Reproducible Research: courseProject2
Jess Li
2024-11-07
Synopsis
This project explores the United States storm database, which documents major storms and weather events, along with their estimated impacts (including fatalities, damages, etc.), from 1950 to 2011. The data, sourced from the National Oceanic and Atmospheric Administration (NOAA), required significant pre-processing due to poor initial reporting. The dataset was cleaned, and values were re-coded to facilitate analysis. Through data visualization, the events most harmful to both public health and the economy were identified. This report includes both the code and the results generated during the project.
Data Processing
Download raw data and load it into R
## download the data file
url <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
download.file(url, "repdata_StormData.csv.bz2")
## load it into R
raw_data <- read.csv("repdata_StormData.csv.bz2", header = TRUE, sep = ",")Basic exploring of the raw data
View all the variables
str(raw_data)## 'data.frame': 902297 obs. of 37 variables:
## $ STATE__ : num 1 1 1 1 1 1 1 1 1 1 ...
## $ BGN_DATE : chr "4/18/1950 0:00:00" "4/18/1950 0:00:00" "2/20/1951 0:00:00" "6/8/1951 0:00:00" ...
## $ BGN_TIME : chr "0130" "0145" "1600" "0900" ...
## $ TIME_ZONE : chr "CST" "CST" "CST" "CST" ...
## $ COUNTY : num 97 3 57 89 43 77 9 123 125 57 ...
## $ COUNTYNAME: chr "MOBILE" "BALDWIN" "FAYETTE" "MADISON" ...
## $ STATE : chr "AL" "AL" "AL" "AL" ...
## $ EVTYPE : chr "TORNADO" "TORNADO" "TORNADO" "TORNADO" ...
## $ BGN_RANGE : num 0 0 0 0 0 0 0 0 0 0 ...
## $ BGN_AZI : chr "" "" "" "" ...
## $ BGN_LOCATI: chr "" "" "" "" ...
## $ END_DATE : chr "" "" "" "" ...
## $ END_TIME : chr "" "" "" "" ...
## $ COUNTY_END: num 0 0 0 0 0 0 0 0 0 0 ...
## $ COUNTYENDN: logi NA NA NA NA NA NA ...
## $ END_RANGE : num 0 0 0 0 0 0 0 0 0 0 ...
## $ END_AZI : chr "" "" "" "" ...
## $ END_LOCATI: chr "" "" "" "" ...
## $ LENGTH : num 14 2 0.1 0 0 1.5 1.5 0 3.3 2.3 ...
## $ WIDTH : num 100 150 123 100 150 177 33 33 100 100 ...
## $ F : int 3 2 2 2 2 2 2 1 3 3 ...
## $ MAG : num 0 0 0 0 0 0 0 0 0 0 ...
## $ FATALITIES: num 0 0 0 0 0 0 0 0 1 0 ...
## $ INJURIES : num 15 0 2 2 2 6 1 0 14 0 ...
## $ PROPDMG : num 25 2.5 25 2.5 2.5 2.5 2.5 2.5 25 25 ...
## $ PROPDMGEXP: chr "K" "K" "K" "K" ...
## $ CROPDMG : num 0 0 0 0 0 0 0 0 0 0 ...
## $ CROPDMGEXP: chr "" "" "" "" ...
## $ WFO : chr "" "" "" "" ...
## $ STATEOFFIC: chr "" "" "" "" ...
## $ ZONENAMES : chr "" "" "" "" ...
## $ LATITUDE : num 3040 3042 3340 3458 3412 ...
## $ LONGITUDE : num 8812 8755 8742 8626 8642 ...
## $ LATITUDE_E: num 3051 0 0 0 0 ...
## $ LONGITUDE_: num 8806 0 0 0 0 ...
## $ REMARKS : chr "" "" "" "" ...
## $ REFNUM : num 1 2 3 4 5 6 7 8 9 10 ...From the output and the documentation from the link: https://d396qusza40orc.cloudfront.net/repdata%2Fpeer2_doc%2Fpd01016005curr.pdf The following variables are relevant to the questions of interest. questions. - Event begin date (BGN_DATE) - Event type (EVTYPE) - State name (STATE): potentially used to explore the geographical pattern - Fatalities (FATALITIES) : related to impact on population health - Injuries (INJURIES): related to impact on population health - Property damages (PROPDMG): related to economic consequences - Property damages exponents (PROPDMGEXP) - Crop damages(CROPDMG): related to economic consequences - Crop damages exponents(CROPDMGEXP): related to economic consequences Subset the data that contains the above variables only
## load library
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, union## subset data
our_data <- raw_data %>%
select(BGN_DATE,EVTYPE,STATE,FATALITIES,INJURIES,
PROPDMG,PROPDMGEXP,CROPDMG,CROPDMGEXP)Further exploring: event type
Unique Event type number: there are way too many types of event (>900), we need to clean them
length(unique(our_data$EVTYPE)) ## [1] 985There are 48 types of event documented in the given document. We should consider how to transfer or redefine this column of EVTYPE based on the 48 pre-defined types.
library(stringr) # load library
## unify the cases
our_data$EVTYPE <- tolower(our_data$EVTYPE)
## correct misspell or abbreviation
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("avalance", "avalanche") %>% # avalanches
str_replace_all("erosin", "erosion") %>%
str_replace_all("coastalstorm", "coastal storm ") %>%
str_replace_all("cstl", "coastal") %>%
str_replace_all("devel", "devil") %>%
str_replace_all("sno$", "snow$") %>%
str_replace_all("tstm", "thunderstorm") %>%
str_replace_all("thundeerstorm", "thunderstorm") %>%
str_replace_all("tunderstorm", "thunderstorm") %>%
str_replace_all("tstmw", "thunderstorm wind") %>%
str_replace_all("temps", "temperatures") %>%
str_replace_all("wintry", "wintery") %>%
str_replace_all("hvy", "heavy") %>%
str_replace_all("flooding", "flood")
## transform all related tyeps into the 48 types
## 1 Astronomical Low Tide
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("astro.*low.*tide$", "Astronomical Low Tide")
## 2 Avalanche
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^avalan.*$", "Avalanche")
## 3 Blizzard
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^blizzard.*$", "Blizzard") %>%
str_replace_all("blow.*snow.*$", "Blizzard")
## 4 Coastal Flood
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^coastal.*flood.*$", "Coastal Flood") %>%
str_replace_all(".*coastal.*flood.*$", "Coastal Flood") %>%
str_replace_all("^beach.*flood.*$", "Coastal Flood") %>%
str_replace_all("^beach.*erosion*$", "Coastal Flood") %>%
str_replace_all("^erosion.*$", "Coastal Flood")
## 5 Cold/Wind Chill
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all(".*cold.*chill.*", "Cold/Wind Chill") %>%
str_replace_all(".*wind.*chill.*", "Cold/Wind Chill") %>%
str_replace_all(".*cold.*wind.*", "Cold/Wind Chill") %>%
str_replace_all("cold", "Cold/Wind Chill") %>%
str_replace_all("cold air funnel.*", "Cold/Wind Chill") %>%
str_replace_all("cold temperature.*", "Cold/Wind Chill") %>%
str_replace_all("cold weather", "Cold/Wind Chill")
## 6 Debris Flow
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all(".*debr.*flow.*", "Debris Flow") %>%
str_replace_all(".*rock.*", "Debris Flow") %>%
str_replace_all(".*mudflow.*", "Debris Flow")
## 7 Dense Fog
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all(".*fog.*", "Dense Fog")
## 8 Dense Smoke
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all(".*smoke.*", "Dense Smoke")
## 9 Drought
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^drought.*", "Drought") %>%
str_replace_all("^dry.*", "Drought") %>%
str_replace_all("^dryness.*", "Drought") %>%
str_replace_all(".*dry$", "Drought")
## 10 Dust Devil
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^dust devil.*", "Dust Devil")
## 11 Dust Storm
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^dust storm.*", "Dust Storm") %>%
str_replace_all("^duststorm.*", "Dust Storm") %>%
str_replace_all("^blowing dust", "Dust Storm")
## 12 Excessive Heat
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^excessive heat.*", "Excessive Heat") %>%
str_replace_all("^extreme heat.*", "Excessive Heat")
## 13 Extreme Cold/Wind Chill
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^extreme.*chill.*", "Extreme Cold/Wind Chill") %>%
str_replace_all("^extreme cold.*", "Extreme Cold/Wind Chill") %>%
str_replace_all("excessive/extended cold", "Extreme Cold/Wind Chill") %>%
str_replace_all("extreme/record cold", "Extreme Cold/Wind Chill")
## 14 Flash Flood
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^flash flood.*", "Flash Flood") %>%
str_replace_all(".*flash.*flood.*", "Flash Flood") %>%
str_replace_all(".*flood.*flash.*", "Flash Flood")
## 15 Flood
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^flood$", "Flood") %>%
str_replace_all("floods", "Flood")%>%
str_replace_all("^flooding$", "Flood") %>%
str_replace_all(".*flood.*rain.*", "Flood")
## 16 Freezing Fog
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^freezing fog$", "Freezing Fog")
## 17 Frost/Freeze
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^frost.*freeze$", "Frost/Freeze") %>%
str_replace_all("freeze$", "Frost/Freeze") %>%
str_replace_all("^frost.*", "Frost/Freeze") %>%
str_replace_all("^freezing rain.*", "Frost/Freeze") %>%
str_replace_all("^freezing spray*", "Frost/Freeze") %>%
str_replace_all("^freezing drizzle.*$", "Frost/Freeze") %>%
str_replace_all("cold and frost", "Frost/Freeze") %>%
str_replace_all("damaging|early freeze|frost", "Frost/Freeze")
## 18 Funnel Cloud
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^funnel cloud$", "Funnel Cloud") %>%
str_replace_all("^funnel.*", "Funnel Cloud")
## 19 Hail
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^hail.*", "Hail") %>%
str_replace_all(".*hail$", "Hail")
## 20 Heat
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^heat.*", "Heat")
## 21 Heavy Rain
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^heavy rain.*", "Heavy Rain") %>%
str_replace_all("^heavy precipitation$", "Heavy Rain") %>%
str_replace_all("^heavy shower.*", "Heavy Rain")
## 22 Heavy Snow
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^heavy snow.*", "Heavy Snow")
## 23 High Surf
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^heavy surf.*", "High Surf") %>%
str_replace_all("^high surf.*", "High Surf") %>%
str_replace_all("^high seas.*", "High Surf") %>%
str_replace_all("^high wave.*", "High Surf") %>%
str_replace_all("^high water.*", "High Surf") %>%
str_replace_all("high surf advisory", "High Surf")
## 24 High Wind
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^heavy winds.*", "High Surf") %>%
str_replace_all("^high wind.*", "High Surf")
## 25 Hurricane/Typhoon
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^hurricane.*", "Hurricane/Typhoon") %>%
str_replace_all("^typhoon$", "Hurricane/Typhoon")
## 26 Ice Storm
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^ice.*storm", "Ice Storm") %>%
str_replace_all("^ice.*snow.*$", "Ice Storm")
## 27 Lakeshore Flood
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^lake.*flood", "Lakeshore Flood") %>%
str_replace_all("^lakeshore flood", "Lakeshore Flood")
## 28 Lake-Effect Snow
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^lake effect snow", "Lake-Effect Snow") %>%
str_replace_all("^lake-effect snow", "Lake-Effect Snow")
## 29 Lightning
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^lightning.*", "Lightning")
## 30 Marine Hail
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^marine hail*", "Marine Hail")
## 31 Marine High Wind
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^marine high wind*", "Marine High Wind")
## 32 Marine Strong Wind
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^marine strong wind*", "Marine Strong Wind")
## 33 Marine Thunderstorm Wind
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^marine thunderstorm wind*", "Marine Thunderstorm Wind")
## 34 Rip Current
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^rip current.*", "Rip Current")
## 35 Seiche
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^seiche", "Seiche")
## 36 Sleet
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^sleet.*", "Sleet") %>%
str_replace_all("snow.*sleet.*", "Sleet")
## 37 Storm Surge/Tide
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^storm surge/tide", "Storm Surge/Tide")
## 38 Strong Wind
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^strong wind.*", "Strong Wind")
## 39 Thunderstorm Wind
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("thunderstorm wind.*$", "Thunderstorm Wind") %>%
str_replace_all("^thunderstorm w inds.*", "Thunderstorm Wind") %>%
str_replace_all("^thunderstorm winds.*", "Thunderstorm Wind") %>%
str_replace_all("severe thunderestorm.*", "Thunderstorm Wind")
## 40 Tornado
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^tornado.*", "Tornado")
## 41 Tropical Depression
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^tropical depression.*", "Tropical Depression")
## 42 Tropical Storm
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^tropical storm.*", "Tropical Storm")
## 43 Tsunami
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^tsunami.*", "Tsunami")
## 44 Volcanic Ash
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^volcanic ash.*", "Volcanic Ash")
## 45 Waterspout
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("waterspout", "Waterspout")
## 46 Wildfire
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^wildfire.*", "Wildfire")
## 47 Winter Storm
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^winter storm.*", "Winter Storm")
## 48 Winter Weather
our_data$EVTYPE <- our_data$EVTYPE %>%
str_replace_all("^winter weather.*", "Winter Weather")After this, there are still many left (~400) that is hard to be assigned to the 48 types of events that is in record. We kept only the 48 types for later analysis.
library(dplyr) # load library
## select only the 48 types of events
events_in_record <- c("Astronomical Low Tide","Avalanche",
"Blizzard","Coastal Flood",
"Cold/Wind Chill","Debris Flow",
"Dense Fog","Dense Smoke",
"Drought","Dust Devil",
"Dust Storm","Excessive Heat",
"Extreme Cold/Wind Chill","Flash Flood",
"Flood","Freezing Fog",
"Frost/Freeze","Funnel Cloud",
"Hail","Heat","Heavy Rain",
"Heavy Snow","High Surf","High Wind",
"Hurricane/Typhoon","Ice Storm",
"Lake-Effect Snow",
"Lakeshore Flood","Lightning",
"Marine Hail", "Marine High Wind",
"Marine Strong Wind",
"Marine Thunderstorm Wind",
"Rip Current", "Seiche","Sleet",
"Storm Surge/Tide","Strong Wind",
"Thunderstorm Wind","Tornado",
"Tropical Depression","Tropical Storm",
"Tsunami","Volcanic Ash",
"Waterspout","Wildfire","Winter Storm","Winter Weather")
our_data_coded <- our_data %>%
filter(EVTYPE %in% events_in_record)further exploring: check other variables
Summary of the related variables
summary(our_data_coded$FATALITIES)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.0165 0.0000 583.0000summary(our_data_coded$INJURIES)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.1553 0.0000 1700.0000summary(our_data_coded$PROPDMG)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0 0.0 0.0 12.0 0.5 5000.0summary(our_data_coded$PROPDMGEXP)## Length Class Mode
## 891628 character charactersummary(our_data_coded$CROPDMG)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 0.00 0.00 1.52 0.00 990.00summary(our_data_coded$CROPDMGEXP)## Length Class Mode
## 891628 character characterBoth types of damages (property and crop) values are encoded with exponents . Needs to be recalculated into true numbers.
unique(our_data_coded$PROPDMGEXP)## [1] "K" "M" "" "B" "m" "+" "0" "5" "6" "?" "4" "2" "3" "h" "7" "H" "-" "1" "8"unique(our_data_coded$CROPDMGEXP)## [1] "" "M" "K" "m" "B" "?" "0" "k" "2"Values in -EXP columns like H/h, K/k, M/m, B/b are easily interpreted; Others are not clear, so further check what they may represent.
## check different values from 1 to 8, manually change the number
head(filter(our_data_coded,PROPDMGEXP==1)) ## BGN_DATE EVTYPE STATE FATALITIES INJURIES PROPDMG
## 1 5/10/1995 0:00:00 Thunderstorm Wind TN 0 0 0
## 2 5/10/1995 0:00:00 Thunderstorm Wind TN 0 0 0
## 3 5/18/1995 0:00:00 Thunderstorm Wind TN 0 0 0
## 4 5/14/1995 0:00:00 Thunderstorm Wind TN 0 0 0
## 5 5/18/1995 0:00:00 Thunderstorm Wind TN 0 0 0
## 6 5/18/1995 0:00:00 Tornado TN 0 0 0
## PROPDMGEXP CROPDMG CROPDMGEXP
## 1 1 0
## 2 1 0
## 3 1 0
## 4 1 0
## 5 1 0
## 6 1 0## do the same for the other values like "+","-"
head(filter(our_data_coded,PROPDMGEXP=="-")) ## BGN_DATE EVTYPE STATE FATALITIES INJURIES PROPDMG PROPDMGEXP
## 1 12/12/1995 0:00:00 High Surf OR 2 0 15 -
## CROPDMG CROPDMGEXP
## 1 0After checking all the other values, I decided to treat the number (0,1,2…etc.) in -EXP column as the exponent value, for example, if the -EXP column is 2, I will multiple the damage value with 10^2. The “+”, “-”, ?” and “” will be ignored or considered null, i.e., set as NA The other letters will be transformed into the number they represent, for example, H/h stands for hundreds, so it will be transformed into 10^2, i.e., 100.
## transform
library(stringr) # load library
our_data_coded$PROPDMGEXP <- our_data_coded$PROPDMGEXP %>% str_replace_all("0","1") %>%
str_replace_all("1","10") %>% str_replace_all("2","100") %>%
str_replace_all("3","1000") %>% str_replace_all("4","10000") %>%
str_replace_all("5","100000") %>% str_replace_all("6","1000000") %>%
str_replace_all("7","10000000") %>% str_replace_all("8","100000000") %>%
str_replace_all("H","100") %>% str_replace_all("h","100") %>%
str_replace_all("K","1000") %>%
str_replace_all("M","1000000") %>% str_replace_all("m","1000000") %>%
str_replace_all("B","1000000000")
our_data_coded$CROPDMGEXP <- our_data_coded$CROPDMGEXP %>% str_replace_all("0","1") %>%
str_replace_all("2","100") %>%
str_replace_all("K","1000") %>% str_replace_all("k","1000") %>%
str_replace_all("M","1000000") %>% str_replace_all("m","1000000") %>%
str_replace_all("B","1000000000")
unique(our_data_coded$PROPDMGEXP)## [1] "1000" "1000000" "" "1000000000" "+"
## [6] "10" "100000" "?" "10000" "100"
## [11] "10000000" "-" "100000000"unique(our_data_coded$CROPDMGEXP)## [1] "" "1000000" "1000" "1000000000" "?"
## [6] "1" "100"## turn into numeric values, "+""-""?""" will be turned to NA
our_data_coded$PROPDMGEXP <- as.numeric(our_data_coded$PROPDMGEXP)## Warning: NAs introduced by coercionour_data_coded$CROPDMGEXP <- as.numeric(our_data_coded$CROPDMGEXP)## Warning: NAs introduced by coercionsummary(our_data_coded$PROPDMGEXP)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 1.000e+01 1.000e+03 1.000e+03 1.077e+05 1.000e+03 1.000e+09 458081summary(our_data_coded$CROPDMGEXP)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 1.00e+00 1.00e+03 1.00e+03 3.61e+04 1.00e+03 1.00e+09 608638Calculate the true values for the damages
our_data_coded <- our_data_coded %>%
mutate(PROPDMG_T = PROPDMG*PROPDMGEXP) %>%
mutate(CROPDMG_T = CROPDMG*CROPDMGEXP)
summary(our_data_coded$PROPDMG_T)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.000e+00 0.000e+00 1.000e+03 8.627e+05 1.000e+04 1.150e+11 458081summary(our_data_coded$CROPDMG_T)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.000e+00 0.000e+00 0.000e+00 1.481e+05 0.000e+00 5.000e+09 608638Results: to address the two following main questions
- Across the United States, which types of events are most harmful with respect to population health?
library(stringr) # load library
library(ggplot2)
library(dplyr)
## summarize the health-realted variable: fatality, injury
## and calculate the total of both types
health <- our_data_coded %>%
group_by(EVTYPE) %>%
summarize(sum_fata = sum(FATALITIES,na.rm=TRUE),
sum_inj = sum(INJURIES,na.rm=TRUE),
mean_fata = mean(FATALITIES),
mean_inj = mean(INJURIES)) %>%
mutate(sum_fata_inj = sum_fata+sum_inj)
## plot
## 1) the top 10 based on the total number of fatality and injury
## 2) the top 10 based on only fatality
## 3) the top 10 based on only injury
## plot 1
health1 <- arrange(health,desc(sum_fata_inj))
g1<- ggplot(data=health1[1:10,],aes(x=reorder(EVTYPE,sum_fata_inj),y=sum_fata_inj)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Total number fatality and injury") + xlab("") +
labs(title="Top 10 Most Harmful Events for public health",
subtitle="Based on total number of fatality and injury") +
theme(axis.text=element_text(size=8),
axis.title=element_text(size=8,face="bold"),
title=element_text(size=9,face="bold"))
## plot 2
health2 <- arrange(health,desc(sum_fata))
g2<- ggplot(data=health2[1:10,],aes(x=reorder(EVTYPE,sum_fata),y=sum_fata)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Number of fatality") + xlab("") +
labs(subtitle="Based on fatality only") +
theme(axis.text=element_text(size=8),
axis.title=element_text(size=8,face="bold"),
title=element_text(size=9,face="bold"))
## plot 3
health3 <- arrange(health,desc(sum_inj))
g3<- ggplot(data=health3[1:10,],aes(x=reorder(EVTYPE,sum_inj),y=sum_inj)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Number of injury") + xlab("") +
labs(subtitle="Based on injury only") +
theme(axis.text=element_text(size=8),
axis.title=element_text(size=8,face="bold"),
title=element_text(size=9,face="bold"))
## plot them in one chart
# install.packages("multipanelfigure") ##need to run once
library(multipanelfigure)## Warning: package 'multipanelfigure' was built under R version 4.3.1health_plot <- multi_panel_figure(columns = 2, rows = 2, panel_label_type = "none")
health_plot %<>%
fill_panel(g1, column = 1:2, row = 1) %<>%
fill_panel(g2, column = 1, row = 2) %<>%
fill_panel(g3, column = 2, row = 2)
health_plot
From the above figure, we can clearly find that the Tornado is the most
harmful event that causes the most fatality and injuries during the year
from 1950 to 2011. Thunderstorm wind, excessive heat and flood can be
considered also very harmful events considered their impact on public
health.
- Across the United States, which types of events have the greatest economic consequences?
library(stringr) # load library
## summarize the economic variable: property damage, crop damage
## and calculate the total of both types
economy <- our_data_coded %>%
group_by(EVTYPE) %>%
summarize(sum_prodm = sum(PROPDMG_T,na.rm=TRUE),
sum_cropdm = sum(CROPDMG_T,na.rm=TRUE),
mean_prodm = mean(PROPDMG_T,na.rm=TRUE),
mean_cropdm = mean(CROPDMG_T,na.rm=TRUE)) %>%
mutate(sum_pro_crop = sum_prodm+sum_cropdm)
## plot
## 1) the top 10 based on the damages of both property and crop
## 2) the top 10 based on only property damage
## 3) the top 10 based on only crop damage
## plot 1
economy1 <- arrange(economy,desc(sum_pro_crop))
g4<- ggplot(data=economy1[1:10,],aes(x=reorder(EVTYPE,sum_pro_crop),y=sum_pro_crop)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Property and crop damage") + xlab("") +
labs(title="Top 10 Most Harmful Events for economy", subtitle="Based on both property and crop damage") +
theme(axis.text=element_text(size=8),
axis.title=element_text(size=8,face="bold"),
title=element_text(size=9,face="bold"))
## plot 2
economy2 <- arrange(economy,desc(sum_prodm))
g5<- ggplot(data=economy2[1:10,],aes(x=reorder(EVTYPE,sum_prodm),y=sum_prodm)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Property damage") + xlab("") +
labs(subtitle="Based on property damage only") +
theme(axis.text=element_text(size=8),
axis.title=element_text(size=8,face="bold"),
title=element_text(size=9,face="bold"))
## plot 3
economy3 <- arrange(economy,desc(sum_cropdm))
g6<- ggplot(data=economy3[1:10,],aes(x=reorder(EVTYPE,sum_cropdm),y=sum_cropdm)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Crop damage") + xlab("") +
labs(subtitle="Based on crop damage only") +
theme(axis.text=element_text(size=8),
axis.title=element_text(size=8,face="bold"),
title=element_text(size=9,face="bold"))
## plot them in one chart
# install.packages("multipanelfigure") ##need to run once
library(multipanelfigure)
economy_plot <- multi_panel_figure(columns = 2, rows = 2, panel_label_type = "none")
economy_plot %<>%
fill_panel(g4, column = 1:2, row = 1) %<>%
fill_panel(g5, column = 1, row = 2) %<>%
fill_panel(g6, column = 2, row = 2)
economy_plot
From the perspective of economy, Flood can been concluded as the most
harmful event during the year from 1950 to 2011. It has caused the
largest damages especially the property damages. Drought, on the other
hand, causes more crop damages than flood but it’s impact on property is
limited. Hurricane/Typhoon is the runner-up and Tornado is in the third
place when considering the total damages.