Here are the solutions to the Challenges in the main tutorial
Plotting Ocean Data With R
We first need to load the tidyverse and akima libraries:
library(tidyverse)## -- Attaching packages ------------------------------------------------------------------------------------------------------------------------- tidyverse 1.3.0 --
## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.3 v dplyr 1.0.2
## v tidyr 1.1.2 v stringr 1.4.0
## v readr 1.3.1 v forcats 0.5.0
## -- Conflicts ---------------------------------------------------------------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
library(akima)## Warning: package 'akima' was built under R version 4.0.3
and the data:
fieldData <- read_csv('data/DamariscottaRiverData.csv')## Parsed with column specification:
## cols(
## date = col_double(),
## station = col_double(),
## depth_m = col_double(),
## year = col_double(),
## month = col_double(),
## day = col_double(),
## temperature_degC = col_double(),
## salinity_psu = col_double(),
## density_kg_m3 = col_double(),
## PAR = col_double(),
## fluorescence_mg_m3 = col_double(),
## oxygenConc_umol_kg = col_double(),
## oxygenSaturation_percent = col_double(),
## latitude = col_double()
## )
1.a. Create a scatter plot of ‘temperature_degC’ by ‘fluorescence_mg_m3’ and color the points by ‘station’.
ggplot(data = fieldData, mapping = aes(x = temperature_degC, y = fluorescence_mg_m3)) +
geom_point(aes(color=station)) 
1.b. Do the same as 1.a. but convert the station values to factors. What’s the difference between the two plots?
ggplot(data = fieldData, mapping = aes(x = temperature_degC, y = fluorescence_mg_m3)) +
geom_point(aes(color=factor(station))) 
In the first plot, the colors are on a continuous colorscale. In the second plot, the colors are discrete, separate colors. When the data was read into R, the station values were read in as numeric values, so R plotted them on a continuous color scale. But the station number can really be thought of as discrete data - they didn’t need to be numbers, they could have been letters, or names. The station numbers are separate from each other - not part of a continuous scale. So, to plot them as four separate entities, we need to tell R to treat the station column as factors (as it would have automatically done if the stations were named “A”, “B”, “C”, and “D”).
- Create boxplots looking at the distribution of temperature_degC by station (tip: change the station values to factors)
ggplot(data = fieldData, mapping = aes(x = factor(station), y = temperature_degC)) +
geom_boxplot() 
- Plot temperature by depth for samples from 2016, coloring the points by station
datasubset <- fieldData %>% filter(date==20160908)
ggplot(data = datasubset, mapping = aes(x = temperature_degC, y = depth_m)) +
geom_point(aes(color=factor(station))) 
What are the steps we need to take to manipulate our data into a data
frame with three columns: cruise, station and fluorescence averaged over
the top 2 m? Describe the step and say the associated dplyr data frame
manipulation functions you would use to do it.
- Select the rows where the depth is 2 m or less (use
filter) - Separate the data into date and station groups (use
group_by) - Take the average for each group of data (use
summarize)
Use the pipes to connect all three steps together.
- Create a
geom_tileplot for temperature_degC from the cruise that took place on Septh 12th 2017.
cruiseData <- filter(fieldData, date==20160908)
ggplot(cruiseData,aes(x=station,y=depth_m)) +
geom_tile(aes(fill=temperature_degC)) +
labs(fill='temperature (degC') +
scale_y_reverse()
- Are there any examples from your own data that you could plot in this way?
Here’s an example with genomic
data
that was created using geom_tile.
Create a plot like the above that shows temperature interpolated by depth and latitude for the cruise that took place on Sept 12th 2017.
#filter based on cruise date
cruiseData <- filter(fieldData, date==20170912)
#interpolate our data:
interpReference <- interp(cruiseData$latitude, cruiseData$depth_m, cruiseData$temperature_degC)
# making data frame from all combinations of latitude and depth
CruiseDataInterp <- expand.grid(latitude=interpReference$x,
depth = interpReference$y) %>%
mutate(temperature = as.vector(interpReference$z))
ggplot(CruiseDataInterp, aes(x=latitude, y=depth)) +
geom_tile(aes(fill = temperature)) +
geom_contour(aes(z = temperature),color="white") +
geom_point(data = cruiseData, aes(x=latitude, y=depth_m),color="black") + #adding in the measurement locations
scale_y_reverse() +
scale_x_reverse() + #so station1 is on the left and station4 is on the right
labs(y="Depth (m)", fill="temperature (degC)") +
scale_fill_distiller(palette="Greens",direction=1)## Warning: Removed 546 rows containing non-finite values (stat_contour).
Create a plot like the above that shows fluorescence interpolated by
depth and latitude for every cruise in 2016 (one plot per cruise). To
get a list of all the dates of cruises in 2016 do
unique(data2016$date) and look at the object printed to the console.
Reminder of the function:
cruiseInterpolationPlot <- function(dataframe, variable, cruiseDate, colorlabel){
# return a heatmap style plot for a variable interpolated over latitude and
# depth
# Inputs: dataframe = raw data frame of the DamariscottaRiverData
# variable = data frame column name (as a string) of the variable to
# interpolate
# cruiseData = numeric value of date in yyyymmdd (to compare with the
# dataframe$date column)
# colorlabel = label for the color bar (as a string)
cruiseData <- dataframe %>% filter(date == cruiseDate)
#interpolate our data:
interpReference <- interp(cruiseData$latitude, cruiseData$depth_m, cruiseData[[variable]])
newdf <- expand.grid(latitude=interpReference$x,
depth_m = interpReference$y) %>%
mutate(varname = as.vector(interpReference$z))
p <- ggplot(newdf, aes(x=latitude, y=depth_m)) +
geom_tile(aes(fill = varname)) +
geom_contour(aes(z = varname),color="white") +
geom_point(data = cruiseData, aes(x=latitude, y=depth_m),color="black") + #adding in the measurement locations
scale_y_reverse() +
scale_x_reverse() + #so station1 is on the left and station4 is on the right
labs(y="Depth (m)", fill=colorlabel, title = cruiseDate) +
scale_fill_distiller(palette="Greens",direction=1)
return(p)
}data2016 <- fieldData %>% filter(year == 2016)
unique(data2016$date)## [1] 20160908 20160920 20161004 20161019 20161101
cruiseInterpolationPlot(fieldData,'fluorescence_mg_m3',20160920,'fluorescence')## Warning: Removed 552 rows containing non-finite values (stat_contour).
cruiseInterpolationPlot(fieldData,'fluorescence_mg_m3',20161004,'fluorescence')## Warning: Removed 491 rows containing non-finite values (stat_contour).
cruiseInterpolationPlot(fieldData,'fluorescence_mg_m3',20161019,'fluorescence')## Warning: Removed 522 rows containing non-finite values (stat_contour).
cruiseInterpolationPlot(fieldData,'fluorescence_mg_m3',20161101,'fluorescence')## Warning: Removed 572 rows containing non-finite values (stat_contour).
Adapt the cruiseInterpolationPlot function such that the x, y and
color bar limits are determined from input arguments and re-plot all the
fluorescence data for each cruise in 2016. (Hint: use the xlim and
ylim functions with ggplot and include the limits argument in the
scale_fill_distiller function.)
cruiseInterpolationPlot <- function(dataframe, variable, cruiseDate, colorlabel, xlims, ylims, clims){
# return a heatmap style plot for a variable interpolated over latitude and
# depth
# Inputs: dataframe = raw data frame of the DamariscottaRiverData
# variable = data frame column name (as a string) of the variable to
# interpolate
# cruiseData = numeric value of date in yyyymmdd (to compare with the
# dataframe$date column)
# colorlabel = label for the color bar (as a string)
# xlims = vector with min and max values for x-axis
# ylims = vector with min and max values for y-axis
# clims = vector with min and max values for colorbar
cruiseData <- dataframe %>% filter(date == cruiseDate)
#interpolate our data:
interpReference <- interp(cruiseData$latitude, cruiseData$depth_m, cruiseData[[variable]])
newdf <- expand.grid(latitude=interpReference$x,
depth_m = interpReference$y) %>%
mutate(varname = as.vector(interpReference$z))
p <- ggplot(newdf, aes(x=latitude, y=depth_m)) +
geom_tile(aes(fill = varname)) +
geom_contour(aes(z = varname),color="white") +
geom_point(data = cruiseData, aes(x=latitude, y=depth_m),color="black") +
labs(y="Depth (m)", fill=colorlabel, title = cruiseDate) +
scale_fill_distiller(palette="Greens",direction=1, limits=clims) +
xlim(xlims) +
ylim(ylims)
return(p)
}data2016 <- fieldData %>% filter(year == 2016)
alldates <- unique(data2016$date)
for (dd in alldates){
print(cruiseInterpolationPlot(fieldData,'fluorescence_mg_m3',dd,
'fluorescence', c(43.91,43.75), c(105,0), c(0,12)))
}## Warning: Removed 557 rows containing non-finite values (stat_contour).
## Warning: Removed 552 rows containing non-finite values (stat_contour).
## Warning: Removed 491 rows containing non-finite values (stat_contour).
## Warning: Removed 522 rows containing non-finite values (stat_contour).
## Warning: Removed 572 rows containing non-finite values (stat_contour).
