Effect of Storm Events in the United States
L Sasser
2/8/2021
Preparing the property and crop damage data set
# Note that the PROPDMG and CROPDMG do not give the full damage number, the exponent in PROPDMGEXP and CROPDMGEXP need to be considered
economic <- noaa %>%
select(EVTYPE, PROPDMG, CROPDMG, PROPDMGEXP, CROPDMGEXP)
unique(economic$PROPDMGEXP)## [1] "K" "M" "" "B" "m" "+" "0" "5" "6" "?" "4" "2" "3" "h" "7" "H" "-" "1" "8"unique(economic$CROPDMGEXP)## [1] "" "M" "K" "m" "B" "?" "0" "k" "2"Combine the exponent values with the appropriate damage to determine actual value
economic$PROPDMGEXP[economic$PROPDMGEXP %in% c("", "+", "?", "-")] <- 1
economic$PROPDMGEXP[economic$PROPDMGEXP %in% c("B")] <- 9
economic$PROPDMGEXP[economic$PROPDMGEXP %in% c("M", "m")] <- 6
economic$PROPDMGEXP[economic$PROPDMGEXP %in% c("K")] <- 3
economic$PROPDMGEXP[economic$PROPDMGEXP %in% c("H", "h")] <- 2
economic$CROPDMGEXP[economic$CROPDMGEXP %in% c("", "?", "-")] <- 1
economic$CROPDMGEXP[economic$CROPDMGEXP %in% c("B")] <- 9
economic$CROPDMGEXP[economic$CROPDMGEXP %in% c("M", "m")] <- 6
economic$CROPDMGEXP[economic$CROPDMGEXP %in% c("K", "k")] <- 3
economic$CROPDMGEXP[economic$CROPDMGEXP %in% c("H")] <- 2
economic$property <- economic$PROPDMG * (10 ^ as.numeric(economic$PROPDMGEXP))
economic$crop <- economic$CROPDMG * (10 ^ as.numeric(economic$CROPDMGEXP))
economic$damage <- economic$property + economic$cropResults:
Show the top health related aspects of storm events in the US.
# Create a plot of the fatalities and injuries caused by the top 10 injurious storm related events.
chart1 <- health_tidy %>%
ggplot(aes(x= reorder(EVTYPE, + total), y = total, fill = type)) +
geom_bar(stat= "identity", position = "dodge") +
labs(x = "Event Type", y = "Total Harmful Numbers",
title = "Total injuries and fatalities from the top 10 weather\nrelated events in the United States",
fill = "Severity") +
scale_fill_discrete(labels = c("Fatalities", "Injuries")) +
theme(axis.text.x = element_text(angle = 0)) +
coord_flip()
chart1
# Tornadoes cause the most injuries and fatalities in the US by a large margin, followed by excessive heat.
Show the weather related events that result in the most economic impact
# Create a plot of the amount of property damage caused by the top 10 injurious storm related events.
eco_prop <- economic %>%
dplyr::group_by(EVTYPE) %>%
summarise(totalprop = sum(property)) %>%
arrange(desc(totalprop)) %>%
slice(1:10) %>%
ggplot(aes(x = reorder(EVTYPE, +totalprop), y = totalprop)) +
geom_bar(stat = "identity") +
labs(x = "Event Type", y = "Total Damage (USD)",
title = "Top 10 property damage causing weather related\neventsin the United States") +
theme(axis.text.x = element_text(angle = 0)) +
coord_flip()## `summarise()` ungrouping output (override with `.groups` argument)eco_prop
# Floods cause the most property damage in the US, according the NOAA data collection.
# Create a plot of the amount of property damage caused by the top 10 injurious storm related events.
eco_crop <- economic %>%
dplyr::group_by(EVTYPE) %>%
summarise(totalcrop = sum(crop)) %>%
arrange(desc(totalcrop)) %>%
slice(1:10) %>%
ggplot(aes(x = reorder(EVTYPE, +totalcrop), y = totalcrop)) +
geom_bar(stat = "identity") +
labs(x = "Event Type", y = "Total Damage (USD)",
title = "Top 10 crop damage causing weather\nrelated events in the United States") +
theme(axis.text.x = element_text(angle = 0)) +
coord_flip()## `summarise()` ungrouping output (override with `.groups` argument)eco_crop
# Droughts are the leading cause of crop damage in the US, according to NOAA data collection.