Summary: Storms and other severe weather events can cause both public health and economic problems for communities and municipalities. Many severe events can result in fatalities, injuries, and property damage, and preventing such outcomes to the extent possible is a key concern.

This project involves exploring the U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database. This database tracks characteristics of major storms and weather events in the United States, including when and where they occur, as well as estimates of any fatalities, injuries, and property damage.

Data processing

Load libraries:

library(ggplot2)
library(reshape2)

Load data:

Load the dataset:

storms <- read.csv("repdata_data_StormData.csv", na.strings="")

Pre-process data:

Before actually start the data process, it is better to irrelevant columns in the data frame and also clean the data. We also need to calculate numeric value of the damage and it is done by factoring the exponent.

storms_processed <- storms

# Clean CROPDMG
not.na <- !is.na(storms$CROPDMGEXP)
B <- storms$CROPDMGEXP=="b"|storms$CROPDMGEXP=="B"
M <- storms$CROPDMGEXP=="m"|storms$CROPDMGEXP=="M"
K <- storms$CROPDMGEXP=="k"|storms$CROPDMGEXP=="K"

storms_processed$CROPDMG[not.na&K] = storms$CROPDMG[not.na&K]*1000
storms_processed$CROPDMG[not.na&M] = storms$CROPDMG[not.na&M]*1000000
storms_processed$CROPDMG[not.na&B] = storms$CROPDMG[not.na&B]*1000000000

# Clean PROPDMG

not.na <- !is.na(storms$PROPDMGEXP)
B <- storms$PROPDMGEXP=="b"|storms$PROPDMGEXP=="B"
M <- storms$PROPDMGEXP=="m"|storms$PROPDMGEXP=="M"
K <- storms$PROPDMGEXP=="k"|storms$PROPDMGEXP=="K"

storms_processed$PROPDMG[not.na&K] = storms$PROPDMG[not.na&K]*1000
storms_processed$PROPDMG[not.na&M] = storms$PROPDMG[not.na&M]*1000000
storms_processed$PROPDMG[not.na&B] = storms$PROPDMG[not.na&B]*1000000000

Summarize damages in a single variable called “Total_damage”:

storms_processed$Total_damage <- storms_processed$PROPDMG + storms_processed$CROPDMG

Summarize the injuries, fatalities and total damage variables:

storms_summary <- aggregate(cbind(INJURIES, FATALITIES, Total_damage) ~ EVTYPE, data=storms_processed, FUN=sum)

Results

Economic losses

The Total damage variable corresponds to the effect of each event.

damage_sorted <- storms_summary[with(storms_summary,order(-Total_damage)),c("EVTYPE", "Total_damage")]
head(damage_sorted,3)

##                EVTYPE Total_damage
## 170             FLOOD 150319678257
## 411 HURRICANE/TYPHOON  71913712800
## 834           TORNADO  57352114049

Floods are the most damaging events.

ggplot(head(damage_sorted,10), aes(reorder(EVTYPE,Total_damage), Total_damage)) + 
        geom_bar(stat = "identity",fill="#FF9999", colour="black") +
        coord_flip() +
        xlab("EVENT") +
        ylab("TOTAL DAMAGE ($)") +
        ggtitle("EVENTS WITH GREATEST ECONOMIC CONSEQUENCES")

Health effects

The impact on health is summarized below:

top_injuries <- storms_summary[with(storms_summary,order(-INJURIES)),c("EVTYPE", "FATALITIES", "INJURIES")]
head(top_injuries,5)

##             EVTYPE FATALITIES INJURIES
## 834        TORNADO       5633    91346
## 856      TSTM WIND        504     6957
## 170          FLOOD        470     6789
## 130 EXCESSIVE HEAT       1903     6525
## 464      LIGHTNING        816     5230

top_fatalities <- storms_summary[with(storms_summary,order(-FATALITIES)),c("EVTYPE", "FATALITIES", "INJURIES")]
head(top_fatalities,5)

##             EVTYPE FATALITIES INJURIES
## 834        TORNADO       5633    91346
## 130 EXCESSIVE HEAT       1903     6525
## 153    FLASH FLOOD        978     1777
## 275           HEAT        937     2100
## 464      LIGHTNING        816     5230

Tornados are the most devasting event.

top_injuries_10 <- head(top_injuries, 10)
most_inj_10 <- melt(top_injuries_10, id.vars="EVTYPE")
ggplot(most_inj_10, aes(reorder(EVTYPE,value), value, fill = variable)) +
        geom_bar(stat="identity") +
        xlab("EVENT TYPE") +
        ylab("# VICTIMS") +
        coord_flip() +
        ggtitle("NUMBER OF INJURIES AND FATALITIES")

Conclusion

Tornados and floods cause most of human health and economical impact.

Study of weather events in the US

March 2015