.Rhistory

Resp.Data$time_Sec  <- period_to_seconds(hms(substr((strptime(sub(".*2021/", "", Resp.Data$Date..Time..DD.MM.YYYY.HH.MM.SS.), "%I:%M:%S %p")) , 12,19))) # time - use 'sub' to call target time of the raw date time after 'year/' + strptime' convert to 24 hr clock + 'period_to_seconds' converts the hms to seconds
Resp.Data$seconds   <- (Resp.Data$time_Sec - Resp.Data$time_Sec[1])    # secs - calc the sec time series
Resp.Data$minutes   <- (Resp.Data$time_Sec - Resp.Data$time_Sec[1])/60 # mins - calc the minute time series
temperature_C       <- as.numeric(Resp.Data$CH1.temp...C.[1])
barromP_kPa         <- as.numeric(Resp.Data$Barometric.pressure..hPa.[1]) / 10
salinity.pp.thou    <- as.numeric(Resp.Data$Salinity....[1])
Resp.Data           <- Resp.Data %>% # use 'dplyr'
dplyr::filter(!Phase %in% 'Flush') %>% # remove the initial rows labeled flush
dplyr::select(c(date, seconds, minutes, contains(".O2...air.sat"))) # call unique column names for the 8 Channels
colnames(Resp.Data)[c(4:(ncol(Resp.Data)))] <- substr( ( colnames(Resp.Data)[c(4:(ncol(Resp.Data)))] ), 1,3) # clean these column names to make things easier - first 3 characters
# Truncate!
# the loligo recoreded values every second, this slows the model dramatically with >2000 values for each Channel, call every 30 seconds to speed this up
# discuss with collaborators on this truncated approach
Resp.Data_30sec = Resp.Data[seq(1, nrow(Resp.Data), 30), ]
# II.B Loop #2 - within each filename, call each Channel (% O2 sat.) to measure respiration rates, output plots and cumulative spreadsheet
for(j in 4:(ncol(Resp.Data_30sec))){
Resp_loop         <- na.omit(Resp.Data_30sec[,c(3,4)]) # noticed some random rows have 'NaN' - so I will loop the min and Channels to ommit Nas before proceeding
Resp_loop$mgL     <- DO.unit.convert(as.numeric(Resp_loop[,2]),  # DO in percent air sat to be converted to mgL
DO.units.in = "pct", DO.units.out ="mg/L",
bar.units.in = "kPa", bar.press = barromP_kPa, bar.units.out = "kpa",
temp.C = temperature_C,
salinity.units = "pp.thou", salinity = salinity.pp.thou)
if (nrow(Resp_loop) < 1) {
resp.table$Date                <- Resp.Data_30sec[1,1]
resp.table$Channel             <- colnames(Resp_loop)[2]
resp.table[3:ncol(resp.table)] <- 'NA'
df       <- data.frame(resp.table) # name dataframe for this single row
df_total <- rbind(df_total,df) #bind to a cumulative list dataframe
print(df_total) # print to monitor progress
} else {
model <- rankLocReg(
xall    = as.numeric(Resp_loop[, 1]),
yall    = as.numeric(Resp_loop[, 3]), # call x as the minute timeseries and y as the % air saturation of the particular Channel
alpha   = a,  # alpha was assigned earlier as 0.4 by the authors default suggestions - review Olito et al. and their github page for details
method  = "pc",
verbose = TRUE)
sum.table <- summary(model)
resp.table$Date       <- Resp.Data_30sec[1,1]
resp.table$Channel    <- colnames(Resp_loop)[2]
resp.table$Lpc        <- sum.table$Lcompare[3,6] # Lpc slope - call the column 'b1'
resp.table$Leq        <- sum.table$Lcompare[2,6] # Leq slope - call the column 'b1'
resp.table$Lz         <- sum.table$Lcompare[1,6] # Lz slope  - call the column 'b1'
resp.table$alpha      <- a
resp.table$Notes      <- file.names.table[i,2]
resp.table$Filename   <- file.names.table[i,1]
df       <- data.frame(resp.table) # name dataframe for this single row
df_total <- rbind(df_total,df) #bind to a cumulative list dataframe
print(df_total) # print to monitor progress
# save plots every inside loop and name by date_run_vialposition
pdf(paste0("C:/Users/samjg/Documents/Github_repositories/Cvriginica_multistressor/Output/Respiration/plots_alpha0.4_increm30sec/",file.names.table[i,2],"_",colnames(Resp_loop)[2],"_regression.pdf"))
plot(model)
dev.off()
} # end of if else statement
} # end of inside for loop
} # end of outside for loop
file.names.table[i,1]
Resp.Data_30sec
Resp_loop
# bring in the respiration file names
file.names.full        <- basename(list.files(path = path.p, pattern = "txt$", recursive = TRUE)) #list all csv file names in the folder and subfolders
file.names             <- file.names.full[c(3:4,6:7,10:11, 13,15, 19:20)] # call all target files (excluding tests and empty files) - open in notepad++ to make sure you have the correct files
file.names # check
file.names.table       <- data.frame(file.names)
file.names.table$about <- c('20210319_new_sensor_7', '20210319', '20210430_LOWtemp_HIGHsal',
'20210430_LOWtemp_LOWsal', '20210430_raw', '20210507_HIGHtemp_LOWsal',
'20210507_LOWtemp_LOWsal', '20210507_HIGHtemp_HIGHsal', '20210507_LOWtemp_HIGHsal_b', '20210507_LOWtemp_HIGHsal')
# ANALYSIS  :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
# Objective: use LoLinr to run all respiration rates in a non-bias and autonomous fashion
# Ouputs: there will be two main resources from this script:
#   (1) cumulative spreasheet of all respiration rate value for each Channel on each day
#   (2) a folder of plots from the LoLinR script to a plots folder - this will allow troubleshooting and sanity checks
# I. Call the cumulative dataframe that we will write to in the for loop below
df_total             <- data.frame() # start dataframe
resp.table           <- data.frame(matrix(nrow = 1, ncol = 8)) # create dataframe to save cumunalitively during for loop
colnames(resp.table) <- c('Date', 'Channel', 'Lpc', 'Leq' , 'Lz', 'alpha', 'Notes', 'Filename') # names for comuns in the for loop
# II.A Loop #1 - call filenames one at a time for analysis
for(i in 1:nrow(file.names.table)) { # for every file in list start at the first and run this following function
Resp.Data           <- read.delim2(file = paste(path.p,'/',file.names.table[i,1], sep=''), header = TRUE,skip = 37) #reads in the data files
Resp.Data$date      <- paste((sub("2021.*", "", Resp.Data$Date..Time..DD.MM.YYYY.HH.MM.SS.)), '2021', sep='') #  date - use 'sub' to call everything before 2021, add back 2021 using paste
Resp.Data$time_Sec  <- period_to_seconds(hms(substr((strptime(sub(".*2021/", "", Resp.Data$Date..Time..DD.MM.YYYY.HH.MM.SS.), "%I:%M:%S %p")) , 12,19))) # time - use 'sub' to call target time of the raw date time after 'year/' + strptime' convert to 24 hr clock + 'period_to_seconds' converts the hms to seconds
Resp.Data$seconds   <- (Resp.Data$time_Sec - Resp.Data$time_Sec[1])    # secs - calc the sec time series
Resp.Data$minutes   <- (Resp.Data$time_Sec - Resp.Data$time_Sec[1])/60 # mins - calc the minute time series
temperature_C       <- as.numeric(Resp.Data$CH1.temp...C.[1])
barromP_kPa         <- as.numeric(Resp.Data$Barometric.pressure..hPa.[1]) / 10
salinity.pp.thou    <- as.numeric(Resp.Data$Salinity....[1])
Resp.Data           <- Resp.Data %>% # use 'dplyr'
dplyr::filter(!Phase %in% 'Flush') %>% # remove the initial rows labeled flush
dplyr::select(c(date, seconds, minutes, contains(".O2...air.sat"))) # call unique column names for the 8 Channels
colnames(Resp.Data)[c(4:(ncol(Resp.Data)))] <- substr( ( colnames(Resp.Data)[c(4:(ncol(Resp.Data)))] ), 1,3) # clean these column names to make things easier - first 3 characters
# Truncate!
# the loligo recoreded values every second, this slows the model dramatically with >2000 values for each Channel, call every 30 seconds to speed this up
# discuss with collaborators on this truncated approach
Resp.Data_30sec = Resp.Data[seq(1, nrow(Resp.Data), 30), ]
# II.B Loop #2 - within each filename, call each Channel (% O2 sat.) to measure respiration rates, output plots and cumulative spreadsheet
for(j in 4:(ncol(Resp.Data_30sec))){
Resp_loop         <- na.omit(Resp.Data_30sec[,c(3,j)]) # noticed some random rows have 'NaN' - so I will loop the min and Channels to ommit Nas before proceeding
Resp_loop$mgL     <- DO.unit.convert(as.numeric(Resp_loop[,2]),  # DO in percent air sat to be converted to mgL
DO.units.in = "pct", DO.units.out ="mg/L",
bar.units.in = "kPa", bar.press = barromP_kPa, bar.units.out = "kpa",
temp.C = temperature_C,
salinity.units = "pp.thou", salinity = salinity.pp.thou)
if (nrow(Resp_loop) < 1) {
resp.table$Date                <- Resp.Data_30sec[1,1]
resp.table$Channel             <- colnames(Resp_loop)[2]
resp.table[3:ncol(resp.table)] <- 'NA'
df       <- data.frame(resp.table) # name dataframe for this single row
df_total <- rbind(df_total,df) #bind to a cumulative list dataframe
print(df_total) # print to monitor progress
} else {
model <- rankLocReg(
xall    = as.numeric(Resp_loop[, 1]),
yall    = as.numeric(Resp_loop[, 3]), # call x as the minute timeseries and y as the % air saturation of the particular Channel
alpha   = a,  # alpha was assigned earlier as 0.4 by the authors default suggestions - review Olito et al. and their github page for details
method  = "pc",
verbose = TRUE)
sum.table <- summary(model)
resp.table$Date       <- Resp.Data_30sec[1,1]
resp.table$Channel    <- colnames(Resp_loop)[2]
resp.table$Lpc        <- sum.table$Lcompare[3,6] # Lpc slope - call the column 'b1'
resp.table$Leq        <- sum.table$Lcompare[2,6] # Leq slope - call the column 'b1'
resp.table$Lz         <- sum.table$Lcompare[1,6] # Lz slope  - call the column 'b1'
resp.table$alpha      <- a
resp.table$Notes      <- file.names.table[i,2]
resp.table$Filename   <- file.names.table[i,1]
df       <- data.frame(resp.table) # name dataframe for this single row
df_total <- rbind(df_total,df) #bind to a cumulative list dataframe
print(df_total) # print to monitor progress
# save plots every inside loop and name by date_run_vialposition
pdf(paste0("C:/Users/samjg/Documents/Github_repositories/Cvriginica_multistressor/Output/Respiration/plots_alpha0.4_increm30sec/",file.names.table[i,2],"_",colnames(Resp_loop)[2],"_regression.pdf"))
plot(model)
dev.off()
} # end of if else statement
} # end of inside for loop
} # end of outside for loop
df_total
df_total
file.names.table
# merge with the preexisiting table?
cumulative_resp_table <- read.csv(file=ouputNAME, header=T) #call the pre existing cumulative table
new_table             <- rbind(cumulative_resp_table, df_total) # bind the new table from the for loop to the pre exisiting table
write.table(new_table,ouputNAME,sep=",", row.names=FALSE)  # write out to the path names outputNAME
library(dplyr)
library(ggplot2)
# SET WORKING DIRECTORY :::::::::::::::::::::::::::::::::::::::::::::::
setwd("C:/Users/samjg/Documents/Github_repositories/Cvriginica_multistressor")
# LOAD DATA :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
exp_metadata <- read.csv(file="Data/ExperimentMetadata.csv", header=T) # treatment assignments to 'Chamber_Tank'
resp.ref     <- read.csv(file="Data/Respiration/Reference_master.csv", header=T) # reference for the respirometry data - contains the 'Chamber_Tank' for each sensor channel (whether an animal or a blank)
resp.data    <- read.csv(file="Output/Respiration/Cumulative_resp_alpha0.4.csv", header=T) # read the calculate raw rates from 'resp_LoLin' script - contains the calculated rate (not normalized for blanks) for each sensor-channel
# merge the exp_metadata with the resp.data
resp.ref_merged                 <- merge(exp_metadata, resp.ref, by = 'Chamber_tank', all=TRUE) # all TRUE allows us to keep the blanks
resp.data_merged                <- merge(resp.data, resp.ref_merged, by = c('Date', 'Channel','Notes','Filename')) # out master file moving forward....
resp.data_merged$TempCarbSal    <- paste(resp.data_merged$Temp, resp.data_merged$pCO2, resp.data_merged$Salinity, sep ='')
dates.runs <- resp.data_merged %>%  # call table
dplyr::distinct(Date, Notes) # call all unique values for date run and sw condition
dates.runs <- na.omit(dates.runs)
# call dataframe and build table to rbind in for loop
blanks_total <- data.frame() # start dataframe
blanks.table <- data.frame(matrix(nrow = 1,ncol = 5)) # make a table template
colnames(blanks.table)<-c('Date', 'Notes', 'BLANK.mean.Lpc', 'BLANK.mean.Leq' , 'BLANK.mean.Lz') # names for comuns in the for loop
# for loop. objective = obtian a mean value for all blanks specific to date, run #, seawater treatment
for(i in 1:nrow(dates.runs)) {
data <- resp.data_merged %>%
dplyr::select(Date, Chamber_tank, Notes, Lpc,  Leq, Lz) %>%
dplyr::filter(!is.na(Lpc)) %>% # ommits empty resp channels (if any)
dplyr::filter(Notes == dates.runs[i,2])
blanks <- data %>%
dplyr::filter(Chamber_tank == "blank") %>%
dplyr::summarise(mean_Lpc = mean(abs(Lpc)),
mean_Leq = mean(abs(Leq)),
mean_Lz = mean(abs(Lz)))
blanks.table$Date           <- dates.runs[i,1] # all files have date in the form of yyyymmdd at the start of each csv name
#blanks.table$RUN <- dates.runs[i,2] # assign the run to the number in the title for the trials completed that day
blanks.table$Notes      <- dates.runs[i,2]
blanks.table$BLANK.mean.Lpc <- blanks[1,1]
blanks.table$BLANK.mean.Leq <- blanks[1,2]
blanks.table$BLANK.mean.Lz  <- blanks[1,3]
# blanks.table$alpha <- data[1,9] # set at start of script - reresents the proportion of data for final estimate of slopes (Lpc, Leq, Lz)
df <- data.frame(blanks.table) # name dataframe for this singl e row
blanks_total <- rbind(blanks_total,df) #bind to a cumulative list dataframe
print(blanks_total) # print to monitor progress
}
blanks_total # view blanks table
Resp.Master <- merge(resp.data_merged, blanks_total, by=c("Date", "Notes")) %>% # NOTE: this repeats for every distinct length value
dplyr::filter(!Chamber_tank =='blank') %>%
dplyr::mutate(resp_norm = Lpc - BLANK.mean.Lpc) # ommits respiration rate values showing an increase in O2 over time
# NOTE: look at the following table to troubleshoot if needed
Resp.outliers <- Resp.Master %>% dplyr::filter(resp_norm > 0)
# calculate resp rates
vial.vol <- 2 # milliliters (ml)
Resp.Master$resp_norm
Resp.Master_OM <- Resp.Master %>% dplyr::filter(!resp_norm > 0)
Resp.Master_OM
Resp.Master_OM$resp_percO2sat_L_hr <- (abs(Resp.Master_OM$resp_norm)*(vial.vol/1000)*(60)) # correct for Liters and hours (currently in minutes fro the resp.lolin.R script)
Resp.Master_OM$resp_percO2sat_L_hr
Resp.Master_OM$resp_ug_L_hr <- ( (abs(Resp.Master_OM$resp_norm)/1000)*(vial.vol/1000)*(60)) # correct for Liters and hours (currently in minutes fro the resp.lolin.R script)
Resp.Master_OM$resp_ug_L_hr
Resp.Master_OM$resp_ug_L_hr <- ( (abs(Resp.Master_OM$resp_norm)*1000)*(vial.vol/1000)*(60)) # correct for Liters and hours (currently in minutes fro the resp.lolin.R script)
Resp.Master_OM$resp_ug_L_hr
(abs(Resp.Master_OM$resp_norm)*1000)*(vial.vol/1000)
(vial.vol/1000)
(abs(Resp.Master_OM$resp_norm)*1000)
Resp.Master_OM$resp_norm
(abs(Resp.Master_OM$resp_norm)*1000)*(vial.vol/1000)
Resp.Master_OM$resp_ug_L_hr <- ( (abs(Resp.Master_OM$resp_norm)*1000)*(vial.vol/1000)*(60)) # correct for Liters and hours (currently in minutes fro the resp.lolin.R script)
Resp.Master_OM$resp_ug_L_hr
# mean sd rates
mean(Resp.Master_OM$resp_ug_L_hr)
sd(Resp.Master_OM$resp_ug_L_hr)
Resp.Master_OM
Reap_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% 5/7/2021)
Reap_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% 5/07/2021)
Reap_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% '5/07/2021')
Resp_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% '5/07/2021')
Resp_0507
Resp.Master_OM
Resp_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% '5/7/2021')
Resp_0507
# model effect of treatment on resp rate 20210507
aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
# model effect of treatment on resp rate 20210507
mode <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
# model effect of treatment on resp rate 20210507
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
summary(mod)
ggboxplot(Resp_0507, x = "Salinity", y = "resp_ug_L_hr",  ylab = "µg.L.hr.indiv",  fill = "Treat",
palette = c("#FC4E07", "#00AFBB"),add = "jitter", title = "Pre-conditioning", xlab = "Initial pCO2 Treatment")
library(ggpubr)
ggboxplot(Resp_0507, x = "Salinity", y = "resp_ug_L_hr",  ylab = "µg.L.hr.indiv",  fill = "Treat",
palette = c("#FC4E07", "#00AFBB"),add = "jitter", title = "Pre-conditioning", xlab = "Initial pCO2 Treatment")
ggboxplot(Resp_0507, x = "Salinity", y = "resp_ug_L_hr",  ylab = "µg.L.hr.indiv",  fill = "Salinity",
palette = c("#FC4E07", "#00AFBB"),add = "jitter", title = "Pre-conditioning", xlab = "Initial pCO2 Treatment")
Resp_0507
Resp_0507_FIGS <- Resp_0507 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0507_melt <- melt(Resp_0507_FIGS, id=c("resp_ug_L_hr"))
install.packages('reshape2')
library('reshape2')
Resp_0507_melt <- melt(Resp_0507_FIGS, id=c("resp_ug_L_hr"))
Resp_0507_melt
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, color = value)) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable)
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, color = value)) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free")
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, color = value)) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, color = value)) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw()
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, color = value)) +
theme(panel.grid=element_blank())
Resp_0507_melt
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, fill = value)) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
ggplot(Resp_0507_melt, aes(x=value, y=resp_ug_L_hr, group = variable,fill = value)) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
ggplot(Resp_0507_FIGS, aes(value, resp_ug_L_hr , fill = factor(value))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0507_FIGS
ggplot(Resp_0507_melt, aes(value, resp_ug_L_hr , fill = factor(value))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0507_melt
Resp_0507_FIGS
ggplot(Resp_0507_FIGS, aes(Temp , resp_ug_L_hr , fill = factor(Temp )))
ggplot(Resp_0507_FIGS, aes(Temp , resp_ug_L_hr , fill = factor(Temp ))) +
theme(panel.grid=element_blank())
ggplot(Resp_0507_FIGS, aes(Temp , resp_ug_L_hr , fill = factor(Temp ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_line(size=0.2, alpha=0.1)
ggplot(Resp_0507_FIGS, aes(Temp , resp_ug_L_hr , fill = factor(Temp ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1)
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
# model effect of treatment on resp rate 20210507
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
summary(mod)
Resp_0507_FIGS
Resp_0507_FIGS_2 <- Resp_0507_FIGS %>% dplyr::mutate(fll.treatment <- paste(Temp, pCO2, Salinity,sep=''))
Resp_0507_FIGS_2
Resp_0507_FIGS_2 <- Resp_0507_FIGS %>% dplyr::mutate(fll.treatment <- (paste(Temp, pCO2, Salinity,sep='')))
Resp_0507_FIGS_2
Resp_0507_FIGS_2 <- Resp_0507_FIGS %>% dplyr::mutate(fll.treatment = (paste(Temp, pCO2, Salinity,sep='')))
Resp_0507_FIGS_2
Resp_0507_FIGS_2 <- Resp_0507_FIGS %>% dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep='')))
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
# scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
ggplot(Resp_0507_FIGS_2, aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Resp_0507
Resp_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% '5/7/2021') %>% dplyr::filter(!resp_ug_L_hr > 20)
# model effect of treatment on resp rate 20210507
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
summary(mod)
Resp_0507_FIGS <- Resp_0507 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0507_melt <- melt(Resp_0507_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0507_FIGS_2 <- Resp_0507_FIGS %>% dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep='')))
ggplot(Resp_0507_FIGS_2, aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Plot <- Resp_0507_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHH", "HLH", "LHH",'LLH')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
library('forcats')
Plot <- Resp_0507_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHH", "HLH", "LHH",'LLH')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
full.treatment
Plot <- Resp_0507_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHL", "HLL", "LHL",'LLL')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Plot <- Resp_0507_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHL", "LHL",'LLL')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Plot
Plot <- Resp_0507_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHL", "LHL",'LLL')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Plot
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
# calculate resp rates
vial.vol <- 0.08 # milliliters (ml)
Resp.Master_OM$resp_ug_L_hr <- ( (abs(Resp.Master_OM$resp_norm)*1000)*(vial.vol/1000)*(60)) # correct for Liters and hours (currently in minutes fro the resp.lolin.R script)
# mean sd rates
mean(Resp.Master_OM$resp_ug_L_hr)
sd(Resp.Master_OM$resp_ug_L_hr)
Resp_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% '5/7/2021') %>% dplyr::filter(!resp_ug_L_hr > 20)
# model effect of treatment on resp rate 20210507
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
summary(mod)
Resp_0507_FIGS <- Resp_0507 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0507_melt <- melt(Resp_0507_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0507 <- Resp.Master_OM %>% dplyr::filter(Date %in% '5/7/2021') %>% dplyr::filter(!resp_ug_L_hr >1)
# model effect of treatment on resp rate 20210507
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0507))
summary(mod)
Resp_0507_FIGS <- Resp_0507 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0507_melt <- melt(Resp_0507_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0507_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0430 <- Resp.Master_OM %>% dplyr::filter(Date %in% '4/30/2021')
Resp_0430
# model effect of treatment on resp rate 20210507
Resp_0430 <- Resp.Master_OM %>% dplyr::filter(Date %in% '4/30/2021') %>% dplyr::filter(!resp_ug_L_hr >1)
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0430))
summary(mod)
Resp_0430_FIGS <- Resp_0507 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0430_melt <- melt(Resp_0430_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0430_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0430_FIGS <- Resp_0430 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0430_melt <- melt(Resp_0430_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0430_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Resp_0430 <- Resp.Master_OM %>% dplyr::filter(Date %in% '4/30/2021')
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0430))
summary(mod)
Resp_0430_melt <- melt(Resp_0430_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0430_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
# model effect of treatment on resp rate 20210507
Resp_0430 <- Resp.Master_OM %>% dplyr::filter(Date %in% '4/30/2021') %>% dplyr::filter(!resp_ug_L_hr >1)
mod <- aov(lm(resp_ug_L_hr~Temp*pCO2*Salinity,data=Resp_0430))
summary(mod)
Resp_0430_FIGS <- Resp_0430 %>% dplyr::select(c('resp_ug_L_hr', 'Temp', 'pCO2', 'Salinity'))
Resp_0430_melt <- melt(Resp_0430_FIGS, id=c("resp_ug_L_hr"))
ggplot(Resp_0430_melt, aes(value , resp_ug_L_hr , fill = factor(value ))) +
theme(panel.grid=element_blank()) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw() +
facet_wrap(~variable, scales = "free_y")
Plot <- Resp_0430_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHL", "LHL",'LLL')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Plot
Plot <- Resp_0430_FIGS %>%
dplyr::mutate(full.treatment = (paste(Temp, pCO2, Salinity,sep=''))) %>%
dplyr::mutate(full.treatment = fct_relevel(full.treatment,
"HHH", "HLH", "LHH",'LLH',
"HHL", "HLL", "LHL", 'LLL')) %>%
ggplot(aes(full.treatment, resp_ug_L_hr , fill = factor(full.treatment))) +
scale_color_manual(values=c("#56B4E9","#D55E00")) +
geom_point(shape = 21, size = 2, position = position_jitterdodge(jitter.width = 0.5))+
geom_boxplot(size=0.2, alpha=0.1) +
theme_bw()
Plot