NOAA Storm Database analysis

Synopsis

This is an anlysis of the U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database. It addressed major storms and weather events in the United States from 1950 to 2011 in terms of injuries, and property damage. As results, Tornado is the most harmful for population health, and Flood have the greatest economic consequences.

Data Processing

#load packages
library(dplyr)  

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

#download data
download.file("http://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2","storm.data")                              

# read data
storm.data <- read.csv("storm.data") 

#select variables
storm<-storm.data %>% select(
  EVTYPE, FATALITIES, INJURIES,#directly related varibales
  contains("DMG")#for damange related variables
                 )

#create prop.damage in proper units
storm<-storm%>%mutate(prop.dmg=case_when(
  PROPDMGEXP == "H" ~ PROPDMG * 10^2,
  PROPDMGEXP == "K" ~ PROPDMG * 10^3,
  PROPDMGEXP == "M" ~ PROPDMG * 10^6,
  PROPDMGEXP == "B" ~ PROPDMG * 10^9
))
#view it
storm%>%group_by(prop.dmg)%>%count()

## # A tibble: 1,655 × 2
## # Groups:   prop.dmg [1,655]
##    prop.dmg      n
##       <dbl>  <int>
##  1        0 197195
##  2       10    850
##  3       20     46
##  4       30   1599
##  5       40      2
##  6       50    557
##  7       60     21
##  8       70      3
##  9      100   1426
## 10      110     27
## # … with 1,645 more rows

#create crop.damage in proper units
storm<-storm%>%mutate(crop.dmg=case_when(
  CROPDMGEXP == "H" ~ CROPDMG * 10^2,
  CROPDMGEXP == "K" ~ CROPDMG * 10^3,
  CROPDMGEXP == "M" ~ CROPDMG * 10^6,
  CROPDMGEXP == "B" ~ CROPDMG * 10^9
))

#view it
storm%>%group_by(crop.dmg)%>%count()

## # A tibble: 524 × 2
## # Groups:   crop.dmg [524]
##    crop.dmg      n
##       <dbl>  <int>
##  1        0 261773
##  2       10      1
##  3       50    156
##  4      100     13
##  5      150      2
##  6      200     25
##  7      240      1
##  8      250      3
##  9      280      1
## 10      300      7
## # … with 514 more rows

Results

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

health<-storm%>% group_by(EVTYPE) %>%
  summarise(sum.pop=sum(FATALITIES+ INJURIES))%>%
  arrange(desc(sum.pop))

library(ggplot2)
health[1:5,]%>%ggplot(aes(x = EVTYPE, y = sum.pop))+
    geom_bar(stat = "identity", fill = "blue", las = 3)+ 
    xlab("Event Type") + ylab("Fatalities") + ggtitle("Number of fatalities by top 10 Weather Events")

## Warning: Ignoring unknown parameters: las

From the barplot, Tornado has the highest fatalties and injuries number.

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

econ<-storm%>% group_by(EVTYPE) %>%
  summarise(sum.econ=sum(prop.dmg + crop.dmg))%>%
  arrange(desc(sum.econ))

library(ggplot2)
econ[1:5,]%>%ggplot(aes(x = EVTYPE, y = sum.econ))+
    geom_bar(stat = "identity", fill = "blue", las = 3)+
  theme(axis.text.x = element_text(angle = 90, hjust = 1))+ 
    xlab("Event Type") + ylab("Damages ($)") + ggtitle("Property & Crop Damages by top 10 Weather Events")

## Warning: Ignoring unknown parameters: las

From the barplot, Flood has the highest negative economical impact (i.e., loss of money).