Tidyverse Activity

Overview

Tidyverse allows you to easily manipulate and analyze data in a way that both readable and maintainable.

Activity Objective

For this activity, you’ll need the following libraries.

library( readr )
library( knitr )
library( lubridate )
library( tidyverse )
library( kableExtra )

and the url for the raw data set.

url <- \https://docs.google.com/spreadsheets/d/1Mk1YGH9LqjF7drJE-td1G_JkdADOU0eMlrP01WFBT8s/pub?gid=0&single=true&output=csv\

For review, here are links to the slides and narrative on this topic.

---

Activity Questions

1. Using tidyverse load in the data and format it properly including:

• Make columns for Date representing a POSIX date object.
• Create a Weekday column as an appropriately ordered factor.
• Convert all data columns that are in Imperial Units (F, mph, in, etc.) to proper SI coordinates.
• Drop any duplicate or irrelevant columns of data to minimize memory footprint.
• Remove any rows that have missing data
• Retain data columns for Date, Weekday, Wind Speed, Wind Direction, Rain, Air Temperature, Water Temperature, PH, Salinity, and Water Depth.

format <- \%m/%d/%Y %I:%M:%S %p\
days <- c(\Monday\,\Tuesday\,\Wednesday\,\Thursday\, \Friday\,\Saturday\, \Sunday\)

read_csv( url ) %>%
  mutate( Date = parse_date_time( DateTime,
                                  orders=format,
                                  tz=\EST\) ) %>%
  mutate( Weekday = factor( weekdays( Date ),
                            ordered=TRUE,
                            levels=days) ) %>%
  mutate( AirTemp = (AirTempF - 32) * 5/9,
          Rain = Rain_in * 2.54,
          `Wind Speed` = 1.60934 * WindSpeed_mph ) %>%
  select( Date, 
          Weekday, 
          Rain,
          `Wind Speed`, 
          `Wind Direction` = WindDir,
          `Air Temperature (°C)` = AirTemp,
          `Water Temperature (°C)` = H2O_TempC,
          PH, 
          Salinity = Salinity_ppt, 
          `Water Depth (m)` = Depth_m ) %>%
  filter( !is.na( PH) ) -> rice 
summary( rice )

      Date                          Weekday          Rain         
 Min.   :2014-01-01 00:00:00   Monday   :1152   Min.   :0.000000  
 1st Qu.:2014-01-22 08:33:45   Tuesday  :1152   1st Qu.:0.000000  
 Median :2014-02-12 16:52:30   Wednesday:1248   Median :0.000000  
 Mean   :2014-02-12 16:49:35   Thursday :1191   Mean   :0.002137  
 3rd Qu.:2014-03-06 01:11:15   Friday   :1151   3rd Qu.:0.000000  
 Max.   :2014-03-27 09:30:00   Saturday :1152   Max.   :0.881380  
                               Sunday   :1152                     
   Wind Speed     Wind Direction   Air Temperature (°C) Water Temperature (°C)
 Min.   : 0.000   Min.   :  0.00   Min.   :-15.6950     Min.   :-0.140        
 1st Qu.: 3.970   1st Qu.: 37.31   1st Qu.: -0.2542     1st Qu.: 3.930        
 Median : 6.583   Median :137.30   Median :  3.0194     Median : 5.450        
 Mean   : 8.764   Mean   :146.20   Mean   :  3.7748     Mean   : 5.529        
 3rd Qu.:11.735   3rd Qu.:249.97   3rd Qu.:  8.0056     3rd Qu.: 7.410        
 Max.   :49.326   Max.   :360.00   Max.   : 23.8167     Max.   :13.300        
                                                                              
       PH          Salinity      Water Depth (m)
 Min.   :6.43   Min.   :0.0000   Min.   :3.705  
 1st Qu.:7.50   1st Qu.:0.0700   1st Qu.:4.451  
 Median :7.58   Median :0.0800   Median :4.684  
 Mean   :7.60   Mean   :0.0759   Mean   :4.677  
 3rd Qu.:7.69   3rd Qu.:0.0800   3rd Qu.:4.913  
 Max.   :9.00   Max.   :0.1000   Max.   :5.454  
                                                

2. On April 23, 1971, the Carpenters released the song “Rainy Days and Mondays Always Get Me Down”. Here is a youtube for those who may be too young to know this one. From the data set, make a plot of total rain for the month of February by each day. Are there significantly different amounts of rain by weekday?

rice %>%
  filter( Date > ymd(\20140201\)  ) %>%
  filter( Date < ymd(\20140301\) ) %>% 
  filter( Rain > 0 ) %>%
  ggplot( aes(Weekday, Rain ) ) + 
  geom_boxplot() + 
  theme( axis.text.x = element_text(angle = 45, hjust=1 )) + 
  scale_y_log10()

3. Is there a prevailing wind direction for this field station?

rice %>%
  ggplot( aes(`Wind Direction`) ) + 
  geom_histogram( bins=24) +
  coord_polar( start=0 ) + 
  theme_bw()

4. This field station is located in the tidal freshwater region of the Chesapeake Bay in Virginia, USA. As such, there is a tidal surge. Make a table of minimum and maximum tide depth for each day of the week that contained Valentines Day (14 February 2014).

rice %>%
  filter( Date < ymd(\2014-02-17\),
          Date >= ymd(\2014-02-10\)) %>%
  group_by( Weekday ) %>% 
  summarize( High = max(`Water Depth (m)`),
             Low = min( `Water Depth (m)`) ) %>%
  kable( digits=2,
         caption = \Tides measured at the Rice Rivers Center (as depth in meters) for each day of the week containing valentines day in 2014.\) %>%
  kable_styling( full_width = FALSE ) %>%
  add_header_above(c(\\,\Measured Tide Depth (m)\=2))