Synopsis

This analysis is based on the data available on U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database. It addresses the following questions:

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

Data Processing

This section involves the following processes:
- Loading the data - Preprocessing the data - Defining the scope
- Subsetting the data
- Cleaning the data
- Further grouping the data for plotting the result

1. Loading the data

url <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
download.file(url, "data.csv.bz2")
unzip("data.csv.bz2")

## Warning in unzip("data.csv.bz2"): error 1 in extracting from zip file

data <- read.csv("data.csv.bz2")

2. Preprocessing the data

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

library(lubridate)

## 
## Attaching package: 'lubridate'

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

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

library(stringdist)
library(ggplot2)
library(directlabels)
library(colorspace)


## Define the 50 US states to be analysised and the 48 events based on Storm Data Event Table
Ustates <- c("AL", "AK", "AZ", "AR", "CA", "CO", "CT", "DE", "FL", "GA",
             "HI", "ID", "IL", "IN", "IA", "KS", "KY", "LA", "ME", "MD",
             "MA", "MI", "MN", "MS", "MO", "MT", "NE", "NV", "NH", "NJ",
             "NM", "NY", "NC", "ND", "OH", "OK", "OR", "PA", "RI", "SC",
             "SD", "TN", "TX", "UT", "VT", "VA", "WA", "WV", "WI", "WY")

eventlist <- c("Astronomical Low Tide", "Avalanche", "Blizzard",
                "Coastal Flood", "Cold/Wind Chill", "Debris Flow",
                "Dense Fog", "Dense Smoke", "Drought",
                "Dust Devil", "Dust Storm","Excessive Heat",
                "Extreme Cold/Wind Chill", "Flash Flood", "Flood",
                "Frost/Freeze", "Funnel Cloud", "Freezing Fog", "Hail",
                "Heat", "Heavy Rain", "Heavy Snow",
                "High Surf", "High Wind", "Hurricane (Typhoon)",
                "Ice Storm", "Lake-Effect Snow", "Lakeshore Flood",
                "Lightning", "Marine Hail", "Marine High Wind",
                "Marine Strong Wind", "Marine Thunderstorm Wind", "Rip Current",
                "Seiche", "Sleet", "Storm Surge/Tide",
                "Strong Wind", "Thunderstorm Wind", "Tornado",
                "Tropical Depression", "Tropical Storm", "Tsunami",
                "Volcanic Ash", "Waterspout", "Wildfire",
                "Winter Storm", "Winter Weather")

data$BGN_DATE <- as.Date(data$BGN_DATE, "%m/%d/%Y")

## Subset the data from 1 Jan 1996 across 50 US states for all event types (since only limited events were recorded before 1996)

newdata<- subset(data, 
                 BGN_DATE > "1996-01-01" & STATE %in% Ustates, 
                 select = c(BGN_DATE, EVTYPE, 
                            FATALITIES, INJURIES, 
                            PROPDMG, PROPDMGEXP, 
                            CROPDMG, CROPDMGEXP))

## Format the Crop & Prop Damage Exp columns to get the actual numeric damage
## Note: Alphabetical characters used to signify magnitude include “K” for thousands, “M” for millions, and “B” for billions.
newdata$PROPDMGEXP <- factor(newdata$PROPDMGEXP, 
                        levels = c("K", "M", "B", "", "0"), 
                        labels = c(10^3, 10^6, 10^9, 1, 1))
newdata$PROPDMGEXP <- as.numeric(as.character(newdata$PROPDMGEXP))
newdata$TTLPROPDMG <- newdata$PROPDMG * newdata$PROPDMGEXP

newdata$CROPDMGEXP <- factor(newdata$CROPDMGEXP, 
                        levels = c("K", "M", "B", "", "0"), 
                        labels = c(10^3, 10^6, 10^9, 1, 1))
newdata$CROPDMGEXP <- as.numeric(as.character(newdata$CROPDMGEXP))
newdata$TTLCROPDMG <- newdata$CROPDMG * newdata$CROPDMGEXP

## Further subset the data set before cleaning it
newdata <- newdata %>% 
              group_by(EVTYPE, YEAR = year(BGN_DATE)) %>%
              summarise(FATALITIES=sum(FATALITIES), 
                        INJURIES=sum(INJURIES), 
                        TTLPROPDMG=sum(TTLPROPDMG), 
                        TTLCROPDMG=sum(TTLCROPDMG)) %>%
              filter(FATALITIES+INJURIES+TTLPROPDMG+TTLCROPDMG!=0)

## Clean the EVTYPE column so that its values align with the 48 defined event types
newdata$EVTYPE <- tolower(newdata$EVTYPE)
newdata$EVTYPE <- gsub("tstm", "thunderstorm", newdata$EVTYPE)

newdata$EVMATCH <- eventlist[amatch(newdata$EVTYPE, tolower(eventlist), method="jw", maxDist = 0.45)]

plot1data <- newdata %>% 
              group_by(EventType = EVMATCH, YEAR) %>%
              summarise(FATALITIES=sum(FATALITIES), 
                        INJURIES=sum(INJURIES), 
                        TTLPROPDMG=sum(TTLPROPDMG), 
                        TTLCROPDMG=sum(TTLCROPDMG))

plot2data <- plot1data %>%
              group_by(EventType) %>%
              summarise(HEALTHDMG = sum(FATALITIES) + sum(INJURIES),
                        ECONDMG = sum(TTLPROPDMG) + sum(TTLCROPDMG)) %>%
              filter(ECONDMG > mean(ECONDMG))

Results

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

The plot below shows the total number of fatalities and injuries under severe weather events in the US from 1996 to 2011.

ggplot(plot1data, 
       aes(x=YEAR, 
           y=FATALITIES+INJURIES, 
           color=EventType)) + 
geom_line() + 
labs(title = "Damage of Severe Weather Events on Population Health in the US (1996-2011)",
     x = "Year", y = "Total number of fatalities and injuries") +
theme(legend.position = "bottom") + 
geom_dl(aes(label = EventType), method = "top.points")

The event that is the most harmful to population health are Tornado.

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

The plot below shows the total economic damage (i.e.for both crops and proporties) under severe weather events in the US from 1996 to 2011. To simplify the plot, only event types with totals more than the mean damage are shown.

with(plot2data, 
     barplot(ECONDMG, by = EventType, 
             xlab="Event type", 
             ylab = "Total economic damage", 
             main = "Damage of Key Weather Events on Economic in the US (1996-2011)", 
             names.arg=EventType,
             col = "blue")
     )

## Warning in plot.window(xlim, ylim, log = log, ...): "by" is not a graphical
## parameter

## Warning in axis(if (horiz) 2 else 1, at = at.l, labels = names.arg, lty =
## axis.lty, : "by" is not a graphical parameter

## Warning in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...): "by" is
## not a graphical parameter

## Warning in axis(if (horiz) 1 else 2, cex.axis = cex.axis, ...): "by" is not a
## graphical parameter

The event that has the greatest economic consequences is Flood.

Data Analysis on Severe Weather Events in the US