1. Using the ebird data set, calculate the total number of species
seen each month of each year in each location. Then plot the number of
species seen each month with the color of the points indicating year and
facet this plot by location.
require(tidyverse)
library(dplyr)
# import the ebird data set from the Data folder
ebird <- read_csv("../Data/MBT_ebird.csv")
# select the location, year, month, and count variables from the ebird dataset
ebird_test <- as_tibble(select(ebird, location, year, month, count))
# calculate the total number of species seen each month
# create year, month, and location variables
YEAR <- as.character(sort(unique(ebird$year)))
MONTH <- as.numeric(sort(unique(ebird$month)))
LOC <- as.character(sort(unique(ebird$location)))
# create an empty tibble to hold the data
ebird_tibble <- tibble(location = character(),
year = character(),
month = character(),
species_obs = numeric(),
)
# create a counter to control the LOC, MONTH, and YEAR variables
l=1
m=1
y=1
# loop through each month, in each year, in each location and write the total observed birds to a new tibble
while (l < (length(LOC)+1)) { # location loop that stops once it reaches the last location
y=1 # restart the year counter
while (y < (length(YEAR)+1)) { # year loop where year starts over at 2023
m=1 # restart the month counter
while (m < (length(MONTH)+1)) { # month loop where month starts over at 13
obs <- filter(ebird_test, location == LOC[l] & year == YEAR[y] & month == MONTH[m]) # filter the selected data
species_obs <- sum(obs$count) # create the observation count variable
#print(species_obs) #confirn the loops are working
if (species_obs == 0) { ## IF the counts = 0
m=m+1 # increase the counter OTHERWISE IT CREATES AN INFINITE LOOP
next # skip to the next iteration
} # close the if condition loop
# change the month number to the corresponding character "Jan", "Dec", etc.
if (MONTH[m] > 1){
if (MONTH[m] > 2){
if (MONTH[m] > 3){
if (MONTH[m] > 4){
if (MONTH[m] > 5){
if (MONTH[m] > 6){
if (MONTH[m] > 7){
if (MONTH[m] > 8){
if (MONTH[m] > 9){
if (MONTH[m] > 10){
if (MONTH[m] > 11){
MON <- "Dec"
} else {MON <- "Nov"}
} else {MON <- "Oct"}
} else {MON <- "Sep"}
} else {MON <- "Aug"}
} else {MON <- "Jul"}
} else {MON <- "Jun"}
} else {MON <- "May"}
} else {MON <- "Apr"}
} else {MON <- "Mar"}
} else {MON <- "Feb"}
} else {MON <- "Jan"}
# write each iteration as a new row in the tibble
ebird_tibble <- add_row(ebird_tibble, location = LOC[l],
month = as.character(MON),
year = YEAR[y],
species_obs = species_obs) # with ONLY nonzero values
m=m+1 # increase the month counter
} # close the month loop
y=y+1 # increase the year counter
} # close the year loop
l=l+1 # increase the location counter
} #close the location loop
# view the first 15 lines of the data
print(ebird_tibble, n=15)
## # A tibble: 118 × 4
## location year month species_obs
## <chr> <chr> <chr> <dbl>
## 1 US-AR 2021 Jan 58
## 2 US-AR 2022 Feb 30
## 3 US-AZ 2009 Aug 25
## 4 US-AZ 2015 Apr 123
## 5 US-AZ 2022 Jan 34
## 6 US-CA 2003 Oct 3
## 7 US-CA 2003 Nov 17
## 8 US-CA 2003 Dec 29
## 9 US-CA 2004 Jan 299
## 10 US-CA 2004 Feb 396
## 11 US-CA 2013 Jan 1
## 12 US-CA 2014 Apr 189
## 13 US-CA 2017 Jan 4
## 14 US-CA 2021 Nov 335
## 15 US-CT 2004 Nov 11
## # … with 103 more rows
# order the months chronologically
ebird_tibble <- ebird_tibble %>%
mutate(month = factor(month, levels = c("Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec")))
# plot the data with points colored by year and faceted by location
p1 <- ggplot(data=ebird_tibble) +
aes(y=species_obs, x=month) +
geom_point(aes(color=year), size = 2)
p1 + facet_wrap(~location, scales = "free_y", drop = T) +
theme(axis.text.x = element_text(angle = 60))
2. Plot a comparison of mass by treatment including the individual
observations, the mean, and standard error of the mean. Use point color
or shape to indicate the sex.
# require tidyverse
require(tidyverse)
# import the data from Assignment 5
Assign5 <- read_csv("Results/Assign5_Merge.csv")
glimpse(Assign5)
## Rows: 16
## Columns: 6
## $ Sample <chr> "Sample3", "Sample4", "Sample5", "Sample6", "Sample7", "Sa…
## $ Sex <chr> "Male", "Male", "Male", "Male", "Male", "Male", "Male", "F…
## $ Experiment <chr> "Control", "Control", "Control", "Treatment", "Treatment",…
## $ body_length <dbl> 4.5366323, 1.0883156, 3.5529698, 8.1889194, 8.4469190, 0.8…
## $ age <dbl> 11, 8, 10, 4, 8, 11, 3, 10, 5, 3, 3, 1, 7, 0, 4, 6
## $ mass <dbl> 8.459178, 3.871743, 3.119359, 9.379790, 7.139136, 7.406650…
# create a boxplot comparing mass by treatment groups with color distinguishing sex
p2 <- ggplot(data=Assign5) +
aes(y=mass, x=Experiment, fill=Sex) +
geom_boxplot()
p2
3. Generate a scatter plot of age and mass, indicate treatment with
point shape or color, and fit separate regression lines (without CI) to
each treatment.
# create the scatter plot of age by mass with point color indicating treatment and regression lines
p3 <- ggplot(data=Assign5) +
aes(y=mass, x=age, color=Experiment) +
geom_point(size =2) +
geom_smooth(method="lm", se=F) # without CI
p3
## `geom_smooth()` using formula 'y ~ x'
4. Combine the plots from 2 and 3 using patchwork tag each panel
with and number or letter and include a title for the overall plot.
#install.packages("ggthemes")
#install.packages("patchwork")
library(ggplot2)
library(ggthemes)
library(patchwork)
# combine the plots using patchwork
p2 + p3 + plot_layout(ncol=1) +
plot_annotation(
title = 'Treatment Groups',
caption = 'made with patchwork',
theme = theme(plot.title = element_text(size = 16)),
tag_levels = 'a'
)
## `geom_smooth()` using formula 'y ~ x'
