Setup

library(tidyverse)

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──

## ✓ ggplot2 3.3.5     ✓ purrr   0.3.4
## ✓ tibble  3.1.5     ✓ dplyr   1.0.7
## ✓ tidyr   1.1.4     ✓ stringr 1.4.0
## ✓ readr   2.0.2     ✓ forcats 0.5.1

## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

library(lubridate)

## 
## Attaching package: 'lubridate'

## The following objects are masked from 'package:base':
## 
##     date, intersect, setdiff, union

load("oly_airport.Rdata")

Exercise 1

Look at today’s weather in the weather channel. What are TMAX, TMIN, and the probability of rain. How do these values compare with the historical data for this day and month?

Solution

As I write these notes it is February 15, 2022. From the weather channel, TMIN = 39, TMAX = 49, and the probability of rain is 3%. Start with summary(). Just use PRCP, TMAX, and TMIN.

feb15 = oly_airport %>% 
  filter(dy == 15 & mo == 2) %>% 
  select(PRCP,TMAX,TMIN)
summary(feb15)

##       PRCP             TMAX            TMIN      
##  Min.   :0.0000   Min.   :25.00   Min.   :10.00  
##  1st Qu.:0.0000   1st Qu.:45.00   1st Qu.:29.00  
##  Median :0.1050   Median :49.00   Median :33.50  
##  Mean   :0.2516   Mean   :48.56   Mean   :33.42  
##  3rd Qu.:0.3700   3rd Qu.:53.00   3rd Qu.:38.25  
##  Max.   :2.1800   Max.   :62.00   Max.   :49.00

We can see that TMAX is right on the mean and median. However, TMIN is about six degrees above the central values. To compare the probability of rain, we need to do another computation to see the historical value. Do this.

Solution

mean(feb15$PRCP > 0)

## [1] 0.725

So 72.5% of the time there is measurable precipitation on February 15. Our current value of 3% is very low.

Forecast

How do you think the weather one month in the future will be. How will it compare with today? Start with a summary().

Solution

mar15 = oly_airport %>% 
  filter(dy == 15 & mo == 3) %>% 
  select(PRCP,TMAX,TMIN) 
  
summary(mar15)

##       PRCP             TMAX           TMIN      
##  Min.   :0.0000   Min.   :43.0   Min.   :24.00  
##  1st Qu.:0.0000   1st Qu.:49.0   1st Qu.:30.00  
##  Median :0.0400   Median :53.0   Median :34.00  
##  Mean   :0.1950   Mean   :53.4   Mean   :34.44  
##  3rd Qu.:0.2225   3rd Qu.:58.0   3rd Qu.:38.25  
##  Max.   :2.0800   Max.   :73.0   Max.   :47.00

mean(mar15$PRCP > 0)

## [1] 0.6375

Here’s what I see.

1. The probability of rain has fallen by about 8%.
2. The mean of TMAX has increased by about 4 degrees.
3. The mean of TMIN has increased by about 1 degree.

Above 100

On what days has the maximum daily temperature exceeded 100 degrees? How many? When?

Solution

oly_airport %>% 
  filter(TMAX > 100) %>% 
  arrange(DATE) %>% 
  select(DATE,TMAX)

## # A tibble: 13 × 2
##    DATE        TMAX
##    <date>     <dbl>
##  1 1941-07-15   103
##  2 1941-07-16   103
##  3 1942-06-30   101
##  4 1942-07-01   102
##  5 1981-08-09   104
##  6 1981-08-10   102
##  7 1994-07-20   102
##  8 2006-07-21   101
##  9 2009-07-28   101
## 10 2009-07-29   104
## 11 2021-06-26   102
## 12 2021-06-27   105
## 13 2021-06-28   110

Below Zero

Repeat the exercise above for days with a TMIN below zero.

Solution

oly_airport %>% 
  filter(TMIN < 0) %>% 
  arrange(DATE) %>% 
  select(DATE,TMIN)

## # A tibble: 12 × 2
##    DATE        TMIN
##    <date>     <dbl>
##  1 1950-02-01    -1
##  2 1955-11-15    -1
##  3 1972-01-27    -7
##  4 1972-01-28    -4
##  5 1972-02-02    -1
##  6 1972-12-08    -3
##  7 1972-12-10    -1
##  8 1978-12-31    -5
##  9 1979-01-01    -8
## 10 1983-12-22    -3
## 11 1983-12-23    -7
## 12 1998-12-22    -1

Heavy Rain

What were the ten heaviest days of rainfall in Olympia. Sort them by date.

Solution

oly_airport %>% 
  arrange(PRCP) %>% 
  tail(10) %>% 
  arrange(DATE) %>% 
  select(DATE, PRCP)

## # A tibble: 10 × 2
##    DATE        PRCP
##    <date>     <dbl>
##  1 1951-02-09  3.64
##  2 1956-12-09  3.5 
##  3 1962-11-19  4.33
##  4 1990-01-09  3.82
##  5 1990-11-24  4.08
##  6 2001-11-14  3.64
##  7 2003-10-20  4.12
##  8 2006-11-06  4.31
##  9 2009-01-07  4.82
## 10 2022-01-06  3.99

Bad Day in January

How frequently do we see a minimum temperature below 32 combined with rain in January.

Solution

oly_airport %>% 
  filter(mo == 1) %>% 
  mutate(bad = TMIN < 32 & PRCP > 0) %>% 
  summarise(mean(bad))

## # A tibble: 1 × 1
##   `mean(bad)`
##         <dbl>
## 1       0.169

Repeat for February

Solution

oly_airport %>% 
  filter(mo == 2) %>% 
  mutate(bad = TMIN < 32 & PRCP > 0) %>% 
  summarise(mean(bad))

## # A tibble: 1 × 1
##   `mean(bad)`
##         <dbl>
## 1       0.146

March?

Solution

oly_airport %>% 
  filter(mo == 3) %>% 
  mutate(bad = TMIN < 32 & PRCP > 0) %>% 
  summarise(mean(bad))

## # A tibble: 1 × 1
##   `mean(bad)`
##         <dbl>
## 1       0.115

Rainiest Day

Which days of the year have the highest probability of rain? List them in order of probability.

Solution

oly_airport %>% 
  group_by(mo, dy) %>% 
  summarize(p_rain = mean(PRCP > 0)) %>% 
  ungroup() %>% 
  arrange(p_rain) %>% 
  tail(10)

## `summarise()` has grouped output by 'mo'. You can override using the `.groups` argument.

## # A tibble: 10 × 3
##    mo    dy    p_rain
##    <fct> <fct>  <dbl>
##  1 12    9      0.716
##  2 12    15     0.716
##  3 12    18     0.716
##  4 12    19     0.716
##  5 2     15     0.725
##  6 12    10     0.728
##  7 11    24     0.741
##  8 12    2      0.741
##  9 2     29     0.75 
## 10 12    20     0.765