Analysis of the Severe Weather Events

Synopsis

The U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database was explored to find out which weather events had severe effect over the public health and economy. This analysis can help the government for preparing for the severe weather events.

Data Processing

Loading the Data

storm <- read.csv("./repdata%2Fdata%2FStormData.csv", na.strings = c("", "NA"))

Aggregating and Ordering the total injuries and fatalities across USA by different weather events.

injurydata <- aggregate(INJURIES ~ EVTYPE, data = storm, sum)
injurydata <- injurydata[order(-injurydata$INJURIES),]

fatalitydata <- aggregate(FATALITIES ~ EVTYPE, data = storm, sum)
fatalitydata <- fatalitydata[order(-fatalitydata$FATALITIES),]

Pre-processing to calculate actual amount of damages(in $)

summary(storm$PROPDMGEXP)

##      -      ?      +      0      1      2      3      4      5      6 
##      1      8      5    216     25     13      4      4     28      4 
##      7      8      B      h      H      K      m      M   NA's 
##      5      1     40      1      6 424665      7  11330 465934

storm$PROPDMGEXP <- as.character(tolower(storm$PROPDMGEXP))
storm$PROPDMGEXP[storm$PROPDMGEXP == "-"] <- 0
storm$PROPDMGEXP[storm$PROPDMGEXP == "?"] <- 0
storm$PROPDMGEXP[storm$PROPDMGEXP == "+"] <- 0
storm$PROPDMGEXP[storm$PROPDMGEXP == "b"] <- 9
storm$PROPDMGEXP[storm$PROPDMGEXP == "k"] <- 3
storm$PROPDMGEXP[storm$PROPDMGEXP == "m"] <- 6
storm$PROPDMGEXP[storm$PROPDMGEXP == "h"] <- 2
storm$PROPDMGEXP <- as.numeric(storm$PROPDMGEXP)

summary(storm$CROPDMGEXP)

##      ?      0      2      B      k      K      m      M   NA's 
##      7     19      1      9     21 281832      1   1994 618413

storm$CROPDMGEXP <- as.character(tolower(storm$CROPDMGEXP))
storm$CROPDMGEXP[storm$CROPDMGEXP == "?"] <- 0
storm$CROPDMGEXP[storm$CROPDMGEXP == "b"] <- 9
storm$CROPDMGEXP[storm$CROPDMGEXP == "k"] <- 3
storm$CROPDMGEXP[storm$CROPDMGEXP == "m"] <- 6
storm$CROPDMGEXP <- as.numeric(storm$CROPDMGEXP)

Calculating the actual damage (in $)

storm$PROPDMG <- storm$PROPDMG*10^storm$PROPDMGEXP
storm$CROPDMG <- storm$CROPDMG*10^storm$CROPDMGEXP

#assigning zero for NAs
storm$PROPDMG[which(is.na(storm$PROPDMG))] <- 0
storm$CROPDMG[which(is.na(storm$CROPDMG))] <- 0

Total damages (in $) due to the different weather events.

Economicdata <- aggregate(PROPDMG + CROPDMG ~ EVTYPE, data = storm, sum)
names(Economicdata)[2] <- "Overalldmg"
Economicdata <- Economicdata[order(-Economicdata$Overalldmg),]

Results

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

library(ggplot2)
library(gridExtra)

head(injurydata)

##             EVTYPE INJURIES
## 834        TORNADO    91346
## 856      TSTM WIND     6957
## 170          FLOOD     6789
## 130 EXCESSIVE HEAT     6525
## 464      LIGHTNING     5230
## 275           HEAT     2100

head(fatalitydata)

##             EVTYPE FATALITIES
## 834        TORNADO       5633
## 130 EXCESSIVE HEAT       1903
## 153    FLASH FLOOD        978
## 275           HEAT        937
## 464      LIGHTNING        816
## 856      TSTM WIND        504

p1 <- ggplot(head(injurydata), aes(EVTYPE, INJURIES)) + 
      geom_bar(stat = "identity") +
      xlab("Weather Events") +
      ylab("Number of Injuries") + 
      ggtitle("Most harmful events with respect to population health")

p2 <- ggplot(head(fatalitydata), aes(EVTYPE, FATALITIES)) + 
      geom_bar(stat = "identity") + 
      xlab("Weather Events") +
      ylab("Number of Fatalities")

grid.arrange(p1, p2, ncol=1)

From the above plots, we can see the top 6 weather events with most harm to human health. Also, it is clear that the TORNADO is the most harmful event with respect to population health.

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

head(Economicdata)

##                EVTYPE   Overalldmg
## 170             FLOOD 150319678250
## 411 HURRICANE/TYPHOON  71913712800
## 834           TORNADO  57362333944
## 670       STORM SURGE  43323541000
## 244              HAIL  18761221926
## 153       FLASH FLOOD  18243990872

ggplot(head(Economicdata), aes(EVTYPE, Overalldmg)) + 
      geom_bar(stat = "identity") + 
      xlab("Weather Events") +
      ylab("Total damages (in Dollars)") + 
      ggtitle("Most harmful events with the greatest economic consequences") + 
      theme(axis.text.x=element_text(size=7))

From the above plots, we can see the top 6 weather events that has the greatest economis consequences.