpha=0.6667, lambda=0.2
## + Fold5.Rep1: alpha=0.7778, lambda=0.2
## - Fold5.Rep1: alpha=0.7778, lambda=0.2
## + Fold5.Rep1: alpha=0.8889, lambda=0.2
## - Fold5.Rep1: alpha=0.8889, lambda=0.2
## + Fold5.Rep1: alpha=1.0000, lambda=0.2
## - Fold5.Rep1: alpha=1.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.0000, lambda=0.2
## - Fold1.Rep2: alpha=0.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.1111, lambda=0.2
## - Fold1.Rep2: alpha=0.1111, lambda=0.2
## + Fold1.Rep2: alpha=0.2222, lambda=0.2
## - Fold1.Rep2: alpha=0.2222, lambda=0.2
## + Fold1.Rep2: alpha=0.3333, lambda=0.2
## - Fold1.Rep2: alpha=0.3333, lambda=0.2
## + Fold1.Rep2: alpha=0.4444, lambda=0.2
## - Fold1.Rep2: alpha=0.4444, lambda=0.2
## + Fold1.Rep2: alpha=0.5556, lambda=0.2
## - Fold1.Rep2: alpha=0.5556, lambda=0.2
## + Fold1.Rep2: alpha=0.6667, lambda=0.2
## - Fold1.Rep2: alpha=0.6667, lambda=0.2
## + Fold1.Rep2: alpha=0.7778, lambda=0.2
## - Fold1.Rep2: alpha=0.7778, lambda=0.2
## + Fold1.Rep2: alpha=0.8889, lambda=0.2
## - Fold1.Rep2: alpha=0.8889, lambda=0.2
## + Fold1.Rep2: alpha=1.0000, lambda=0.2
## - Fold1.Rep2: alpha=1.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.0000, lambda=0.2
## - Fold2.Rep2: alpha=0.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.1111, lambda=0.2
## - Fold2.Rep2: alpha=0.1111, lambda=0.2
## + Fold2.Rep2: alpha=0.2222, lambda=0.2
## - Fold2.Rep2: alpha=0.2222, lambda=0.2
## + Fold2.Rep2: alpha=0.3333, lambda=0.2
## - Fold2.Rep2: alpha=0.3333, lambda=0.2
## + Fold2.Rep2: alpha=0.4444, lambda=0.2
## - Fold2.Rep2: alpha=0.4444, lambda=0.2
## + Fold2.Rep2: alpha=0.5556, lambda=0.2
## - Fold2.Rep2: alpha=0.5556, lambda=0.2
## + Fold2.Rep2: alpha=0.6667, lambda=0.2
## - Fold2.Rep2: alpha=0.6667, lambda=0.2
## + Fold2.Rep2: alpha=0.7778, lambda=0.2
## - Fold2.Rep2: alpha=0.7778, lambda=0.2
## + Fold2.Rep2: alpha=0.8889, lambda=0.2
## - Fold2.Rep2: alpha=0.8889, lambda=0.2
## + Fold2.Rep2: alpha=1.0000, lambda=0.2
## - Fold2.Rep2: alpha=1.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.0000, lambda=0.2
## - Fold3.Rep2: alpha=0.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.1111, lambda=0.2
## - Fold3.Rep2: alpha=0.1111, lambda=0.2
## + Fold3.Rep2: alpha=0.2222, lambda=0.2
## - Fold3.Rep2: alpha=0.2222, lambda=0.2
## + Fold3.Rep2: alpha=0.3333, lambda=0.2
## - Fold3.Rep2: alpha=0.3333, lambda=0.2
## + Fold3.Rep2: alpha=0.4444, lambda=0.2
## - Fold3.Rep2: alpha=0.4444, lambda=0.2
## + Fold3.Rep2: alpha=0.5556, lambda=0.2
## - Fold3.Rep2: alpha=0.5556, lambda=0.2
## + Fold3.Rep2: alpha=0.6667, lambda=0.2
## - Fold3.Rep2: alpha=0.6667, lambda=0.2
## + Fold3.Rep2: alpha=0.7778, lambda=0.2
## - Fold3.Rep2: alpha=0.7778, lambda=0.2
## + Fold3.Rep2: alpha=0.8889, lambda=0.2
## - Fold3.Rep2: alpha=0.8889, lambda=0.2
## + Fold3.Rep2: alpha=1.0000, lambda=0.2
## - Fold3.Rep2: alpha=1.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.0000, lambda=0.2
## - Fold4.Rep2: alpha=0.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.1111, lambda=0.2
## - Fold4.Rep2: alpha=0.1111, lambda=0.2
## + Fold4.Rep2: alpha=0.2222, lambda=0.2
## - Fold4.Rep2: alpha=0.2222, lambda=0.2
## + Fold4.Rep2: alpha=0.3333, lambda=0.2
## - Fold4.Rep2: alpha=0.3333, lambda=0.2
## + Fold4.Rep2: alpha=0.4444, lambda=0.2
## - Fold4.Rep2: alpha=0.4444, lambda=0.2
## + Fold4.Rep2: alpha=0.5556, lambda=0.2
## - Fold4.Rep2: alpha=0.5556, lambda=0.2
## + Fold4.Rep2: alpha=0.6667, lambda=0.2
## - Fold4.Rep2: alpha=0.6667, lambda=0.2
## + Fold4.Rep2: alpha=0.7778, lambda=0.2
## - Fold4.Rep2: alpha=0.7778, lambda=0.2
## + Fold4.Rep2: alpha=0.8889, lambda=0.2
## - Fold4.Rep2: alpha=0.8889, lambda=0.2
## + Fold4.Rep2: alpha=1.0000, lambda=0.2
## - Fold4.Rep2: alpha=1.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.0000, lambda=0.2
## - Fold5.Rep2: alpha=0.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.1111, lambda=0.2
## - Fold5.Rep2: alpha=0.1111, lambda=0.2
## + Fold5.Rep2: alpha=0.2222, lambda=0.2
## - Fold5.Rep2: alpha=0.2222, lambda=0.2
## + Fold5.Rep2: alpha=0.3333, lambda=0.2
## - Fold5.Rep2: alpha=0.3333, lambda=0.2
## + Fold5.Rep2: alpha=0.4444, lambda=0.2
## - Fold5.Rep2: alpha=0.4444, lambda=0.2
## + Fold5.Rep2: alpha=0.5556, lambda=0.2
## - Fold5.Rep2: alpha=0.5556, lambda=0.2
## + Fold5.Rep2: alpha=0.6667, lambda=0.2
## - Fold5.Rep2: alpha=0.6667, lambda=0.2
## + Fold5.Rep2: alpha=0.7778, lambda=0.2
## - Fold5.Rep2: alpha=0.7778, lambda=0.2
## + Fold5.Rep2: alpha=0.8889, lambda=0.2
## - Fold5.Rep2: alpha=0.8889, lambda=0.2
## + Fold5.Rep2: alpha=1.0000, lambda=0.2
## - Fold5.Rep2: alpha=1.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.0000, lambda=0.2
## - Fold1.Rep3: alpha=0.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.1111, lambda=0.2
## - Fold1.Rep3: alpha=0.1111, lambda=0.2
## + Fold1.Rep3: alpha=0.2222, lambda=0.2
## - Fold1.Rep3: alpha=0.2222, lambda=0.2
## + Fold1.Rep3: alpha=0.3333, lambda=0.2
## - Fold1.Rep3: alpha=0.3333, lambda=0.2
## + Fold1.Rep3: alpha=0.4444, lambda=0.2
## - Fold1.Rep3: alpha=0.4444, lambda=0.2
## + Fold1.Rep3: alpha=0.5556, lambda=0.2
## - Fold1.Rep3: alpha=0.5556, lambda=0.2
## + Fold1.Rep3: alpha=0.6667, lambda=0.2
## - Fold1.Rep3: alpha=0.6667, lambda=0.2
## + Fold1.Rep3: alpha=0.7778, lambda=0.2
## - Fold1.Rep3: alpha=0.7778, lambda=0.2
## + Fold1.Rep3: alpha=0.8889, lambda=0.2
## - Fold1.Rep3: alpha=0.8889, lambda=0.2
## + Fold1.Rep3: alpha=1.0000, lambda=0.2
## - Fold1.Rep3: alpha=1.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.0000, lambda=0.2
## - Fold2.Rep3: alpha=0.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.1111, lambda=0.2
## - Fold2.Rep3: alpha=0.1111, lambda=0.2
## + Fold2.Rep3: alpha=0.2222, lambda=0.2
## - Fold2.Rep3: alpha=0.2222, lambda=0.2
## + Fold2.Rep3: alpha=0.3333, lambda=0.2
## - Fold2.Rep3: alpha=0.3333, lambda=0.2
## + Fold2.Rep3: alpha=0.4444, lambda=0.2
## - Fold2.Rep3: alpha=0.4444, lambda=0.2
## + Fold2.Rep3: alpha=0.5556, lambda=0.2
## - Fold2.Rep3: alpha=0.5556, lambda=0.2
## + Fold2.Rep3: alpha=0.6667, lambda=0.2
## - Fold2.Rep3: alpha=0.6667, lambda=0.2
## + Fold2.Rep3: alpha=0.7778, lambda=0.2
## - Fold2.Rep3: alpha=0.7778, lambda=0.2
## + Fold2.Rep3: alpha=0.8889, lambda=0.2
## - Fold2.Rep3: alpha=0.8889, lambda=0.2
## + Fold2.Rep3: alpha=1.0000, lambda=0.2
## - Fold2.Rep3: alpha=1.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.0000, lambda=0.2
## - Fold3.Rep3: alpha=0.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.1111, lambda=0.2
## - Fold3.Rep3: alpha=0.1111, lambda=0.2
## + Fold3.Rep3: alpha=0.2222, lambda=0.2
## - Fold3.Rep3: alpha=0.2222, lambda=0.2
## + Fold3.Rep3: alpha=0.3333, lambda=0.2
## - Fold3.Rep3: alpha=0.3333, lambda=0.2
## + Fold3.Rep3: alpha=0.4444, lambda=0.2
## - Fold3.Rep3: alpha=0.4444, lambda=0.2
## + Fold3.Rep3: alpha=0.5556, lambda=0.2
## - Fold3.Rep3: alpha=0.5556, lambda=0.2
## + Fold3.Rep3: alpha=0.6667, lambda=0.2
## - Fold3.Rep3: alpha=0.6667, lambda=0.2
## + Fold3.Rep3: alpha=0.7778, lambda=0.2
## - Fold3.Rep3: alpha=0.7778, lambda=0.2
## + Fold3.Rep3: alpha=0.8889, lambda=0.2
## - Fold3.Rep3: alpha=0.8889, lambda=0.2
## + Fold3.Rep3: alpha=1.0000, lambda=0.2
## - Fold3.Rep3: alpha=1.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.0000, lambda=0.2
## - Fold4.Rep3: alpha=0.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.1111, lambda=0.2
## - Fold4.Rep3: alpha=0.1111, lambda=0.2
## + Fold4.Rep3: alpha=0.2222, lambda=0.2
## - Fold4.Rep3: alpha=0.2222, lambda=0.2
## + Fold4.Rep3: alpha=0.3333, lambda=0.2
## - Fold4.Rep3: alpha=0.3333, lambda=0.2
## + Fold4.Rep3: alpha=0.4444, lambda=0.2
## - Fold4.Rep3: alpha=0.4444, lambda=0.2
## + Fold4.Rep3: alpha=0.5556, lambda=0.2
## - Fold4.Rep3: alpha=0.5556, lambda=0.2
## + Fold4.Rep3: alpha=0.6667, lambda=0.2
## - Fold4.Rep3: alpha=0.6667, lambda=0.2
## + Fold4.Rep3: alpha=0.7778, lambda=0.2
## - Fold4.Rep3: alpha=0.7778, lambda=0.2
## + Fold4.Rep3: alpha=0.8889, lambda=0.2
## - Fold4.Rep3: alpha=0.8889, lambda=0.2
## + Fold4.Rep3: alpha=1.0000, lambda=0.2
## - Fold4.Rep3: alpha=1.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.0000, lambda=0.2
## - Fold5.Rep3: alpha=0.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.1111, lambda=0.2
## - Fold5.Rep3: alpha=0.1111, lambda=0.2
## + Fold5.Rep3: alpha=0.2222, lambda=0.2
## - Fold5.Rep3: alpha=0.2222, lambda=0.2
## + Fold5.Rep3: alpha=0.3333, lambda=0.2
## - Fold5.Rep3: alpha=0.3333, lambda=0.2
## + Fold5.Rep3: alpha=0.4444, lambda=0.2
## - Fold5.Rep3: alpha=0.4444, lambda=0.2
## + Fold5.Rep3: alpha=0.5556, lambda=0.2
## - Fold5.Rep3: alpha=0.5556, lambda=0.2
## + Fold5.Rep3: alpha=0.6667, lambda=0.2
## - Fold5.Rep3: alpha=0.6667, lambda=0.2
## + Fold5.Rep3: alpha=0.7778, lambda=0.2
## - Fold5.Rep3: alpha=0.7778, lambda=0.2
## + Fold5.Rep3: alpha=0.8889, lambda=0.2
## - Fold5.Rep3: alpha=0.8889, lambda=0.2
## + Fold5.Rep3: alpha=1.0000, lambda=0.2
## - Fold5.Rep3: alpha=1.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.0000, lambda=0.2
## - Fold1.Rep4: alpha=0.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.1111, lambda=0.2
## - Fold1.Rep4: alpha=0.1111, lambda=0.2
## + Fold1.Rep4: alpha=0.2222, lambda=0.2
## - Fold1.Rep4: alpha=0.2222, lambda=0.2
## + Fold1.Rep4: alpha=0.3333, lambda=0.2
## - Fold1.Rep4: alpha=0.3333, lambda=0.2
## + Fold1.Rep4: alpha=0.4444, lambda=0.2
## - Fold1.Rep4: alpha=0.4444, lambda=0.2
## + Fold1.Rep4: alpha=0.5556, lambda=0.2
## - Fold1.Rep4: alpha=0.5556, lambda=0.2
## + Fold1.Rep4: alpha=0.6667, lambda=0.2
## - Fold1.Rep4: alpha=0.6667, lambda=0.2
## + Fold1.Rep4: alpha=0.7778, lambda=0.2
## - Fold1.Rep4: alpha=0.7778, lambda=0.2
## + Fold1.Rep4: alpha=0.8889, lambda=0.2
## - Fold1.Rep4: alpha=0.8889, lambda=0.2
## + Fold1.Rep4: alpha=1.0000, lambda=0.2
## - Fold1.Rep4: alpha=1.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.0000, lambda=0.2
## - Fold2.Rep4: alpha=0.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.1111, lambda=0.2
## - Fold2.Rep4: alpha=0.1111, lambda=0.2
## + Fold2.Rep4: alpha=0.2222, lambda=0.2
## - Fold2.Rep4: alpha=0.2222, lambda=0.2
## + Fold2.Rep4: alpha=0.3333, lambda=0.2
## - Fold2.Rep4: alpha=0.3333, lambda=0.2
## + Fold2.Rep4: alpha=0.4444, lambda=0.2
## - Fold2.Rep4: alpha=0.4444, lambda=0.2
## + Fold2.Rep4: alpha=0.5556, lambda=0.2
## - Fold2.Rep4: alpha=0.5556, lambda=0.2
## + Fold2.Rep4: alpha=0.6667, lambda=0.2
## - Fold2.Rep4: alpha=0.6667, lambda=0.2
## + Fold2.Rep4: alpha=0.7778, lambda=0.2
## - Fold2.Rep4: alpha=0.7778, lambda=0.2
## + Fold2.Rep4: alpha=0.8889, lambda=0.2
## - Fold2.Rep4: alpha=0.8889, lambda=0.2
## + Fold2.Rep4: alpha=1.0000, lambda=0.2
## - Fold2.Rep4: alpha=1.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.0000, lambda=0.2
## - Fold3.Rep4: alpha=0.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.1111, lambda=0.2
## - Fold3.Rep4: alpha=0.1111, lambda=0.2
## + Fold3.Rep4: alpha=0.2222, lambda=0.2
## - Fold3.Rep4: alpha=0.2222, lambda=0.2
## + Fold3.Rep4: alpha=0.3333, lambda=0.2
## - Fold3.Rep4: alpha=0.3333, lambda=0.2
## + Fold3.Rep4: alpha=0.4444, lambda=0.2
## - Fold3.Rep4: alpha=0.4444, lambda=0.2
## + Fold3.Rep4: alpha=0.5556, lambda=0.2
## - Fold3.Rep4: alpha=0.5556, lambda=0.2
## + Fold3.Rep4: alpha=0.6667, lambda=0.2
## - Fold3.Rep4: alpha=0.6667, lambda=0.2
## + Fold3.Rep4: alpha=0.7778, lambda=0.2
## - Fold3.Rep4: alpha=0.7778, lambda=0.2
## + Fold3.Rep4: alpha=0.8889, lambda=0.2
## - Fold3.Rep4: alpha=0.8889, lambda=0.2
## + Fold3.Rep4: alpha=1.0000, lambda=0.2
## - Fold3.Rep4: alpha=1.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.0000, lambda=0.2
## - Fold4.Rep4: alpha=0.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.1111, lambda=0.2
## - Fold4.Rep4: alpha=0.1111, lambda=0.2
## + Fold4.Rep4: alpha=0.2222, lambda=0.2
## - Fold4.Rep4: alpha=0.2222, lambda=0.2
## + Fold4.Rep4: alpha=0.3333, lambda=0.2
## - Fold4.Rep4: alpha=0.3333, lambda=0.2
## + Fold4.Rep4: alpha=0.4444, lambda=0.2
## - Fold4.Rep4: alpha=0.4444, lambda=0.2
## + Fold4.Rep4: alpha=0.5556, lambda=0.2
## - Fold4.Rep4: alpha=0.5556, lambda=0.2
## + Fold4.Rep4: alpha=0.6667, lambda=0.2
## - Fold4.Rep4: alpha=0.6667, lambda=0.2
## + Fold4.Rep4: alpha=0.7778, lambda=0.2
## - Fold4.Rep4: alpha=0.7778, lambda=0.2
## + Fold4.Rep4: alpha=0.8889, lambda=0.2
## - Fold4.Rep4: alpha=0.8889, lambda=0.2
## + Fold4.Rep4: alpha=1.0000, lambda=0.2
## - Fold4.Rep4: alpha=1.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.0000, lambda=0.2
## - Fold5.Rep4: alpha=0.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.1111, lambda=0.2
## - Fold5.Rep4: alpha=0.1111, lambda=0.2
## + Fold5.Rep4: alpha=0.2222, lambda=0.2
## - Fold5.Rep4: alpha=0.2222, lambda=0.2
## + Fold5.Rep4: alpha=0.3333, lambda=0.2
## - Fold5.Rep4: alpha=0.3333, lambda=0.2
## + Fold5.Rep4: alpha=0.4444, lambda=0.2
## - Fold5.Rep4: alpha=0.4444, lambda=0.2
## + Fold5.Rep4: alpha=0.5556, lambda=0.2
## - Fold5.Rep4: alpha=0.5556, lambda=0.2
## + Fold5.Rep4: alpha=0.6667, lambda=0.2
## - Fold5.Rep4: alpha=0.6667, lambda=0.2
## + Fold5.Rep4: alpha=0.7778, lambda=0.2
## - Fold5.Rep4: alpha=0.7778, lambda=0.2
## + Fold5.Rep4: alpha=0.8889, lambda=0.2
## - Fold5.Rep4: alpha=0.8889, lambda=0.2
## + Fold5.Rep4: alpha=1.0000, lambda=0.2
## - Fold5.Rep4: alpha=1.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.0000, lambda=0.2
## - Fold1.Rep5: alpha=0.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.1111, lambda=0.2
## - Fold1.Rep5: alpha=0.1111, lambda=0.2
## + Fold1.Rep5: alpha=0.2222, lambda=0.2
## - Fold1.Rep5: alpha=0.2222, lambda=0.2
## + Fold1.Rep5: alpha=0.3333, lambda=0.2
## - Fold1.Rep5: alpha=0.3333, lambda=0.2
## + Fold1.Rep5: alpha=0.4444, lambda=0.2
## - Fold1.Rep5: alpha=0.4444, lambda=0.2
## + Fold1.Rep5: alpha=0.5556, lambda=0.2
## - Fold1.Rep5: alpha=0.5556, lambda=0.2
## + Fold1.Rep5: alpha=0.6667, lambda=0.2
## - Fold1.Rep5: alpha=0.6667, lambda=0.2
## + Fold1.Rep5: alpha=0.7778, lambda=0.2
## - Fold1.Rep5: alpha=0.7778, lambda=0.2
## + Fold1.Rep5: alpha=0.8889, lambda=0.2
## - Fold1.Rep5: alpha=0.8889, lambda=0.2
## + Fold1.Rep5: alpha=1.0000, lambda=0.2
## - Fold1.Rep5: alpha=1.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.0000, lambda=0.2
## - Fold2.Rep5: alpha=0.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.1111, lambda=0.2
## - Fold2.Rep5: alpha=0.1111, lambda=0.2
## + Fold2.Rep5: alpha=0.2222, lambda=0.2
## - Fold2.Rep5: alpha=0.2222, lambda=0.2
## + Fold2.Rep5: alpha=0.3333, lambda=0.2
## - Fold2.Rep5: alpha=0.3333, lambda=0.2
## + Fold2.Rep5: alpha=0.4444, lambda=0.2
## - Fold2.Rep5: alpha=0.4444, lambda=0.2
## + Fold2.Rep5: alpha=0.5556, lambda=0.2
## - Fold2.Rep5: alpha=0.5556, lambda=0.2
## + Fold2.Rep5: alpha=0.6667, lambda=0.2
## - Fold2.Rep5: alpha=0.6667, lambda=0.2
## + Fold2.Rep5: alpha=0.7778, lambda=0.2
## - Fold2.Rep5: alpha=0.7778, lambda=0.2
## + Fold2.Rep5: alpha=0.8889, lambda=0.2
## - Fold2.Rep5: alpha=0.8889, lambda=0.2
## + Fold2.Rep5: alpha=1.0000, lambda=0.2
## - Fold2.Rep5: alpha=1.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.0000, lambda=0.2
## - Fold3.Rep5: alpha=0.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.1111, lambda=0.2
## - Fold3.Rep5: alpha=0.1111, lambda=0.2
## + Fold3.Rep5: alpha=0.2222, lambda=0.2
## - Fold3.Rep5: alpha=0.2222, lambda=0.2
## + Fold3.Rep5: alpha=0.3333, lambda=0.2
## - Fold3.Rep5: alpha=0.3333, lambda=0.2
## + Fold3.Rep5: alpha=0.4444, lambda=0.2
## - Fold3.Rep5: alpha=0.4444, lambda=0.2
## + Fold3.Rep5: alpha=0.5556, lambda=0.2
## - Fold3.Rep5: alpha=0.5556, lambda=0.2
## + Fold3.Rep5: alpha=0.6667, lambda=0.2
## - Fold3.Rep5: alpha=0.6667, lambda=0.2
## + Fold3.Rep5: alpha=0.7778, lambda=0.2
## - Fold3.Rep5: alpha=0.7778, lambda=0.2
## + Fold3.Rep5: alpha=0.8889, lambda=0.2
## - Fold3.Rep5: alpha=0.8889, lambda=0.2
## + Fold3.Rep5: alpha=1.0000, lambda=0.2
## - Fold3.Rep5: alpha=1.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.0000, lambda=0.2
## - Fold4.Rep5: alpha=0.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.1111, lambda=0.2
## - Fold4.Rep5: alpha=0.1111, lambda=0.2
## + Fold4.Rep5: alpha=0.2222, lambda=0.2
## - Fold4.Rep5: alpha=0.2222, lambda=0.2
## + Fold4.Rep5: alpha=0.3333, lambda=0.2
## - Fold4.Rep5: alpha=0.3333, lambda=0.2
## + Fold4.Rep5: alpha=0.4444, lambda=0.2
## - Fold4.Rep5: alpha=0.4444, lambda=0.2
## + Fold4.Rep5: alpha=0.5556, lambda=0.2
## - Fold4.Rep5: alpha=0.5556, lambda=0.2
## + Fold4.Rep5: alpha=0.6667, lambda=0.2
## - Fold4.Rep5: alpha=0.6667, lambda=0.2
## + Fold4.Rep5: alpha=0.7778, lambda=0.2
## - Fold4.Rep5: alpha=0.7778, lambda=0.2
## + Fold4.Rep5: alpha=0.8889, lambda=0.2
## - Fold4.Rep5: alpha=0.8889, lambda=0.2
## + Fold4.Rep5: alpha=1.0000, lambda=0.2
## - Fold4.Rep5: alpha=1.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.0000, lambda=0.2
## - Fold5.Rep5: alpha=0.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.1111, lambda=0.2
## - Fold5.Rep5: alpha=0.1111, lambda=0.2
## + Fold5.Rep5: alpha=0.2222, lambda=0.2
## - Fold5.Rep5: alpha=0.2222, lambda=0.2
## + Fold5.Rep5: alpha=0.3333, lambda=0.2
## - Fold5.Rep5: alpha=0.3333, lambda=0.2
## + Fold5.Rep5: alpha=0.4444, lambda=0.2
## - Fold5.Rep5: alpha=0.4444, lambda=0.2
## + Fold5.Rep5: alpha=0.5556, lambda=0.2
## - Fold5.Rep5: alpha=0.5556, lambda=0.2
## + Fold5.Rep5: alpha=0.6667, lambda=0.2
## - Fold5.Rep5: alpha=0.6667, lambda=0.2
## + Fold5.Rep5: alpha=0.7778, lambda=0.2
## - Fold5.Rep5: alpha=0.7778, lambda=0.2
## + Fold5.Rep5: alpha=0.8889, lambda=0.2
## - Fold5.Rep5: alpha=0.8889, lambda=0.2
## + Fold5.Rep5: alpha=1.0000, lambda=0.2
## - Fold5.Rep5: alpha=1.0000, lambda=0.2
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 0.778, lambda = 0.1 on full training set
print(en_b)
## glmnet
##
## 9 samples
## 15 predictors
##
## No pre-processing
## Resampling: Cross-Validated (5 fold, repeated 5 times)
## Summary of sample sizes: 7, 8, 7, 7, 7, 7, ...
## Resampling results across tuning parameters:
##
## alpha lambda RMSE Rsquared MAE
## 0.0000000 0.000100 0.3240204 1 0.2853706
## 0.0000000 0.050075 0.3240204 1 0.2853706
## 0.0000000 0.100050 0.3240204 1 0.2853706
## 0.0000000 0.150025 0.3240204 1 0.2853706
## 0.0000000 0.200000 0.3240204 1 0.2853706
## 0.1111111 0.000100 0.2819756 1 0.2672011
## 0.1111111 0.050075 0.2811649 1 0.2640328
## 0.1111111 0.100050 0.2805939 1 0.2592482
## 0.1111111 0.150025 0.2780674 1 0.2558157
## 0.1111111 0.200000 0.2752600 1 0.2522081
## 0.2222222 0.000100 0.2809041 1 0.2657937
## 0.2222222 0.050075 0.2771808 1 0.2572174
## 0.2222222 0.100050 0.2664525 1 0.2448157
## 0.2222222 0.150025 0.2586974 1 0.2378401
## 0.2222222 0.200000 0.2544444 1 0.2329466
## 0.3333333 0.000100 0.2782128 1 0.2624318
## 0.3333333 0.050075 0.2696229 1 0.2468498
## 0.3333333 0.100050 0.2542299 1 0.2328748
## 0.3333333 0.150025 0.2479350 1 0.2248910
## 0.3333333 0.200000 0.2485499 1 0.2240723
## 0.4444444 0.000100 0.2774664 1 0.2573549
## 0.4444444 0.050075 0.2624569 1 0.2372101
## 0.4444444 0.100050 0.2479621 1 0.2260924
## 0.4444444 0.150025 0.2402050 1 0.2168968
## 0.4444444 0.200000 0.2486434 1 0.2198841
## 0.5555556 0.000100 0.2816538 1 0.2617566
## 0.5555556 0.050075 0.2659894 1 0.2406232
## 0.5555556 0.100050 0.2415642 1 0.2192756
## 0.5555556 0.150025 0.2389819 1 0.2133765
## 0.5555556 0.200000 0.2578007 1 0.2264976
## 0.6666667 0.000100 0.2890458 1 0.2703562
## 0.6666667 0.050075 0.2747661 1 0.2468887
## 0.6666667 0.100050 0.2361213 1 0.2125531
## 0.6666667 0.150025 0.2489030 1 0.2208648
## 0.6666667 0.200000 0.2735498 1 0.2400684
## 0.7777778 0.000100 0.2966698 1 0.2758776
## 0.7777778 0.050075 0.2681299 1 0.2416270
## 0.7777778 0.100050 0.2338884 1 0.2096062
## 0.7777778 0.150025 0.2634257 1 0.2329960
## 0.7777778 0.200000 0.2903918 1 0.2557196
## 0.8888889 0.000100 0.2978482 1 0.2764814
## 0.8888889 0.050075 0.2641560 1 0.2385991
## 0.8888889 0.100050 0.2449850 1 0.2194059
## 0.8888889 0.150025 0.2795829 1 0.2466691
## 0.8888889 0.200000 0.2986662 1 0.2635765
## 1.0000000 0.000100 0.2977368 1 0.2746219
## 1.0000000 0.050075 0.2634028 1 0.2380975
## 1.0000000 0.100050 0.2584236 1 0.2307093
## 1.0000000 0.150025 0.2883897 1 0.2549802
## 1.0000000 0.200000 0.3034788 1 0.2681929
##
## RMSE was used to select the optimal model using the smallest value.
## The final values used for the model were alpha = 0.7777778 and lambda = 0.10005.
plot(varImp(en_b, scale = TRUE))
# Coefficients
coefs <- coef(en_b$finalModel, en_b$bestTune$lambda)
coefs.zero <- coefs[which(coefs == 0),]
coef_order <- order(abs(coefs), decreasing = TRUE)
head(as.matrix(coefs[coef_order,]), 10)
## [,1]
## (Intercept) 1.3215079
## Avg_Cobble -0.5761352
## Avg_Gravel 0.2752391
## Avg_Sand 0.2086779
## Avg_Flow 0.0000000
## Flow_Variance 0.0000000
## Avg_Boulder 0.0000000
## Avg_Bedrock 0.0000000
## Avg_Organic 0.0000000
## Mean_Slope_Degrees 0.0000000
### --- DRIFT BIOMASS MODEL --- ###
drinet <- merged[,-c(1,2,4,5)]
train <- drinet[split_ind == 1,]
test <- drinet[split_ind == 2,]
set.seed(1234)
en_d <- train(Drift ~ .,
train,
method = 'glmnet',
tuneGrid = expand.grid(alpha = seq(0, 1, length = 10),
lambda = seq(0.0001, 0.2, length = 5)),
trControl = custom)
## + Fold1.Rep1: alpha=0.0000, lambda=0.2
## - Fold1.Rep1: alpha=0.0000, lambda=0.2
## + Fold1.Rep1: alpha=0.1111, lambda=0.2
## - Fold1.Rep1: alpha=0.1111, lambda=0.2
## + Fold1.Rep1: alpha=0.2222, lambda=0.2
## - Fold1.Rep1: alpha=0.2222, lambda=0.2
## + Fold1.Rep1: alpha=0.3333, lambda=0.2
## - Fold1.Rep1: alpha=0.3333, lambda=0.2
## + Fold1.Rep1: alpha=0.4444, lambda=0.2
## - Fold1.Rep1: alpha=0.4444, lambda=0.2
## + Fold1.Rep1: alpha=0.5556, lambda=0.2
## - Fold1.Rep1: alpha=0.5556, lambda=0.2
## + Fold1.Rep1: alpha=0.6667, lambda=0.2
## - Fold1.Rep1: alpha=0.6667, lambda=0.2
## + Fold1.Rep1: alpha=0.7778, lambda=0.2
## - Fold1.Rep1: alpha=0.7778, lambda=0.2
## + Fold1.Rep1: alpha=0.8889, lambda=0.2
## - Fold1.Rep1: alpha=0.8889, lambda=0.2
## + Fold1.Rep1: alpha=1.0000, lambda=0.2
## - Fold1.Rep1: alpha=1.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.0000, lambda=0.2
## - Fold2.Rep1: alpha=0.0000, lambda=0.2
## + Fold2.Rep1: alpha=0.1111, lambda=0.2
## - Fold2.Rep1: alpha=0.1111, lambda=0.2
## + Fold2.Rep1: alpha=0.2222, lambda=0.2
## - Fold2.Rep1: alpha=0.2222, lambda=0.2
## + Fold2.Rep1: alpha=0.3333, lambda=0.2
## - Fold2.Rep1: alpha=0.3333, lambda=0.2
## + Fold2.Rep1: alpha=0.4444, lambda=0.2
## - Fold2.Rep1: alpha=0.4444, lambda=0.2
## + Fold2.Rep1: alpha=0.5556, lambda=0.2
## - Fold2.Rep1: alpha=0.5556, lambda=0.2
## + Fold2.Rep1: alpha=0.6667, lambda=0.2
## - Fold2.Rep1: alpha=0.6667, lambda=0.2
## + Fold2.Rep1: alpha=0.7778, lambda=0.2
## - Fold2.Rep1: alpha=0.7778, lambda=0.2
## + Fold2.Rep1: alpha=0.8889, lambda=0.2
## - Fold2.Rep1: alpha=0.8889, lambda=0.2
## + Fold2.Rep1: alpha=1.0000, lambda=0.2
## - Fold2.Rep1: alpha=1.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.0000, lambda=0.2
## - Fold3.Rep1: alpha=0.0000, lambda=0.2
## + Fold3.Rep1: alpha=0.1111, lambda=0.2
## - Fold3.Rep1: alpha=0.1111, lambda=0.2
## + Fold3.Rep1: alpha=0.2222, lambda=0.2
## - Fold3.Rep1: alpha=0.2222, lambda=0.2
## + Fold3.Rep1: alpha=0.3333, lambda=0.2
## - Fold3.Rep1: alpha=0.3333, lambda=0.2
## + Fold3.Rep1: alpha=0.4444, lambda=0.2
## - Fold3.Rep1: alpha=0.4444, lambda=0.2
## + Fold3.Rep1: alpha=0.5556, lambda=0.2
## - Fold3.Rep1: alpha=0.5556, lambda=0.2
## + Fold3.Rep1: alpha=0.6667, lambda=0.2
## - Fold3.Rep1: alpha=0.6667, lambda=0.2
## + Fold3.Rep1: alpha=0.7778, lambda=0.2
## - Fold3.Rep1: alpha=0.7778, lambda=0.2
## + Fold3.Rep1: alpha=0.8889, lambda=0.2
## - Fold3.Rep1: alpha=0.8889, lambda=0.2
## + Fold3.Rep1: alpha=1.0000, lambda=0.2
## - Fold3.Rep1: alpha=1.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.0000, lambda=0.2
## - Fold4.Rep1: alpha=0.0000, lambda=0.2
## + Fold4.Rep1: alpha=0.1111, lambda=0.2
## - Fold4.Rep1: alpha=0.1111, lambda=0.2
## + Fold4.Rep1: alpha=0.2222, lambda=0.2
## - Fold4.Rep1: alpha=0.2222, lambda=0.2
## + Fold4.Rep1: alpha=0.3333, lambda=0.2
## - Fold4.Rep1: alpha=0.3333, lambda=0.2
## + Fold4.Rep1: alpha=0.4444, lambda=0.2
## - Fold4.Rep1: alpha=0.4444, lambda=0.2
## + Fold4.Rep1: alpha=0.5556, lambda=0.2
## - Fold4.Rep1: alpha=0.5556, lambda=0.2
## + Fold4.Rep1: alpha=0.6667, lambda=0.2
## - Fold4.Rep1: alpha=0.6667, lambda=0.2
## + Fold4.Rep1: alpha=0.7778, lambda=0.2
## - Fold4.Rep1: alpha=0.7778, lambda=0.2
## + Fold4.Rep1: alpha=0.8889, lambda=0.2
## - Fold4.Rep1: alpha=0.8889, lambda=0.2
## + Fold4.Rep1: alpha=1.0000, lambda=0.2
## - Fold4.Rep1: alpha=1.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.0000, lambda=0.2
## - Fold5.Rep1: alpha=0.0000, lambda=0.2
## + Fold5.Rep1: alpha=0.1111, lambda=0.2
## - Fold5.Rep1: alpha=0.1111, lambda=0.2
## + Fold5.Rep1: alpha=0.2222, lambda=0.2
## - Fold5.Rep1: alpha=0.2222, lambda=0.2
## + Fold5.Rep1: alpha=0.3333, lambda=0.2
## - Fold5.Rep1: alpha=0.3333, lambda=0.2
## + Fold5.Rep1: alpha=0.4444, lambda=0.2
## - Fold5.Rep1: alpha=0.4444, lambda=0.2
## + Fold5.Rep1: alpha=0.5556, lambda=0.2
## - Fold5.Rep1: alpha=0.5556, lambda=0.2
## + Fold5.Rep1: alpha=0.6667, lambda=0.2
## - Fold5.Rep1: alpha=0.6667, lambda=0.2
## + Fold5.Rep1: alpha=0.7778, lambda=0.2
## - Fold5.Rep1: alpha=0.7778, lambda=0.2
## + Fold5.Rep1: alpha=0.8889, lambda=0.2
## - Fold5.Rep1: alpha=0.8889, lambda=0.2
## + Fold5.Rep1: alpha=1.0000, lambda=0.2
## - Fold5.Rep1: alpha=1.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.0000, lambda=0.2
## - Fold1.Rep2: alpha=0.0000, lambda=0.2
## + Fold1.Rep2: alpha=0.1111, lambda=0.2
## - Fold1.Rep2: alpha=0.1111, lambda=0.2
## + Fold1.Rep2: alpha=0.2222, lambda=0.2
## - Fold1.Rep2: alpha=0.2222, lambda=0.2
## + Fold1.Rep2: alpha=0.3333, lambda=0.2
## - Fold1.Rep2: alpha=0.3333, lambda=0.2
## + Fold1.Rep2: alpha=0.4444, lambda=0.2
## - Fold1.Rep2: alpha=0.4444, lambda=0.2
## + Fold1.Rep2: alpha=0.5556, lambda=0.2
## - Fold1.Rep2: alpha=0.5556, lambda=0.2
## + Fold1.Rep2: alpha=0.6667, lambda=0.2
## - Fold1.Rep2: alpha=0.6667, lambda=0.2
## + Fold1.Rep2: alpha=0.7778, lambda=0.2
## - Fold1.Rep2: alpha=0.7778, lambda=0.2
## + Fold1.Rep2: alpha=0.8889, lambda=0.2
## - Fold1.Rep2: alpha=0.8889, lambda=0.2
## + Fold1.Rep2: alpha=1.0000, lambda=0.2
## - Fold1.Rep2: alpha=1.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.0000, lambda=0.2
## - Fold2.Rep2: alpha=0.0000, lambda=0.2
## + Fold2.Rep2: alpha=0.1111, lambda=0.2
## - Fold2.Rep2: alpha=0.1111, lambda=0.2
## + Fold2.Rep2: alpha=0.2222, lambda=0.2
## - Fold2.Rep2: alpha=0.2222, lambda=0.2
## + Fold2.Rep2: alpha=0.3333, lambda=0.2
## - Fold2.Rep2: alpha=0.3333, lambda=0.2
## + Fold2.Rep2: alpha=0.4444, lambda=0.2
## - Fold2.Rep2: alpha=0.4444, lambda=0.2
## + Fold2.Rep2: alpha=0.5556, lambda=0.2
## - Fold2.Rep2: alpha=0.5556, lambda=0.2
## + Fold2.Rep2: alpha=0.6667, lambda=0.2
## - Fold2.Rep2: alpha=0.6667, lambda=0.2
## + Fold2.Rep2: alpha=0.7778, lambda=0.2
## - Fold2.Rep2: alpha=0.7778, lambda=0.2
## + Fold2.Rep2: alpha=0.8889, lambda=0.2
## - Fold2.Rep2: alpha=0.8889, lambda=0.2
## + Fold2.Rep2: alpha=1.0000, lambda=0.2
## - Fold2.Rep2: alpha=1.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.0000, lambda=0.2
## - Fold3.Rep2: alpha=0.0000, lambda=0.2
## + Fold3.Rep2: alpha=0.1111, lambda=0.2
## - Fold3.Rep2: alpha=0.1111, lambda=0.2
## + Fold3.Rep2: alpha=0.2222, lambda=0.2
## - Fold3.Rep2: alpha=0.2222, lambda=0.2
## + Fold3.Rep2: alpha=0.3333, lambda=0.2
## - Fold3.Rep2: alpha=0.3333, lambda=0.2
## + Fold3.Rep2: alpha=0.4444, lambda=0.2
## - Fold3.Rep2: alpha=0.4444, lambda=0.2
## + Fold3.Rep2: alpha=0.5556, lambda=0.2
## - Fold3.Rep2: alpha=0.5556, lambda=0.2
## + Fold3.Rep2: alpha=0.6667, lambda=0.2
## - Fold3.Rep2: alpha=0.6667, lambda=0.2
## + Fold3.Rep2: alpha=0.7778, lambda=0.2
## - Fold3.Rep2: alpha=0.7778, lambda=0.2
## + Fold3.Rep2: alpha=0.8889, lambda=0.2
## - Fold3.Rep2: alpha=0.8889, lambda=0.2
## + Fold3.Rep2: alpha=1.0000, lambda=0.2
## - Fold3.Rep2: alpha=1.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.0000, lambda=0.2
## - Fold4.Rep2: alpha=0.0000, lambda=0.2
## + Fold4.Rep2: alpha=0.1111, lambda=0.2
## - Fold4.Rep2: alpha=0.1111, lambda=0.2
## + Fold4.Rep2: alpha=0.2222, lambda=0.2
## - Fold4.Rep2: alpha=0.2222, lambda=0.2
## + Fold4.Rep2: alpha=0.3333, lambda=0.2
## - Fold4.Rep2: alpha=0.3333, lambda=0.2
## + Fold4.Rep2: alpha=0.4444, lambda=0.2
## - Fold4.Rep2: alpha=0.4444, lambda=0.2
## + Fold4.Rep2: alpha=0.5556, lambda=0.2
## - Fold4.Rep2: alpha=0.5556, lambda=0.2
## + Fold4.Rep2: alpha=0.6667, lambda=0.2
## - Fold4.Rep2: alpha=0.6667, lambda=0.2
## + Fold4.Rep2: alpha=0.7778, lambda=0.2
## - Fold4.Rep2: alpha=0.7778, lambda=0.2
## + Fold4.Rep2: alpha=0.8889, lambda=0.2
## - Fold4.Rep2: alpha=0.8889, lambda=0.2
## + Fold4.Rep2: alpha=1.0000, lambda=0.2
## - Fold4.Rep2: alpha=1.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.0000, lambda=0.2
## - Fold5.Rep2: alpha=0.0000, lambda=0.2
## + Fold5.Rep2: alpha=0.1111, lambda=0.2
## - Fold5.Rep2: alpha=0.1111, lambda=0.2
## + Fold5.Rep2: alpha=0.2222, lambda=0.2
## - Fold5.Rep2: alpha=0.2222, lambda=0.2
## + Fold5.Rep2: alpha=0.3333, lambda=0.2
## - Fold5.Rep2: alpha=0.3333, lambda=0.2
## + Fold5.Rep2: alpha=0.4444, lambda=0.2
## - Fold5.Rep2: alpha=0.4444, lambda=0.2
## + Fold5.Rep2: alpha=0.5556, lambda=0.2
## - Fold5.Rep2: alpha=0.5556, lambda=0.2
## + Fold5.Rep2: alpha=0.6667, lambda=0.2
## - Fold5.Rep2: alpha=0.6667, lambda=0.2
## + Fold5.Rep2: alpha=0.7778, lambda=0.2
## - Fold5.Rep2: alpha=0.7778, lambda=0.2
## + Fold5.Rep2: alpha=0.8889, lambda=0.2
## - Fold5.Rep2: alpha=0.8889, lambda=0.2
## + Fold5.Rep2: alpha=1.0000, lambda=0.2
## - Fold5.Rep2: alpha=1.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.0000, lambda=0.2
## - Fold1.Rep3: alpha=0.0000, lambda=0.2
## + Fold1.Rep3: alpha=0.1111, lambda=0.2
## - Fold1.Rep3: alpha=0.1111, lambda=0.2
## + Fold1.Rep3: alpha=0.2222, lambda=0.2
## - Fold1.Rep3: alpha=0.2222, lambda=0.2
## + Fold1.Rep3: alpha=0.3333, lambda=0.2
## - Fold1.Rep3: alpha=0.3333, lambda=0.2
## + Fold1.Rep3: alpha=0.4444, lambda=0.2
## - Fold1.Rep3: alpha=0.4444, lambda=0.2
## + Fold1.Rep3: alpha=0.5556, lambda=0.2
## - Fold1.Rep3: alpha=0.5556, lambda=0.2
## + Fold1.Rep3: alpha=0.6667, lambda=0.2
## - Fold1.Rep3: alpha=0.6667, lambda=0.2
## + Fold1.Rep3: alpha=0.7778, lambda=0.2
## - Fold1.Rep3: alpha=0.7778, lambda=0.2
## + Fold1.Rep3: alpha=0.8889, lambda=0.2
## - Fold1.Rep3: alpha=0.8889, lambda=0.2
## + Fold1.Rep3: alpha=1.0000, lambda=0.2
## - Fold1.Rep3: alpha=1.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.0000, lambda=0.2
## - Fold2.Rep3: alpha=0.0000, lambda=0.2
## + Fold2.Rep3: alpha=0.1111, lambda=0.2
## - Fold2.Rep3: alpha=0.1111, lambda=0.2
## + Fold2.Rep3: alpha=0.2222, lambda=0.2
## - Fold2.Rep3: alpha=0.2222, lambda=0.2
## + Fold2.Rep3: alpha=0.3333, lambda=0.2
## - Fold2.Rep3: alpha=0.3333, lambda=0.2
## + Fold2.Rep3: alpha=0.4444, lambda=0.2
## - Fold2.Rep3: alpha=0.4444, lambda=0.2
## + Fold2.Rep3: alpha=0.5556, lambda=0.2
## - Fold2.Rep3: alpha=0.5556, lambda=0.2
## + Fold2.Rep3: alpha=0.6667, lambda=0.2
## - Fold2.Rep3: alpha=0.6667, lambda=0.2
## + Fold2.Rep3: alpha=0.7778, lambda=0.2
## - Fold2.Rep3: alpha=0.7778, lambda=0.2
## + Fold2.Rep3: alpha=0.8889, lambda=0.2
## - Fold2.Rep3: alpha=0.8889, lambda=0.2
## + Fold2.Rep3: alpha=1.0000, lambda=0.2
## - Fold2.Rep3: alpha=1.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.0000, lambda=0.2
## - Fold3.Rep3: alpha=0.0000, lambda=0.2
## + Fold3.Rep3: alpha=0.1111, lambda=0.2
## - Fold3.Rep3: alpha=0.1111, lambda=0.2
## + Fold3.Rep3: alpha=0.2222, lambda=0.2
## - Fold3.Rep3: alpha=0.2222, lambda=0.2
## + Fold3.Rep3: alpha=0.3333, lambda=0.2
## - Fold3.Rep3: alpha=0.3333, lambda=0.2
## + Fold3.Rep3: alpha=0.4444, lambda=0.2
## - Fold3.Rep3: alpha=0.4444, lambda=0.2
## + Fold3.Rep3: alpha=0.5556, lambda=0.2
## - Fold3.Rep3: alpha=0.5556, lambda=0.2
## + Fold3.Rep3: alpha=0.6667, lambda=0.2
## - Fold3.Rep3: alpha=0.6667, lambda=0.2
## + Fold3.Rep3: alpha=0.7778, lambda=0.2
## - Fold3.Rep3: alpha=0.7778, lambda=0.2
## + Fold3.Rep3: alpha=0.8889, lambda=0.2
## - Fold3.Rep3: alpha=0.8889, lambda=0.2
## + Fold3.Rep3: alpha=1.0000, lambda=0.2
## - Fold3.Rep3: alpha=1.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.0000, lambda=0.2
## - Fold4.Rep3: alpha=0.0000, lambda=0.2
## + Fold4.Rep3: alpha=0.1111, lambda=0.2
## - Fold4.Rep3: alpha=0.1111, lambda=0.2
## + Fold4.Rep3: alpha=0.2222, lambda=0.2
## - Fold4.Rep3: alpha=0.2222, lambda=0.2
## + Fold4.Rep3: alpha=0.3333, lambda=0.2
## - Fold4.Rep3: alpha=0.3333, lambda=0.2
## + Fold4.Rep3: alpha=0.4444, lambda=0.2
## - Fold4.Rep3: alpha=0.4444, lambda=0.2
## + Fold4.Rep3: alpha=0.5556, lambda=0.2
## - Fold4.Rep3: alpha=0.5556, lambda=0.2
## + Fold4.Rep3: alpha=0.6667, lambda=0.2
## - Fold4.Rep3: alpha=0.6667, lambda=0.2
## + Fold4.Rep3: alpha=0.7778, lambda=0.2
## - Fold4.Rep3: alpha=0.7778, lambda=0.2
## + Fold4.Rep3: alpha=0.8889, lambda=0.2
## - Fold4.Rep3: alpha=0.8889, lambda=0.2
## + Fold4.Rep3: alpha=1.0000, lambda=0.2
## - Fold4.Rep3: alpha=1.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.0000, lambda=0.2
## - Fold5.Rep3: alpha=0.0000, lambda=0.2
## + Fold5.Rep3: alpha=0.1111, lambda=0.2
## - Fold5.Rep3: alpha=0.1111, lambda=0.2
## + Fold5.Rep3: alpha=0.2222, lambda=0.2
## - Fold5.Rep3: alpha=0.2222, lambda=0.2
## + Fold5.Rep3: alpha=0.3333, lambda=0.2
## - Fold5.Rep3: alpha=0.3333, lambda=0.2
## + Fold5.Rep3: alpha=0.4444, lambda=0.2
## - Fold5.Rep3: alpha=0.4444, lambda=0.2
## + Fold5.Rep3: alpha=0.5556, lambda=0.2
## - Fold5.Rep3: alpha=0.5556, lambda=0.2
## + Fold5.Rep3: alpha=0.6667, lambda=0.2
## - Fold5.Rep3: alpha=0.6667, lambda=0.2
## + Fold5.Rep3: alpha=0.7778, lambda=0.2
## - Fold5.Rep3: alpha=0.7778, lambda=0.2
## + Fold5.Rep3: alpha=0.8889, lambda=0.2
## - Fold5.Rep3: alpha=0.8889, lambda=0.2
## + Fold5.Rep3: alpha=1.0000, lambda=0.2
## - Fold5.Rep3: alpha=1.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.0000, lambda=0.2
## - Fold1.Rep4: alpha=0.0000, lambda=0.2
## + Fold1.Rep4: alpha=0.1111, lambda=0.2
## - Fold1.Rep4: alpha=0.1111, lambda=0.2
## + Fold1.Rep4: alpha=0.2222, lambda=0.2
## - Fold1.Rep4: alpha=0.2222, lambda=0.2
## + Fold1.Rep4: alpha=0.3333, lambda=0.2
## - Fold1.Rep4: alpha=0.3333, lambda=0.2
## + Fold1.Rep4: alpha=0.4444, lambda=0.2
## - Fold1.Rep4: alpha=0.4444, lambda=0.2
## + Fold1.Rep4: alpha=0.5556, lambda=0.2
## - Fold1.Rep4: alpha=0.5556, lambda=0.2
## + Fold1.Rep4: alpha=0.6667, lambda=0.2
## - Fold1.Rep4: alpha=0.6667, lambda=0.2
## + Fold1.Rep4: alpha=0.7778, lambda=0.2
## - Fold1.Rep4: alpha=0.7778, lambda=0.2
## + Fold1.Rep4: alpha=0.8889, lambda=0.2
## - Fold1.Rep4: alpha=0.8889, lambda=0.2
## + Fold1.Rep4: alpha=1.0000, lambda=0.2
## - Fold1.Rep4: alpha=1.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.0000, lambda=0.2
## - Fold2.Rep4: alpha=0.0000, lambda=0.2
## + Fold2.Rep4: alpha=0.1111, lambda=0.2
## - Fold2.Rep4: alpha=0.1111, lambda=0.2
## + Fold2.Rep4: alpha=0.2222, lambda=0.2
## - Fold2.Rep4: alpha=0.2222, lambda=0.2
## + Fold2.Rep4: alpha=0.3333, lambda=0.2
## - Fold2.Rep4: alpha=0.3333, lambda=0.2
## + Fold2.Rep4: alpha=0.4444, lambda=0.2
## - Fold2.Rep4: alpha=0.4444, lambda=0.2
## + Fold2.Rep4: alpha=0.5556, lambda=0.2
## - Fold2.Rep4: alpha=0.5556, lambda=0.2
## + Fold2.Rep4: alpha=0.6667, lambda=0.2
## - Fold2.Rep4: alpha=0.6667, lambda=0.2
## + Fold2.Rep4: alpha=0.7778, lambda=0.2
## - Fold2.Rep4: alpha=0.7778, lambda=0.2
## + Fold2.Rep4: alpha=0.8889, lambda=0.2
## - Fold2.Rep4: alpha=0.8889, lambda=0.2
## + Fold2.Rep4: alpha=1.0000, lambda=0.2
## - Fold2.Rep4: alpha=1.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.0000, lambda=0.2
## - Fold3.Rep4: alpha=0.0000, lambda=0.2
## + Fold3.Rep4: alpha=0.1111, lambda=0.2
## - Fold3.Rep4: alpha=0.1111, lambda=0.2
## + Fold3.Rep4: alpha=0.2222, lambda=0.2
## - Fold3.Rep4: alpha=0.2222, lambda=0.2
## + Fold3.Rep4: alpha=0.3333, lambda=0.2
## - Fold3.Rep4: alpha=0.3333, lambda=0.2
## + Fold3.Rep4: alpha=0.4444, lambda=0.2
## - Fold3.Rep4: alpha=0.4444, lambda=0.2
## + Fold3.Rep4: alpha=0.5556, lambda=0.2
## - Fold3.Rep4: alpha=0.5556, lambda=0.2
## + Fold3.Rep4: alpha=0.6667, lambda=0.2
## - Fold3.Rep4: alpha=0.6667, lambda=0.2
## + Fold3.Rep4: alpha=0.7778, lambda=0.2
## - Fold3.Rep4: alpha=0.7778, lambda=0.2
## + Fold3.Rep4: alpha=0.8889, lambda=0.2
## - Fold3.Rep4: alpha=0.8889, lambda=0.2
## + Fold3.Rep4: alpha=1.0000, lambda=0.2
## - Fold3.Rep4: alpha=1.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.0000, lambda=0.2
## - Fold4.Rep4: alpha=0.0000, lambda=0.2
## + Fold4.Rep4: alpha=0.1111, lambda=0.2
## - Fold4.Rep4: alpha=0.1111, lambda=0.2
## + Fold4.Rep4: alpha=0.2222, lambda=0.2
## - Fold4.Rep4: alpha=0.2222, lambda=0.2
## + Fold4.Rep4: alpha=0.3333, lambda=0.2
## - Fold4.Rep4: alpha=0.3333, lambda=0.2
## + Fold4.Rep4: alpha=0.4444, lambda=0.2
## - Fold4.Rep4: alpha=0.4444, lambda=0.2
## + Fold4.Rep4: alpha=0.5556, lambda=0.2
## - Fold4.Rep4: alpha=0.5556, lambda=0.2
## + Fold4.Rep4: alpha=0.6667, lambda=0.2
## - Fold4.Rep4: alpha=0.6667, lambda=0.2
## + Fold4.Rep4: alpha=0.7778, lambda=0.2
## - Fold4.Rep4: alpha=0.7778, lambda=0.2
## + Fold4.Rep4: alpha=0.8889, lambda=0.2
## - Fold4.Rep4: alpha=0.8889, lambda=0.2
## + Fold4.Rep4: alpha=1.0000, lambda=0.2
## - Fold4.Rep4: alpha=1.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.0000, lambda=0.2
## - Fold5.Rep4: alpha=0.0000, lambda=0.2
## + Fold5.Rep4: alpha=0.1111, lambda=0.2
## - Fold5.Rep4: alpha=0.1111, lambda=0.2
## + Fold5.Rep4: alpha=0.2222, lambda=0.2
## - Fold5.Rep4: alpha=0.2222, lambda=0.2
## + Fold5.Rep4: alpha=0.3333, lambda=0.2
## - Fold5.Rep4: alpha=0.3333, lambda=0.2
## + Fold5.Rep4: alpha=0.4444, lambda=0.2
## - Fold5.Rep4: alpha=0.4444, lambda=0.2
## + Fold5.Rep4: alpha=0.5556, lambda=0.2
## - Fold5.Rep4: alpha=0.5556, lambda=0.2
## + Fold5.Rep4: alpha=0.6667, lambda=0.2
## - Fold5.Rep4: alpha=0.6667, lambda=0.2
## + Fold5.Rep4: alpha=0.7778, lambda=0.2
## - Fold5.Rep4: alpha=0.7778, lambda=0.2
## + Fold5.Rep4: alpha=0.8889, lambda=0.2
## - Fold5.Rep4: alpha=0.8889, lambda=0.2
## + Fold5.Rep4: alpha=1.0000, lambda=0.2
## - Fold5.Rep4: alpha=1.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.0000, lambda=0.2
## - Fold1.Rep5: alpha=0.0000, lambda=0.2
## + Fold1.Rep5: alpha=0.1111, lambda=0.2
## - Fold1.Rep5: alpha=0.1111, lambda=0.2
## + Fold1.Rep5: alpha=0.2222, lambda=0.2
## - Fold1.Rep5: alpha=0.2222, lambda=0.2
## + Fold1.Rep5: alpha=0.3333, lambda=0.2
## - Fold1.Rep5: alpha=0.3333, lambda=0.2
## + Fold1.Rep5: alpha=0.4444, lambda=0.2
## - Fold1.Rep5: alpha=0.4444, lambda=0.2
## + Fold1.Rep5: alpha=0.5556, lambda=0.2
## - Fold1.Rep5: alpha=0.5556, lambda=0.2
## + Fold1.Rep5: alpha=0.6667, lambda=0.2
## - Fold1.Rep5: alpha=0.6667, lambda=0.2
## + Fold1.Rep5: alpha=0.7778, lambda=0.2
## - Fold1.Rep5: alpha=0.7778, lambda=0.2
## + Fold1.Rep5: alpha=0.8889, lambda=0.2
## - Fold1.Rep5: alpha=0.8889, lambda=0.2
## + Fold1.Rep5: alpha=1.0000, lambda=0.2
## - Fold1.Rep5: alpha=1.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.0000, lambda=0.2
## - Fold2.Rep5: alpha=0.0000, lambda=0.2
## + Fold2.Rep5: alpha=0.1111, lambda=0.2
## - Fold2.Rep5: alpha=0.1111, lambda=0.2
## + Fold2.Rep5: alpha=0.2222, lambda=0.2
## - Fold2.Rep5: alpha=0.2222, lambda=0.2
## + Fold2.Rep5: alpha=0.3333, lambda=0.2
## - Fold2.Rep5: alpha=0.3333, lambda=0.2
## + Fold2.Rep5: alpha=0.4444, lambda=0.2
## - Fold2.Rep5: alpha=0.4444, lambda=0.2
## + Fold2.Rep5: alpha=0.5556, lambda=0.2
## - Fold2.Rep5: alpha=0.5556, lambda=0.2
## + Fold2.Rep5: alpha=0.6667, lambda=0.2
## - Fold2.Rep5: alpha=0.6667, lambda=0.2
## + Fold2.Rep5: alpha=0.7778, lambda=0.2
## - Fold2.Rep5: alpha=0.7778, lambda=0.2
## + Fold2.Rep5: alpha=0.8889, lambda=0.2
## - Fold2.Rep5: alpha=0.8889, lambda=0.2
## + Fold2.Rep5: alpha=1.0000, lambda=0.2
## - Fold2.Rep5: alpha=1.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.0000, lambda=0.2
## - Fold3.Rep5: alpha=0.0000, lambda=0.2
## + Fold3.Rep5: alpha=0.1111, lambda=0.2
## - Fold3.Rep5: alpha=0.1111, lambda=0.2
## + Fold3.Rep5: alpha=0.2222, lambda=0.2
## - Fold3.Rep5: alpha=0.2222, lambda=0.2
## + Fold3.Rep5: alpha=0.3333, lambda=0.2
## - Fold3.Rep5: alpha=0.3333, lambda=0.2
## + Fold3.Rep5: alpha=0.4444, lambda=0.2
## - Fold3.Rep5: alpha=0.4444, lambda=0.2
## + Fold3.Rep5: alpha=0.5556, lambda=0.2
## - Fold3.Rep5: alpha=0.5556, lambda=0.2
## + Fold3.Rep5: alpha=0.6667, lambda=0.2
## - Fold3.Rep5: alpha=0.6667, lambda=0.2
## + Fold3.Rep5: alpha=0.7778, lambda=0.2
## - Fold3.Rep5: alpha=0.7778, lambda=0.2
## + Fold3.Rep5: alpha=0.8889, lambda=0.2
## - Fold3.Rep5: alpha=0.8889, lambda=0.2
## + Fold3.Rep5: alpha=1.0000, lambda=0.2
## - Fold3.Rep5: alpha=1.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.0000, lambda=0.2
## - Fold4.Rep5: alpha=0.0000, lambda=0.2
## + Fold4.Rep5: alpha=0.1111, lambda=0.2
## - Fold4.Rep5: alpha=0.1111, lambda=0.2
## + Fold4.Rep5: alpha=0.2222, lambda=0.2
## - Fold4.Rep5: alpha=0.2222, lambda=0.2
## + Fold4.Rep5: alpha=0.3333, lambda=0.2
## - Fold4.Rep5: alpha=0.3333, lambda=0.2
## + Fold4.Rep5: alpha=0.4444, lambda=0.2
## - Fold4.Rep5: alpha=0.4444, lambda=0.2
## + Fold4.Rep5: alpha=0.5556, lambda=0.2
## - Fold4.Rep5: alpha=0.5556, lambda=0.2
## + Fold4.Rep5: alpha=0.6667, lambda=0.2
## - Fold4.Rep5: alpha=0.6667, lambda=0.2
## + Fold4.Rep5: alpha=0.7778, lambda=0.2
## - Fold4.Rep5: alpha=0.7778, lambda=0.2
## + Fold4.Rep5: alpha=0.8889, lambda=0.2
## - Fold4.Rep5: alpha=0.8889, lambda=0.2
## + Fold4.Rep5: alpha=1.0000, lambda=0.2
## - Fold4.Rep5: alpha=1.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.0000, lambda=0.2
## - Fold5.Rep5: alpha=0.0000, lambda=0.2
## + Fold5.Rep5: alpha=0.1111, lambda=0.2
## - Fold5.Rep5: alpha=0.1111, lambda=0.2
## + Fold5.Rep5: alpha=0.2222, lambda=0.2
## - Fold5.Rep5: alpha=0.2222, lambda=0.2
## + Fold5.Rep5: alpha=0.3333, lambda=0.2
## - Fold5.Rep5: alpha=0.3333, lambda=0.2
## + Fold5.Rep5: alpha=0.4444, lambda=0.2
## - Fold5.Rep5: alpha=0.4444, lambda=0.2
## + Fold5.Rep5: alpha=0.5556, lambda=0.2
## - Fold5.Rep5: alpha=0.5556, lambda=0.2
## + Fold5.Rep5: alpha=0.6667, lambda=0.2
## - Fold5.Rep5: alpha=0.6667, lambda=0.2
## + Fold5.Rep5: alpha=0.7778, lambda=0.2
## - Fold5.Rep5: alpha=0.7778, lambda=0.2
## + Fold5.Rep5: alpha=0.8889, lambda=0.2
## - Fold5.Rep5: alpha=0.8889, lambda=0.2
## + Fold5.Rep5: alpha=1.0000, lambda=0.2
## - Fold5.Rep5: alpha=1.0000, lambda=0.2
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 1, lambda = 0.1 on full training set
print(en_d)
## glmnet
##
## 9 samples
## 15 predictors
##
## No pre-processing
## Resampling: Cross-Validated (5 fold, repeated 5 times)
## Summary of sample sizes: 7, 8, 7, 7, 7, 7, ...
## Resampling results across tuning parameters:
##
## alpha lambda RMSE Rsquared MAE
## 0.0000000 0.000100 0.3018466 1 0.2621042
## 0.0000000 0.050075 0.3018466 1 0.2621042
## 0.0000000 0.100050 0.3018466 1 0.2621042
## 0.0000000 0.150025 0.3018466 1 0.2621042
## 0.0000000 0.200000 0.3018466 1 0.2621042
## 0.1111111 0.000100 0.4079107 1 0.3693623
## 0.1111111 0.050075 0.3853183 1 0.3467703
## 0.1111111 0.100050 0.3580800 1 0.3188674
## 0.1111111 0.150025 0.3396780 1 0.3000203
## 0.1111111 0.200000 0.3279803 1 0.2898342
## 0.2222222 0.000100 0.4158578 1 0.3781358
## 0.2222222 0.050075 0.3683127 1 0.3318370
## 0.2222222 0.100050 0.3340651 1 0.2976492
## 0.2222222 0.150025 0.3165231 1 0.2811197
## 0.2222222 0.200000 0.3066727 1 0.2695501
## 0.3333333 0.000100 0.4210263 1 0.3847318
## 0.3333333 0.050075 0.3470938 1 0.3118891
## 0.3333333 0.100050 0.3114716 1 0.2770359
## 0.3333333 0.150025 0.3023396 1 0.2674801
## 0.3333333 0.200000 0.2967204 1 0.2615725
## 0.4444444 0.000100 0.4273850 1 0.3924805
## 0.4444444 0.050075 0.3261126 1 0.2936150
## 0.4444444 0.100050 0.2962810 1 0.2635316
## 0.4444444 0.150025 0.2892884 1 0.2570612
## 0.4444444 0.200000 0.2899346 1 0.2566599
## 0.5555556 0.000100 0.4308227 1 0.3979816
## 0.5555556 0.050075 0.3053717 1 0.2750192
## 0.5555556 0.100050 0.2834180 1 0.2529481
## 0.5555556 0.150025 0.2826094 1 0.2512320
## 0.5555556 0.200000 0.2838240 1 0.2505237
## 0.6666667 0.000100 0.4317863 1 0.3986584
## 0.6666667 0.050075 0.2909196 1 0.2582465
## 0.6666667 0.100050 0.2728236 1 0.2434354
## 0.6666667 0.150025 0.2745284 1 0.2429465
## 0.6666667 0.200000 0.2768853 1 0.2469416
## 0.7777778 0.000100 0.4385506 1 0.4042271
## 0.7777778 0.050075 0.2759798 1 0.2402731
## 0.7777778 0.100050 0.2596748 1 0.2320422
## 0.7777778 0.150025 0.2645247 1 0.2360092
## 0.7777778 0.200000 0.2667466 1 0.2381834
## 0.8888889 0.000100 0.4469232 1 0.4104420
## 0.8888889 0.050075 0.2761240 1 0.2449359
## 0.8888889 0.100050 0.2480228 1 0.2211811
## 0.8888889 0.150025 0.2568940 1 0.2310975
## 0.8888889 0.200000 0.2653948 1 0.2367666
## 1.0000000 0.000100 0.4602056 1 0.4220002
## 1.0000000 0.050075 0.2663598 1 0.2354491
## 1.0000000 0.100050 0.2405888 1 0.2152169
## 1.0000000 0.150025 0.2522479 1 0.2272323
## 1.0000000 0.200000 0.2673058 1 0.2381457
##
## RMSE was used to select the optimal model using the smallest value.
## The final values used for the model were alpha = 1 and lambda = 0.10005.
plot(varImp(en_d, scale = TRUE))
# Coefficients
coefs <- coef(en_d$finalModel, en_d$bestTune$lambda)
coefs.zero <- coefs[which(coefs == 0),]
coef_order <- order(abs(coefs), decreasing = TRUE)
head(as.matrix(coefs[coef_order,]), 10)
## [,1]
## (Intercept) 0.9962072
## Avg_Sand 0.6980398
## Avg_Flow 0.0000000
## Flow_Variance 0.0000000
## Avg_Gravel 0.0000000
## Avg_Cobble 0.0000000
## Avg_Boulder 0.0000000
## Avg_Bedrock 0.0000000
## Avg_Organic 0.0000000
## Mean_Slope_Degrees 0.0000000
#RDA analysis
rda_env <- as.data.frame(env_scaled[,-c(1,3)])
rda_env1 <- lapply(rda_env, as.numeric)
rda_env <- na.omit(rda_env)
rda_biomass <- as.data.frame(for_clust[, -c(1)])
cor_matrix_env <- cor(rda_env)
corrplot(cor_matrix_env,
method = "color", # Color-based visualization
type = "upper", # Display upper triangle to avoid redundancy
tl.col = "black", # Text label color
tl.srt = 45, # Rotate text labels for better readability
col = colorRampPalette(c("blue", "white", "red"))(200), # Color scale (blue for negative, red for positive)
addCoef.col = "black", # Add correlation coefficients on the plot
diag = FALSE) # Remove the diagonal as it's always 1 (perfect correlation)
#Removing, total length, reach length, distance to headwaters, and boulder to improve model due to collinearity
rda_result1 <- rda(rda_biomass ~ Avg_Flow + Avg_Sand + Avg_Cobble + Avg_Gravel + Avg_Bedrock + Avg_Organic + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
summary(rda_result1)
##
## Call:
## rda(formula = rda_biomass ~ Avg_Flow + Avg_Sand + Avg_Cobble + Avg_Gravel + Avg_Bedrock + Avg_Organic + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
##
## Partitioning of variance:
## Inertia Proportion
## Total 0.3573 1.0000
## Constrained 0.2097 0.5869
## Unconstrained 0.1476 0.4131
##
## Eigenvalues, and their contribution to the variance
##
## Importance of components:
## RDA1 RDA2 RDA3 RDA4 PC1 PC2 PC3
## Eigenvalue 0.1228 0.06855 0.01180 0.006549 0.07579 0.06204 0.008995
## Proportion Explained 0.3437 0.19185 0.03303 0.018328 0.21209 0.17361 0.025173
## Cumulative Proportion 0.3437 0.53558 0.56861 0.586938 0.79903 0.97264 0.997812
## PC4
## Eigenvalue 0.0007818
## Proportion Explained 0.0021880
## Cumulative Proportion 1.0000000
##
## Accumulated constrained eigenvalues
## Importance of components:
## RDA1 RDA2 RDA3 RDA4
## Eigenvalue 0.1228 0.06855 0.01180 0.006549
## Proportion Explained 0.5856 0.32686 0.05627 0.031227
## Cumulative Proportion 0.5856 0.91250 0.96877 1.000000
# Get the summary of the RDA results
rda_summary1 <- summary(rda_result1)
# Access the scores for each variable (explained by RDA1 and RDA2)
explanatory_scores1 <- rda_summary1$species
biplot_scores1 <- scores(rda_result1, display = "bp")
# View the RDA1 and RDA2 scores for each variable
print(explanatory_scores1)
## RDA1 RDA2 RDA3 RDA4 PC1
## Benthic -0.4483168 0.3695983 -0.196402365 -0.0009572963 -0.0145337
## Drift -0.5373464 0.2376426 0.195353557 -0.0566125905 -0.3072169
## Fish 0.5352608 0.3514003 -0.002373459 -0.1345227378 -0.3818513
## Periphyton 0.2663705 0.3953242 0.068297590 0.1545028399 -0.5313528
## PC2
## Benthic -0.39791675
## Drift -0.42622368
## Fish -0.04881877
## Periphyton 0.29240054
rda_result1$CCA$biplot
## RDA1 RDA2 RDA3 RDA4
## Avg_Flow -0.03170125 0.6536434 -0.02107042 -0.301932192
## Avg_Sand -0.40469023 0.1696484 -0.30244201 -0.229385145
## Avg_Cobble 0.53051217 -0.1861885 0.54395708 -0.175714888
## Avg_Gravel -0.35884343 0.4993419 -0.21560312 -0.058080944
## Avg_Bedrock -0.17224580 -0.5107016 -0.12999230 0.026879231
## Avg_Organic -0.23401655 -0.1583259 -0.07028935 0.075607200
## Mean_Slope_Degrees 0.06155451 -0.5004119 -0.31513246 0.004100659
## Dist_Pour 0.06234654 0.3056049 0.56239915 -0.560803136
## Dist_Head 0.31421881 0.6199239 -0.05376866 -0.413030874
## Relative_position 0.20381983 0.1236047 -0.47819552 0.329960873
## Percent_occupied -0.40788546 -0.4472868 -0.11181006 0.518950245
rda_result1$CCA$v
## RDA1 RDA2 RDA3 RDA4
## Benthic -0.4870747 0.5374929 -0.688359933 -0.004504069
## Drift -0.5838010 0.3455946 0.684684024 -0.266361625
## Fish 0.5815351 0.5110283 -0.008318606 -0.632928024
## Periphyton 0.2893987 0.5749051 0.239372495 0.726934188
rda_result1$CCA$u
## RDA1 RDA2 RDA3 RDA4
## row1 -0.21138899 -0.344484529 -0.2472420 -0.07658550
## row2 0.18459751 -0.077433964 0.1011064 0.06256028
## row3 0.37064035 -0.077292946 -0.1982373 -0.22955477
## row4 0.13380192 -0.183502520 0.3797046 -0.13651876
## row5 -0.14775774 0.422597736 0.2234676 -0.21407699
## row6 0.12918815 0.140144711 0.2297294 -0.15864812
## row7 0.23410422 -0.045563572 -0.3042305 -0.39776854
## row8 0.28502672 0.179747948 -0.2566591 0.40737299
## row9 0.07388433 0.149794017 -0.1388560 0.25698915
## row10 -0.03490215 -0.395146190 0.1589141 0.47016520
## row11 0.27354363 -0.002961474 -0.1513152 0.26696058
## row12 -0.08851214 0.185897238 0.3538624 0.18855209
## row13 -0.05272083 -0.070477938 0.3268264 0.02380564
## row14 -0.64361898 -0.089689993 -0.1825884 0.10298529
## row15 -0.07546313 0.108310282 0.1206975 -0.17643623
## row16 -0.04822695 -0.040638259 0.0707420 -0.18444929
## row17 -0.14651964 -0.350146864 -0.1605113 -0.22993585
## row18 -0.23567627 0.490846317 -0.3254105 0.02458285
#reduce model further/explore if reduction is possible
sqrt(vif.cca(rda_result1))
## Avg_Flow Avg_Sand Avg_Cobble Avg_Gravel
## 3.752504 1.883210 3.865884 3.119525
## Avg_Bedrock Avg_Organic Mean_Slope_Degrees Dist_Pour
## 2.237205 2.201307 1.710499 4.828406
## Dist_Head Relative_position Percent_occupied
## 6.228138 4.965958 2.818134
anova.cca(rda_result1)#global RDA significance
## Permutation test for rda under reduced model
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = rda_biomass ~ Avg_Flow + Avg_Sand + Avg_Cobble + Avg_Gravel + Avg_Bedrock + Avg_Organic + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## Model 11 0.20973 0.7751 0.742
## Residual 6 0.14760
anova.cca(rda_result1, by = "axis") # significance of axes
## Permutation test for rda under reduced model
## Forward tests for axes
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = rda_biomass ~ Avg_Flow + Avg_Sand + Avg_Cobble + Avg_Gravel + Avg_Bedrock + Avg_Organic + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## RDA1 1 0.122826 10.8180 0.687
## RDA2 1 0.068553 6.0379 0.942
## RDA3 1 0.011802 1.0395 1.000
## RDA4 1 0.006549 0.5768
## Residual 13 0.147599
anova.cca(rda_result1, by = "terms") #significance of terms
## Permutation test for rda under reduced model
## Terms added sequentially (first to last)
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = rda_biomass ~ Avg_Flow + Avg_Sand + Avg_Cobble + Avg_Gravel + Avg_Bedrock + Avg_Organic + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## Avg_Flow 1 0.030015 1.2201 0.348
## Avg_Sand 1 0.027126 1.1027 0.368
## Avg_Cobble 1 0.018410 0.7484 0.527
## Avg_Gravel 1 0.013705 0.5571 0.641
## Avg_Bedrock 1 0.013967 0.5678 0.639
## Avg_Organic 1 0.021678 0.8812 0.460
## Mean_Slope_Degrees 1 0.010148 0.4125 0.774
## Dist_Pour 1 0.005825 0.2368 0.888
## Dist_Head 1 0.040541 1.6480 0.213
## Relative_position 1 0.009013 0.3664 0.787
## Percent_occupied 1 0.019302 0.7846 0.518
## Residual 6 0.147599
#Trying ordistep() which uses p-value as a stopping criterion
step.forward <- ordistep(rda(rda_biomass ~ 1, data=rda_env1), scope=formula(rda_result1), direction="forward", pstep=1000)
##
## Start: rda_biomass ~ 1
##
## Df AIC F Pr(>F)
## + Avg_Cobble 1 -17.726 2.0532 0.130
## + Dist_Head 1 -17.668 1.9955 0.145
## + Percent_occupied 1 -17.468 1.7959 0.150
## + Avg_Gravel 1 -17.323 1.6541 0.185
## + Avg_Flow 1 -17.132 1.4672 0.230
## + Avg_Sand 1 -16.778 1.1270 0.350
## + Avg_Bedrock 1 -16.682 1.0359 0.390
## + Mean_Slope_Degrees 1 -16.526 0.8887 0.395
## + Dist_Pour 1 -16.203 0.5883 0.630
## + Relative_position 1 -16.041 0.4399 0.750
## + Avg_Organic 1 -15.988 0.3918 0.755
# Define basin colors (you can adjust this list based on your actual basin names)
site_colors <- c(
"AA1_REACH" = "darkorange3", "AA2_REACH" = "darkorange4",
"AC_REACH" = "slateblue3", "AP_REACH" = "darkcyan",
"AZ_REACH" = "gold3", "BA1_REACH" = "firebrick3",
"BA2_REACH" = "green4", "BC_REACH" = "purple",
"BZ_REACH" = "darkgreen", "CA1_REACH" = "darkolivegreen4",
"CA2_REACH" = "mediumblue", "CP_REACH" = "yellowgreen",
"CZ2_REACH" = "darkred", "DA1_REACH" = "seagreen",
"DA2_REACH" = "indianred", "DC_REACH" = "orangered",
"DP_REACH" = "chocolate", "DZ_REACH" = "violet"
)
# Vector of colors for each site based on bAsin
colors <- sapply(env_scaled$Basin, function(Basin) site_colors[Basin])
# Extract site coordinates (coordinates of site scores)
site_scores1 <- rda_result1$CCA$u
par(mar = c(5, 5, 2, 7))
# Plot RDA result, scaling 2 (displaying effects of explanatory variables)
plot(rda_result1, xlim = c(-1, 1), ylim = c(-1,1), scaling = 2, main = "RDA - Scaling = 2", type = "n")
# Plot species labels
text(rda_result1, display = "species", col = "firebrick3", cex = 1)
# Plot sites with colors corresponding to basins
points(site_scores1, col = colors, pch = 20, cex = 1.)
# Plot biplot arrows (explanatory variables)
text(rda_result1, scaling = 2, display = "bp", col = "darkblue", cex = 1)
# Add the legend to the plot, positioned outside
legend("right", legend = names(site_colors), col = site_colors, pch = 19, title = "Basin",
inset = -0.2, xpd = TRUE, cex = 0.7)
rda_result2 <- rda(rda_biomass ~ Avg_Flow + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
summary(rda_result2)
##
## Call:
## rda(formula = rda_biomass ~ Avg_Flow + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
##
## Partitioning of variance:
## Inertia Proportion
## Total 0.3573 1.0000
## Constrained 0.1271 0.3557
## Unconstrained 0.2302 0.6443
##
## Eigenvalues, and their contribution to the variance
##
## Importance of components:
## RDA1 RDA2 RDA3 RDA4 PC1 PC2 PC3
## Eigenvalue 0.07456 0.04206 0.006889 0.003577 0.1082 0.09487 0.01840
## Proportion Explained 0.20866 0.11772 0.019280 0.010010 0.3027 0.26551 0.05149
## Cumulative Proportion 0.20866 0.32638 0.345660 0.355670 0.6583 0.92384 0.97533
## PC4
## Eigenvalue 0.008814
## Proportion Explained 0.024665
## Cumulative Proportion 1.000000
##
## Accumulated constrained eigenvalues
## Importance of components:
## RDA1 RDA2 RDA3 RDA4
## Eigenvalue 0.07456 0.04206 0.006889 0.003577
## Proportion Explained 0.58667 0.33098 0.054209 0.028145
## Cumulative Proportion 0.58667 0.91765 0.971855 1.000000
# Get the summary of the RDA results
rda_summary2 <- summary(rda_result2)
# Access the scores for each variable (explained by RDA1 and RDA2)
explanatory_scores2 <- rda_summary2$species
biplot_scores2 <- scores(rda_result2, display = "bp")
# View the RDA1 and RDA2 scores for each variable
print(explanatory_scores2)
## RDA1 RDA2 RDA3 RDA4 PC1 PC2
## Benthic 0.2860379 -0.2653565 -0.16828029 0.007521157 -0.4006270 0.4179902
## Drift 0.4401563 -0.2415147 0.13429969 0.032453225 -0.2465405 0.5466802
## Fish -0.4523481 -0.2887791 0.01108955 0.087751864 0.3887168 0.2930666
## Periphyton -0.1846746 -0.2792890 0.03228372 -0.125943466 0.6112154 0.3081027
rda_result2$CCA$biplot
## RDA1 RDA2 RDA3 RDA4
## Avg_Flow 0.008001701 -0.8388008 -0.20142998 0.1938876
## Mean_Slope_Degrees -0.068603785 0.6525343 -0.24143783 0.1980816
## Dist_Pour -0.080189772 -0.4273673 0.70977831 0.5391080
## Dist_Head -0.434572225 -0.7746998 -0.09516631 0.2561819
## Relative_position -0.277630207 -0.1094919 -0.60495568 -0.4696942
## Percent_occupied 0.542384037 0.5609244 -0.21239479 -0.4029779
rda_result2$CCA$v
## RDA1 RDA2 RDA3 RDA4
## Benthic 0.3988635 -0.4926365 -0.77196139 0.0478832
## Drift 0.6137728 -0.4483740 0.61608034 0.2066124
## Fish -0.6307735 -0.5361207 0.05087168 0.5586693
## Periphyton -0.2575182 -0.5185023 0.14809689 -0.8018149
rda_result2$CCA$u
## RDA1 RDA2 RDA3 RDA4
## row1 -0.194362619 0.29471347 -0.01829024 -0.285462452
## row2 -0.125291289 0.20671338 0.03528298 0.234165395
## row3 -0.221996348 0.15826910 0.06081694 0.248893946
## row4 -0.186136729 -0.09304450 0.44740051 -0.075706044
## row5 0.150712346 -0.48837164 0.18766781 0.140916994
## row6 0.053073880 0.05654218 0.01997357 -0.038398959
## row7 -0.435438370 -0.12111428 -0.40048615 0.245345080
## row8 -0.373076868 0.23822194 -0.08452013 -0.212036968
## row9 -0.073447035 -0.08346520 0.10484241 -0.293917641
## row10 0.282077401 0.33524618 0.03829178 -0.383200287
## row11 -0.017363776 -0.23730407 -0.46473935 -0.272604529
## row12 0.047659926 -0.23030157 0.22056040 -0.215222743
## row13 0.009060125 0.07272707 0.20260956 0.185595112
## row14 0.591781168 0.17743083 -0.25255673 -0.003898403
## row15 0.088546291 -0.12132158 0.33247063 0.120749864
## row16 0.158488290 -0.05951718 -0.05081201 0.061520118
## row17 0.183285800 0.28955038 -0.07806656 0.511688624
## row18 0.062427809 -0.39497452 -0.30044542 0.031572892
#reduce model further/explore if reduction is possible
sqrt(vif.cca(rda_result2))
## Avg_Flow Mean_Slope_Degrees Dist_Pour Dist_Head
## 2.676554 1.664104 3.811689 4.984668
## Relative_position Percent_occupied
## 3.925248 2.423043
anova.cca(rda_result2)#global RDA significance
## Permutation test for rda under reduced model
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = rda_biomass ~ Avg_Flow + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## Model 6 0.12709 1.012 0.47
## Residual 11 0.23024
anova.cca(rda_result2, by = "axis") # significance of axes
## Permutation test for rda under reduced model
## Forward tests for axes
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = rda_biomass ~ Avg_Flow + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## RDA1 1 0.074560 4.2099 0.498
## RDA2 1 0.042065 2.3751 0.828
## RDA3 1 0.006889 0.3890 1.000
## RDA4 1 0.003577 0.2020
## Residual 13 0.230238
anova.cca(rda_result2, by = "terms") #significance of terms
## Permutation test for rda under reduced model
## Terms added sequentially (first to last)
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = rda_biomass ~ Avg_Flow + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = rda_env1, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## Avg_Flow 1 0.030015 1.4340 0.236
## Mean_Slope_Degrees 1 0.006484 0.3098 0.830
## Dist_Pour 1 0.006464 0.3088 0.837
## Dist_Head 1 0.065564 3.1324 0.040 *
## Relative_position 1 0.003510 0.1677 0.937
## Percent_occupied 1 0.015055 0.7193 0.531
## Residual 11 0.230238
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Trying ordistep() which uses p-value as a stopping criterion
step.forward <- ordistep(rda(rda_biomass ~ 1, data=rda_env1), scope=formula(rda_result2), direction="forward", pstep=1000)
##
## Start: rda_biomass ~ 1
##
## Df AIC F Pr(>F)
## + Dist_Head 1 -17.668 1.9955 0.140
## + Percent_occupied 1 -17.468 1.7959 0.140
## + Avg_Flow 1 -17.132 1.4672 0.240
## + Mean_Slope_Degrees 1 -16.526 0.8887 0.430
## + Dist_Pour 1 -16.203 0.5883 0.635
## + Relative_position 1 -16.041 0.4399 0.730
# Extract site coordinates (coordinates of site scores)
site_scores2 <- rda_result2$CCA$u
par(mar = c(5, 5, 2, 7))
# Plot RDA result, scaling 2 (displaying effects of explanatory variables)
plot(rda_result2, xlim = c(-1, 1), ylim = c(-1,1), scaling = 2, main = "RDA - Scaling = 2", type = "n")
# Plot species labels
text(rda_result2, display = "species", col = "firebrick3", cex = 1)
# Plot sites with colors corresponding to basins
points(site_scores2, col = colors, pch = 20, cex = 1.)
# Plot biplot arrows (explanatory variables)
text(rda_result2, scaling = 2, display = "bp", col = "darkblue", cex = 1)
# Add the legend to the plot, positioned outside
legend("right", legend = names(site_colors), col = site_colors, pch = 19, title = "Basin",
inset = -0.2, xpd = TRUE, cex = 0.7)
# Define response and predictor variables
response_var <- colnames(merged)[2:5] # Columns 2, 3, 4, and 5 are response variables
env_var <- colnames(merged)[6:20] # Columns 6 to 20 are environmental variables
# Empty data frame to store results
results_table <- data.frame(Response_Variable = character(),
Predictor_Variable = character(),
Coefficient = numeric(),
Std_Error = numeric(),
t_value = numeric(),
p_value = numeric(),
RSquared = numeric(), # Add a column for R^2
stringsAsFactors = FALSE)
# Perform regression analysis for each response variable and predictor variable
for (response in response_var) {
for (env in env_var) {
model <- lm(merged[[response]] ~ merged[[env]])
# Results stored in the table
results_table <- rbind(results_table, data.frame(Response_Variable = response,
Predictor_Variable = env,
Coefficient = coef(model)[2],
Std_Error = summary(model)$coefficients[2, 2],
t_value = summary(model)$coefficients[2, 3],
p_value = summary(model)$coefficients[2, 4],
RSquared = summary(model)$r.squared))
}
}
# Filter significant results (p-value < 0.05)
significant_results <- results_table[results_table$p_value < 0.05, ]
# Highlighting the table based on conditions
library(DT)
## Warning: package 'DT' was built under R version 4.4.2
# Hide column names
rownames(results_table) <- NULL
# Highlight R^2 > 0.3 and p-values < 0.05
datatable(results_table, options = list(
columnDefs = list(
list(
targets = 7, # R_squared column index
render = JS("function(data, type, row, meta) {",
"if (parseFloat(data) > 0.3) {",
"return '<span style=\"color: red;\">' + data + '</span>';",
"} else {",
"return data;}",
"}")
),
list(
targets = 6, # p_value column index
render = JS("function(data, type, row, meta) {",
"if (parseFloat(data) < 0.05) {",
"return '<span style=\"color: blue;\">' + data + '</span>';",
"} else {",
"return data;}",
"}")
)
),
dom = 'tip',
pageLength = 10,
lengthMenu = c(10, 20, 50),
scrollX = TRUE,
width = "80%"
))
FFG_df <- benthic_absummary %>% filter(Feeding.Group != "PA")
# Remove the '%' sign and convert the column to numeric
FFG_df$Relative_Abundance <- as.numeric(gsub("%", "", FFG_df$Relative_Abundance)) / 100
FFG_df <- FFG_df %>%
group_by(Basin, Feeding.Group) %>%
summarise(Relative_Abundance = sum(Relative_Abundance), .groups = "drop")
df_wide <- FFG_df %>%
pivot_wider(names_from = Feeding.Group, values_from = Relative_Abundance, values_fill = 0)
expected_cols <- c("DSH", "HSH", "CF", "CG", "PR", "MH", "OM", "SC")
df_wide[setdiff(expected_cols, names(df_wide))] <- 0
df_wide <- df_wide %>%
mutate(
PR = ifelse((DSH + CG + CF) == 0, NA, (SC + HSH) / (DSH + CG + CF)),
Plants_Algae = ifelse(SC == 0, NA, HSH / SC),
CPOM_FPOM = ifelse(CG + CF == 0, NA, DSH / (CG + CF)),
FPOM_CPOM = ifelse(CG == 0, NA, CF / CG),
Channel_Stability = ifelse((DSH + CG) == 0, NA, (SC + HSH + CF) / (DSH + CG)),
Predator_Control = ifelse((SC + DSH + HSH + CF + CG) == 0, NA, PR / (SC + DSH + HSH + CF + CG))
)
df_wide <- df_wide %>%
mutate(
PR_Flag = case_when(PR > 0.75 ~ "Autotrophic (Above 0.75)", TRUE ~ "Heterotrophic (Below 0.75)"),
Plants_Algae_Flag = case_when(Plants_Algae > 0.50 ~ "Vascular Plants Dominant (Above 0.50)", TRUE ~ "Algae Dominant (Below 0.50)"),
CPOM_FPOM_Flag = case_when(CPOM_FPOM > 0.50 ~ "High Shredder Dependency (Above 0.25)", TRUE ~ "Low Shredder Dependency (Below 0.25)"),
FPOM_CPOM_Flag = case_when(FPOM_CPOM > 0.50 ~ "High FPOM Transport (Above 0.50)", TRUE ~ "Normal FPOM (Below 0.50)"),
Channel_Stability_Flag = case_when(Channel_Stability > 0.50 ~ "Stable Channel (Above 0.50)", TRUE ~ "Unstable Channel (Below 0.50)"),
Predator_Control_Flag = case_when(Predator_Control > 0.50 ~ "Predator Dominated (Above 0.50)", TRUE ~ "Prey Dominated (Below 0.50)")
)
# Create a concise table with flagging results
df_flags <- df_wide %>%
select(Basin, PR_Flag, Plants_Algae_Flag, CPOM_FPOM_Flag, FPOM_CPOM_Flag, Channel_Stability_Flag, Predator_Control_Flag)
# View the table
print(df_flags)
## # A tibble: 20 × 7
## Basin PR_Flag Plants_Algae_Flag CPOM_FPOM_Flag FPOM_CPOM_Flag
## <chr> <chr> <chr> <chr> <chr>
## 1 AA1_REACH Heterotrophic (Bel… Algae Dominant (… High Shredder… Normal FPOM (…
## 2 AA2_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 3 AC_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 4 AP_REACH Autotrophic (Above… Algae Dominant (… Low Shredder … Normal FPOM (…
## 5 AZ_REACH Autotrophic (Above… Algae Dominant (… Low Shredder … Normal FPOM (…
## 6 BA1_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 7 BA2_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … High FPOM Tra…
## 8 BC_REACH Autotrophic (Above… Algae Dominant (… Low Shredder … Normal FPOM (…
## 9 BP_REACH Autotrophic (Above… Algae Dominant (… Low Shredder … Normal FPOM (…
## 10 BZ_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 11 CA1_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 12 CA2_REACH Autotrophic (Above… Algae Dominant (… Low Shredder … Normal FPOM (…
## 13 CC_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 14 CP_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 15 CZ2_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 16 DA1_REACH Heterotrophic (Bel… Algae Dominant (… High Shredder… Normal FPOM (…
## 17 DA2_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 18 DC_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 19 DP_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## 20 DZ_REACH Heterotrophic (Bel… Algae Dominant (… Low Shredder … Normal FPOM (…
## # ℹ 2 more variables: Channel_Stability_Flag <chr>, Predator_Control_Flag <chr>
condition_biomass1 <- read.csv("reach_condition.csv")
# Shapiro-Wilk test for normality (for each response variable)
shapiro.test(condition_biomass1$Fish)
##
## Shapiro-Wilk normality test
##
## data: condition_biomass1$Fish
## W = 0.95456, p-value = 0.501
shapiro.test(condition_biomass1$Periphyton)
##
## Shapiro-Wilk normality test
##
## data: condition_biomass1$Periphyton
## W = 0.94668, p-value = 0.3753
shapiro.test(condition_biomass1$Benthic)
##
## Shapiro-Wilk normality test
##
## data: condition_biomass1$Benthic
## W = 0.8098, p-value = 0.002096
shapiro.test(condition_biomass1$Drift)
##
## Shapiro-Wilk normality test
##
## data: condition_biomass1$Drift
## W = 0.85601, p-value = 0.01055
condition_biomass <- condition_biomass1[, -c(1, 3, 8, 9)]
condition_biomass$PR_Flag <- as.factor(condition_biomass$PR_Flag)
condition_biomass$CPOM_FPOM_Flag <- as.factor(condition_biomass$CPOM_FPOM_Flag)
condition_biomass$FPOM_CPOM_Flag <- as.factor(condition_biomass$FPOM_CPOM_Flag)
condition_biomass$Channel_Stability_Flag <- as.factor(condition_biomass$Channel_Stability_Flag)
condition_biomass$Predator_Control_Flag <- as.factor(condition_biomass$Predator_Control_Flag)
manova1 <- manova(cbind(Fish, Periphyton, Benthic, Drift) ~ PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag, data = condition_biomass)
summary(manova1, test = "Pillai")
## Df Pillai approx F num Df den Df Pr(>F)
## PR_Flag 1 0.28096 0.97683 4 10 0.4622
## CPOM_FPOM_Flag 1 0.40604 1.70905 4 10 0.2240
## FPOM_CPOM_Flag 1 0.25102 0.83786 4 10 0.5315
## Channel_Stability_Flag 1 0.36633 1.44526 4 10 0.2893
## Residuals 13
#ANOVAS
summary(aov(Fish ~ PR_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## PR_Flag 1 0.0978 0.09780 1.141 0.301
## Residuals 16 1.3711 0.08569
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ PR_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## PR_Flag 1 0.0014 0.00139 0.014 0.907
## Residuals 16 1.5724 0.09827
## 2 observations deleted due to missingness
summary(aov(Benthic ~ PR_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## PR_Flag 1 0.1064 0.10641 1.33 0.266
## Residuals 16 1.2804 0.08003
## 2 observations deleted due to missingness
summary(aov(Drift ~ PR_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## PR_Flag 1 0.0532 0.05316 0.534 0.475
## Residuals 16 1.5920 0.09950
## 2 observations deleted due to missingness
summary(aov(Fish ~ CPOM_FPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## CPOM_FPOM_Flag 1 0.3246 0.3246 4.538 0.049 *
## Residuals 16 1.1443 0.0715
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2 observations deleted due to missingness
summary(aov(Fish ~ FPOM_CPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## FPOM_CPOM_Flag 1 0.0884 0.08835 1.024 0.327
## Residuals 16 1.3805 0.08628
## 2 observations deleted due to missingness
summary(aov(Fish ~ Channel_Stability_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Channel_Stability_Flag 1 0.6298 0.6298 12.01 0.00319 **
## Residuals 16 0.8391 0.0524
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2 observations deleted due to missingness
summary(aov(Fish ~ Predator_Control_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Predator_Control_Flag 1 0.4376 0.4376 6.79 0.0191 *
## Residuals 16 1.0313 0.0645
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2 observations deleted due to missingness
summary(aov(Fish ~ Mean_Slope_Degrees, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Mean_Slope_Degrees 1 0.0502 0.05019 0.566 0.463
## Residuals 16 1.4187 0.08867
## 2 observations deleted due to missingness
summary(aov(Fish ~ Dist_Pour, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Pour 1 0.1138 0.11379 1.344 0.263
## Residuals 16 1.3551 0.08469
## 2 observations deleted due to missingness
summary(aov(Fish ~ Dist_Head, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Head 1 0.4809 0.4809 7.788 0.0131 *
## Residuals 16 0.9880 0.0617
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2 observations deleted due to missingness
summary(aov(Fish ~ Relative_position, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Relative_position 1 0.0294 0.02943 0.327 0.575
## Residuals 16 1.4395 0.08997
## 2 observations deleted due to missingness
summary(aov(Fish ~ Percent_occupied, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Percent_occupied 1 0.4882 0.4882 7.964 0.0123 *
## Residuals 16 0.9807 0.0613
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ CPOM_FPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## CPOM_FPOM_Flag 1 0.1421 0.14211 1.588 0.226
## Residuals 16 1.4317 0.08948
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ FPOM_CPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## FPOM_CPOM_Flag 1 0.000 0.00001 0 0.994
## Residuals 16 1.574 0.09836
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Channel_Stability_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Channel_Stability_Flag 1 0.1858 0.18585 2.142 0.163
## Residuals 16 1.3879 0.08675
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Predator_Control_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Predator_Control_Flag 1 0.1872 0.18718 2.16 0.161
## Residuals 16 1.3866 0.08666
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Mean_Slope_Degrees, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Mean_Slope_Degrees 1 0.1009 0.10088 1.096 0.311
## Residuals 16 1.4729 0.09206
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Dist_Pour, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Pour 1 0.0196 0.01960 0.202 0.659
## Residuals 16 1.5542 0.09714
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Dist_Head, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Head 1 0.1683 0.16826 1.915 0.185
## Residuals 16 1.4055 0.08784
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Relative_position, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Relative_position 1 0.0364 0.03637 0.378 0.547
## Residuals 16 1.5374 0.09609
## 2 observations deleted due to missingness
summary(aov(Periphyton ~ Percent_occupied, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Percent_occupied 1 0.1117 0.11174 1.223 0.285
## Residuals 16 1.4620 0.09138
## 2 observations deleted due to missingness
summary(aov(Benthic ~ CPOM_FPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## CPOM_FPOM_Flag 1 0.1618 0.16185 2.114 0.165
## Residuals 16 1.2250 0.07656
## 2 observations deleted due to missingness
summary(aov(Benthic ~ FPOM_CPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## FPOM_CPOM_Flag 1 0.0062 0.00621 0.072 0.792
## Residuals 16 1.3806 0.08629
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Channel_Stability_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Channel_Stability_Flag 1 0.1929 0.19287 2.585 0.127
## Residuals 16 1.1939 0.07462
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Predator_Control_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Predator_Control_Flag 1 0.1645 0.1645 2.153 0.162
## Residuals 16 1.2223 0.0764
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Mean_Slope_Degrees, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Mean_Slope_Degrees 1 0.0559 0.05594 0.673 0.424
## Residuals 16 1.3309 0.08318
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Dist_Pour, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Pour 1 0.0015 0.00153 0.018 0.896
## Residuals 16 1.3853 0.08658
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Dist_Head, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Head 1 0.0243 0.02426 0.285 0.601
## Residuals 16 1.3626 0.08516
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Relative_position, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Relative_position 1 0.0057 0.00566 0.066 0.801
## Residuals 16 1.3812 0.08632
## 2 observations deleted due to missingness
summary(aov(Benthic ~ Percent_occupied, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Percent_occupied 1 0.0038 0.00375 0.043 0.838
## Residuals 16 1.3831 0.08644
## 2 observations deleted due to missingness
summary(aov(Drift ~ CPOM_FPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## CPOM_FPOM_Flag 1 0.2371 0.2371 2.694 0.12
## Residuals 16 1.4081 0.0880
## 2 observations deleted due to missingness
summary(aov(Drift ~ FPOM_CPOM_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## FPOM_CPOM_Flag 1 0.0712 0.07121 0.724 0.407
## Residuals 16 1.5739 0.09837
## 2 observations deleted due to missingness
summary(aov(Drift ~ Channel_Stability_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Channel_Stability_Flag 1 0.0605 0.06048 0.611 0.446
## Residuals 16 1.5846 0.09904
## 2 observations deleted due to missingness
summary(aov(Drift ~ Predator_Control_Flag, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Predator_Control_Flag 1 0.0155 0.01548 0.152 0.702
## Residuals 16 1.6296 0.10185
## 2 observations deleted due to missingness
summary(aov(Drift ~ Mean_Slope_Degrees, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Mean_Slope_Degrees 1 0.1127 0.11265 1.176 0.294
## Residuals 16 1.5325 0.09578
## 2 observations deleted due to missingness
summary(aov(Drift ~ Dist_Pour, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Pour 1 0.0805 0.08051 0.823 0.378
## Residuals 16 1.5646 0.09779
## 2 observations deleted due to missingness
summary(aov(Drift ~ Dist_Head, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Dist_Head 1 0.0002 0.00018 0.002 0.967
## Residuals 16 1.6449 0.10281
## 2 observations deleted due to missingness
summary(aov(Drift ~ Relative_position, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Relative_position 1 0.0911 0.09109 0.938 0.347
## Residuals 16 1.5540 0.09713
## 2 observations deleted due to missingness
summary(aov(Drift ~ Percent_occupied, data = condition_biomass))
## Df Sum Sq Mean Sq F value Pr(>F)
## Percent_occupied 1 0.0094 0.00937 0.092 0.766
## Residuals 16 1.6357 0.10223
## 2 observations deleted due to missingness
mancova <- manova(cbind(Fish, Periphyton, Benthic, Drift) ~ PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag + Mean_Slope_Degrees + Dist_Pour + Dist_Head + Relative_position + Percent_occupied, data = condition_biomass)
summary(mancova, test = "Pillai")
## Df Pillai approx F num Df den Df Pr(>F)
## PR_Flag 1 0.48963 1.19922 4 5 0.4139
## CPOM_FPOM_Flag 1 0.58741 1.77963 4 5 0.2699
## FPOM_CPOM_Flag 1 0.43938 0.97969 4 5 0.4938
## Channel_Stability_Flag 1 0.45069 1.02559 4 5 0.4756
## Mean_Slope_Degrees 1 0.35813 0.69745 4 5 0.6256
## Dist_Pour 1 0.33791 0.63797 4 5 0.6580
## Dist_Head 1 0.50847 1.29310 4 5 0.3847
## Relative_position 1 0.18576 0.28518 4 5 0.8761
## Percent_occupied 1 0.61406 1.98886 4 5 0.2346
## Residuals 8
#NO SIGNIFICANT RELATIONSHIPS
# Model for Fish biomass
gam_fish <- gam(Fish ~ s(Mean_Slope_Degrees, k = 10) + s(Dist_Pour, k = 10) + s(Dist_Head, k = 10) + s(Relative_position, k= 10) + s(Percent_occupied, k=5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag, data = condition_biomass, family = gaussian)
# Model for Periphyton biomass
gam_periphyton <- gam(Periphyton ~ s(Mean_Slope_Degrees, k = 5) + s(Dist_Pour, k = 5) + s(Dist_Head, k = 5) + s(Relative_position, k= 5) + s(Percent_occupied, k=5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag, data = condition_biomass, family = gaussian)
# Model for Benthic biomass
gam_benthic <- gam(Benthic ~ s(Mean_Slope_Degrees, k = 5) + s(Dist_Pour, k = 5) + s(Dist_Head, k = 5) + s(Relative_position, k= 5) + s(Percent_occupied, k=5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag, data = condition_biomass, family = gaussian)
# Model for Drift biomass
gam_drift <- gam(Drift ~ s(Mean_Slope_Degrees, k = 5) + s(Dist_Pour, k = 5) + s(Dist_Head, k = 5) + s(Relative_position, k= 5) + s(Percent_occupied, k=5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag, data = condition_biomass, family = gaussian)
# Checking the summaries and assumptions for each model
# Summary of the Fish biomass model
summary(gam_fish)
##
## Family: gaussian
## Link function: identity
##
## Formula:
## Fish ~ s(Mean_Slope_Degrees, k = 10) + s(Dist_Pour, k = 10) +
## s(Dist_Head, k = 10) + s(Relative_position, k = 10) + s(Percent_occupied,
## k = 5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag +
## Predator_Control_Flag
##
## Parametric coefficients:
## Estimate Std. Error t value
## (Intercept) 1.17909 0.33168 3.555
## PR_FlagHeterotrophic (Below 0.75) 0.01717 0.31314 0.055
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) 0.20747 0.24815 0.836
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.14992 0.14064 1.066
## Channel_Stability_FlagUnstable Channel (Below 0.50) -0.56652 0.16073 -3.525
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.46265 0.20188 2.292
## Pr(>|t|)
## (Intercept) 0.0258 *
## PR_FlagHeterotrophic (Below 0.75) 0.9590
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) 0.4525
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.3494
## Channel_Stability_FlagUnstable Channel (Below 0.50) 0.0265 *
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.0872 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Approximate significance of smooth terms:
## edf Ref.df F p-value
## s(Mean_Slope_Degrees) 1.000 1.000 4.819 0.0932 .
## s(Dist_Pour) 4.041 4.645 2.484 0.1865
## s(Dist_Head) 1.000 1.000 5.280 0.0831 .
## s(Relative_position) 1.000 1.000 0.293 0.6168
## s(Percent_occupied) 2.159 2.447 5.061 0.0805 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Rank: 45/46
## R-sq.(adj) = 0.769 Deviance explained = 94.8%
## GCV = 0.094479 Scale est. = 0.019946 n = 18
gam.check(gam_fish)
##
## Method: GCV Optimizer: magic
## Smoothing parameter selection converged after 31 iterations.
## The RMS GCV score gradient at convergence was 0.00000009283181 .
## The Hessian was positive definite.
## Model rank = 45 / 46
##
## Basis dimension (k) checking results. Low p-value (k-index<1) may
## indicate that k is too low, especially if edf is close to k'.
##
## k' edf k-index p-value
## s(Mean_Slope_Degrees) 9.00 1.00 1.46 0.96
## s(Dist_Pour) 9.00 4.04 1.69 0.99
## s(Dist_Head) 9.00 1.00 1.69 1.00
## s(Relative_position) 9.00 1.00 1.36 0.90
## s(Percent_occupied) 4.00 2.16 1.16 0.72
shapiro.test(gam_fish$residuals)
##
## Shapiro-Wilk normality test
##
## data: gam_fish$residuals
## W = 0.92947, p-value = 0.1901
# Summary of the Periphyton biomass model
summary(gam_periphyton)
##
## Family: gaussian
## Link function: identity
##
## Formula:
## Periphyton ~ s(Mean_Slope_Degrees, k = 5) + s(Dist_Pour, k = 5) +
## s(Dist_Head, k = 5) + s(Relative_position, k = 5) + s(Percent_occupied,
## k = 5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag +
## Predator_Control_Flag
##
## Parametric coefficients:
## Estimate Std. Error t value
## (Intercept) 1.52561 0.30737 4.963
## PR_FlagHeterotrophic (Below 0.75) -0.58878 0.27430 -2.147
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) -0.01111 0.22032 -0.050
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.30882 0.13368 2.310
## Channel_Stability_FlagUnstable Channel (Below 0.50) -0.54442 0.14860 -3.664
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.67237 0.22077 3.046
## Pr(>|t|)
## (Intercept) 0.0147 *
## PR_FlagHeterotrophic (Below 0.75) 0.1185
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) 0.9629
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.1015
## Channel_Stability_FlagUnstable Channel (Below 0.50) 0.0335 *
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.0536 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Approximate significance of smooth terms:
## edf Ref.df F p-value
## s(Mean_Slope_Degrees) 1.502 1.732 3.074 0.236
## s(Dist_Pour) 3.250 3.614 2.918 0.170
## s(Dist_Head) 1.640 1.834 0.703 0.475
## s(Relative_position) 2.520 2.808 5.912 0.078 .
## s(Percent_occupied) 1.000 1.000 0.495 0.532
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Rank: 25/26
## R-sq.(adj) = 0.859 Deviance explained = 97.4%
## GCV = 0.075843 Scale est. = 0.01301 n = 18
gam.check(gam_periphyton)
##
## Method: GCV Optimizer: magic
## Smoothing parameter selection converged after 19 iterations.
## The RMS GCV score gradient at convergence was 0.00000007960886 .
## The Hessian was positive definite.
## Model rank = 25 / 26
##
## Basis dimension (k) checking results. Low p-value (k-index<1) may
## indicate that k is too low, especially if edf is close to k'.
##
## k' edf k-index p-value
## s(Mean_Slope_Degrees) 4.00 1.50 1.42 0.92
## s(Dist_Pour) 4.00 3.25 1.45 0.95
## s(Dist_Head) 4.00 1.64 1.42 0.92
## s(Relative_position) 4.00 2.52 1.59 1.00
## s(Percent_occupied) 4.00 1.00 1.56 0.98
shapiro.test(gam_periphyton$residuals)
##
## Shapiro-Wilk normality test
##
## data: gam_periphyton$residuals
## W = 0.933, p-value = 0.2192
# Summary of the Benthic biomass model
summary(gam_benthic)
##
## Family: gaussian
## Link function: identity
##
## Formula:
## Benthic ~ s(Mean_Slope_Degrees, k = 5) + s(Dist_Pour, k = 5) +
## s(Dist_Head, k = 5) + s(Relative_position, k = 5) + s(Percent_occupied,
## k = 5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag +
## Predator_Control_Flag
##
## Parametric coefficients:
## Estimate Std. Error t value
## (Intercept) 0.86426 0.24308 3.556
## PR_FlagHeterotrophic (Below 0.75) 0.14067 0.16280 0.864
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) -0.36598 0.19656 -1.862
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.57120 0.17012 3.358
## Channel_Stability_FlagUnstable Channel (Below 0.50) -0.01361 0.20488 -0.066
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.27946 0.18732 1.492
## Pr(>|t|)
## (Intercept) 0.00886 **
## PR_FlagHeterotrophic (Below 0.75) 0.41542
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) 0.10374
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.01163 *
## Channel_Stability_FlagUnstable Channel (Below 0.50) 0.94884
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.17820
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Approximate significance of smooth terms:
## edf Ref.df F p-value
## s(Mean_Slope_Degrees) 1.000 1.000 0.835 0.3912
## s(Dist_Pour) 1.000 1.000 3.045 0.1245
## s(Dist_Head) 1.802 2.195 3.770 0.0823 .
## s(Relative_position) 1.000 1.000 3.749 0.0940 .
## s(Percent_occupied) 1.000 1.000 4.221 0.0790 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Rank: 25/26
## R-sq.(adj) = 0.444 Deviance explained = 76.5%
## GCV = 0.11332 Scale est. = 0.045317 n = 18
gam.check(gam_benthic)
##
## Method: GCV Optimizer: magic
## Smoothing parameter selection converged after 14 iterations.
## The RMS GCV score gradient at convergence was 0.00000003721144 .
## The Hessian was positive definite.
## Model rank = 25 / 26
##
## Basis dimension (k) checking results. Low p-value (k-index<1) may
## indicate that k is too low, especially if edf is close to k'.
##
## k' edf k-index p-value
## s(Mean_Slope_Degrees) 4.0 1.0 1.28 0.79
## s(Dist_Pour) 4.0 1.0 0.94 0.36
## s(Dist_Head) 4.0 1.8 1.40 0.93
## s(Relative_position) 4.0 1.0 1.16 0.61
## s(Percent_occupied) 4.0 1.0 1.38 0.94
shapiro.test(gam_benthic$residuals)
##
## Shapiro-Wilk normality test
##
## data: gam_benthic$residuals
## W = 0.86414, p-value = 0.01427
# Summary of the Drift biomass model
summary(gam_drift)
##
## Family: gaussian
## Link function: identity
##
## Formula:
## Drift ~ s(Mean_Slope_Degrees, k = 5) + s(Dist_Pour, k = 5) +
## s(Dist_Head, k = 5) + s(Relative_position, k = 5) + s(Percent_occupied,
## k = 5) + PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag +
## Predator_Control_Flag
##
## Parametric coefficients:
## Estimate Std. Error t value
## (Intercept) 1.1845 0.3024 3.917
## PR_FlagHeterotrophic (Below 0.75) 0.2107 0.2014 1.046
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) -0.5431 0.2462 -2.206
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.5050 0.2123 2.379
## Channel_Stability_FlagUnstable Channel (Below 0.50) -0.3679 0.2553 -1.441
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.3116 0.2297 1.357
## Pr(>|t|)
## (Intercept) 0.00493 **
## PR_FlagHeterotrophic (Below 0.75) 0.32768
## CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25) 0.06028 .
## FPOM_CPOM_FlagNormal FPOM (Below 0.50) 0.04629 *
## Channel_Stability_FlagUnstable Channel (Below 0.50) 0.18954
## Predator_Control_FlagPrey Dominated (Below 0.50) 0.21379
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Approximate significance of smooth terms:
## edf Ref.df F p-value
## s(Mean_Slope_Degrees) 1.000 1.00 2.667 0.1411
## s(Dist_Pour) 1.000 1.00 0.019 0.8934
## s(Dist_Head) 1.419 1.71 0.990 0.4005
## s(Relative_position) 1.000 1.00 0.681 0.4331
## s(Percent_occupied) 1.000 1.00 3.920 0.0831 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Rank: 25/26
## R-sq.(adj) = 0.258 Deviance explained = 66.9%
## GCV = 0.17055 Scale est. = 0.07183 n = 18
gam.check(gam_drift)
##
## Method: GCV Optimizer: magic
## Smoothing parameter selection converged after 14 iterations.
## The RMS GCV score gradient at convergence was 0.000000037265 .
## The Hessian was positive definite.
## Model rank = 25 / 26
##
## Basis dimension (k) checking results. Low p-value (k-index<1) may
## indicate that k is too low, especially if edf is close to k'.
##
## k' edf k-index p-value
## s(Mean_Slope_Degrees) 4.00 1.00 1.17 0.650
## s(Dist_Pour) 4.00 1.00 0.74 0.095 .
## s(Dist_Head) 4.00 1.42 1.00 0.430
## s(Relative_position) 4.00 1.00 1.03 0.460
## s(Percent_occupied) 4.00 1.00 0.84 0.200
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
shapiro.test(gam_drift$residuals)
##
## Shapiro-Wilk normality test
##
## data: gam_drift$residuals
## W = 0.93783, p-value = 0.2659
condition_biomass_clean <- na.omit(condition_biomass)
response <- condition_biomass_clean[, c("Fish", "Periphyton", "Benthic", "Drift")]
explanatory <- condition_biomass_clean[, c(1:5)]
exp_binary <- model.matrix(~ PR_Flag + CPOM_FPOM_Flag + FPOM_CPOM_Flag + Channel_Stability_Flag + Predator_Control_Flag - 1, data = explanatory)
exp_binary <- as.data.frame(exp_binary)
response <- as.data.frame(response)
condition_rda <- rda(response ~ ., data = exp_binary, grouping = rda_env$Basin)
##
## Some constraints or conditions were aliased because they were redundant. This
## can happen if terms are linearly dependent (collinear):
## '`PR_FlagHeterotrophic (Below 0.75)`',
## '`Predator_Control_FlagPrey Dominated (Below 0.50)`'
anova(condition_rda, by = "terms", permutations = 999)
## Permutation test for rda under reduced model
## Terms added sequentially (first to last)
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = response ~ `PR_FlagAutotrophic (Above 0.75)` + `PR_FlagHeterotrophic (Below 0.75)` + `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` + `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` + `Channel_Stability_FlagUnstable Channel (Below 0.50)` + `Predator_Control_FlagPrey Dominated (Below 0.50)`, data = exp_binary, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## `PR_FlagAutotrophic (Above 0.75)` 1 0.015221 0.7654 0.538
## `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` 1 0.043321 2.1784 0.102
## `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` 1 0.008884 0.4467 0.606
## `Channel_Stability_FlagUnstable Channel (Below 0.50)` 1 0.031373 1.5776 0.188
## Residual 13 0.258530
summary(condition_rda)
##
## Call:
## rda(formula = response ~ `PR_FlagAutotrophic (Above 0.75)` + `PR_FlagHeterotrophic (Below 0.75)` + `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` + `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` + `Channel_Stability_FlagUnstable Channel (Below 0.50)` + `Predator_Control_FlagPrey Dominated (Below 0.50)`, data = exp_binary, grouping = rda_env$Basin)
##
## Partitioning of variance:
## Inertia Proportion
## Total 0.3573 1.0000
## Constrained 0.0988 0.2765
## Unconstrained 0.2585 0.7235
##
## Eigenvalues, and their contribution to the variance
##
## Importance of components:
## RDA1 RDA2 RDA3 RDA4 PC1 PC2 PC3
## Eigenvalue 0.08305 0.01240 0.002063 0.001282 0.1301 0.09262 0.01920
## Proportion Explained 0.23243 0.03471 0.005775 0.003587 0.3641 0.25920 0.05374
## Cumulative Proportion 0.23243 0.26713 0.272906 0.276493 0.6406 0.89981 0.95355
## PC4
## Eigenvalue 0.01660
## Proportion Explained 0.04645
## Cumulative Proportion 1.00000
##
## Accumulated constrained eigenvalues
## Importance of components:
## RDA1 RDA2 RDA3 RDA4
## Eigenvalue 0.08305 0.0124 0.002063 0.001282
## Proportion Explained 0.84062 0.1255 0.020885 0.012975
## Cumulative Proportion 0.84062 0.9661 0.987025 1.000000
# Get the summary of the RDA results
rda_summary2 <- summary(condition_rda)
# Access the scores for each variable (explained by RDA1 and RDA2)
explan_condition <- rda_summary2$species
biplot_scores2 <- scores(condition_rda, display = "bp")
# View the RDA1 and RDA2 scores for each variable
print(explan_condition)
## RDA1 RDA2 RDA3 RDA4 PC1
## Fish 0.5433552 -0.05269302 -0.05930159 0.042582055 -0.2597892
## Periphyton 0.3181973 -0.15975378 0.08567874 -0.008977313 -0.2627197
## Benthic -0.3188273 -0.04835442 0.02805880 0.080882754 -0.5747066
## Drift -0.2733430 -0.23431198 -0.05087016 -0.020146884 -0.6562257
## PC2
## Fish 0.3799758
## Periphyton 0.6335570
## Benthic -0.2197752
## Drift -0.2115968
condition_rda$CCA$biplot
## RDA1 RDA2
## `PR_FlagAutotrophic (Above 0.75)` 0.3878409 0.3527564
## `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` 0.7681939 0.3223728
## `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` -0.2895138 -0.3744074
## `Channel_Stability_FlagUnstable Channel (Below 0.50)` -0.8624438 0.2370271
## RDA3 RDA4
## `PR_FlagAutotrophic (Above 0.75)` -0.5719269 -0.63090567
## `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` 0.5367472 -0.13362747
## `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` 0.2707745 -0.83826134
## `Channel_Stability_FlagUnstable Channel (Below 0.50)` 0.4448151 0.04635023
condition_rda$CCA$v
## RDA1 RDA2 RDA3 RDA4
## Fish 0.7178980 -0.1801662 -0.4970778 0.45285351
## Periphyton 0.4204123 -0.5462248 0.7181762 -0.09547232
## Benthic -0.4212447 -0.1653318 0.2351945 0.86017546
## Drift -0.3611494 -0.8011517 -0.4264038 -0.21425897
condition_rda$CCA$u
## RDA1 RDA2 RDA3 RDA4
## 1 -0.5121293 -0.2149152 -0.3578315 0.08908498
## 2 0.2160189 -0.3722182 0.0820836 0.04996143
## 3 0.2160189 -0.3722182 0.0820836 0.04996143
## 4 0.1710218 0.1555510 -0.2521961 -0.27820325
## 5 0.1710218 0.1555510 -0.2521961 -0.27820325
## 6 -0.1150373 0.1332344 0.2370369 -0.01140661
## 7 0.2813568 0.3638586 -0.2631455 0.81464359
## 8 0.1710218 0.1555510 -0.2521961 -0.27820325
## 10 -0.1150373 0.1332344 0.2370369 -0.01140661
## 11 -0.1150373 0.1332344 0.2370369 -0.01140661
## 12 0.1710218 0.1555510 -0.2521961 -0.27820325
## 14 0.2160189 -0.3722182 0.0820836 0.04996143
## 15 -0.1150373 0.1332344 0.2370369 -0.01140661
## 16 -0.5121293 -0.2149152 -0.3578315 0.08908498
## 17 -0.1150373 0.1332344 0.2370369 -0.01140661
## 18 0.2160189 -0.3722182 0.0820836 0.04996143
## 19 -0.1150373 0.1332344 0.2370369 -0.01140661
## 20 -0.1150373 0.1332344 0.2370369 -0.01140661
#reduce model further/explore if reduction is possible
sqrt(vif.cca(condition_rda))
## `PR_FlagAutotrophic (Above 0.75)`
## 1.247219
## `PR_FlagHeterotrophic (Below 0.75)`
## NA
## `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)`
## 1.069045
## `FPOM_CPOM_FlagNormal FPOM (Below 0.50)`
## 1.086534
## `Channel_Stability_FlagUnstable Channel (Below 0.50)`
## 1.329608
## `Predator_Control_FlagPrey Dominated (Below 0.50)`
## NA
anova.cca(condition_rda)#global RDA significance
## Permutation test for rda under reduced model
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = response ~ `PR_FlagAutotrophic (Above 0.75)` + `PR_FlagHeterotrophic (Below 0.75)` + `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` + `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` + `Channel_Stability_FlagUnstable Channel (Below 0.50)` + `Predator_Control_FlagPrey Dominated (Below 0.50)`, data = exp_binary, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## Model 4 0.098799 1.242 0.275
## Residual 13 0.258530
anova.cca(condition_rda, by = "axis") # significance of axes
## Permutation test for rda under reduced model
## Forward tests for axes
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = response ~ `PR_FlagAutotrophic (Above 0.75)` + `PR_FlagHeterotrophic (Below 0.75)` + `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` + `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` + `Channel_Stability_FlagUnstable Channel (Below 0.50)` + `Predator_Control_FlagPrey Dominated (Below 0.50)`, data = exp_binary, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## RDA1 1 0.083052 4.1762 0.189
## RDA2 1 0.012401 0.6236 0.971
## RDA3 1 0.002063 0.1038 1.000
## RDA4 1 0.001282 0.0645
## Residual 13 0.258530
anova.cca(condition_rda)
## Permutation test for rda under reduced model
## Permutation: free
## Number of permutations: 999
##
## Model: rda(formula = response ~ `PR_FlagAutotrophic (Above 0.75)` + `PR_FlagHeterotrophic (Below 0.75)` + `CPOM_FPOM_FlagLow Shredder Dependency (Below 0.25)` + `FPOM_CPOM_FlagNormal FPOM (Below 0.50)` + `Channel_Stability_FlagUnstable Channel (Below 0.50)` + `Predator_Control_FlagPrey Dominated (Below 0.50)`, data = exp_binary, grouping = rda_env$Basin)
## Df Variance F Pr(>F)
## Model 4 0.098799 1.242 0.292
## Residual 13 0.258530
#Define basin colors (you can adjust this list based on your actual basin names)
site_colors <- c(
"AA1_REACH" = "darkorange3", "AA2_REACH" = "darkorange4",
"AC_REACH" = "slateblue3", "AP_REACH" = "darkcyan",
"AZ_REACH" = "gold3", "BA1_REACH" = "firebrick3",
"BA2_REACH" = "green4", "BC_REACH" = "purple",
"BZ_REACH" = "darkgreen", "CA1_REACH" = "darkolivegreen4",
"CA2_REACH" = "mediumblue", "CP_REACH" = "yellowgreen",
"CZ2_REACH" = "darkred", "DA1_REACH" = "seagreen",
"DA2_REACH" = "indianred", "DC_REACH" = "orangered",
"DP_REACH" = "chocolate", "DZ_REACH" = "violet"
)
# Vector of colors for each site based on bAsin
colors <- sapply(env_scaled$Basin, function(Basin) site_colors[Basin])
# Extract site coordinates (coordinates of site scores)
site_scores <- condition_rda$CCA$u
par(mar = c(5, 5, 2, 7))
# Plot RDA result, scaling 2 (displaying effects of explanatory variables)
plot(condition_rda, xlim = c(-1, 1), ylim = c(-1,1), scaling = 2, main = "RDA - Scaling = 2", type = "n")
# Plot species labels
text(condition_rda, display = "species", col = "firebrick3", cex = 1)
# Plot sites with colors corresponding to basins
points(site_scores, col = colors, pch = 20, cex = 1.)
# Plot biplot arrows (explanatory variables)
text(condition_rda, scaling = 2, display = "bp", col = "darkblue", cex = 1)
# Add the legend to the plot, positioned outside
legend("right", legend = names(site_colors), col = site_colors, pch = 19, title = "Basin",
inset = -0.4, xpd = TRUE, cex = 0.7)
condition_biomass1 <- read.csv("reach_condition.csv")
condition_biomass1 <- condition_biomass1[-c(9,13),-c(8,9)]
# Select numerical variables
numerical_data <- condition_biomass1 %>%
select(Mean_Slope_Degrees, Dist_Pour, Dist_Head, Relative_position, Percent_occupied)
# Convert categorical variables into binary format (dummy variables)
categorical_data <- condition_biomass1 %>%
select(PR_Flag, CPOM_FPOM_Flag, FPOM_CPOM_Flag, Channel_Stability_Flag, Predator_Control_Flag) %>%
mutate(across(everything(), as.factor)) # Convert all to factor if not already
# Create dummy variables for categorical factors
categorical_binary <- model.matrix(~ . - 1, data = categorical_data)
# Combine numerical and binary categorical data into one dataset
combined_data <- cbind(numerical_data, categorical_binary)
# Remove rows with NA values
combined_data <- drop_na(combined_data)
#shorten names
colnames(combined_data)[6] <- "Auto"
colnames(combined_data)[7] <- "Hetero"
colnames(combined_data)[8] <- "Shred_Dep"
colnames(combined_data)[9] <- "Normal_FPOM"
colnames(combined_data)[10] <- "Stable"
colnames(combined_data)[11] <- "Prey_Dom"
# Compute Gower's distance (excluding Basin)
dist_matrix_condition <- daisy(combined_data, metric = "gower")
## Warning in daisy(combined_data, metric = "gower"): binary variable(s) 6, 7, 8,
## 9, 10, 11 treated as interval scaled
# Perform hierarchical clustering (Ward's method)
condition_clust <- hclust(dist_matrix_condition, method = "ward.D2")
# Plot the dendrogram
plot(condition_clust, main = "Dendrogram of Clustering", cex = 0.7)
final_condition <- cbind(combined_data, Basin = condition_biomass1$Basin)
# Add cluster assignments to the original data (including Basin names)
final_condition <- final_condition %>%
mutate(cluster = cutree(condition_clust, k = 5))
# View the table of cluster assignments
table(final_condition$cluster)
##
## 1 2 3 4 5
## 3 4 4 6 1
# Print the final data with Basin names and cluster assignments
print(final_condition)
## Mean_Slope_Degrees Dist_Pour Dist_Head Relative_position Percent_occupied
## 1 0.5627990 0.09405779 0.30081412 0.8652373 0.30231646
## 2 0.8278380 0.56497136 0.51868785 0.5699650 0.18226414
## 3 0.8374997 0.65261370 0.58848006 0.5408931 0.12180508
## 4 0.3305135 0.99707250 0.48775562 0.2085846 0.06105181
## 5 0.3409368 1.00000000 1.00000000 0.5699650 0.00000000
## 6 0.3123774 0.51169256 0.24743058 0.3322185 0.24243322
## 7 0.9065602 0.43921008 0.93335283 0.8515649 0.06105181
## 8 0.7491686 0.17680785 0.41551067 0.8378557 0.30231646
## 9 0.1185268 0.51506735 0.32368682 0.4227938 0.18226414
## 10 1.0000000 0.00000000 0.07878532 0.8378557 1.00000000
## 11 0.1826329 0.01487185 0.62956884 1.0000000 0.18226414
## 12 0.0000000 0.71330529 0.48569967 0.4078205 0.06105181
## 13 0.4479005 0.77502576 0.45266012 0.3322185 0.06105181
## 14 0.8701152 0.14326609 0.00000000 0.4377192 1.00000000
## 15 0.3757263 0.94002705 0.60314221 0.3474385 0.06105181
## 16 0.3698918 0.45304981 0.59677976 0.7127662 0.12180508
## 17 0.4735465 0.66698235 0.11603491 0.0000000 0.12180508
## 18 0.2227579 0.44129387 0.88232499 0.8378557 0.00000000
## Auto Hetero Shred_Dep Normal_FPOM Stable Prey_Dom Basin cluster
## 1 0 1 0 1 1 1 AA1_REACH 1
## 2 0 1 1 1 0 0 AA2_REACH 2
## 3 0 1 1 1 0 0 AC_REACH 2
## 4 1 0 1 1 0 0 AP_REACH 3
## 5 1 0 1 1 0 0 AZ_REACH 3
## 6 0 1 1 1 1 1 BA1_REACH 4
## 7 0 1 1 0 0 1 BA2_REACH 5
## 8 1 0 1 1 0 0 BC_REACH 3
## 9 0 1 1 1 1 1 BZ_REACH 4
## 10 0 1 1 1 1 1 CA1_REACH 1
## 11 1 0 1 1 0 0 CA2_REACH 3
## 12 0 1 1 1 0 0 CP_REACH 2
## 13 0 1 1 1 1 1 CZ2_REACH 4
## 14 0 1 0 1 1 1 DA1_REACH 1
## 15 0 1 1 1 1 1 DA2_REACH 4
## 16 0 1 1 1 0 0 DC_REACH 2
## 17 0 1 1 1 1 1 DP_REACH 4
## 18 0 1 1 1 1 1 DZ_REACH 4
# Create the clusters using cutree
clusters <- cutree(condition_clust, k = 5)
# Plot the dendrogram and add rectangles around clusters
plot(condition_clust, labels = final_condition$Basin, main = "Clustering Dendrogram by Basin", cex = 0.7)
rect.hclust(condition_clust, k = 5, border = 2:5) # Use the same 'k' value
# Prepare the data for the heatmap (remove non-numeric columns like Basin and cluster)
cluster_matrix_condition <- final_condition %>%
select(-Basin, -cluster) %>%
as.data.frame()
# Convert character columns to factors
cluster_matrix_condition[] <- lapply(cluster_matrix_condition, function(x) {
if (is.character(x)) {
return(factor(x))
} else {
return(x)
}
})
# Calculate the distance matrix for the heatmap
condition_matrix_heatmap <- dist(cluster_matrix_condition, method = "euclidean")
final_condition_heatmap <- hclust(condition_matrix_heatmap, method = "ward.D2")
# Create the heatmap
pheatmap(cluster_matrix_condition,
clustering_method = "ward.D2",
cluster_rows = final_condition_heatmap,
cluster_cols = TRUE,
show_rownames = TRUE,
show_colnames = TRUE,
main = "Heatmap of Clustering Results",
labels_row = final_condition$Basin) # Custom labels based on Basin names
