Synopsis

This report studies the impact of storm related events in the United States from 1955 to 2011. The storm related events are studied considering the impact to population health and economy.

The following questions are addressed in this study.

Which types of events are most harmful with respect to population health across the United States?
Which types of events have the greatest economic consequences across the United States?

This study shows that:

Tornadoes have caused the largest number of fatalities and injuries among all storm related events in the United States.
Floods have caused the greatest property damage among all storm related events in the United States.
Droughts have caused the largest crop damage among all storm related events in the United States.

Data Processing

The data in this study is collected by U.S. National Oceanic and Atmospheric Administration’s (NOAA). The NOAA storm 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.

The NOAA storm data in this study is stored in a compressed comma-separated-value file. The data is loaded into R using the read.table() command.

Reading the Storm Data Dataset

if(!file.exists("./repdata-data-StormData.csv.bz2")){
  file.create("./repdata-data-StormData.csv.bz2")
  URL <- "http://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
  destfile <- "./repdata-data-StormData.csv.bz2"
  download.file(URL, destfile)
}

stormData <- read.table("repdata-data-StormData.csv.bz2", header = TRUE, sep=",")

Performing Necessary Transformations

# Transformation of Data

# The values in PROPDMG and CROPDMG columns should be multiplied by a multiplier based on the 
# values in PROPDMGEXP, and CROPDMGEXP columns respectively. The multiplier codes are defined
# below:
#
# PROPDMGEXP, and CROPDMGEXP column codes
# ------------------------------------------
#   H,h = hundreds = 100 
#   K,k = kilos = thousands = 1,000 
#   M,m = millions = 1,000,000 
#   B,b = billions = 1,000,000,000 
#   (+) = 1 
#   (-) = 0 
#   (?) = 0 
#   blank/empty character = 0 
#   numeric 0..8 = 10 
# ------------------------------------------

# Adding a column as multiplier to PROPDMG column and populating it based on the PROPDMGEXP
stormData[stormData$PROPDMGEXP == "-" | stormData$PROPDMGEXP =="?", "PROPDMG_MULTIPLIER"] <- 0.0
stormData[stormData$PROPDMGEXP == "+",                              "PROPDMG_MULTIPLIER"] <- 1.0
stormData[stormData$PROPDMGEXP %in% "0":"8",                        "PROPDMG_MULTIPLIER"] <- 10.0
stormData[stormData$PROPDMGEXP == "H" | stormData$PROPDMGEXP =="h", "PROPDMG_MULTIPLIER"] <- 100.0
stormData[stormData$PROPDMGEXP == "K" | stormData$PROPDMGEXP =="k", "PROPDMG_MULTIPLIER"] <- 1000.0
stormData[stormData$PROPDMGEXP == "M" | stormData$PROPDMGEXP =="m", "PROPDMG_MULTIPLIER"] <- 1000000.0
stormData[stormData$PROPDMGEXP == "B" | stormData$PROPDMGEXP =="b", "PROPDMG_MULTIPLIER"] <- 1000000000.0
stormData[is.na(stormData$PROPDMG_MULTIPLIER),                      "PROPDMG_MULTIPLIER"] <- 0.0

# Adding a column as multiplier to CROPDMG column and populating it based on the CROPDMGEXP
stormData[stormData$CROPDMGEXP == "-" | stormData$CROPDMGEXP =="?", "CROPDMG_MULTIPLIER"] <- 0.0
stormData[stormData$CROPDMGEXP == "+",                              "CROPDMG_MULTIPLIER"] <- 1.0
stormData[stormData$CROPDMGEXP %in% "0":"8",                        "CROPDMG_MULTIPLIER"] <- 10.0
stormData[stormData$CROPDMGEXP == "H" | stormData$CROPDMGEXP =="h", "CROPDMG_MULTIPLIER"] <- 100.0
stormData[stormData$CROPDMGEXP == "K" | stormData$CROPDMGEXP =="k", "CROPDMG_MULTIPLIER"] <- 1000.0
stormData[stormData$CROPDMGEXP == "M" | stormData$CROPDMGEXP =="m", "CROPDMG_MULTIPLIER"] <- 1000000.0
stormData[stormData$CROPDMGEXP == "B" | stormData$CROPDMGEXP =="b", "CROPDMG_MULTIPLIER"] <- 1000000000.0
stormData[is.na(stormData$CROPDMG_MULTIPLIER),                      "CROPDMG_MULTIPLIER"] <- 0.0

Results

Fatalities, Injuries, Property Damages, and Crop Damages caused by the storms have been studied in this section. The results are summarized using the R code below.

library(ggplot2)
library(dplyr)
library(gridExtra)

# Preparing summaries of impact of different events on population health and economy.

# FATALITIES
eventFatalities <- stormData %>%
                    group_by(EVTYPE) %>%
                    summarize(total = sum(FATALITIES)) %>%
                    arrange(desc(total))

# INJURIES
eventInjuries <- stormData %>%
                    group_by(EVTYPE) %>%
                    summarize(total = sum(INJURIES)) %>%
                    arrange(desc(total))

# PROPERTY DAMAGE
eventPropertyDamage <- stormData %>%
                  group_by(EVTYPE) %>%
                  summarize(total = sum(PROPDMG * PROPDMG_MULTIPLIER)) %>%
                  arrange(desc(total))

# CROP DAMAGE
eventCropDamage <- stormData %>%
                  group_by(EVTYPE) %>%
                  summarize(total = sum(CROPDMG * CROPDMG_MULTIPLIER)) %>%
                  arrange(desc(total))

Plots for Fatalities, Injuries, Property Damages, and Crop Damages are prepared in the following R code block.

# Preparing to plot data

plotFatalities <-  ggplot(eventFatalities[c(1:10), ] ) +
  aes(reorder(EVTYPE, -total), total) +
  geom_bar(stat="identity", fill="blue") + 
  theme_bw() + 
  xlab("Event") + 
  ylab("Fatalities") + 
  theme(axis.text.x = element_text(angle = 90))

plotInjuries <-  ggplot(eventInjuries[c(1:10), ] ) +
  aes(reorder(EVTYPE, -total), total) +
  geom_bar(stat="identity", fill="blue") + 
  theme_bw() + 
  xlab("Event") + 
  ylab("Injuries") + 
  theme(axis.text.x = element_text(angle = 90))

plotProperty <-  ggplot(eventPropertyDamage[c(1:10), ] ) +
  aes(reorder(EVTYPE, -total), total/1e9) +
  geom_bar(stat="identity", fill="blue") + 
  theme_bw() + 
  xlab("Event") + 
  ylab("Property Damage - $ Billion") + 
  theme(axis.text.x = element_text(angle = 90))

plotCrop <-  ggplot(eventCropDamage[c(1:10), ] ) +
  aes(reorder(EVTYPE, -total), total/1e9) +
  geom_bar(stat="identity", fill="blue") + 
  theme_bw() + 
  xlab("Event") + 
  ylab("Crop Damage - $ Billion") + 
  theme(axis.text.x = element_text(angle = 90))

Top 10 Most Harmful Events to Population Health

# Plotting Data for Population Health Impact
grid.arrange(plotFatalities, plotInjuries, ncol = 2)

Top 10 Events with Greatest Economic Consequences

# Plotting Data for Economic Impact
grid.arrange(plotProperty, plotCrop, ncol = 2)

Conclusion

Based on the above graphs we conclude:

Tornadoes have caused the largest number of fatalities among all storm related events in the United States.
Tornadoes have caused the largest number of Injuries among all storm related events in the United States.
Floods have caused the greatest property damages among all storm related events in the United States.
Droughts have caused the greatest crop damages among all storm related events in the United States.

Storm Related Events with the Most Impact on Population Health and Economy in the United States