The most harmful weather events during 1950 to 2011

1. Synopsis

TORNADO is the most harmful evnt with respect to population health. FLOOD is the event have the greatest economic consequences.

2 Data Processing

2.1 Read data

mydata <- read.csv("StormData.csv")

2.2 Subset data

Remove harmless events, and remove “PROPDMGEXP” with value of missing, +, - ?

exp <- c("0", "1", "2", "3", "4", "5", "6", "7", "8", "B", "h", "H", "K", "m", 
    "M")
mysubset <- subset(mydata, (INJURIES > 0 | FATALITIES > 0 | PROPDMG > 0 | CROPDMG > 
    0) & PROPDMGEXP %in% exp)

2.3 transform “PROPDMGEXP”

mysubset$PROPDMGEXP <- as.character(mysubset$PROPDMGEXP)
mysubset[grepl("M", mysubset$PROPDMGEXP, ignore.case = TRUE), ]$PROPDMGEXP <- 1e+06
mysubset[grepl("B", mysubset$PROPDMGEXP, ignore.case = TRUE), ]$PROPDMGEXP <- 1e+09
mysubset[grepl("H", mysubset$PROPDMGEXP, ignore.case = TRUE), ]$PROPDMGEXP <- 100
mysubset[grepl("K", mysubset$PROPDMGEXP, ignore.case = TRUE), ]$PROPDMGEXP <- 1000

3 Results

3.1 Across the United States, which types of events (as indicated in the EVTYPE variable) are most harmful with respect to population health?

mysubset$EVTYPE <- as.character(mysubset$EVTYPE)
sumharm <- aggregate(INJURIES ~ EVTYPE, data = mysubset, FUN = sum, na.rm = TRUE)
# most harmful evnts with respect to population health
sumharm[sumharm$INJURIES == max(sumharm$INJURIES), c("EVTYPE")]

[1] "TORNADO"

# top five most harmful evnts
sumharm2 <- sumharm[order(-sumharm$INJURIES), ][1:5, ]
library(ggplot2)
ggplot(sumharm2, aes(x = EVTYPE, y = INJURIES)) + geom_bar(stat = "identity", 
    position = position_dodge())

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

mysubset$PROPDMGEXP <- as.numeric(mysubset$PROPDMGEXP)
mysubset$myharm = mysubset$PROPDMG * mysubset$PROPDMGEXP
sumharm <- aggregate(myharm ~ EVTYPE, data = mysubset, FUN = sum, na.rm = TRUE)
# events have the greatest economic consequences
sumharm[sumharm$myharm == max(sumharm$myharm), c("EVTYPE")]

[1] "FLOOD"