Storm Data: Tornados and Floods have the greatest puplic health and economic consequences respectively

Conversion of DMG & EXP columns to actual numbers

For both property and crop damage a number is provided along with an exponent which should be applied to the number i.e. “H” for hundred, “T” for thousand etc.

unique(data$PROPDMGEXP)

##  [1] K M   B m + 0 5 6 ? 4 2 3 h 7 H - 1 8
## Levels:  - ? + 0 1 2 3 4 5 6 7 8 B h H K m M

unique(data$CROPDMGEXP)

## [1]   M K m B ? 0 k 2
## Levels:  ? 0 2 B k K m M

Some of the the exponent keys are provided in lower case so let’s convert them all to uppercase for consistency.

data$PROPDMGEXP <- toupper(data$PROPDMGEXP)
data$CROPDMGEXP <- toupper(data$CROPDMGEXP)
unique(data$PROPDMGEXP)

##  [1] "K" "M" ""  "B" "+" "0" "5" "6" "?" "4" "2" "3" "H" "7" "-" "1" "8"

unique(data$CROPDMGEXP)

## [1] ""  "M" "K" "B" "?" "0" "2"

Then convert property exponent codes to numeric values and create a new column with the value of damage to property for each event.

# Map property damage alphanumeric exponents to numeric values.
PROPDMGEXP <- c("\"\"", "-", "+", "0", "1", "2", "3", "4", "5", "6", "7", "8", "9", "H", "K", "M", "B")
prop_numeric <- c(10^0, 10^0, 10^0, 10^0, 10^1, 10^2, 10^3, 10^4, 10^5, 10^6, 10^7, 10^8, 10^9, 10^2, 10^3, 10^6, 10^9)
prop_dmg <- data.frame(PROPDMGEXP, prop_numeric)
library(dplyr)
data <- full_join(data, prop_dmg, by = "PROPDMGEXP") %>% 
  mutate(Property_Damage = PROPDMG*prop_numeric)
head(data) #sanity check

##   REFNUM  EVTYPE FATALITIES INJURIES PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP
## 1      1 TORNADO          0       15    25.0          K       0           
## 2      2 TORNADO          0        0     2.5          K       0           
## 3      3 TORNADO          0        2    25.0          K       0           
## 4      4 TORNADO          0        2     2.5          K       0           
## 5      5 TORNADO          0        2     2.5          K       0           
## 6      6 TORNADO          0        6     2.5          K       0           
##   prop_numeric Property_Damage
## 1         1000           25000
## 2         1000            2500
## 3         1000           25000
## 4         1000            2500
## 5         1000            2500
## 6         1000            2500

Now let’s do the same to calculate the damage to crops from each event and convert any NAs to zero so we can calculate the total economic damage caused by each event later.

# Map crop damage alphanumeric exponents to numeric values
CROPDMGEXP <- c("\"\"", "?", "0", "2", "K", "M", "B")
crop_numeric <- c(10^0, 10^0, 10^0, 10^2, 10^3, 10^6, 10^9)
crop_dmg <- data.frame(CROPDMGEXP, crop_numeric)
data <- full_join(data, crop_dmg, by = "CROPDMGEXP") %>% 
  mutate(Crop_Damage = CROPDMG*crop_numeric)
data[is.na(data)] <- 0 #replace NAs with 0

Across the United States, which types of events are most harmful with respect to population health?

library(dplyr); library(ggplot2)
A <- data %>% 
  group_by(EVTYPE) %>% 
  summarise(Total_Fatalities = sum(FATALITIES)) %>% 
  arrange(desc(Total_Fatalities)) %>% 
  slice(1:20) %>% 
  ggplot(aes(x=reorder(EVTYPE, Total_Fatalities), y=Total_Fatalities)) +
  geom_bar(stat="identity", fill = "red") +
  coord_flip() + theme_minimal() +
  ggtitle("Top 20 event types causing fatalities across the United States between 1950 and 2011") +
  xlab("Event type") + ylab("Number of fatalities")
B <- data %>% 
  group_by(EVTYPE) %>% 
  summarise(Total_Injuries = sum(INJURIES)) %>% 
  arrange(desc(Total_Injuries)) %>% 
  slice(1:20) %>% 
  ggplot(aes(x=reorder(EVTYPE, Total_Injuries), y=Total_Injuries)) +
  geom_bar(stat="identity", fill = "orange") +
  coord_flip() + theme_minimal() +
  ggtitle("Top 20 event types causing injuries across the United States between 1950 and 2011") +
  xlab("Event type") + ylab("Number of injuries")
ggpubr::ggarrange(A,B, nrow = 2, labels = c("A", "B"))

Across the United States tornados have caused both the most fatalities (A) and the most injuries (B). Excessive heat and heat both figure prominently in the ranks of causes of fataltities and injuries, as do floods and flash floods. Lightning and TSTM wind are also harmful for population health.

Across the United States, which types of events have the greatest economic consequences?

library(dplyr); library(ggplot2)
data %>% 
  mutate(Economic_damage = Property_Damage + Crop_Damage) %>% #add together costs incurred to property and crops
  group_by(EVTYPE) %>% 
  summarise(Total_damage = sum(Economic_damage)) %>% #get total cost for each event type
  arrange(desc(Total_damage)) %>% #sort by Total damage per event type
  slice(1:20) %>% #take the top 20
  ggplot(aes(x=reorder(EVTYPE, Total_damage), y=Total_damage)) + #do a bar plot
  geom_bar(stat="identity", fill = "green") +
  coord_flip() + theme_minimal() +
  ggtitle("Economic consequences of each of the top 20 event types across the US (1950 and 2011)") +
  xlab("Event type") + ylab("Damage in USD")

The event type with the greatest economic consequences, to property and crops combined, is floods. The cost due to floods is over twice the amount over the next most expensive event types, hurricaine/typhoon and tornado.