BI303 - Lab1 - Forest Succession
Zoey Werbin
1/22/2021
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(tidyr)
library(ggplot2)R Markdown
1. Enter data
To match the steps of the lab handout, we will create two tables: one for each site. In R, these are called data frames, and they are used when we have multiple types of data (such as site categories, and numeric values). To create the data frames, we will use the cbind() function, which binds columns together. The table we create is saved to an object using the <- operator. We then use the print() function to view what we have created.
site1 <- cbind.data.frame(site = "site1",
plot = c("plot1", "plot2","plot3"),
plot_area = c(20, 20, 20),
Pinus_strobus = c(13, 17, 11),
Betula_papyrifera = c(16, 19, 23),
Fagus_grandifolia = c(2, 0, 5))
print(site1)## site plot plot_area Pinus_strobus Betula_papyrifera Fagus_grandifolia
## 1 site1 plot1 20 13 16 2
## 2 site1 plot2 20 17 19 0
## 3 site1 plot3 20 11 23 5We do the same for site 2, then combine the data frames from each site:
site2 <- cbind.data.frame(site = "site2",
plot = c("plot1", "plot2","plot3"),
plot_area = c(30, 30, 30),
Pinus_strobus = c(10, 12, 9),
Betula_papyrifera = c(8, 1, 3),
Fagus_grandifolia = c(12, 10, 22))
# Combine data from both sites into one data frame using row-bind (`rbind.data.frame`)
sites <- rbind.data.frame(site1, site2)
print(sites)## site plot plot_area Pinus_strobus Betula_papyrifera Fagus_grandifolia
## 1 site1 plot1 20 13 16 2
## 2 site1 plot2 20 17 19 0
## 3 site1 plot3 20 11 23 5
## 4 site2 plot1 30 10 8 12
## 5 site2 plot2 30 12 1 10
## 6 site2 plot3 30 9 3 222. Determine Density
Next, we determine the density of each species in each plot on a m2 basis by dividing each species count by the area of the plot. To do this in R, we use functions from the dplyr package, which is one of the most widely-used R packages. It uses a ‘pipe’ operator, %>%, that takes the output of one function and ‘pipes’ it into another function. Below, the sites data frame is piped into the group_by function, which groups the data into sites. This is piped into the summarize() function, which we use to calculate the per-site per-species density.
# Calculate density by dividing counts by plot areas
density <- sites %>% group_by(site) %>% summarize(Pinus_strobus_density = Pinus_strobus/plot_area,
Betula_papyrifera_density = Betula_papyrifera/plot_area,
Fagus_grandifolia_density = Fagus_grandifolia/plot_area)## `summarise()` regrouping output by 'site' (override with `.groups` argument)print(density)## # A tibble: 6 x 4
## # Groups: site [2]
## site Pinus_strobus_densi… Betula_papyrifera_dens… Fagus_grandifolia_den…
## <fct> <dbl> <dbl> <dbl>
## 1 site1 0.65 0.8 0.1
## 2 site1 0.85 0.95 0
## 3 site1 0.55 1.15 0.25
## 4 site2 0.333 0.267 0.4
## 5 site2 0.4 0.0333 0.333
## 6 site2 0.3 0.1 0.7333. Calculate Average
4. Calculate Standard Deviation
In this section, we will calculate both the average and the standard deviation. First, we reshape the data into “long” format, which makes it easier to plot. We do this using the pivot_longer() function, which is from the tidyr package.
density_long <- density %>% group_by(site) %>%
pivot_longer(cols = c(Pinus_strobus_density,
Betula_papyrifera_density,
Fagus_grandifolia_density), values_to = "density")
print(density_long)## # A tibble: 18 x 3
## # Groups: site [2]
## site name density
## <fct> <chr> <dbl>
## 1 site1 Pinus_strobus_density 0.65
## 2 site1 Betula_papyrifera_density 0.8
## 3 site1 Fagus_grandifolia_density 0.1
## 4 site1 Pinus_strobus_density 0.85
## 5 site1 Betula_papyrifera_density 0.95
## 6 site1 Fagus_grandifolia_density 0
## 7 site1 Pinus_strobus_density 0.55
## 8 site1 Betula_papyrifera_density 1.15
## 9 site1 Fagus_grandifolia_density 0.25
## 10 site2 Pinus_strobus_density 0.333
## 11 site2 Betula_papyrifera_density 0.267
## 12 site2 Fagus_grandifolia_density 0.4
## 13 site2 Pinus_strobus_density 0.4
## 14 site2 Betula_papyrifera_density 0.0333
## 15 site2 Fagus_grandifolia_density 0.333
## 16 site2 Pinus_strobus_density 0.3
## 17 site2 Betula_papyrifera_density 0.1
## 18 site2 Fagus_grandifolia_density 0.733Now, we use the summarize() function from the dplyr package to calculate mean and standard deviation. If there were any empty cells (or NAs) in the data, this could create errors, so we use the na.rm = T argument just in case.
average_density <- density_long %>% group_by(site, name) %>%
summarize(mean = mean(density, na.rm = T),
sd = sd(density, na.rm = T))## `summarise()` regrouping output by 'site' (override with `.groups` argument)print(average_density)## # A tibble: 6 x 4
## # Groups: site [2]
## site name mean sd
## <fct> <chr> <dbl> <dbl>
## 1 site1 Betula_papyrifera_density 0.967 0.176
## 2 site1 Fagus_grandifolia_density 0.117 0.126
## 3 site1 Pinus_strobus_density 0.683 0.153
## 4 site2 Betula_papyrifera_density 0.133 0.120
## 5 site2 Fagus_grandifolia_density 0.489 0.214
## 6 site2 Pinus_strobus_density 0.344 0.05095. Create chart
In the original lab handout, steps 5 through 7 all create a final bar plot. In R, we can combine these steps using the ggplot2 package, which is the most widely-used for graphics and plots.
final_plot <- ggplot(average_density, aes(x = name, fill = site)) +
geom_bar(stat="identity", aes(y = mean), position=position_dodge()) +
# The line below creates error bars
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width=.2,
position=position_dodge(.9))
final_plot
Finally, we’ll edit the plot to add a title and caption. To recieve extra credit, modify the code below to add a custom title and caption.
final_plot + labs(title = "Main title", caption = "My caption") +
xlab("x-axis label") +
ylab("y-axis label") +
theme(plot.caption = element_text(hjust = 0)) 